JTWell

Antibody

Fully human IgG4

Target

Mature myostatin (GDF8)

Codes

REGN1033 · SAR391786

Status

Investigational — approved nowhere

What it is — and how it works

Trevogrumab (development code REGN1033; SAR391786 in the old Sanofi collaboration) is a fully human IgG4 monoclonal antibody that binds and neutralises myostatin — growth differentiation factor 8 (GDF8), a TGF-β-superfamily ligand secreted mainly by skeletal muscle. In COURAGE it was given subcutaneously, once weekly.

Myostatin is the body's main negative regulator — the brake — on muscle size. Mature myostatin binds activin type II receptors (ActRIIA/ActRIIB) on muscle fibres, driving SMAD2/3 signalling that suppresses protein synthesis and myoblast differentiation. Trevogrumab sequesters the mature ligand before it reaches the receptor, releasing that brake — so during the caloric deficit a GLP-1 drug imposes, more lean tissue survives the weight loss.

The human proof of mechanism came first: in a randomised Phase 1 trial in healthy postmenopausal women (Gonzalez Trotter et al., Nature Communications 2025), a single dose of trevogrumab 6 mg/kg increased MRI thigh muscle volume +4.6% vs placebo at 8 weeks; trevogrumab + garetosmab increased it +7.7%, with fat-mass reductions — the combination beat either antibody alone, which was the rationale for the COURAGE triplet.

One pathway, four points of attack

Agent	Point of attack	What the choice means
Trevogrumab (Regeneron)	Mature GDF8 / myostatin ligand only	Most selective of the Regeneron pair; activin A signalling left intact — a smaller effect than dual blockade, and in COURAGE far better tolerability.
Garetosmab (Regeneron)	Activin A ligand only	Deepened preservation in the triplet — but activin A also works in bone formation, reproduction / FSH biology, skin and mucosa, matching its signature adverse events: nosebleeds, eyelash loss, skin infections / abscesses, muscle spasms. Developed separately for FOP.
Bimagrumab (Lilly)	The ActRII receptor itself (ActRIIA/B)	Blocks every ligand through the receptor at once — myostatin, activins, GDF11 and others. The largest body-composition effects in the field, and the broadest adverse-event signature.
Apitegromab (Scholar Rock)	Pro- / latent myostatin	Binds myostatin before it is activated and prevents activation — the most myostatin-selective mechanism in the class.

Sources: Medscape, EASD 2025 · Mastaitis et al., Nat Commun 2025 · Gonzalez Trotter et al., Nat Commun 2025

The COURAGE trial

COURAGE (NCT06299098) is a randomised, double-blind, placebo-controlled Phase 2 trial in adults with obesity (BMI ≥30), without diabetes — a history of type 1 or type 2 diabetes was an exclusion, as were prior bariatric surgery, hypertrophic cardiomyopathy, and an ECG meeting Cornell voltage criteria for LV hypertrophy. Across both portions: 975 participants, mean age 49, mean BMI 36.7, 64% female. Every arm ran on a backbone of semaglutide titrated to 2.4 mg once weekly. Primary endpoints were percent change in lean mass, fat mass and body weight — lean and fat by DXA — at week 26 (higher-dose portion) and week 52 (lower-dose portion), with a predefined MRI substudy (n=76) in the lower-dose portion.

The registry record — checked live on 26 Oct 2026 — shows the trial active, not recruiting, with no results posted. Everything reported so far comes from sponsor releases and conference / journal-in-press summaries, not from registry-posted results. Regeneron's releases describe a “higher-dose portion” and a “lower-dose portion”; this page uses that portion language, because sources disagree on the part letters.

Source: ClinicalTrials.gov, NCT06299098 (verified live 26 Oct 2026)

Higher doses — the 26-week results

Higher-dose portion (n=599) · Change from baseline at week 26 · Least-squares means (SE)

Arm	Lean mass %Δ (SE)	Lean kg · % of loss	Fat mass %Δ (SE)	Fat kg · % of loss	Body weight %Δ (SE)
Semaglutide alone (n=151)	−6.5 (0.5)	−3.3 kg · 33.0%	−15.7 (0.9)	−6.7 kg · 67.0%	−10.6 (0.5)
+ Trevogrumab 200 mg (n=149)	−3.3 (0.5)	−1.5 kg*** · 16.8%	−17.3 (0.9)	−7.6 kg · 83.2%	−9.9 (0.5)
+ Trevogrumab 400 mg (n=152)	−3.8 (0.5)	−1.9 kg*** · 18.1%	−19.1 (0.9)	−8.5 kg* · 81.9%	−11.1 (0.5)
Triplet + garetosmab (n=147)	−2.0 (0.6)	−0.9 kg*** · 7.4%	−27.1 (1.1)	−11.8 kg*** · 92.6%	−13.4*** (0.6)

*** p<0.001, * p<0.05 vs semaglutide alone · Efficacy estimand (data after discontinuation excluded) · P-values not corrected for multiplicity · Source: Regeneron, GlobeNewswire, 17 Sep 2025 (EASD 2025)

Read it plainly: a third of everything the semaglutide-alone group lost was lean mass. Trevogrumab roughly halved that loss and modestly deepened fat loss — but the doublets produced no meaningful extra total weight loss (≈ −10% to −11%). Only the triplet lost significantly more weight (−13.4%) — and the triplet is the arm that was abandoned, below.

Lower doses — 26 and 52 weeks

Lower-dose portion (n=376) · DXA lean mass, % change from baseline (95% CI)

Arm	Week 26	Week 52
Placebo + semaglutide	−6.7% (−7.9 to −5.5)	−7.3% (−8.8 to −5.8)
Trevogrumab 25 mg + semaglutide	−4.7% (−5.5 to −3.9) · 29.9% preserved	−5.8% (−6.8 to −4.7) · 20.5% preserved
Trevogrumab 75 mg + semaglutide	−3.3% (−4.1 to −2.4) · 50.7% preserved	−4.2% (−5.3 to −3.1) · 42.5% preserved

Preservation is relative to placebo + semaglutide · Source: Regeneron, GlobeNewswire, 1 Oct 2026 (EASD 2026; in press, The Lancet)

At week 52 the treatment difference was +1.5 percentage points for 25 mg — not statistically significant — and +3.1 points for 75 mg — statistically significant. Most lean-mass loss happened in the first six months, the fastest weight-loss phase. Regeneron stated the lower doses “did not meaningfully enhance the weight loss achieved with semaglutide alone” — and published no fat-mass or body-weight table for this portion (see Open Questions).

A pre-specified subgroup with low lean mass at baseline (by the FNIH sarcopenia imaging criterion) was older (mean 53 vs 46 years), lost more lean mass on semaglutide alone, and showed numerically greater preservation with trevogrumab; per the Lancet summary, in this subgroup trevogrumab alone performed as well as the triplet. No subgroup effect sizes were published in the releases.

MRI substudy (n=76, lower-dose portion) · Change in thigh fat-free muscle volume, mean ± SE · Units not stated in the release

Arm	Week 26	Week 52
Placebo + semaglutide	−0.88 ± 0.21	−1.03 ± 0.22
Trevogrumab 25 mg	−0.32 ± 0.14 · 63.6% preserved	−0.29 ± 0.12 · 71.8% preserved
Trevogrumab 75 mg	−0.24 ± 0.13 · 72.7% preserved	−0.32 ± 0.13 · 68.9% preserved

Source: Regeneron, GlobeNewswire, 1 Oct 2026

After the semaglutide stopped

At week 26, everyone in the higher-dose portion stopped semaglutide and was re-randomised to trevogrumab 400 mg alone or placebo alone for a further 26 weeks. Among participants originally on semaglutide + placebo, mean lean-mass change from baseline at week 52 was −2.4% if switched to placebo vs +0.4% if switched to trevogrumab 400 mg — a 2.8-percentage-point difference; on average, the switchers ended marginally above their starting lean mass. But total weight loss shrank from ~10% at week 26 to ~6% at week 52 similarly in all groups: trevogrumab did not prevent weight regain — it changed what the regain was made of. Medscape's EASD 2026 report puts the share of lost lean mass regained higher on trevogrumab 400 mg than placebo in every origin arm (semaglutide + placebo origin: 65% vs 48%).

Sources: Lancet in-press summary via Medical Xpress, 30 Sep 2026 · Medscape, EASD 2026

Safety — trevogrumab itself

Higher-dose portion · Adverse events, % of participants

Measure	Sema alone	+ 200 mg	+ 400 mg	Triplet
≥1 TEAE (interim, week 26)	64.9%	68.2%	68.2%	77.2%
Serious TEAE (interim)	0.7%	0.7%	1.3%	6.7%
AE → discontinuation (complete, week 26)	4.6%	4.7%	10.6%	30.9%
Muscle spasms	~4.6–5%	6.1%	9.3%	40.9%

Sources: Regeneron, 2 Jun 2025 (interim) · Regeneron, 17 Sep 2025 (complete) · Lancet in-press figures are near-identical (triplet SAE ~9%, discontinuation 32%)

In the lower-dose portion over 52 weeks, ≥1 adverse event was reported by 77% on trevogrumab vs 82% on placebo (both + semaglutide), and the most common events (≥10%) were the familiar semaglutide gastrointestinal set — nausea, constipation, diarrhoea, vomiting. Serious AEs / discontinuations (Lancet summary): placebo + semaglutide 5.3% / 15.8%; trevogrumab 25 mg 1.4% / 6.8%; 75 mg 3.3% / 10.5%. Muscle spasms remained the one clearly trevogrumab-attributable signal: 2.7% (25 mg) and 9.2% (75 mg) vs ~5% on semaglutide alone (Medscape, EASD 2026). Most events across arms were mild–moderate.

No serious signal has been attributed to trevogrumab alone in any release or conference report — no unexpected adverse events, no deaths, no cardiac, hepatic or tendon signal — and Medscape's 2026 report found no significant differences in triglycerides, total / HDL / LDL cholesterol or A1C with trevogrumab vs semaglutide alone.

Class concerns to teach alongside the data

• Muscle spasms / cramps — the reproducible on-target complaint across myostatin / activin blockade: trevogrumab ~6–9%, bimagrumab in a majority at high doses in earlier reports, the triplet 41%.
• Tendons (animal data only) — myostatin-null mice grow tendons that are smaller, brittle and hypocellular, with less type I collagen, greater stiffness and lower peak strain (Mendias et al., PNAS 2008), and their muscle is more susceptible to contraction-induced injury: bigger muscle pulling on under-adapted tendon. No tendon events were reported in COURAGE — but no human trial has been powered, or run long enough, to test this.
• Reproductive / endocrine — the activin axis regulates FSH and gonadal function. Regeneron's Phase 1 trial initially enrolled only postmenopausal women because of preclinical testicular-toxicity concerns, adding men “not intending to reproduce” only in a later part. That precaution attaches mainly to activin-A blockade, but long-term reproductive data for chronic myostatin blockade in younger adults do not exist — in a trial population that was 64% women, mean age 49.
• Cardiac — myostatin is expressed in heart muscle, and COURAGE's exclusions (hypertrophic cardiomyopathy, ECG LV-hypertrophy criteria) show the precaution was designed in. The one well-described human with congenital myostatin deficiency was monitored cardiac-wise with no cardiomyopathy or conduction disturbance detected by age 4.5 years — reassuring, but a single child, and too young to rule anything out.

Where it stands

Regulatory status: investigational, not approved for any indication in any country. Regeneron (Oct 2026): the safety and efficacy of trevogrumab “have not been evaluated by any regulatory authority.”

Publication: the higher-dose 26-week data were a late-breaking oral at EASD 2025 (Vienna); the full two-portion data were presented at EASD 2026 (Milan) and stated by Regeneron on 1 Oct 2026 to be in press at The Lancet (Mosenzon, Rosenstock et al.). As of 26 Oct 2026, no published Lancet article could be located — at least one secondary outlet already describes the data as “published in The Lancet,” which is premature against the sponsor's own wording.

Next steps — and it is not Phase 3: Regeneron is “planning to initiate a Phase 2 trial evaluating trevogrumab in combination with GLP-1 receptor agonist-based therapies in older adults with obesity and decreased muscle mass and/or strength.” Reuters adds that the 75 mg once-weekly dose has been selected and the planned trial is in patients over 65. No Phase 3 design, start date, or filing timeline has been announced in any source located; blog and analyst claims of “Phase 3 expected 2027, launch ~2030–31” are speculation, not company guidance.

The first life: trevogrumab began as a sarcopenia drug. In a Phase 2 trial (NCT01963598; n=253, age ≥70, 12 weeks), REGN1033 at 100 mg Q4W, 300 mg Q4W or 300 mg Q2W increased total lean body mass at all doses — placebo-adjusted differences of +1.66%, +1.78% and +2.29% — with MRI thigh-volume increases, fat-mass decreases at 300 mg, and only trends on strength and function tests. It was well tolerated, and the program was not advanced at the time — the classic class pattern, mass up and function unproven — before being revived for GLP-1 combination use. Those figures come from sponsor data filed in patent US12029788B2, a lower-tier source than a journal paper, and are flagged as such. Trevogrumab was never an SMA drug; that is apitegromab and taldefgrobep territory.

The rest of the field

Bimagrumab — Eli Lilly (ex-Novartis / Versanis)

ActRII-receptor blockade, and the most potent body-composition data in the field, now peer-reviewed. In the Phase 2 BELIEVE trial (n=507, obesity without diabetes; Nature Medicine 2026), high-dose bimagrumab (30 mg/kg IV q12w) + semaglutide produced −17.8 kg at week 48 vs −14.2 kg for semaglutide alone, −9.3 kg for bimagrumab alone and −3.3 kg for placebo — extending to −22.1% body weight at week 72 vs −15.7% in the open-label extension, with ~92–93% of combination weight loss from fat, lean mass −2.9% vs −7.4%, and bimagrumab monotherapy actually gaining lean mass (+2.5%) while losing only fat. The price is tolerability: muscle spasms, diarrhoea and acne. Still Phase 2 for obesity; Lilly halted a Phase 2b bimagrumab + tirzepatide trial in obesity with type 2 diabetes in September 2025 for “strategic business reasons” while continuing the non-diabetes program. Earlier, it failed a Phase 2/3 functional endpoint in sporadic inclusion body myositis despite increasing muscle volume.

Apitegromab — Scholar Rock

Anti-latent-myostatin — the most selective mechanism in the class. In the Phase 2 EMBRAZE trial (n=102, on tirzepatide, 24 weeks; Nature Medicine 2026), apitegromab 10 mg/kg IV q4w preserved an additional 1.9 kg of lean mass vs placebo (54.9% relative retention; p=0.001) with similar total weight loss — weight lost was ~85% fat / 15% lean vs ~70% / 30% on tirzepatide alone; adverse events were balanced (76% vs 71%, one serious AE in each arm). Uniquely, it also owns a functional win elsewhere: FDA-approved on 11 September 2026 as Isembyld (apitegromab-mstn) for SMA in patients ≥2 years on SMN-targeted therapy — the first approved muscle-targeted myostatin drug, though not for obesity, where it remains Phase 2. Its 2025 complete response letter concerned a third-party fill-finish facility, not the drug.

Taldefgrobep alfa — Biohaven (BHV-2000)

A small anti-myostatin adnectin–Fc fusion blocking ActRII signalling from myostatin and multiple activins. Its Phase 3 RESILIENT trial in SMA missed its primary functional endpoint (reported late 2024), though treated patients showed body-composition signals (fat −9.7% vs placebo, p<0.0003; muscle mass +2.1% numerically), and a Phase 1 study showed >6% fat-mass reduction with up to +4% lean mass after 29 days. Biohaven has pivoted it to obesity as monotherapy: a Phase 2 proof-of-concept (~150 adults) completed enrolment in March 2026, with topline data expected in 2H 2026 — no obesity efficacy results exist yet. Its selling point is a >700-participant safety database with muscle / GI adverse-event rates reportedly comparable to placebo.

Enobosarm — Veru

An oral SARM, not a myostatin drug — included because it competes for the same “quality weight loss” claim with a daily pill. In the Phase 2b QUALITY trial (n=168, age ≥60, on semaglutide, 16 weeks), enobosarm produced a 71% relative reduction in lean-mass loss vs semaglutide alone (p=0.002) — the 3 mg dose alone at >99% mean relative reduction (p<0.001) — with 27% more fat loss, similar total weight loss, and fewer patients losing clinically significant stair-climb function; in the maintenance extension after semaglutide stopped, enobosarm significantly reduced weight and fat regain. FDA feedback (September 2025) accepted 3 mg as the go-forward dose; the follow-on Phase 2b PLATEAU trial reached full enrolment in 2026. All figures are sponsor topline releases — no peer-reviewed publication yet — and SARM-class long-term safety (liver, lipids, prostate) remains the open question.

Comparison snapshot — not head-to-head

Program	Mechanism	Headline lean-mass result	Extra weight loss vs GLP-1 alone?	Status (Oct 2026)
Trevogrumab + semaglutide	Anti-mature-GDF8	DXA loss ~halved (high dose, 26 wk); 42.5% preserved at 75 mg / 52 wk	No (doublets)	Phase 2 complete; new Phase 2 planned
Bimagrumab + semaglutide	Anti-ActRII receptor	Lean −2.9% vs −7.4%; monotherapy +2.5%	Yes (−22.1% vs −15.7% at 72 wk)	Phase 2; T2D combo trial halted
Apitegromab + tirzepatide	Anti-latent myostatin	1.9 kg less lean loss (54.9%) at 24 wk	No	Phase 2 (obesity); approved for SMA
Taldefgrobep alfa	Adnectin, ActRII ligands	No obesity data yet	Unknown	Phase 2 obesity; topline due 2H 2026
Enobosarm + semaglutide	Oral SARM	71% less lean loss at 16 wk (3 mg >99%)	No (similar total loss)	Phase 2b PLATEAU enrolled

Different populations, GLP-1 backbones and durations — do not rank these directly against each other.

Reality check — this class has been here before

The recurring pattern of myostatin inhibition in humans: lean mass up, function flat.

• Stamulumab (MYO-029, Wyeth / Pfizer) — the first anti-myostatin antibody in humans. Acceptable safety, but no significant improvement in muscle strength or function; development stopped. A buried lesson: the antibody concentration needed for effect in monkeys was ~20× that in mice — animal potency badly over-predicted human potency.
• Domagrozumab (PF-06252616, Pfizer) — increased lean mass ~5% and muscle volume ~4% in healthy humans, then failed the 4-stair-climb test in Duchenne muscular dystrophy; the Phase 2 was terminated for lack of efficacy in 2018.
• Landogrozumab (LY2495655, Lilly) — the partial exception that proves the rule: in sarcopenic adults ≥75 who had fallen, 20 weeks added 0.43 kg of appendicular lean mass with some functional gains in one trial — but it failed its primary endpoint after hip replacement and failed overall survival in pancreatic-cancer cachexia, and was not advanced.
• Bimagrumab itself failed its Phase 2/3 primary functional endpoint (6-minute walk) in sporadic inclusion body myositis despite increasing thigh muscle volume — the same molecule now succeeding on body composition in obesity.

Why mass ≠ function — the caveats, stated plainly

• DXA “lean mass” is not contractile muscle. It includes water, organs, skin and connective tissue, and shifts with hydration and glycogen during rapid weight loss. COURAGE's MRI substudy — actual thigh muscle volume — is the stronger measurement, and it covered ~76 participants, one muscle group; volume still isn't strength.
• Bigger fibres aren't proportionally stronger fibres. Myostatin-null mouse muscle is larger but shows reduced specific force (force per cross-sectional area), a shift toward fast-glycolytic fibres, fewer mitochondria and greater fatigability (Amthor et al., PNAS 2007). Adult antibody blockade in animals yields ~20–30% muscle gains, not the 2–3× knockout phenotype.
• Tendon and support tissues may not keep up (Mendias 2008, above): smaller, stiffer, more brittle tendons in myostatin-null mice, plus greater susceptibility to contraction-induced injury.
• COURAGE itself proves nothing about function. No strength, mobility or patient-reported functional endpoint was reported. EASD 2026 discussant Caterina Conte made exactly this point — lean-mass loss after bariatric surgery and in STEP-UP occurred without measurable loss of endurance function — and the Lancet authors themselves state that whether preservation translates into functional benefit “requires further investigation.”
• The congenital-deficiency phenotype is a single case. The human evidence for “no myostatin = healthy super-muscle” is essentially one child (Schuelke et al., NEJM 2004): a loss-of-function MSTN mutation, gross hypertrophy at birth, unusual strength at 4.5 years — and, at last report, no health problems observed, with cardiac function specifically monitored and normal, but explicitly too young to exclude later problems. Lifelong, complete deficiency from conception is not a model for partial, adult-onset, reversible antibody blockade — in either direction.

Claim check

Claims circulating in media and social coverage of trevogrumab, checked against the primary sources:

• Not supported

  “Trevogrumab builds muscle on Ozempic / Wegovy.”

  In every COURAGE combination arm, patients still lost lean mass (−3.3% to −5.8% on DXA). It preserves a fraction of what would have been lost; net gain appeared only in the post-semaglutide withdrawal phase (+0.4% vs baseline, one switch sequence) and in a single-dose Phase 1 study in healthy postmenopausal women.

• Needs the qualifier

  “It preserves 70% of muscle.”

  Only with all of them attached: ~69–72% preservation of thigh fat-free muscle volume, by MRI, in a predefined substudy of ~76 participants, in the lower-dose portion. Whole-body DXA preservation in the same trial was 20.5–42.5% at 52 weeks.

• Needs the qualifier

  “It halves muscle loss.”

  True only for specific cells: DXA at 26 weeks (75 mg: 50.7%; 200/400 mg: ~49–51% by the interim calculation). At 52 weeks the 75 mg figure is 42.5%, and the 25 mg dose's week-52 effect was not statistically significant.

• Not supported

  “Trevogrumab melts extra fat / boosts GLP-1 weight loss.”

  Not for the doses going forward. Extra fat loss appeared in the higher-dose portion (fat −17.3% / −19.1% vs −15.7%) and dramatically in the abandoned triplet (−27.1%); for the lower doses, Regeneron states weight loss was not meaningfully enhanced and published no fat-mass numbers in October 2026. Doublet total weight loss (≈ −10%) matched semaglutide alone.

• Not supported

  “The triplet is the breakthrough combo.”

  It is discontinued: ~41% muscle spasms, ~9% serious AEs, ~31–32% discontinuations, and the trial's only two deaths in that arm. Regeneron is not advancing it.

• Not supported

  “Trevogrumab killed two trial patients” — or, conversely, “proven completely safe.”

  Both directions overreach. The two deaths were only in the garetosmab-containing triplet arm, in patients with multiple cardiovascular risk factors / established cardiovascular disease, and the sponsor states no causal association was identified. But a 975-person Phase 2 also cannot prove safety: no regulator has reviewed the data, muscle spasms are a real dose-related effect, and discontinuations reached 10.6% at 400 mg.

• Needs the qualifier

  “It restores your muscle after you stop the GLP-1.”

  The +0.4%-of-baseline figure is a group mean for one sequence (semaglutide + placebo → trevogrumab 400 mg); placebo switchers also regained most lean mass (−2.4% vs baseline), and everyone regained weight (net loss ~10% → ~6%). It is a composition-of-regain signal, not restoration-therapy evidence.

• Not supported

  “Preserving lean mass improves strength, function and metabolism.”

  Not measured or reported in COURAGE. Lipid and A1C changes were numeric in 2025 and not significantly different in the 2026 analysis; the authors state functional translation is unproven. This is the class's historical failure point.

• Not supported

  “Phase 3 is starting / approval is coming in 2027.”

  No Phase 3 has been announced. The announced next step is another Phase 2 — older adults, 75 mg once weekly. Any launch-date arithmetic in blogs is speculation.

• Not supported

  “You can buy trevogrumab (peptide) online.”

  Trevogrumab is a ~144 kDa monoclonal antibody given by weekly subcutaneous injection, available only inside trials. It is not a peptide, is not sold for human use, and grey-market listings using its name have no connection to the COURAGE data.

• Needs the qualifier

  “35% of GLP-1 weight loss is muscle.”

  COURAGE's complete analysis says 33% of semaglutide weight loss was lean mass — DXA lean soft tissue, which is not all skeletal muscle. The 34.5% / 35% figure was the June 2025 interim cut, and the lean-mass share varies widely (~15–45%) across incretin studies by drug, dose, population and method.

Open questions

Marked plainly, because they are open:

• Lower-dose fat-mass and body-weight numbers — the 1 Oct 2026 release gives only the lean-mass and MRI tables plus “did not meaningfully enhance weight loss.” Exact values await the Lancet article, currently in press.
• The two deaths — no source located gives ages, sexes, timing on treatment, case narratives, or any FDA / EMA assessment; only the sponsor's “no causal association identified” statement exists publicly. One 2026 secondary report described the triplet arm as “discontinued” mid-trial; the timing and mechanics of that decision were not in a primary source.
• Whether the Lancet paper has actually published — sponsor wording (1 Oct 2026) is “in press”; one secondary outlet (23 Oct 2026) says “published.” No published article was located as of 26 Oct 2026; treated here as in press.
• Phase 3 — none announced in any primary source as of 26 Oct 2026 (absence checked across Regeneron's three releases and Reuters), though a very recent announcement cannot be absolutely excluded.
• REGN1033 in sporadic inclusion body myositis — appears only in a Scholar Rock patent application's background list; no primary Regeneron or registry source was confirmed, so it is not asserted anywhere on this page.
• Part-letter naming — sources variously label the lower-dose portion Part B or Part C; the registry contains Parts A, B and C, and Regeneron's releases avoid letters — hence “portion” language throughout.

Sources

1. Regeneron, GlobeNewswire, 1 Oct 2026 — lower-dose COURAGE 52-week results. globenewswire.com/news-release/2026/10/01/3373155
2. Regeneron, GlobeNewswire, 17 Sep 2025 — complete higher-dose 26-week results. drugs.com/clinical_trials/results-phase-2-courage-trial-22165
3. Regeneron, GlobeNewswire, 2 Jun 2025 — interim results and safety table. biospace.com/press-releases/interim-results-from-ongoing-phase-2-courage-trial
4. ClinicalTrials.gov — COURAGE record, NCT06299098 (verified live 26 Oct 2026: active, not recruiting; no results posted). clinicaltrials.gov/study/NCT06299098
5. Mosenzon, Rosenstock et al., COURAGE, in press, The Lancet — summary via EASD / Medical Xpress, 30 Sep 2026. medicalxpress.com/news/2026-09-trevogrumab-anti-myostatin-mass-semaglutide
6. Gonzalez Trotter et al., Nature Communications 2025;16:4376 — Phase 1, trevogrumab ± garetosmab. doi.org/10.1038/s41467-025-59380-3
7. Mastaitis et al., Nature Communications 2025;16:4377 — mechanism, preclinical. nature.com/articles/s41467-025-59485-9
8. Medscape — EASD 2025 conference report. medscape.com/viewarticle/myostatin-blocker-preserves-muscle-glp-1-treatment-2025a1000qs4
9. Medscape — EASD 2026 conference report. medscape.com/viewarticle/trevogrumab-curbs-muscle-loss-glp-1-aids-regain-after-2026a10011r8
10. Reuters, 30 Sep 2026 — triplet decision; 75 mg dose selected. reuters.com/business/healthcare-pharmaceuticals/regeneron-drug-halves-muscle-loss
11. Di Rocco et al., Nature Medicine 2023 — garetosmab LUMINA-1 in FOP. pmc.ncbi.nlm.nih.gov/articles/PMC10579054
12. Journal of Bone Metabolism 2020 — myostatin-inhibitor class history review. pmc.ncbi.nlm.nih.gov/articles/PMC7571243
13. Mendias et al., PNAS 2008 — tendon findings in myostatin-null mice. explore.openaire.eu/search/publication?pid=10.1073/pnas.0707069105
14. BELIEVE trial, Nature Medicine 2026 — bimagrumab + semaglutide, abstract. ouci.dntb.gov.ua/en/works/7PMddEBe
15. Pharmacy Times — BELIEVE body-composition data (ADA 2025). pharmacytimes.com/view/bimagrumab-semaglutide-combination
16. EMBRAZE trial, Nature Medicine 2026 — apitegromab + tirzepatide. nature.com/articles/s41591-026-04440-4.pdf
17. SMA UK, Sep 2026 — apitegromab (Isembyld) FDA approval for SMA. smauk.org.uk/news/apitegromabs-us-approval
18. Veru, 27 Jan 2025 — enobosarm QUALITY topline release. ir.verupharma.com/news-events/press-releases/detail/225
19. Biohaven, Mar 2026 — taldefgrobep obesity Phase 2 enrolment complete (via Finviz). finviz.com/news/338586
20. US Patent US12029788B2 — REGN1033 sarcopenia data as filed by the sponsor. patents.google.com/patent/US12029788B2
21. Phase 1 peer-review file, Nature Communications 2025 — enrolment precaution (postmenopausal women first). static-content.springer.com (supplementary file)

Molecule

Thiadiazol-3-one

CAS · MW

1261289-04-6 · 380.24

Now called

ATX-304

Status

Investigational — approved nowhere

What it is — and how it works

O-304 (formal name 4-chloro-N-[2-[(4-chlorophenyl)methyl]-2,3-dihydro-3-oxo-1,2,4-thiadiazol-5-yl]benzamide; formula C₁₆H₁₁Cl₂N₃O₂S; PubChem CID 50923806) is a small heterocyclic thiadiazolone from Betagenon AB / Umeå University, described by its developers as a 4th-generation AMPK-activator scaffold, rationally designed from a cellular-screen hit. The free base was given as a suspension; from about 2020 a sodium-salt solid formulation (O304 Na-salt, CAS 2639972-26-0) replaced it, and after 2023 the compound is ATX-304. It is a crystalline solid, DMSO-soluble (10 mg/mL) and very poorly water-soluble (DMSO:PBS 0.03 mg/mL) — a fact that drives much of its pharmacology below.

Source: Cayman Chemical product page · Steneberg et al., JCI Insight 2018 supplement

Protection from dephosphorylation — not allosteric activation

From Steneberg et al., JCI Insight 2018;3(12):e99114: in vitro, O304 dose-dependently suppressed PP2C-mediated dephosphorylation of phospho-Thr172 on human recombinant AMPK α/β/γ trimers — without inhibiting PP2C phosphatase activity itself and without allosterically activating AMPK (negative in the SAMS peptide assay). It protected phospho-AMPK even in the presence of excess ATP, was additive with ADP, and the authors conclude it mimics ADP, not AMP.

The exact binding site is not characterised: the paper explicitly leaves open whether O304 binds the γ-subunit nucleotide sites like ADP or a distinct site, and a 2025 medicinal-chemistry review likewise states the binding site remains unknown.

• Pan-AMPK: it raised phospho-Thr172 AMPK and downstream phospho-Ser79 ACC in many cell types expressing either the β1 or β2 subunit — human skeletal myotubes, hepatocytes, preadipocytes, endothelial cells, Wi-38 fibroblasts — hence “pan-AMPK activator” across α1/α2-containing trimers.
• LKB1-dependent and self-limiting: in LKB1-null HeLa cells O304 could not raise the near-zero basal phospho-AMPK, whereas a CaMKK activator (ionomycin) could. O304 therefore only amplifies AMPK activity where upstream phosphorylation already occurs — the developers' proposed explanation for its cleaner cardiac profile versus ADaM-site activators (see Cardiac, below).
• No energy stress: O304 raised, not lowered, cellular ATP in Wi-38 cells. The 2026 ADA companion poster (#1788-P) re-confirmed AMPK activation independent of ATP changes, with ATP preserved at all concentrations.
• A second action (2023): Norlin et al. showed O304 stimulates mitochondrial uncoupling in C2C12 myotubes (dose-dependent rise in oxygen consumption), generating the metabolic demand that lets increased glucose uptake be utilised rather than stored as glycogen — the basis of the developer's current “dual AMPK / mitochondrial activator” positioning. Cayman now also lists “mitochondrial uncoupling activity.”

How this differs from indirect activators: indirect activators (metformin; AICAR via ZMP in its energy-sensing role) work by changing the cell's AMP/ADP:ATP ratio or mimicking AMP at the γ subunit — they depend on, or imitate, an energy-deficit signal. O304 neither creates an energy deficit nor mimics AMP; it acts downstream of phosphorylation by slowing phospho-Thr172 removal, so its effect is proportional to existing kinase flux and tissue AMPK activity. Versus direct ADaM-site activators (MK-8722, PF-739), it does not force catalytic activity in unphosphorylated AMPK.

Human data — every clinical trial found

All O-304 / ATX-304 trials in the public record

Study	Population / n	Dose & duration	Status / identifier
Phase I SAD / MAD (first-in-man, ~2014)	Young healthy, overweight middle-aged, and T2D subjects (incl. + metformin cohorts); n not published	100–2,400 mg/day, single & multiple ascending doses	Completed; no public registry entry found (adult Phase I trials are not publicly listed in the EU register)
TELLUS — Phase IIa (2016–17)	T2D on metformin ≥3 months, HbA1c 6.5–9.0%, age 18–80; n=65 randomised (33 O304 / 32 placebo)	1,000 mg/day oral suspension (20 mg/mL, 50 mL each morning) × 28 days	Completed. EudraCT 2016-002183-13, protocol O304-2016-02. No NCT number (EU register field: “–”)
Bridging PK / safety (2020)	Healthy volunteers	O304 Na-salt solid, single + multiple dosing, ± dapagliflozin at steady state; open, randomised, parallel-group	Started Aug 2020; no results published found
Phase 1b translational, ATX-304 Na-salt tablets (2023–25)	Obesity with prediabetes (ADA report: 23 obese prediabetic adults, randomised 2:1)	400 mg once daily × 8 weeks + optional 8-week open-label extension (to 16 weeks)	Completed ~May 2025 (NL/DE). EUCT2023-505967-36-00; Dutch registry NL-OMON60115; sponsor listed as Betagenon AB. Reported ADA June 2026

Sources: Steneberg supplement & Cision releases · EU Clinical Trials Register TELLUS results (verified live 26 Oct 2026) · AdisInsight trial profile & Dutch registry · ADA abstract 1782-P

Phase I — summary data only

The 2018 supplement states: no serious adverse events in the Phase I single/multiple ascending-dose study at 100–2,400 mg daily doses or in TELLUS; no dose- or treatment-related increase in AE frequency or intensity; most AEs mild; no clinically significant physical-exam, vital-sign, ECG or laboratory findings. Betagenon press releases add that Phase I covered young healthy and overweight middle-aged subjects and T2D patients in combination with metformin, “did not cause any significant adverse effects, was very well tolerated and showed the expected pharmacokinetics,” and established the dose range for Phase IIa. No peer-reviewed Phase I paper, sample size, or numeric PK parameter set was ever published — any source giving precise Phase I n's or half-life numbers is going beyond the public record.

TELLUS — the only peer-reviewed human trial

Single-centre (Uppsala, Sweden), randomised, double-blind, placebo-controlled, parallel-group, 28 days, all patients on stable metformin monotherapy. Completers: 31/33 (O304) and 28/32 (placebo). Mean age 66.2 vs 64.6 years; 17 women / 48 men overall (registry baseline table).

• HOMA-IR / insulin sensitivity: significant within the O304 group (absolute p=0.0097) but the comparison vs placebo did not reach significance (sponsor release). Twelve days after stopping (day 40), FPG had risen and HOMA-IR relatively increased in the O304 group vs placebo — effects reversed off-treatment.
• Blood pressure: O304 −5.8 mmHg systolic / −3.8 diastolic vs +1.2 / +0.9 on placebo at day 28 (within-group p=0.030 / 0.009; 1-way ANOVA vs placebo p=0.047 / 0.044). No heart-rate change.
• Microvascular perfusion: MRI T2* calf-muscle hyperaemic perfusion increased vs placebo (Δ-T2* p=0.026, 2-way ANOVA), driven by patients with the worst baseline perfusion (TTP shortening p=0.043 in the long-TTP half).
• HbA1c: not an outcome — it was a screening criterion only; 28 days is too short to interpret HbA1c. Body weight: no weight outcome reported. Claims that TELLUS showed weight loss or HbA1c reduction are not supported.
• Cardiac in humans: no ECG abnormalities and no effect on LV mass reported in TELLUS (paper discussion).
• Hypoglycaemia: no hypoglycaemia event appears in the registry AE table and none is reported in the paper — consistent with an insulin-independent, glucose-dependent mechanism, but the trial was small, short, and in metformin-treated patients, so this is “no signal observed,” not proof of absence.

TELLUS · Adverse events (EU registry, verified live 26 Oct 2026)

Measure	O304 (n=33)	Placebo (n=32)
≥1 AE	19/33 (57.6%; 38 events)	23/32 (71.9%; 49 events)
Serious AEs	0	1 (renal colic)
AE-related discontinuations	0	2

Most frequent O304 non-serious AEs: nasopharyngitis 5/33 (15.2%; placebo 34.4%), headache 3/33, diarrhoea 3/33, fatigue 2/33, hard faeces 2/33, muscle spasms 2/33 — a profile with no dose-limiting pattern in a 28-day window. No clinically significant lab, vital-sign or physical-exam changes in either arm.

Sources: Steneberg et al., JCI Insight 2018 · EU Clinical Trials Register, EudraCT 2016-002183-13 results · Betagenon / Cision TELLUS release

The newest human data — Phase 1b in obesity / prediabetes

Abstract #1782-P, Diabetes 75 Suppl 1 (published 5 Jun 2026): n=23 obese prediabetic adults, 2:1 randomisation, 400 mg/day × 8 weeks (+ optional open-label extension to 16 weeks). “Safe and well tolerated”; total adiponectin significantly increased; liver and visceral fat (MRI) significantly reduced, time-dependent; resting metabolic rate “increased up to 33% compared to baseline.”

The sponsor press release (18 Jun 2026) reports RMR +8% (p<0.01) — note the discrepancy with the abstract's “up to 33%” (the release figure appears to be the mean effect; the abstract's phrasing is ambiguous and should not be quoted as a mean). Also: triglycerides ↓ (p<0.01), adiponectin ↑ (p<0.01), liver fat by MRI-PDFF ↓ (p<0.05), visceral adipose tissue ↓ (p<0.05); “minimal weight loss was observed at this exposure level”; TEAEs predominantly mild, similar frequency to placebo; no increase in continuously monitored core body temperature or 24-hour heart rate (the key uncoupler-safety readouts); no new signals in the extension; PK exposure reached the preclinically predicted threshold. No glucose / HbA1c efficacy claim was made for this study.

Pharmacokinetics — the thin part of the record

Published PK is thin — this is a genuine gap in the record. All human dosing has been oral, once daily.

• Human half-life: no numeric value has been published. What is published: plasma steady state was not reached until ~day 14 in TELLUS patients on daily dosing, attributed by the authors to a long plasma half-life. Any specific human half-life figure circulating online (e.g. “~30 hours”) has no primary source in the retrieved record.
• TELLUS exposure (EU registry, verified live; n=31, suspension 1,000 mg/day): median Cmax 55.70 µg/mL (range 28.10–89.30), Cmin 46.60 µg/mL (22.20–74.00), Css 52.74 µg/mL (24.56–82.88), AUCtau 1,265.7 µg·mL⁻¹·h (589.38–1,989.2). The flat Cmax≈Cmin profile corroborates a long half-life / accumulation. Inter-individual spread is ~3-fold.
• Formulation dominates exposure: the original suspension had limited oral bioavailability (sponsor's own wording, 2020); the Na-salt solid markedly improved bioavailability in animals and is the form used since (bridging study; Phase 1b tablets; planned Phase 2). No absolute bioavailability % (F), Tmax, clearance, volume, protein binding, metabolite identity, CYP / elimination pathway, or food-effect data have been published for humans or animals in the retrieved record.
• Rodents (animal data): O304 is “orally available with a long plasma half-life” (40 mg/kg gavage; plasma sampled 4–24 h in mice and rats) and does not cross the blood-brain barrier (rat liver/brain measurement after 3 weeks' dosing) — consistent with Cambrian's 2026 description of ATX-304 as “peripherally restricted.”

Sources: Steneberg et al. 2018, main text & supplement · EU Clinical Trials Register TELLUS results

Preclinical evidence — animal data only

Doses below are animal doses and are not human-equivalent recommendations. The “exercise mimetic / energy expenditure” paper is Steneberg et al. 2018 in JCI Insight (not Cell Metabolism), extended by Ericsson et al. 2021 in Communications Biology (aged mice) and Norlin et al. 2023 (uncoupling).

Steneberg et al. 2018 — DIO and hIAPP-transgenic mice; rats; dogs

• Gavage studies: O304 100 mg/kg/day (alone or + metformin) for 8 weeks in high-fat-diet mice prevented the rise in fasting glucose, fasting insulin and HOMA-IR; the combination was most effective and significantly better than O304 alone on HOMA-IR; metformin alone showed no significant effect in this paradigm. In already-obese / insulin-resistant mice, 4 weeks of O304 + metformin normalised fasting glucose to regular-diet levels (insulin / HOMA-IR improved by week 1, glucose by week 2).
• Diet-admixed dose–response: 0.4 / 0.8 / 2 mg/g diet for 7 weeks dose-dependently raised muscle phospho-AMPK and prevented dysglycaemia / hyperinsulinaemia / insulin resistance. In hIAPPtg mice with established diabetes, switching to 2 mg/g reversed hyperglycaemia, hyperinsulinaemia and insulin resistance.
• Body composition / energy expenditure: at 2 mg/g, O304 caused body-weight and body-fat loss despite increased food intake, including at thermoneutrality (30 °C), with only a non-significant +0.2 °C core-temperature difference. Indirect calorimetry (0.8 mg/g, 11 weeks): VO₂ and energy expenditure significantly increased in light and dark phases; RER fell (shift to fat oxidation). At 0.8 mg/g the glycaemic benefit occurred with no weight / fat loss — glucose effects are not merely secondary to weight loss in mice. Mechanistically: WAT lipolysis (ATGL Ser406 phosphorylation ↑), fatty-acid-oxidation genes ↑ in WAT/BAT, hepatic de-novo lipogenesis ↓ ~45%, liver steatosis prevented and reduced.
• Glucose uptake: insulin-independent, AMPK-dependent 2-deoxyglucose uptake in myotubes; in-vivo FDG-PET uptake ↑ in calf / thigh muscle and heart (2 mg/g, 2 weeks).
• β-cells: reduced islet amyloid in hIAPPtg mice and in isolated islets at high glucose (autophagy-dependent), prevented ER-stress / UPR gene induction, and doubled arginine-stimulated insulin secretion after 11 weeks at 0.8 mg/g — interpreted as β-cell rest / preservation.
• Endurance: weight-matched 14-month-old lean mice gavaged 20 mg/kg/day × 30 days ran significantly farther to exhaustion with a smaller post-exercise lactate rise. Microvascular perfusion in hind-paw ↑ at 40 mg/kg/day × 8 weeks. Dogs (single-dose telemetry, 60 / 180 / 540 mg/kg): acute systolic / diastolic BP reduction.

Ericsson et al. 2021 — aged mice

F1 hybrid mice fed 0.25 or 0.5 mg/g from 6 months (12-month follow-up) or from 12 months (6-month treatment). O304 dose-dependently prevented and reversed age-associated hyperinsulinaemia / insulin resistance, increased muscle FDG uptake, improved echocardiographic stroke volume, cardiac output, end-diastolic volume and ejection fraction, lowered heart rate, and increased treadmill running distance to exhaustion in 18-month-old mice. This is the core “exercise mimetic in aged mice” dataset — mice only; there is no human endurance or exercise-capacity trial.

Norlin et al. 2023 — severe diabetes models

In streptozotocin (insulin-deficient) and db/db mice, O304 stimulated insulin-independent glucose uptake and utilisation in muscle and heart while preventing glycogen accumulation, averted metabolic-inflexibility gene-expression changes, reversed diabetic cardiomyopathy (see Cardiac), and in db/db mice preserved insulin secretion, β-cell mass and pancreatic insulin content.

Other animal work (brief, all preclinical): islet gene-expression / epigenetic protection in DIO mice (Sci Rep 2021); renal ageing / autophagy protection in uninephrectomised aged mice (Zhu et al. 2022, 10 mg/kg/day i.p.); abdominal aortic aneurysm suppression in ApoE⁻/⁻ + angiotensin-II mice (Sun et al. 2024, 200 mg/kg oral); PCSK9 / cholesterol lowering claimed in obese rats in a sponsor abstract. None of these has human corroboration.

Sources: Steneberg et al., JCI Insight 2018 · Ericsson et al., Commun Biol 2021 · Norlin et al., Commun Biol 2023

Where it stands

2014–2017 (Betagenon / Baltic Bio, Umeå): Phase I completed; EU Horizon 2020 project 754268 (AMPK-DIAB) funded TELLUS explicitly to generate a Phase IIa package for out-licensing to a Phase IIb/III partner. That partner never publicly materialised under Betagenon — the program spent ~5 years in partnering limbo, during which Betagenon worked on the formulation problem (limited-bioavailability suspension → Na-salt solid, bridging trial started Aug 2020) and broadened positioning to heart failure, renal disease and insulin resistance.

No public statement was found attributing the stall to a safety or efficacy failure. The stated technical problem was formulation / bioavailability; the stated commercial plan was licensing. The TELLUS data themselves (post-hoc primary signal, modest FPG effect, baseline imbalances) are a plausible commercial headwind, but that is inference, not a stated reason — flagged as such.

14 March 2023 — acquisition, not termination: Cambrian Bio launched Amplifier Therapeutics and acquired Betagenon AB; O304 was renamed ATX-304; Betagenon CEO James Hall leads Amplifier. BioSpace reported the acquisition at ~$26M.

2023–2026 (Amplifier / Cambrian): Phase 1b Na-salt tablet study run in NL/DE (completed ~May 2025), ADA June 2026 readout (above), and two planned Phase 2 studies: REWIRE-1 (higher exposures; muscle function and lipid metabolism) and REWIRE-2 (weight-loss proof-of-concept in obesity), with incretin-combination / weight-regain-prevention positioning based on animal data.

How it compares

O-304 against the field — mechanism, maturity, route

Compound	Target / mechanism	Human evidence & route	Cardiac glycogen / hypertrophy
O-304 / ATX-304	AMPK, anti-dephosphorylation of phospho-Thr172 (ADP-mimetic); + mild mitochondrial uncoupling (2023)	Phase I + Phase IIa (n=65, 28 d) + Phase 1b (n=23, 8 wk); no approval. Oral (suspension → Na-salt tablet)	Glycogen ↓, no hypertrophy in mice / rats (see nuance above)
AICAR (acadesine)	Intracellular ZMP (AMP mimetic at γ subunit): allosteric activation + anti-dephosphorylation + ↑ phosphorylation	Human IV trials in other indications (CABG, oncology); no chronic metabolic development. Practically IV; poor oral bioavailability, short IV half-life, ↑ lactate / uric acid in humans	—
Metformin	Indirect: lysosomal PEN2–v-ATPase → LKB1 at clinical doses; complex I inhibition only at supra-pharmacological mM. Tissue-skewed (liver / gut, transporter-dependent); weak muscle AMPK effect	Approved drug, decades of outcome data. Oral. No hypoglycaemia as monotherapy (established)	—
MK-8722	AMPK, direct allosteric ADaM-site, all 12 complexes; pan, ~10× β1>β2 preference; works without phospho-Thr172	None — rodents + rhesus only; development halted at preclinical stage amid cardiac findings. Oral (preclinical)	Hypertrophy + glycogen ↑ (rodents, monkeys)
PF-739 / PF-249	AMPK, direct allosteric ADaM-site; PF-739 pan (β1+β2), PF-249 β1-selective. PF-739 activates muscle + liver; PF-249 liver only — and PF-249 did not lower glucose in diabetic mice	None — mice + cynomolgus only. Oral (preclinical)	Not the reported liability; chronic glucose effect of the related PF-793 was limited
SLU-PP-332	Not AMPK — pan-ERRα/β/γ nuclear-receptor agonist (PGC-1α transcriptional program); ERRα most potent (EC50 ~98 nM)	None — no human trial, PK, or dose exists. Mouse studies i.p. (50 mg/kg BID); reported to lack oral bioavailability (oral analogue SLU-PP-915 followed)	Improved HF-model cardiac function in mice (2024 Circulation report, secondary sources)

Bottom line of the comparison: O-304 is the only direct AMPK activator with randomised human metabolic data, and its anti-dephosphorylation mechanism is genuinely distinct from every ADaM-site compound. Its human evidence is still only ~88 randomised patients for ≤8 weeks (plus Phase I exposures). SLU-PP-332, despite being marketed alongside it, is a different target class with zero human data. Hypoglycaemia: none reported for O-304 (small trials).

Sources: Steneberg 2018 & discussion · Myers 2017 abstract · Cokorinos et al., Cell Metab 2017 · AICAR reviews (PMC5945561 / PMC9664587) · Ma et al., Nature 2022 via reviews · Billon et al. 2023/2024 via secondary summaries. Note: Steneberg cites this group's compound as “PF-793,” the paper itself reports PF-739 / PF-249.

The grey market — powder and capsules today

• Legitimate research supply: Cayman Chemical and similar suppliers sell the free base as a crystalline solid in milligram quantities (5–100 mg), labelled ≥98% purity, storage −20 °C, “not for human or veterinary use.” Solubility data on those listings (DMSO 10 mg/mL; aqueous ~0.03 mg/mL) match the formulation problem above.
• Bulk supply: Chinese bulk manufacturers list O-304 / ATX-304 powder at “99% HPLC” in 1 kg foil bags / 25 kg drums — commodity-scale synthesis far beyond laboratory demand (Echemi listings).
• Consumer-facing formats: capsules of the sodium salt, 50 mg × 60 (one vendor guide) and 100 mg capsules (a European research-chemical shop, ~€139); combination liquids (e.g. ATX-304 + 5-amino-1MQ blends); and marketplace listings such as “ATX-304, O-304 Peptide Powder… Tested 99.93% Purity, 5 Grams” — O-304 is not a peptide, an elementary identity error in the listing itself.

Purity / identity caveats documented in the landscape

• Two different substances, one name. Free base (CAS 1261289-04-6, MW 380.24) and sodium salt (CAS 2639972-26-0, MW ~402–403) are sold interchangeably as “O-304 / ATX-304.” They differ in mass (~6%), solubility and — per the sponsor's own bridging work — bioavailability. A capsule labelled only “ATX-304 50 mg” does not tell a buyer which form, or whether the stated mass is salt or free-base equivalents.
• Vendor copy errors signal weak identity control. At least one bulk listing for “O-304 sodium salt” describes it as a CXXC5–DVL / Wnt/β-catenin activator for hair regrowth — text belonging to an unrelated compound — and gives an internally inconsistent molecular formula (Echemi LYNA listing). If the mechanism text is copied from another product, the CoA deserves the same scepticism.
• Purity figures (98–99.9%) are vendor-generated CoA claims. No independent, published analytical verification of grey-market O-304 (identity by NMR/MS, assay, impurity profile, salt form) was found in the literature; there is also no published study of grey-market material's stability — relevant for a poorly soluble thiadiazolone sold in capsules at room temperature.
• Dose-form mismatch with the trials. The human data come from a GMP suspension at 1,000 mg/day and GMP Na-salt tablets at 400 mg/day. Grey-market 50–100 mg capsules have no demonstrated equivalence to either, and community dosing protocols are extrapolations, not trial regimens.

Claim check

Claims circulating online about O-304, checked against the primary sources:

• Not supported

  “Published in Cell Metabolism 2018.”

  Wrong journal — JCI Insight 2018;3(12):e99114. The Cell Metabolism AMPK papers of that era are the Pfizer PF-739 / PF-249 work.

• Not supported

  “Trial NCT03351478.”

  That NCT is Lexicon's sotagliflozin trial; TELLUS has no NCT number (EudraCT 2016-002183-13 only). The NCT numbers printed in the 2018 paper's study-approval paragraph (NCT00508287, NCT01167881) likewise do not correspond to TELLUS in the EU register and should not be cited for it.

• Not supported

  “Lowered HbA1c / caused weight loss in diabetics.”

  No HbA1c or weight outcome exists for TELLUS; the 2026 Phase 1b at 400 mg showed minimal weight loss.

• Needs the qualifier

  “Resting metabolic rate increased 33%.”

  The ADA abstract says “up to 33%”; the sponsor's own release reports +8% as the study result. Quote the 8% (or flag both).

• Not supported

  “Treats heart failure.”

  No HF trial and no HF model in the published record; sponsor positioning only. TELLUS excluded NYHA III–IV heart failure.

• Not supported

  “Discontinued / abandoned drug.”

  Superseded by the 2023 acquisition and 2025–26 trials; one pipeline database's “Discontinued” tag refers to the old T2D path.

• Not supported

  “No human safety data / never tested in healthy volunteers.”

  False; Phase I SAD/MAD in healthy subjects at 100–2,400 mg/day is reported (in summary form).

• Not supported

  “Half-life is X hours” (any precise figure).

  No numeric human half-life is published; only “long,” steady state ~day 14, and the flat TELLUS concentration profile.

Open questions

Marked plainly, because they are open:

• Numeric human half-life, absolute oral bioavailability, Tmax, metabolism / elimination pathway — not published in any retrieved primary source; only qualitative statements and TELLUS steady-state concentrations exist.
• Phase I SAD/MAD sample sizes and per-cohort results — never published beyond the supplement's summary safety paragraph and press releases.
• Bridging Na-salt PK study (2020) results — announced as started; no results publication or registry results entry found.
• Phase 1b (EUCT2023-505967-36-00) full dataset — only the ADA abstract + sponsor release are public; registry outcomes are under deferred disclosure (Dutch registry). Effect sizes for liver fat / VAT / triglycerides beyond p-values are not public.
• Whether REWIRE-1 / REWIRE-2 have actually started recruiting (as opposed to “planned,” June 2026) — no registry entries found as of 26 Oct 2026.
• Independent analytical verification of grey-market O-304 purity / identity — none found.
• The 2018 paper's printed NCT numbers could not be reconciled with any O304 trial; the EU register (verified live) records “US NCT number: –” for TELLUS.

Sources

1. Steneberg P, et al. PAN-AMPK activator O304 improves glucose homeostasis and microvascular perfusion in mice and type 2 diabetes patients. JCI Insight 2018;3(12):e99114. pmc.ncbi.nlm.nih.gov/articles/PMC6124394 · Supplement: jci.insight.99114.sd.pdf
2. EU Clinical Trials Register — TELLUS results, EudraCT 2016-002183-13 (verified live, 26 Oct 2026). clinicaltrialsregister.eu/ctr-search/trial/2016-002183-13/results
3. Betagenon / Cision: TELLUS positive results release. news.cision.com/betagenon-ab (TELLUS results) · Phase IIa start / Phase I summary: news.cision.com/betagenon-ab (Phase IIa start) · Bridging Na-salt PK trial, 17 Aug 2020: news.cision.com/betagenon-ab (bridging trial)
4. Ericsson M, et al. AMPK activator O304 improves metabolic and cardiac function, and exercise capacity in aged mice. Commun Biol 2021;4:1306. pmc.ncbi.nlm.nih.gov/articles/PMC8602430
5. Norlin S, et al. O304 ameliorates hyperglycemia in mice by dually promoting muscle glucose effectiveness and preserving β-cell function. Commun Biol 2023;6:877. publications.scilifelab.se/publication/c48c857b0f664fc8802da70c86d0d433
6. Thieroff-Ekerdt R, et al. 1782-P: Phase 1b Study of AMPK/Mitochondrial Activator ATX-304 in Prediabetic Obese Participants. Diabetes 2026;75(Suppl 1). diabetesjournals.org/diabetes (1782-P)
7. Cambrian Bio press release, 18 Jun 2026 — Phase 1b data, REWIRE plans, mechanism poster #1788-P. cambrianbio.com/news-and-publications (ADA 2026)
8. Cambrian Bio, 14 Mar 2023 — Amplifier Therapeutics launch / Betagenon acquisition. cambrianbio.com/news-and-publications (Amplifier launch) · BioSpace coverage incl. ~$26M price: biospace.com/amplifier-launches-to-bring-first-ampk-activator-for-obesity-to-clinic
9. Phase 1b registry traces: AdisInsight trial profile EUCT2023-505967-36-00. adisinsight.springer.com/trials/700362720 · Dutch registry NL-OMON60115: onderzoekmetmensen.nl/en/node/60115/pdf
10. Cayman Chemical — O-304 identity, formula, solubility, purity. caymanchem.com/product/26186/o-304
11. Myers RW, et al. Systemic pan-AMPK activator MK-8722 improves glucose homeostasis but induces cardiac hypertrophy. Science 2017;357:507–511 (abstract). mendeley.com/catalogue (MK-8722)
12. Cokorinos EC, et al. Activation of Skeletal Muscle AMPK Promotes Glucose Disposal and Glucose Lowering in Non-human Primates and Mice (PF-739/PF-249). Cell Metab 2017;25:1147–1159.
13. AMPK / AICAR reviews. pmc.ncbi.nlm.nih.gov/articles/PMC5945561 · pmc.ncbi.nlm.nih.gov/articles/PMC9664587
14. Grey-market listings (26 Oct 2026): Echemi bulk O-304 and O-304 sodium-salt listings (incl. the erroneous Wnt / hair-regrowth description); vendor capsule listings summarised at disguisedalpha.com/research/atx-304-research-guide
15. Synapse / Patsnap pipeline entry listing ATX-304 “Discontinued” (T2D) — conflicting source, see Where it stands.

Molecule

Methenolone — a DHT derivative

Forms

Oral acetate · Depot enanthate

Ester math

100 mg enanthate ≈ 73 mg base

Status

Withdrawn from most markets

Chemistry — what methenolone actually is

Methenolone is 17β-hydroxy-1-methyl-5α-androst-1-en-3-one — dihydrotestosterone modified in two ways. A Δ1 double bond (between C1 and C2) makes the molecule a poor substrate for aromatase — it cannot be converted to estrogen — and reduces its androgenic potency relative to DHT. A 1-methyl group gives the oral form partial resistance to first-pass hepatic breakdown without 17α-alkylation. And because the A-ring is already 5α-reduced, 5α-reductase inhibitors (finasteride / dutasteride) have nothing to act on — they do not reduce methenolone's activity anywhere, including the scalp.

Two forms, two different drugs in practice

Form	Route logic	Historical dosing pattern
Primobolan / Nibal (oral) — acetate ester	Short-acting; acetate hydrolyses fast; oral bioavailability propped up by the 1-methyl group but still substantially degraded on first pass	Daily oral dosing, tens of mg/day clinically
Primobolan Depot / Nibal Injection — enanthate ester	Oil depot; slow release governed by the enanthate ester, like other enanthate esters. Formula C₂₇H₄₂O₃, MW 414.62	IM every 1–3 weeks clinically

Why the injectable is not hepatotoxic by structure: 17α-alkylation exists to let an oral steroid survive the liver — and that same blockade is what makes 17α-alkylated orals (stanozolol, oxandrolone, oxymetholone, methandienone) hepatotoxic and brutal to HDL. An injectable depot doesn't need the trick: esterases free the parent hormone after absorption, bypassing first-pass entirely. So methenolone enanthate carries no structural hepatotoxicity mechanism, and the clinical literature agrees — no appreciable hepatic-stress marker changes at therapeutic doses. One reported case of liver failure / death in an elderly patient on oral methenolone acetate exists, so “zero liver risk” is not claimable — but for the injectable, liver enzymes are not the axis to watch. Lipids and suppression are (below).

Pharmacokinetics — measured data vs folklore

This is the thinnest section of the entire methenolone literature, and it is stated plainly:

• “Half-life 10.5 days” (enanthate) — folklore-grade. The figure circulates through drug-information monographs (e.g. DrugInfoSys) and vendor / profile pages, all citing each other or nothing. No human IM plasma pharmacokinetic study of methenolone enanthate was located. The number is plausible by analogy to other enanthate esters (testosterone enanthate's terminal half-life is ~4.5–7 days in modern assays), but plausible-by-analogy is not measured. Any dosing-interval or front-loading math built on “10.5 days” is built on sand.
• Circulating “Primo PK” tables are contaminated — documented. A current BenchChem “technical guide” to methenolone enanthate pharmacokinetics presents a Cmax / Tmax table by dose (50 / 100 / 150 mg → 2.14 / 4.26 / 5.16 ng/mL, Tmax 30–72 h) whose own footnote states the data come from a study of nandrolone decanoate in healthy men. Anyone citing precise methenolone peak levels from secondary guides is, knowingly or not, citing nandrolone.
• Oral acetate half-life “~4–6 hours” — a secondary-citation figure, consistent with an acetate ester and the urinary-excretion data below, but again not from a located primary plasma-PK paper.
• What IS measured — metabolism of the oral form. Goudreault & Massé (1990), a doping-control metabolism study: after a single 50 mg oral dose of methenolone acetate, the major urinary metabolite is 3α-hydroxy-1-methylen-5α-androstan-17-one (as glucuronide); parent methenolone remains detectable in urine to ~90 hours; only ~1.63% of the dose is excreted unchanged as parent drug (as summarised in secondary reporting of that paper). Two implications: the oral form is extensively metabolised — hence its weak per-mg showing — and detection windows are long relative to the acetate's reputation.
• Injection-site / depot behaviour: nothing methenolone-specific located. Enanthate-in-oil depot kinetics are assumed, not demonstrated, for this compound.

Section verdict: any source that gives methenolone enanthate a precise half-life, peak, or clearance number without a citation is repeating folklore. The honest PK statement is: long-acting enanthate depot, duration consistent with 1–2 week clinical dosing intervals; exact human PK unmeasured in the accessible literature.

Receptor pharmacology

Androgen-receptor binding · Relative binding affinity vs methyltrienolone = 100 · Saartok, Dahlberg & Gustafsson, Endocrinology 1984 (rat tissue assays)

Compound	Skeletal muscle AR	Prostate AR
Methenolone	58%	54%
Testosterone	51%	72%
Nandrolone	77%	82%
DHT	23%	96%

Methenolone binds the muscle androgen receptor at least as avidly as testosterone in this assay. Whatever “weak” means in the Primo folklore, it does not mean weak receptor binding.

• SHBG binding: weak. In the comparative SHBG affinity ordering (mesterolone ≫ DHT > testosterone > … > methenolone > methandienone > stanozolol), methenolone sits near the bottom. Low SHBG affinity means methenolone neither displaces much testosterone from SHBG nor is itself much buffered by SHBG. Community claims that Primo “frees up testosterone by binding SHBG” have the pharmacology backwards — that is Proviron / mesterolone, the strong SHBG binder.
• Aromatization: none. The Δ1 structure is not an aromatase substrate. Methenolone produces no estrogen of its own — and, critically, is not an aromatase inhibitor either (see the estrogen claim, adjudicated below).
• Progestogenic activity: none reported. Methenolone is not a 19-nor and has no progesterone-receptor story. The prolactin axis is not a methenolone axis.
• Anabolic : androgenic ratio 88 : 44–57 (secondary citation — Vida's compilation of Hershberger-assay data: rodent tissue weights, decades old, useful for rough ranking only). On this scale methenolone is roughly testosterone-strength anabolically (T = 100) with about half the androgenic tissue signal — the historical root of the “clean anabolic” reputation.

Clinical history — Schering, Nibal, and what it was actually for

Methenolone esters were introduced by Schering in the early 1960s: the acetate as Primobolan (and as Nibal in some markets), the enanthate depot as Primobolan Depot / Nibal Injection. The approved / clinical uses were the era's standard AAS list:

• Anemia due to bone marrow failure (including aplastic anemia) — the headline indication; methenolone was used to stimulate erythropoiesis when few alternatives existed.
• Wasting / catabolic states — recovery from surgery, chronic infection, prolonged corticosteroid therapy, malnutrition.
• Osteoporosis (postmenopausal / senile) — from the era when AAS were a mainstream osteoporosis tool.
• Sarcopenia and weight-gain support in underweight premature infants and children — a use that reads as alarming now and was unremarkable for 1960s AAS labeling.

Clinical doses (monograph figures — DrugInfoSys)

Indication	Dose
Osteoporosis	Enanthate 100 mg IM every 2 weeks
Progressive breast cancer (period indication)	Enanthate 200 mg IM every 2–3 weeks
Aplastic anemia	Oral acetate 100–150 mg/week

Note the scale: clinical enanthate exposure was ~50–100 mg/week-equivalent — an order of magnitude below physique folklore. The clinical record is period-typical — small, old, mostly uncontrolled studies showing weight gain, nitrogen retention, and hemoglobin responses in the indicated populations. There are no modern RCTs of methenolone for any indication, no dose-ranging trials in healthy men, and no long-term safety cohort. Anyone claiming “clinically proven safe / effective” for physique use is laundering 1960s anemia case-series evidence into a context it never covered.

Withdrawal: Schering withdrew Primobolan Depot from most markets in the 1990s amid doping scrutiny; the oral followed from most markets by the early 2000s. Bayer (which absorbed Schering) remains nearly the sole human-grade producer, with the brand still sold in a handful of countries (Spain, Turkey, Japan, Paraguay, Ecuador among them). Methenolone technically retains FDA-approved status in the US but has not been commercially sold there for decades.

The “weak — needs 600–1000 mg/wk” claim, adjudicated

The claim: Primobolan is a weak anabolic that only “works” at 600–1000 mg/wk (vendor / profile dose ranges typically quoted: 300–800 mg/wk enanthate for men — reported here descriptively only). What the data actually support:

• The receptor is not the bottleneck. Muscle AR affinity ≈ testosterone's (above). There is no binding-pharmacology basis for needing 6–10× testosterone's milligram dose.
• The ester hides ~27% of the label. 600 mg/wk of enanthate ≈ 438 mg/wk of base hormone. Folklore doses quoted in ester weight inflate the apparent requirement.
• The comparison baseline is rigged. Primo is usually compared, per-mg, against aromatizing compounds (testosterone, Dianabol) whose early “gains” include estrogen-mediated water, glycogen and sodium retention. A non-aromatizing compound producing dry, tissue-only change will always look “weaker” on the scale at week 4 while possibly differing far less in actual contractile tissue. The folklore grades Primo on a metric structurally biased against it.
• The counterfeit confound is enormous. Primo is historically the most-substituted injectable on the black market (next section). A generation of users “needing” 800 mg/wk includes an unknown fraction who were running underdosed product, drostanolone, nandrolone, testosterone cypionate, or nothing — titrating their dose against a vial that wasn't what the label said. Dose folklore built on counterfeit product is not dose-response data.
• The oral form genuinely is weak per-mg — extensively metabolised on first pass, only ~1.6% excreted as parent — and oral-era impressions (“you need a lot of Primo”) bleed into injectable folklore where they don't belong.
• The clinical record shows activity at a fraction of the folklore range: measurable anabolic / erythropoietic effects at 50–100 mg/wk-equivalent enanthate in sick, elderly and pediatric populations. That does not mean 100 mg/wk is a physique dose in a trained man on TRT — it does mean the molecule is pharmacologically active at a small fraction of the folklore range, and the burden of proof sits on the “needs 600+” claim, which has no human dose-response study behind it at any dose.

Section verdict: “Weak” is mostly artifact — of ester weight, of water-weight comparison bias, of oral-form bleed-over, and above all of a counterfeit-saturated supply history. Per-mg, expect a testosterone-class anabolic signal without estrogenic scale noise: less dramatic, not less real.

Per-mg expectations vs the rest of the cabinet

• Vs testosterone — similar AR signal, no aromatization, worse LDL signal per the cohort data, costlier, counterfeit-prone.
• Vs Masteron (drostanolone) — Primo is the more anabolic / less androgenic-feeling of the two DHT derivatives (A:A 88 : ~50 vs drostanolone's more androgenic profile); Mast the stronger “hardening” / libido agent per universal community report [community anecdote].
• Vs nandrolone — Primo is dry where nandrolone is wet; no prolactin axis; no joint-comfort reputation.
• Vs EQ (boldenone) — similar “lean, slow” niche, but EQ carries the stronger hematocrit drive and a real (if contested) aromatization-to-E1 story; Primo's lipid signal looks worse, its hematocrit signal (unmeasured) presumably lighter.

The “mild / safe” reputation, adjudicated axis by axis

6.1 · Lipids — the key question, and the answer is not reassuring

Human user-cohort data · Achar, Rostamian & Narayan, Am J Cardiol 2010 (PMC4111565)

Cohort	Exposure	What was found
Lenders et al.	Methenolone 385–690 mg/wk (n=20)	LDL 206 ± 21 mg/dL vs 156 ± 9 in ex-users vs 130 ± 13 in controls — a large, dose-contextual LDL elevation that partly reversed after stopping. The single most decision-relevant human data point on Primo.
Baldo-Enzi et al.	Methenolone enanthate 100–300 mg/wk (n=14, age 27±5)	LDL-column values 129 ± 37 (users) vs 119 ± 17 (controls). The study's separate testosterone cypionate arm (200–300 mg/wk) showed HDL 27 ± 11 vs control 48 ± 6 — the class effect at moderate doses.
Lajarin et al.	Methenolone enanthate 100 mg/wk	The users' value positioned in the HDL column is 14 ± 0.4 mg/dL; the review text independently notes “HDL levels as low as 14 mg/dL have been noted” in AAS users.

Review-wide class findings: AAS decrease HDL 20–70% and raise LDL >20%; abnormalities can appear within ~9 weeks; they appear reversible, normalizing ~5 months after cessation in the cited follow-ups. The mechanism story (secondary citation — Llewellyn) is coherent and matches the cohort data's direction: methenolone “should have a stronger negative effect on the hepatic management of cholesterol than testosterone or nandrolone due to its non-aromatizable nature, but much weaker than c-17 alpha alkylated steroids”; the oral form rates slightly worse than the injectable.

Axis verdict: on HDL / LDL, Primo ranks worse than testosterone and nandrolone, better than the 17α-alkylated orals (winstrol / anavar / anadrol territory). “Mild” does not survive contact with the lipid data. Among the DHT derivatives specifically, Primo and Masteron share the non-aromatizable mechanism and neither has clean human lipid cohorts of its own — but Primo is the one with actual user-cohort numbers, and the numbers are bad at physique doses and non-trivial even at 100 mg/wk.

6.2 · Hematocrit

No methenolone-specific human hematocrit cohort was located (see Open Questions). What can be said: methenolone was prescribed to treat anemia — erythropoiesis stimulation is its documented clinical pharmacology, not a side effect discovered later — and erythrocytosis is a class effect of androgens via EPO / iron-incorporation pathways. Expect some hematocrit pressure, plausibly less per-mg than testosterone or EQ, but “Primo doesn't raise hematocrit” is not a claim the evidence supports — the drug's original job was raising red-cell mass in bone-marrow failure.

6.3 · Prostate

No methenolone-specific human prostate-outcome data located. Pharmacology: prostate AR binding 54% of methyltrienolone vs testosterone's 72% — same order of magnitude, somewhat lower. It is a DHT derivative that is not further 5α-reduced (already 5α) and not attenuated by finasteride. Therapeutic-era use in elderly men (osteoporosis, sarcopenia) without a prostate-safety signal in the period literature is weak reassurance at period doses. PSA monitoring stays the operative control; nothing about Primo argues for relaxing it.

6.4 · Hair — a DHT derivative, full scalp relevance

Methenolone is intrinsically androgenic at the follicle by structure: a DHT derivative whose activity does not depend on 5α-reduction. Finasteride / dutasteride logic does not apply — there is no conversion step to block (the same structural situation as Masteron). Its androgenic rating (~44–57 vs testosterone's 100, Hershberger) and lower prostate-AR binding suggest a somewhat lower per-mg follicular androgenic load than DHT itself or drostanolone — but “lower than Masteron” is a relative statement, not a safe one. In a genetically susceptible scalp, methenolone is a plausible accelerator of miniaturization at physique doses [mechanistic inference from structure + binding data; no human hair-outcome study exists].

6.5 · Mood

No clinical mood data for methenolone exist. Community reputation is “feel-good, calm, no aggression” [community anecdote] — plausibly a contrast effect against trenbolone / 19-nors and against crashed-estrogen states users create with other stacks. Nothing in the receptor pharmacology predicts a distinctive mood signature either way.

6.6 · Fertility / HPTA suppression — “mild” fails here too

At oral methenolone acetate 30–45 mg/day — ordinary clinical doses — more than half of patients showed 15–65% gonadotropin suppression (secondary citation — Llewellyn-cited clinical study), and Llewellyn himself notes the low-suppression distinction may not hold at physique doses. Class logic concurs: a non-aromatizing androgen still feeds back at the hypothalamus / pituitary via the androgen receptor. For someone on TRT this axis is largely moot — endogenous production is already suppressed and managed by design — but it dismantles the general folklore that Primo is “non-suppressive” or PCT-optional in other populations.

6.7 · Liver

Covered under Chemistry: no 17α-alkylation, no hepatic-stress signal at therapeutic injectable doses, one elderly oral-case report preventing an absolute claim. For the injectable enanthate, liver is the least concerning axis.

Practitioner experience — “Big Paul” Barnett

• Overall stance [verbatim transcript]: Primobolan is his #1-ranked compound — “the most well tolerated agent in terms of practical application and clinical studies that are out there.” Core use-case: anti-catabolic protection in a deficit (“you're going to be anti-catabolic on primobolan, which means in a calorie deficit you ain't gonna lose your muscle”) or a lean-gaining phase.
• Doses he reports running / recommending [verbatim transcript]: his cutting-phase construction runs more Primo than test late in a blast — “three to four hundred test and 500 Primobolan at the end of a cycle in the blast phase” — with less total anabolic load needed in a surplus (“in a bulking phase you don't need so many anabolics”). In a separate video summary of his preferred cycle (2022) [secondary summary]: Primo characterized as effective but expensive, Masteron named as the more affordable alternative, and Primo described as tolerating high doses “without significant side effects due to its smooth nature.”
• The “shine” threshold [description-level]: a Q&A episode devotes a timestamped segment (13:48) to “At What Doses Does Primo Start To ‘Shine’?” — the framing itself presupposes a dose below which Primo underwhelms, consistent with the community folklore examined above. His specific number was not accessible in transcript form and is not invented here.
• Primo as an estrogen-management tool [verbatim transcript]: “big mistake that people make on primobolan: they add an AI to it. When you deploy a protocol that has primobolan in it, it's going to decrease the estrogenic activity at the receptor site, therefore no AI is needed when you're on Primo or any other DHT derived steroids.” He ties this to SHBG stewardship and to ratio management: if HCG is increased on a test + Primo protocol, “you're gonna need to up the ratio of Primo to test” to hold estrogenic activity in check.
• His personal ratio [near-verbatim, YouTube Short via secondary mirror]: “A 3:2 ratio of Test to Primo or Test to EQ would typically have my Estradiol in range in the past but not anymore” — at 3:2 test : Primo, his measured estradiol historically sat in range without an AI, while noting this has stopped working as reliably for him.
• Anastrozole tapering when introducing Primo [description-level]: a recent Q&A's topic list includes “tapering anastrozole when introducing Primobolan” — establishing that the taper-off-the-AI-as-Primo-comes-in practice is live advice in his community; his specific taper schedule was not accessible.
• Masteron vs Primo [description-level + secondary summary]: recurring Q&A topics include “Masteron vs Primo for estrogen management” and “Masteron vs Primobolan” (Jul 3, 2026); his consistent framing across accessible material: the two are functional cousins for estrogen management and dryness, with Primo the more anabolic builder and Masteron the cheaper, more purely cosmetic hardener.
• Authenticity / bloodwork: his show devoted an episode to the 2025 UGL lab-testing study (“UGL Steroids Lab-Tested… The Results Are HORRIFYING (2025 Study)”) [description / title-level], and his transcript material repeatedly routes decisions through bloodwork. No Primo-specific lipid / HDL numbers from Paul were accessible; none are attributed to him.

Where Paul and the data agree

• Anti-catabolic-in-a-deficit as Primo's home use — matches the clinical wasting / cachexia history and the compound's fenced slots below.
• Bloodwork-gated titration — his stated method is this site's stated method.
• Counterfeit seriousness — his coverage of the 2025 testing study aligns with the analytical record above.
• Primo > Masteron as a builder, Masteron as hardener — consistent with the A:A and binding profile.

Adjudication — “Primo lowers estrogen,” the claim that needs care

Paul's own transcript wording is more careful than the community slogan: he claims Primo decreases estrogenic activity at the receptor site, and separately reports (Short) that a 3:2 ratio kept his measured estradiol in range. These are different claims. Adjudicated separately:

• “Primo is an aromatase inhibitor” — NOT SUPPORTED. Methenolone is not an aromatase substrate and there is no evidence it meaningfully inhibits aromatase at achievable concentrations. It does not do what anastrozole does. Any protocol that treats Primo as a replacement AI with a predictable dose-response is standing on nothing.
• “Primo reduces estrogenic signaling / symptoms at a given E2” — PLAUSIBLE, mechanism-consistent. Raising the androgen : estrogen ratio with a non-aromatizing androgen changes tissue-level signaling balance (the same reason Masteron is reported to “feel” anti-estrogenic). This is a ratio effect, not an E2-lowering effect — and it does not require serum E2 to move at all.
• “Adding Primo lowered my measured E2” — REPORTED REPEATEDLY in the community, mechanism unresolved. Candidate explanations, in rough order of evidentiary weight: (a) SHBG reduction — DHT-class compounds lower SHBG, shifting free / total relationships and hepatic clearance dynamics, which can move total E2 readings without aromatase being touched; (b) substrate / flux shifts when total androgen load and ratios change; (c) assay artifact — E2 immunoassays are notoriously cross-reactive in the presence of exogenous steroids (the textbook case is trenbolone producing falsely high immunoassay E2); results differ between immunoassay and LC-MS/MS, and much community “data” is immunoassay data. Paul's own panel discipline (LC-MS-grade testing via his TRT provider links) strengthens his n=1 somewhat, but an n=1 with an unresolved mechanism remains an anecdote.
• Practical ruling: Primo must never be assigned the estrogen-management job. The AI ladder documented on this site (raloxifene first for gyno; anastrozole for lab-confirmed high E2 + symptoms; never prophylactic) is built on measured E2 from sensitive LC-MS/MS at defined gates. Layering an unmeasured “Primo will handle the estrogen” assumption on top would contradict the ladder, confound the gate draws, and risk the specific failure mode Paul himself warns about in reverse: stacking Primo and an AI — or crediting Primo with AI work it isn't doing — is how users end up with crashed E2, which also means worse lipids (estrogen is part of HDL support) and worse joints. If Primo is ever trialed, E2 is managed by the ladder and measured at the trial's own before / after panel — full stop.

The “shine” framing vs the dose evidence: Paul's “shine” framing [description-level] and his 500 mg cutting example [verbatim] sit inside the community dose culture adjudicated above as substantially artifactual. The charitable synthesis: an experienced coach titrating verified-feeling product in large, advanced bodybuilders will land at higher absolute doses than a 146 lb man needs for the same per-kg exposure — Paul's clientele context is competitive bodybuilders. His numbers are reported, not endorsed, and no dose is recommended here.

Where it fits — adjudication for the plan this site documents

The consolidated cycle plan fences Primo as a future-cut (lean retention) or long, gentle cruise-plus candidate — with Phase 0 discipline and the tren-acetate probe window forbidding any new compound inside another compound's trial, and a trial principle of minimum effective dose, one variable at a time, scored against preset bail lines with a before / after panel for any trial. Both fenced slots are the right ones:

• Future cut / lean retention: Primo's best-evidenced niche — clinical anti-wasting history, non-aromatizing dry anabolism, Paul's independent practitioner convergence. It competes with anavar for final-cut-weeks in the plan's fencing; the trade is: anavar is oral, shorter, joint-kinder per the plan's prior analysis but a 17α-alkylated HDL hit; Primo is injectable, longer-acting, liver-quiet, but carries the worse LDL cohort data, the counterfeit problem, and a months-long commitment per trial. Neither is “the safe one” — they fail on different axes, and the panels adjudicate which failure mode is affordable.
• Gentle cruise-plus: structurally coherent (long ester, no aromatization to manage, tolerability reputation) — with one hard caveat: a cruise-plus is by definition long, and the lipid signal is the thing that accumulates quietly over long exposures. A cruise-plus Primo slot is only defensible with lipid panels at the same gate discipline as blasts — arguably it needs more lipid surveillance, not less, precisely because everything else about it feels benign.
• Explicitly not Primo's slots: inside the tren probe window (unknown-variable violation), as an estrogen-management tool (above), as a bulk driver (dry per-mg economics lose to test / EQ / nandrolone in a surplus), and as a “safe” filler at arbitrary dose.

Watched axes, mapped

• HDL / LDL: Primo presses exactly here, and presses LDL too. Ezetimibe (the plan's Gate-3-triggered guard) addresses LDL absorption — relevant — but does not restore HDL. Translation: a Primo trial's before / after panel must include the full lipid panel + ApoB, and an HDL crash or LDL surge is a trial-ending scorecard event, not a “watch it” event, because there is no corrective owned for the HDL half.
• Hematocrit: expect some pressure (the drug treated anemia); the >52% bail line and hydration / dose / clinician sequence apply. Primo does not get EQ-level suspicion, but it does not get a pass either.
• Shoulders: Primo is neutral-to-dry — no nandrolone-style joint-comfort reputation, no fluid cushion (non-aromatizing), but also no winstrol-style active joint drying or collagen insult. The ranking inside the cabinet stands: nandrolone family = the only joint-positive claim; anavar = joint-kinder oral; Primo = neutral; winstrol = worst. One indirect risk: a Primo stack run with over-suppressed E2 would worsen joint feel — another reason E2 stays on the ladder, not on Primo.
• Sleep / mood / HR / BP scorecard axes: no Primo-specific signal in the literature; scored as usual. Primo's long ester cuts both ways — a bad reaction exits slowly (weeks, like Deca / EQ) — which is why a Primo trial belongs in a clean window with no competing changes.

Panel requirement: any future Primo trial requires its own before / after panel, specified to this compound: lipids + ApoB (primary axis), CBC / hematocrit, total / free T + sensitive E2 (to keep the estrogen question honest — measured, not assumed), PSA, SHBG (Primo's SHBG effect is part of the E2-folklore mechanism and is cheap to check), plus baseline hair photos before day one — and a Janoshik identity + concentration test on the actual vial first. Without that test, the “trial” could silently become a different compound's trial — in the nandrolone case, a second 19-nor — inside a plan that sequences 19-nors deliberately. This precondition is stronger for Primo than for anything else in the cabinet.

Claim check

Claims circulating in profile pages and community coverage of Primobolan, checked against the sources above:

• Not supported

  “Methenolone enanthate's half-life is 10.5 days.”

  No human IM plasma pharmacokinetic study of methenolone enanthate was located. The figure circulates through monographs and vendor pages citing each other or nothing — plausible by analogy to other enanthate esters, but folklore-grade. Precise peak-level tables in circulation are, by their own footnote, nandrolone decanoate data.

• Needs the qualifier

  “Primo is weak — you need 600–1000 mg/wk.”

  Mostly artifact: ester weight hides ~27% of the label, aromatizing comparators win the scale with water, oral-form weakness bleeds into injectable folklore, and a counterfeit-saturated supply history means much of the “dose-response” was titrated against vials that weren't methenolone. Its muscle AR affinity ≈ testosterone's, and clinical effects appear at 50–100 mg/wk-equivalent. There is no human dose-response study behind the folklore range at any dose.

• Not supported

  “Primo is the mild, safe one — it doesn't touch lipids.”

  Human user cohorts say otherwise: LDL 206 ± 21 mg/dL at 385–690 mg/wk vs 130 ± 13 in controls (partly reversible after stopping), and an HDL value of 14 ± 0.4 mg/dL recorded in the 100 mg/wk row of the same review (table-alignment caveat noted). On lipids Primo ranks worse than testosterone and nandrolone, better than the 17α-alkylated orals. “Mild” survives only for androgenic feel and the injectable's liver profile.

• Not supported

  “Primo is non-suppressive — PCT-optional.”

  At oral acetate 30–45 mg/day — ordinary clinical doses — more than half of patients showed 15–65% gonadotropin suppression, and the profile literature itself notes the distinction may not hold at physique doses.

• Not supported

  “Primo lowers estrogen — run it instead of an AI.”

  Methenolone is not an aromatase inhibitor; there is no evidence it meaningfully inhibits aromatase. What survives scrutiny is a ratio / signaling effect at a given E2, plus community reports of lower measured E2 with unresolved mechanism (SHBG reduction, flux shifts, immunoassay cross-reactivity). Estrogen management stays with measured E2 and the AI ladder.

• Not supported

  “Primo binds SHBG and frees up your testosterone.”

  Backwards. Methenolone's SHBG affinity is weak — near the bottom of the comparative ordering. The strong SHBG binder in that story is mesterolone (Proviron).

• Unverified lore

  “Primobolan was Arnold Schwarzenegger's favorite steroid.”

  Repeated by essentially every profile page; no primary source located in which Schwarzenegger confirms methenolone use, doses, or preference. His publicly acknowledged use centers on Dianabol. It is evidence about Primo's brand, not about Primo.

• Needs the qualifier

  “The vial says Primobolan, so it's Primobolan.”

  Historically the most-substituted injectable in the class: seized / analysed “Primo” has contained drostanolone + nandrolone (no methenolone at all) and testosterone cypionate, and 15 of 42 black-market products in one GC-MS series lacked the expected ingredient. Genuine, accurately dosed product also demonstrably exists in third-party lab results — which is exactly why batch-level testing, not the label, is the evidence.

Open questions

Marked plainly, because they are open:

• Any primary human pharmacokinetic study of methenolone enanthate (plasma levels, true half-life) — none located. The “10.5-day” figure is secondary-source circulation only.
• Big Paul's exact “shine” threshold (Q&A segment at 13:48) and his anastrozole-taper specifics — episode audio / transcripts were not accessible; only descriptions, timestamps, and his other verbatim material could be used. Nothing was invented to fill this.
• The Baldo-Enzi / Lajarin lipid table cells as extracted from PMC4111565 have imperfect column alignment; the Lenders LDL finding and the review's class-level ranges are solid, but individual HDL cell attributions should be re-checked against the published PDF before being quoted to third parties.
• Schwarzenegger's alleged Primobolan use / doses — no primary source located; treated as lore throughout.
• Methenolone-specific human data on hematocrit dose-response, prostate outcomes, hair outcomes, and mood — none located; those sections rely on class pharmacology and binding data, flagged as such.
• Whether the Alpha Rise vial in this cabinet contains methenolone, and at what concentration — unknowable without third-party testing.

Sources

1. Achar S, Rostamian A, Narayan SM. Cardiac and metabolic effects of anabolic-androgenic steroid abuse on lipids, blood pressure, left ventricular dimensions, and hemostasis. Am J Cardiol 2010. pmc.ncbi.nlm.nih.gov/articles/PMC4111565
2. Saartok T, Dahlberg E, Gustafsson JÅ. Relative binding affinity of anabolic-androgenic steroids. Endocrinology 1984 (PMID 6539197). pubmed.ncbi.nlm.nih.gov/6539197
3. Coopman V, Cordonnier J. Counterfeit drugs and pharmaceutical preparations seized from the black market. Annales de Toxicologie Analytique 2012. ata-journal.org/articles/ata/pdf/2012/02/ata120011.pdf
4. Piatkowski T, et al. Lead astray: the hidden contaminants in the Australian anabolic-androgenic steroid market. Drug and Alcohol Review 2025. researchgate.net/publication/393888095
5. Musshoff F, Daldrup T, Pellegrini M, et al. A simple toxicological analysis of anabolic steroid preparations from the black market. researchgate.net/publication/273613278
6. DrugInfoSys — methenolone monograph (dosing; half-life as circulated). druginfosys.com/drug.aspx?drugCode=2129
7. Metenolone — Wikipedia (clinical history, formulations, market status). en.wikipedia.org/wiki/Metenolone
8. Primobolan Depot profile — BP Medical (secondary profile literature). bpmedical.info/product/primobolan-depot-methenolone-enanthate
9. Primobolan (methenolone acetate) profile — Wikibolics (secondary profile literature, incl. Llewellyn-derived material). wikibolics.com/anabolics/drug-profiles (Primobolan)
10. AnabolicLab — HYB methenolone enanthate lab results (Janoshik #57378). anaboliclab.com/lab_reports/hyb-methenolone-enanthate · Optitropin methenolone enanthate (Janoshik #48458). anaboliclab.com/lab_reports/optitropin-methenolone-enanthate
11. Hilma BioCare methenolone enanthate HPLC results (SIMEC via AnabolicLab, 2018). thinksteroids.com/community (Hilma BioCare SIMEC 2018)
12. Barnett P (Big Paul), Anabolic Bodybuilding — “My Top 3 Anabolic Steroids For Ultimate Gains,” YouTube transcript (verbatim practitioner material). youtubetotranscript.com/transcript?v=JWKwDlilvv8
13. Barnett P, Anabolic Bodybuilding — “Best Primobolan Cycle? Advice Before Starting Gear, Thoughts On T3 & Clen, Metformin Lowers IGF-1?” (episode description / timestamps only). podcast.app (Best Primobolan Cycle?) · also on Audible: audible.ca (Best Primobolan Cycle?)
14. Barnett P, Anabolic Bodybuilding — recent Q&A (anastrozole taper with Primobolan; Masteron vs Primo for estrogen management; description-level). podtail.com/podcast/anabolic-bodybuilding
15. Barnett P, Anabolic Bodybuilding — “Masteron Lowers Side-Effects From Deca? Using GLPs For Bulking?…” Jul 3, 2026 (description-level). audible.in/podcast/Anabolic-Bodybuilding
16. Barnett P, Anabolic Bodybuilding — Podbean episode feed (DHT-derived compounds / hard-dry look episode; description-level). anabolicbodybuilding.podbean.com
17. Barnett P, Anabolic Bodybuilding — “My Favorite Cycle for Real Muscle Gains” (2022; secondary summary). videohighlight.com/v/WEabgJfBXVQ

Class

SARM — oral, non-steroidal

Also called

MK-2866 · Enobosarm

Developer

GTx — muscle wasting / cachexia

Status

Never FDA-approved · WADA-banned

What it is — and how it works

Ostarine is a selective androgen receptor modulator (SARM): an oral, non-steroidal agonist of the androgen receptor, engineered for tissue selectivity — anabolic in muscle and bone, quieter in prostate and skin. That engineering intent is real. What the marketing leaves off is the qualifier the evidence attaches to it: selectivity is relative and dose-dependent, not absolute. Push the dose and the “selective” part erodes — which matters, because community doses are a multiple of the studied ones (below).

GTx developed it as enobosarm, aimed at muscle wasting / cachexia — involuntary muscle loss in serious illness, where an oral anabolic without a steroid's full side-effect load would be genuinely valuable. It never cleared the bar for approval. What it became instead is two things at once: a grey-market “research chemical” sold to physique users, and — under Veru — a candidate for muscle preservation alongside GLP-1 weight-loss drugs, the angle that now carries its development.

The evidence — lean mass up, function missed, then a second act

The POWER trials — why enobosarm was never approved

In GTx's Phase 3 POWER trials in cancer cachexia, enobosarm did the body-composition half of its job: lean mass increased. But the trials' functional endpoints missed — the patients did not get measurably stronger or more capable in the ways the endpoints required — and approval failed. It is the same pattern this library keeps meeting across the muscle-preservation field: mass and function are different claims, and only one of them was proven. Anyone citing the POWER programme as “clinically proven” is citing a programme that ended in a failed approval.

QUALITY — enobosarm on top of semaglutide

The live path is Veru's Phase 2b QUALITY trial: enobosarm paired with semaglutide in older adults losing weight on a GLP-1 drug. Enobosarm reduced lean-mass loss vs semaglutide alone — the sponsor's headline figures:

QUALITY · Phase 2b · n=168, age ≥60, on semaglutide, 16 weeks · Sponsor topline figures

Measure	Result
Lean-mass loss vs semaglutide alone	71% relative reduction (p=0.002)
3 mg dose, lean-mass loss	>99% mean relative reduction (p<0.001)
Fat loss	27% more than semaglutide alone
Total weight loss	Similar to semaglutide alone
Function	Fewer patients losing clinically significant stair-climb function

Source: Veru, QUALITY topline release, 27 Jan 2025

In the maintenance extension, after semaglutide stopped, enobosarm significantly reduced weight and fat regain. FDA feedback in September 2025 accepted 3 mg as the go-forward dose, and the follow-on Phase 2b PLATEAU trial reached full enrolment in 2026. Two hedges travel with every one of those numbers, and they are not fine print: all QUALITY figures are sponsor topline releases — there is no peer-reviewed publication yet — and SARM-class long-term safety (liver, lipids, prostate) remains the open question over the whole programme.

Harms and catches

• HPTA suppression is real and dose-dependent. Despite “non-suppressive” marketing, ostarine shuts down endogenous testosterone in natural users. This is the best-documented rebuttal to the SARM sales pitch: the compound reaches the hypothalamic–pituitary axis like any other androgen-receptor agonist, and suppression deepens with dose — in a dose range where community use starts well above the trial range.
• HDL suppression — the class's signature lipid effect. Documented in trials, and described in the research file as the SARM class's most reliable lipid effect. For a reader tracking a panel, this is the marker ostarine moves most dependably, and it moves the wrong way.
• Liver — mostly mild, with case reports behind it. Mild transaminase elevations are common. Case reports of drug-induced liver injury (DILI) exist — often tied to mislabeled or contaminated products, which is its own problem (next point) rather than an exoneration: a user cannot separate “ostarine's liver risk” from “whatever was actually in the bottle.”
• The bottle problem. SARM products are notoriously inaccurate — independent analyses repeatedly find other SARMs and prohormones in bottles sold as one compound. Every grey-market ostarine “experience report,” positive or negative, carries an unquantified chance of being about a different drug entirely.
• No aromatization — with a secondary catch. Ostarine does not convert to estrogen. But in natural users, estradiol falls secondarily anyway, because testosterone suppression removes the substrate estrogen is made from. “Doesn't aromatize” is not “leaves your hormones alone”; it is a different route to a low-estrogen state, with the joint, mood, libido and lipid consequences that follow one.
• Unknown long-term safety at recreational doses. No human data exist beyond the trial windows, and none at community doses at all. That is not a statement that long-term harm is proven — it is a statement that the experiment has never been run under observation.

Where it fits — adjudication for the plan this site documents

No slot — and the reasoning is worth stating, because ostarine is the compound readers ask about most.

• Nothing it offers is missing. Testosterone already delivers everything ostarine offers — androgen-receptor-driven anabolism — with known pharmacology and this site's own measured responses behind it. Adding a less-characterised AR agonist on top of a characterised one buys no new capability.
• The specific collision is HDL. HDL suppression is ostarine's most reliable effect, and HDL is a watched line in the plan this site documents — an HDL below 40 is a bail line in the current trenbolone probe, and the next lipid gate is a decision point built around exactly these markers. Adding the class whose signature harm is HDL suppression would contaminate both readings: a bad HDL number would no longer say which compound caused it.
• Its real value here is literacy. SARMs are among the most-asked audience topics, and the enobosarm + GLP-1 lean-mass-preservation story maps directly onto the journey this site documents — muscle kept while a GLP-1 drug takes the weight off. Ostarine is education content, not a shelf item: filed, profiled, and deliberately not on hand and not planned.

Claim check

Claims circulating in SARM marketing and community coverage of ostarine, checked against the research file:

• Not supported

  “SARMs are non-suppressive — no shutdown, no PCT needed.”

  HPTA suppression on ostarine is real and dose-dependent; it shuts down endogenous testosterone in natural users. The claim survives only at the level of marketing copy.

• Needs the qualifier

  “It's tissue-selective, so prostate and skin are spared.”

  Selectivity is relative and dose-dependent, not absolute — and community doses (10–25 mg/day) run far above the 1–3 mg range where the selectivity was characterised.

• Not supported

  “Ostarine was approved for muscle wasting.”

  Never FDA-approved for anything. In the Phase 3 POWER trials lean mass increased but the functional endpoints missed, and approval failed.

• Not supported

  “It doesn't touch your lipids.”

  HDL suppression is documented in trials and is the SARM class's most reliable lipid effect.

• Needs the qualifier

  “No aromatization, so no hormone side effects.”

  Ostarine itself does not convert to estrogen — but in natural users estradiol falls secondarily through testosterone suppression. No conversion is not no consequence.

• Needs the qualifier

  “The bottle says ostarine, so it's ostarine.”

  Independent analyses repeatedly find other SARMs and prohormones in SARM products. The DILI case reports are often tied to exactly this — mislabeled or contaminated bottles.

• Not supported

  “10–25 mg a day is the studied dose.”

  Human trial data sit at 1–3 mg. The 10–25 mg/day, 8–12-week community pattern is extrapolated far beyond the studied range, with no human safety data behind it.

Open questions

Marked plainly, because they are open:

• Long-term safety at recreational doses — unknown. No human data exist beyond the trial windows, and none at 10–25 mg/day at all.
• QUALITY in peer review — as of this research, every QUALITY figure is a sponsor topline release; no peer-reviewed publication has appeared, and the PLATEAU follow-on (fully enrolled in 2026) has reported no results in the sources behind this page.
• Whether lean-mass preservation becomes an approval — the GLP-1 combination path is live development, not an approved indication; enobosarm remains approved nowhere, for anything.
• What is in any given grey-market bottle — unknowable without independent testing; the published analyses of SARM products justify treating the label as a claim, not a fact.

Sources

1. Veru — QUALITY Phase 2b topline release, 27 Jan 2025 (enobosarm + semaglutide: 71% relative reduction in lean-mass loss; 3 mg dose; stair-climb function; maintenance extension). ir.verupharma.com/news-events/press-releases/detail/225
2. GTx enobosarm development programme — Phase 3 POWER trials in cancer cachexia (lean mass increased; functional endpoints missed; approval failed), as summarised in the JTWell compound profile, filed 26 Oct 2026.
3. SARM product-accuracy analyses (other SARMs / prohormones found in labelled products) and DILI case reports, as summarised in the JTWell compound profile, filed 26 Oct 2026.

Molecule

19-nortestosterone

Ester

Decanoate · terminal t½ 7–12 days

US label

Anemia of renal insufficiency only

Status

Schedule III · US brand discontinued 2002

What it is — and how it works

Nandrolone is 19-nortestosterone (estr-4-en-17β-ol-3-one): testosterone without the methyl group at carbon 19, first described by Birch in 1950. It is an androgen-receptor agonist — in skeletal muscle, AR activation drives protein synthesis and nitrogen retention — and it also stimulates erythropoietin production and bone-marrow erythropoiesis, which is the basis of its anemia indication, and acts on androgen receptors in bone. It has very low affinity for SHBG (~5% of testosterone's), so a larger fraction is unbound than for testosterone at the same total level.

Anabolic : androgenic — three numbers circulate, from three assay systems

Figure	System	What it means
Myotrophic-androgenic index ≈ 11 (testosterone = 1)	Classic rat Hershberger-type assay — levator ani (muscle) vs seminal vesicle / prostate weight	Several-fold more anabolic than androgenic relative to testosterone, in rats.
Anabolic potency 3.29–4.92 · androgenic 0.31–0.41	Ratios vs testosterone propionate, as quoted in the 2020 systematic review	Same qualitative direction, different scaling.
125 : 37	The “anabolic:androgenic rating” in bodybuilding references (e.g. Llewellyn's Anabolics)	Community / reference-book figure, same rat-assay lineage. None of the three is a measured ratio in humans — do not treat any single one as a human constant.

Sources: Tatem et al., Transl Androl Urol 2020 · Patanè et al., Medicina 2020

The three metabolic differences that define it

• Aromatization — real, but at ~20% of testosterone's rate. Nandrolone does aromatize to estradiol, so gynecomastia and estrogenic effects are possible but less common than with equivalent testosterone. Two caveats the community usually omits: Pan & Kovac's review notes the aromatization data are largely from animal models, with no specific in-vivo human aromatization studies described — and nandrolone itself binds and fully agonizes estrogen receptor-α, so its estrogenic activity is not fully captured by the “20%” number and is not fully blockable by an aromatase inhibitor.
• Progestogenic activity. Nandrolone is classified pharmacologically as a progestogen as well as an androgen: it binds and activates the progesterone receptor. It does not convert into progesterone — it is itself a progestin-like ligand. This activity is believed to contribute to its unusually strong HPTA suppression relative to its estrogen conversion, and to mammary / sexual side effects when combined with estrogen. Controlled human quantification of how much of any given side effect is progesterone-receptor mediated does not exist.
• 5α-reduction to DHN — the single most important difference from testosterone. Testosterone is 5α-reduced to DHT, a more potent AR agonist — amplification exactly in prostate, skin and hair follicles. Nandrolone is 5α-reduced to 5α-dihydronandrolone (DHN), a much weaker AR agonist — inactivation in those same tissues. Skeletal muscle expresses negligible 5α-reductase, so muscle sees full-strength nandrolone while prostate / scalp / skin see mostly weak DHN. That tissue split is the accepted explanation for less prostate growth, less acne and less scalp hair loss than testosterone at comparable anabolic doses — reduced, not absent.

Pharmacokinetics — the ester is the drug's clock

The decanoate ester is hydrolyzed rapidly once in blood (hydrolysis half-life ≤1 hour; combined hydrolysis + distribution + elimination of free nandrolone ≈ 4.3 hours). The long action is entirely depot-limited (“flip-flop” kinetics): release from the IM oil depot is the rate-limiting step, with a depot-release half-life of ~6 days, a terminal half-life of 7–12 days, and a duration of action of ~2–3 weeks per injection.

• Single-dose peaks (healthy men, Bagchus et al. 2005): 2.14 ng/mL (50 mg), 4.26 ng/mL (100 mg), 5.16 ng/mL (150 mg); time to peak 30 h (50 / 100 mg) to 72 h (150 mg); exposure rises proportionally with dose.
• Steady state on weekly dosing takes ~5 half-lives ≈ 5–8 weeks — why both clinicians and bodybuilders say Deca “kicks in” slowly, and why short Deca cycles end around the time levels plateau.

Deca vs NPP — same drug, different exit ramp

Nandrolone phenylpropionate (NPP, brand name Durabolin) actually came first (Organon, 1959); decanoate (1962) was the longer-acting successor. Once the ester is cleaved the molecule is identical — every difference below is pharmacokinetic, not pharmacodynamic.

Decanoate vs phenylpropionate

Parameter	Decanoate (Deca)	Phenylpropionate (NPP)
Elimination half-life	6–12 days (terminal 7–12 d)	~2.7 days (some secondary sources say 4–5 d)
Duration of action	2–3 weeks	5–7 days (~1 week)
Performance-use frequency [Community]	1–2×/week	Every other day to 3×/week
Time to steady state	~5–8 weeks	~2 weeks
Practical minimum cycle [Community]	12–16 weeks	8–10 weeks
Testosterone suppression after one 100 mg dose [PK/PD model]	~11–17.5 days, by site / volume	~5.8 days
Urine detection	Many months; 12–18 months cited with chronic use	Also many months — shorter, not short. Same metabolite (19-NA)

The real practical difference is manageability: with NPP, stopping actually stops exposure within ~1–2 weeks; with Deca the same decision plays out over a month or more, on a 7–12-day half-life, with no way to accelerate clearance. NPP's advantage is an exit ramp, not a different effect — at the cost of roughly 3× the injection burden and more peak–trough fluctuation if under-injected.

Medical history — a narrow, old drug

First FDA approval came on 5 October 1962 (NDA 13-132). Under the later DESI review the FDA concluded it was effective for the anemia of renal insufficiency. Broader 1975-era labeling (“probably effective” adjunct in senile / postmenopausal osteoporosis and pituitary-deficient dwarfism; “possibly effective” for lean-mass retention, advanced breast cancer, some anemias) was progressively narrowed from the 1970s onward, and recombinant erythropoietin largely replaced it for renal anemia in the late 1980s. On 21 May 2002 Organon notified the FDA it was no longer marketing Deca-Durabolin in the US; the FDA's later determination was that it was not withdrawn for reasons of safety or effectiveness. It is a Schedule III controlled substance in the US.

Management of the anemia of renal insufficiency — men 100–200 mg/week, women 50–100 mg/week IM; children 2–13 y: 25–50 mg every 3–4 weeks; stop if no hematologic improvement in 6 months. “Has not been shown to be safe and effective for the enhancement of athletic performance.”— DailyMed / Watson label, rev. March 2006 — the sole current US indication

Osteoporosis — real RCT support, outside the US label

Still on-label in the UK, Brazil and Canada (historical monograph) at 50 mg IM every 3 weeks; off-label in the US. A 2025 systematic review / meta-analysis of 7 trials, 293 postmenopausal women found nandrolone reduced fracture risk (moderate certainty), produced modest BMD gains (low certainty) and larger forearm bone-mineral-content gains (moderate certainty), reduced pain (moderate certainty) and increased muscle mass (moderate certainty); virilizing events were more frequent but mostly mild (low certainty). The authors call it a potential adjuvant for selected women, not a first-line therapy — and note the trials are old and small.

HIV wasting — lean mass at a fraction of bodybuilding doses

• 150 mg IM every 2 weeks × 12 weeks (HIV-infected men, 5–15% weight loss): lean body mass +1.6 kg vs +0.4 kg placebo; not significantly different from rhGH, with fewer adverse effects / discontinuations than rhGH.
• 150 mg every 2 weeks × 12 weeks, n=303 HIV-positive men: fat-free mass +1.34 kg (95% CI 0.60–2.08) and weight +1.48 kg vs placebo; weight gain also significantly greater than a testosterone arm.
• Dose-ranging work found lean-mass / weight gains essentially only at 100–150 mg every 2 weeks. Also studied in small trials in COPD wasting, dialysis, and burns / trauma / surgery recovery — none a US label indication.

Modern TRT-clinic practice

Because the US brand is discontinued, US prescribing is via compounding pharmacies. The contemporary pattern documented in the andrology literature is low-dose nandrolone added to testosterone replacement — typically 50–150 mg/week — with stated aims of joint-pain relief, added lean mass / bone support, and anabolic effect without escalating the testosterone dose. It is off-label, evidence is pilot-level, and long-term safety data in this population do not exist — Pan & Kovac and Tatem both say so explicitly.

Joint effects — the evidence, in descending order of quality

• 1 · Osteoporotic bone pain [Clinical, moderate certainty]. Two double-blind placebo-controlled trials in postmenopausal osteoporosis showed reduced bone pain; the 2025 meta-analysis rates pain reduction as moderate-certainty. This is bone pain in osteoporosis, at 50 mg every 3 weeks — not athletic joint pain.
• 2 · Patellofemoral pain syndrome [Clinical, old and small]. Darracott et al.: weekly IM nandrolone for 6 weeks improved symptoms and patellar bone density vs placebo; a secondary summary puts it at 20/23 improved on NPP vs 1/20 on placebo. Small, dated, never replicated with modern methods.
• 3 · Total knee arthroplasty [Clinical, very small]. Hohmann et al., double-blind RCT, n=10: better quadriceps strength and composite Knee Society Scores at 6 weeks, 6 and 12 months. The KSS contains a pain component but the pain subscore was never reported separately — so this study cannot be cited as joint-pain evidence.
• 4 · Rheumatoid arthritis [Clinical, negative]. Bird et al. 1987: 50 mg every 3 weeks for 2 years in postmenopausal women with RA — no significant difference in joint or bone pain vs control. Whatever nandrolone does, it did not relieve inflammatory joint pain in the best-duration trial.
• 5 · The TRT joint-pain pilot [Clinical, uncontrolled]. Hypogonadal men already on testosterone cypionate with significant joint pain were given nandrolone at half their testosterone dose (median 110 mg/week, IQR 100–150), reassessed at ~8 weeks (median 62 days). Of 48 men who completed the baseline survey, only 18 (37.5%) responded to follow-up; of those, 13 (72.2%) reported marked improvement, average pain-score reduction 52% among responders, and 5 (27.8%) reduced longstanding pain medication. No adverse events were noted. No placebo, no blinding, 62.5% lost to follow-up, subjective scale only. This is the study almost all TRT-clinic joint marketing ultimately rests on.

Doses reported — a ladder, not a recommendation

Ranges below are what the label, the trials and community practice report, with the evidence tier attached. Risk discussion should always specify which rung it is talking about.

Reported dose ranges by setting

Setting	Dose	Evidence tier
Anemia of renal insufficiency (US label)	Men 100–200 mg/wk; women 50–100 mg/wk. Stop if no response by 6 months; intermittent if possible	Clinical — label
Postmenopausal osteoporosis (UK / BR / CA label)	50 mg every 3 weeks, chronic and monitored	Clinical — label / RCTs
HIV wasting (RCTs)	150 mg every 2 weeks (≈75 mg/wk), 12–24 weeks; +1.3–1.6 kg lean mass	Clinical — RCT
TRT adjunct (“low-dose Deca”)	50–150 mg/wk, most commonly ~100 mg/wk; Tatem pilot: nandrolone = ½ the testosterone dose, alongside 100–200 mg/wk testosterone	Clinical pilot + Community
Bodybuilding — beginner	200–300 mg/wk + testosterone 300–500 mg/wk, 10–12 weeks (many sources say 12–16, given slow steady state)	Community
Bodybuilding — typical / intermediate	300–400 mg/wk (classic texts: “usual range 300–600 mg/wk”), 12–16 weeks	Community
Bodybuilding — advanced / high	400–600 mg/wk; >600 mg/wk reported, with diminishing returns vs suppression / lipid cost even by pro-use sources	Community
Women — performance use	~50 mg/wk, with high virilization risk even here; shorter ester historically preferred so it can be stopped	Community

Ratios are folklore-tier. The only ratio with any clinical provenance is the Tatem pilot's testosterone : nandrolone = 2 : 1. Community practice clusters around testosterone ≥ nandrolone — commonly 1:1 to 2:1 — on the theory that adequate testosterone / DHT / estradiol protects libido and erections; a vocal minority runs the reverse and reports fewer estrogenic problems. No trial has ever compared ratios. TRT-adjunct use, notably, is not “cycled” — it is continuous, which is precisely why its long-term safety gap matters more, not less.

Risks — with evidence quality

• HPTA suppression — profound and rapid, driven by AR + progestogenic negative feedback despite low estrogen conversion. A single 100 mg injection suppressed testosterone ~11–17 days in PK/PD modelling. Secondary-source figures (Llewellyn, primary studies not directly opened): 100 mg/wk × 6 wk → serum T −57%; 300 mg/wk → −70%. Testicular atrophy, oligospermia / azoospermia and infertility are on the label. Recovery is usually months, not weeks — classic texts state 2–6 months without intervention — and no nandrolone-specific prospective recovery trial exists; timelines are extrapolated.
• Lipids — HDL ↓, LDL ↑; the best-documented metabolic harm. An RCT at 200 mg/wk × 8 weeks altered apolipoproteins / lipoprotein(a). A secondary source reports 600 mg/wk × 10 wk → HDL −26%, slightly worse than equal testosterone enanthate; the 600 mg/wk HIV study found HDL −8–10 points but no worsening of total / LDL cholesterol or triglycerides. Changes revert on discontinuation (label). Injectable nandrolone is far less lipid-toxic than oral 17α-alkylated steroids.
• Cardiovascular — a class signal, not a nandrolone outcome dataset. AAS abuse cohorts show LVH, impaired diastolic function, cardiomyopathy, hypertension and sudden death, and animal nandrolone studies at ~20× clinical doses produce cardiomyopathy / cardiac collagen deposition. But the double-blind RCT of 200 mg/wk × 8 weeks in bodybuilders found no change in blood pressure or cardiac size / function in that window. No long-term cardiovascular outcome data (MI, stroke, mortality) exist at any dose, and polypharmacy confounds all abuse cohorts.
• Hematocrit — the therapeutic effect overshooting. Raising red-cell mass is the anemia indication. The high-dose HIV study recorded hemoglobin +1.0 to +1.4 g/dL. The effect is additive with testosterone — the combination is more erythropoietic than either alone — and thrombosis risk is the concern at high hematocrit.
• Mood / psychological. A systematic review of the abuse literature found psychiatric / mood effects in 9% of included studies — aggressiveness, anxiety, insomnia; the label lists habituation, excitation, insomnia, depression. Human causal data are thin and confounded, and the hypogonadal crash after a cycle is itself depressogenic. TRT-community reports run both ways, including men stopping even 100 mg/wk for mood.
• Virilization (women) — some changes irreversible. Voice deepening, hirsutism, acne, clitoral enlargement, menstrual disruption. At 50 mg every 3 weeks the meta-analysis found events more frequent than placebo but mostly mild — and the label warning is explicit that some virilizing changes are irreversible even after prompt discontinuation; voice and clitoral changes are the typically irreversible ones.
• Gynecomastia occurs despite low aromatization — via residual estradiol (especially from co-administered testosterone), nandrolone's own ER-α agonism, and progestogenic mammary effects. An AI can address the estrogenic component; it does not address the progestogenic component.
• Prostate — relative, preclinical, not “safe.” DHN is weak in prostate and rat ventral-prostate stimulation is weaker than testosterone's. Still, the label contraindicates known / suspected prostate cancer, warns geriatric men may be at increased risk of prostatic hypertrophy / carcinoma, and PSA can rise on androgen therapy generally.
• Liver — low-risk by structure, not zero paperwork. Injectable nandrolone is not 17α-alkylated and is not considered meaningfully hepatotoxic at clinical doses; the label nonetheless carries class warnings (peliosis hepatis, hepatic tumors with long-term androgen therapy) and advises periodic LFTs. Small ALT rises (~1.5×) appeared in one high-dose HIV trial, attributed partly to IM injections themselves.
• Other label effects: edema / fluid retention (caution in cardiac / renal disease; can worsen heart failure), hypercalcemia, decreased glucose tolerance, premature epiphyseal closure in children, acne, injection-site reactions.

“Deca dick” and prolactin — the standard story is oversimplified

Sexual dysfunction is real as a reported phenomenon; its mechanism is multifactorial and not settled. The clinical reviewers' proposed mechanisms: (1) nandrolone suppresses endogenous testosterone, so DHT falls — and DHT is required for nitric-oxide-mediated erection; (2) nandrolone's own androgenic activity in relevant tissues is weak (the DHN problem again), so it does not substitute for testosterone / DHT sexually even while total “androgen” load is high; (3) estrogen imbalance in either direction. Pan & Kovac note the WHO database lists impotence as an adverse effect but that no consistent reports of ED exist in the formal literature; the phenomenon is documented anecdotally, mainly with nandrolone used without testosterone, and risk rises consistently in reports with nandrolone dose, nandrolone:testosterone ratio, and cycle length — a consistent anecdotal gradient, not a measured dose–response curve.

On prolactin: elevated prolactin genuinely causes low libido, ED and anorgasmia, and 19-nor compounds are assumed to raise it. However, nandrolone does not convert to progesterone, progesterone itself inhibits prolactin in the cited physiology literature, and estrogen (from the testosterone in the stack) is a known prolactin-raiser. Critical reviews conclude controlled human evidence tying nandrolone itself to prolactin elevation is unconvincing at studied doses. Check prolactin if symptomatic — don't assume it.

Stack logic and interactions

• Why testosterone is the classic base — mechanistically sound. Nandrolone alone leaves a man without adequate DHT (erections, libido) and little estradiol (libido, mood, bone, joint comfort, lipids all need some estradiol), while suppressing his own production. Testosterone restores DHT and estradiol while nandrolone adds anabolic load with less prostate / scalp androgenicity per unit of anabolism. Hence “never run Deca alone” as community doctrine — the doctrine is sound physiology; the specific ratios are not evidence-based.
• Two 19-nors together (e.g. with trenbolone) — discouraged, but as a heuristic. Overlapping reputations for sexual dysfunction and mood effects, additive suppression and lipid harm, and — the most honest reason — if something goes wrong you cannot tell which drug is responsible. There is no pharmacologic contraindication and no study of the combination; commercial blends exist and people use it. “Never mix two 19-nors” is harm-reduction practice for self-experimenters, not a rule of pharmacology. The same logic applies, more weakly, to trestolone (MENT).
• Aromatase inhibitors act only on the testosterone. On a testosterone + nandrolone stack, estradiol comes overwhelmingly from the testosterone; an AI does nothing to progesterone-receptor activity or direct ER-α agonism — so “the AI isn't controlling my Deca gyno” is pharmacologically predictable. And over-suppression of estradiol itself causes joint pain, low libido and mood symptoms — an aggressive AI can recreate the exact symptoms nandrolone was added to treat.
• Prolactin-control agents — measure first. Cabergoline (typically 0.25 mg once or twice weekly in community practice) is proven to normalize prolactin in hyperprolactinemia and has never been trialed in nandrolone users; it carries nausea, hypotension, impulse-control disorders, and (at high cumulative doses, mainly in Parkinson's treatment) cardiac valvulopathy. Vitamin B6 / P5P (typically 100–200 mg/day in community practice) has only weak, supplement-level support, and high-dose B6 itself causes peripheral neuropathy with chronic use. The hierarchy in better harm-reduction sources: measure prolactin → look for other causes → reduce / stop the offending dose → dopamine agonist only for confirmed elevation with symptoms, under medical care.
• Finasteride / dutasteride: as above — blocks DHN formation and thereby increases nandrolone's androgenic effects in prostate / skin / scalp, the reverse of its effect with testosterone.
• Label interactions: anabolic steroids increase sensitivity to oral anticoagulants (monitor prothrombin time; dose often reduced when nandrolone starts or stops); they can alter glucose tolerance, so insulin / oral hypoglycemic doses may need adjustment; corticosteroids / ACTH add edema.
• A lab trap: nandrolone cross-reacts modestly in testosterone immunoassays (~2.1% in the Roche Elecsys assay studied). When testosterone values don't fit the clinical picture on a nandrolone regimen, LC-MS/MS is the reliable method.

Monitoring — borrowed thresholds, stated as borrowed

There is no nandrolone-specific monitoring guideline. Practice uses the testosterone-therapy framework plus the label's own lab instructions, plus prolactin because of the 19-nor reputation.

Markers, timing and red-flag thresholds in common use

Marker	Why / timing	Thresholds in use
CBC — hematocrit & hemoglobin	On-target erythrocytosis, additive with testosterone. Baseline, ~3 months, then every 6–12 months	Do not start if Hct >48–50%; stop / reduce and evaluate if >54%. Harm-reduction sources act earlier, at >52%
Lipid panel (HDL, LDL, triglycerides)	HDL ↓ / LDL ↑ is a label warning. Baseline, 3–6 months, annually	HDL <40 mg/dL, LDL >130–160 mg/dL (risk-dependent), or a large HDL fall from baseline → dose / cessation discussion
Blood pressure	Fluid retention + erythrocytosis + class effect. Baseline and every visit; home BP in community practice	Sustained ≥140/90 mmHg → address before continuing
PSA (+ prostate exam per age / risk)	Androgen therapy generally. Baseline (age >40), 3–6 months, annually	PSA >4.0 ng/mL, or a confirmed rise >1.4 ng/mL over baseline, or new nodule → stop and refer to urology
Estradiol (sensitive / LC-MS)	Mostly reflects the testosterone component. Baseline, after dose changes, ~3 months	No nandrolone-specific range; TRT panels commonly cite ~20–50 pg/mL — treat symptoms + level together, never the number alone
Prolactin	19-nor reputation; symptoms are the trigger. Baseline if obtainable; check when symptomatic	Above lab upper limit (commonly ~15–20 ng/mL) on a repeat morning draw → investigate causes before any cabergoline
Testosterone (LC-MS/MS preferred), LH / FSH, SHBG	Documents suppression; LH / FSH is the recovery marker after stopping	Immunoassay values that don't fit the picture → re-test by LC-MS/MS
Liver enzymes, glucose, symptoms	Label advises periodic LFTs; glucose tolerance can change. Baseline, annually; symptom review every visit	ALT / AST >2–3× upper limit → stop and evaluate. New edema / dyspnea, major mood change, or planned conception → reassess

Claim check

Claims circulating about nandrolone, checked against the research file:

• Not supported

  “Deca repairs joints and regenerates cartilage.”

  No human evidence of cartilage regeneration, joint-space preservation, or structural disease modification exists. The defensible claim is symptomatic relief of non-inflammatory pain in some men, mechanism unknown.

• Not supported

  “It boosts collagen synthesis by 250% and increases synovial fluid.”

  No primary human study supporting either percentage was found, and no study measured an increase in synovial fluid volume in humans. Animal tendon data raise the opposite worry — stiffer, dysplastic tendon.

• Not supported

  “Deca dick is prolactin — just take cabergoline.”

  Controlled human evidence tying nandrolone itself to prolactin elevation is unconvincing, and cabergoline has never been trialed in nandrolone users. The reviewers' mechanisms run through DHT loss, weak tissue androgenicity, and estrogen balance.

• Needs the qualifier

  “It doesn't aromatize, so there are no estrogen issues.”

  It aromatizes at ~20% of testosterone's rate (largely animal data), and nandrolone itself fully agonizes estrogen receptor-α — activity an aromatase inhibitor cannot fully block. Gynecomastia occurs.

• Needs the qualifier

  “It's prostate- and hair-safe.”

  Reduced, not absent: the DHN mechanism is a relative, preclinical claim. The label still contraindicates prostate cancer and warns on prostatic hypertrophy / carcinoma in older men.

• Not supported

  “Finasteride will tame Deca's androgenic sides, like it does testosterone's.”

  The pharmacology runs the other way: blocking 5α-reduction prevents DHN formation and raises nandrolone's androgenic activity in prostate, skin and scalp.

• Needs the qualifier

  “Never mix two 19-nors — it's a pharmacologic contraindication.”

  There is no pharmacologic contraindication and no study of the combination. Avoiding a second 19-nor is prudent troubleshooting practice — attribution and additive burden — not proven pharmacology.

• Not supported

  “Switch to NPP and you're clear for testing in a few weeks.”

  Same metabolite, 19-norandrosterone. Even a single 150 mg decanoate dose was detectable at 6 months in 83% of subjects; NPP's window is shorter, not short.

• Not supported

  “Low-dose Deca on TRT is proven safe long-term.”

  There are no prospective long-term data on mortality, cardiovascular events, prostate cancer, cognition or fertility recovery for prescribed nandrolone at any dose — and none at all for the modern TRT-adjunct pattern.

Open questions

Marked plainly, because they are open:

• Long-term safety of continuous low-dose use — no prospective data exist for >1 year of 50–150 mg/week alongside TRT, to verify or refute safety.
• Collagen and synovial-fluid claims — no primary human source for any specific collagen-synthesis percentage, and no human measurement of synovial fluid volume on nandrolone, was found.
• Prolactin — no controlled human demonstration that nandrolone raises prolactin, and no trial that cabergoline or P5P improves nandrolone-related sexual dysfunction.
• Two secondary figures — “HDL −26% at 600 mg/week” and “testosterone −57% at 100 mg/week / −70% at 300 mg/week” both trace to Llewellyn's Anabolics citing primary studies not directly opened in this research; directionally consistent with the RCT literature, quoted as secondary.
• NPP half-life — sources give 2.7 days (pharmacology references) vs 4–5 days (secondary / community); the 2.7-day figure is better sourced, not settled.
• Current US compounded availability — whether discontinuation and compounded access are uniform across all states / clinics in 2026 was not verified by a live pharmacy check; regulatory sources are the label / Federal Register record and clinic literature.

Sources

1. Tatem AJ et al. Nandrolone decanoate relieves joint pain in hypogonadal men: a novel prospective pilot study and review of the literature. Transl Androl Urol 2020. pmc.ncbi.nlm.nih.gov/articles/PMC7108994
2. DailyMed — Nandrolone Decanoate Injection USP label (Watson, rev. March 2006). dailymed.nlm.nih.gov (nandrolone decanoate label)
3. Patanè FG et al. Nandrolone Decanoate: Use, Abuse and Side Effects. Medicina 2020;56:606. pmc.ncbi.nlm.nih.gov/articles/PMC7696474
4. Camara LC et al. Nandrolone Decanoate for Postmenopausal Osteoporosis: A Systematic Review and Meta-Analysis of Randomized Trials. Cureus 2025. pmc.ncbi.nlm.nih.gov/articles/PMC12748007
5. Pan MM, Kovac JR. Beyond testosterone cypionate: evidence behind the use of nandrolone in male health and wellness. Transl Androl Urol 2016. pmc.ncbi.nlm.nih.gov/articles/PMC4837307
6. Wijnand HP et al. Pharmacokinetic parameters of nandrolone after IM nandrolone decanoate. Acta Endocrinol Suppl 1985. pubmed.ncbi.nlm.nih.gov/3865478
7. FDA / Federal Register — Deca-Durabolin discontinuation determination (approved 5 Oct 1962; marketing ceased 2002; not a safety withdrawal). federalregister.gov/documents/full_text/html/2010/08/10/2010-19698.html
8. Single-dose detection studies — 19-norandrosterone at 6 months (Bagchus et al.) and 9 months by GC/C-IRMS. sciencedirect.com/science/article/abs/pii/S0039128X16000350 · WADA summary: wada-ama.org (GC/C-IRMS nandrolone study)
9. Hartgens F et al. Effects of AAS on apolipoproteins and lipoprotein(a). Br J Sports Med 2004; and prospective echocardiographic assessment of AAS effects. pubmed.ncbi.nlm.nih.gov/15155420 · pubmed.ncbi.nlm.nih.gov/12868045
10. Jones IA et al. Anabolic steroids and tendons: review of mechanical, structural and biologic effects; and tendon narrative review. pmc.ncbi.nlm.nih.gov/articles/PMC12579764
11. Testosterone immunoassay cross-reactivity (Krasowski et al.). link.springer.com/article/10.1186/1472-6890-14-33/tables/5
12. Monitoring thresholds — testosterone-guideline summaries and harm-reduction panel (Hct, PSA, lipids, prolactin, estradiol ranges). pmc.ncbi.nlm.nih.gov/articles/PMC5690890 · teletest.ca/blog/blood-work-for-steroid-users
13. Progestin / prolactin critique. steroidal.com/steroids-news/progesterone-prolactin-clearing-confusion
14. Community practice, cited as community tier — dosing and ratios (Llewellyn Anabolics profile), TRT-adjunct experience, two-19-nor discussion. anaboliccartel.com/deca-durabolin-nandrolone-decanoate · excelmale.com/threads/prescribed-200mg-of-nandrolone.33392 · t-nation.com/t/mixing-two-19-nors/281771

Molecule

Δ¹-testosterone

Ester

Undecylenate · ~63% base by weight

Status

Veterinary only · US Schedule III

Human trials

None — no efficacy or PK trial

What it is — one double bond away from testosterone

Boldenone (androsta-1,4-dien-17β-ol-3-one; CAS 846-48-0) is testosterone with one structural change: an added double bond between C1 and C2. It is also the non-17α-alkylated parent of metandienone (Dianabol) — metandienone is 17α-methyl-boldenone — which is why boldenone itself is not an oral-type hepatotoxin by structure. It has never been marketed unesterified; the drug product is the C17β undecylenate ester.

• Ester math the community skips: the undecylenate ester weighs ~452 g/mol against 286 g/mol for boldenone base, so 250 mg of boldenone undecylenate delivers only ~158 mg of boldenone base (~63%). Milligram-for-milligram comparisons against testosterone silently under-dose boldenone on top of any potency difference.
• Androgen-receptor activity: hydrolysed boldenone is an AR agonist. Compiled in-vitro tables put its relative AR binding at ~50–75% of the reference androgen — a secondary-citation figure, consistent with Llewellyn's statement that it binds AR more weakly than testosterone but binds SHBG less, partly offsetting that in vivo. Kicman's standard review makes the more durable point: anabolic / androgenic dissociation is partial and tissue-dependent, and human skeletal muscle has no detectable 5α-reductase activity, so the parent steroid itself acts in muscle.
• Progestogenic activity — little to none. Progesterone-receptor relative binding is listed as <1%, against substantial PR binding for the nandrolone / trenbolone class. That is one of its better-supported differences from Deca — though it still rests on in-vitro binding data, not human outcomes.

Sources: Kicman AT, Br J Pharmacol 2008 · Relative-affinities compilation (Wikipedia template) · Llewellyn Anabolics profile, via Wikibolics

The 100:50 ratio — a rat number wearing a human suit

The endlessly repeated rating — anabolic 100 (testosterone = 100), androgenic 50 — traces to Llewellyn's Anabolics and Vida's 1969 compilation of rodent data. No primary paper reporting this exact boldenone experiment was located. The ratios come from the Hershberger assay: castrated immature male rats, dosed ~10 days, levator ani weight (“anabolic”) against ventral prostate / seminal vesicle weight (“androgenic”). The result shifts with dose, route, ester, duration, strain and endpoint — and it measures a rat perineal muscle and rat prostate, not human skeletal muscle, scalp, larynx or CNS. Read 100:50 as “moderately dissociated, roughly testosterone-like anabolism with less rat-prostate effect” — not as a human prediction. The veterinary label's own softer language is “marked anabolic properties and a minimal amount of androgenic activity.”

Aromatisation — the murkiest section in the whole file

• “~50% the rate of testosterone” — unverifiable. It appears in Llewellyn and every derivative profile, but no in-vitro aromatase-kinetics paper, human study or animal study reporting a 50% rate was found. It should be retired as a measurement and kept only as “Llewellyn said so.”
• What it would aromatise to: aromatase converts a 17β-hydroxy androgen to estradiol and a 17-keto androgen to estrone. Boldenone is 17β-hydroxy, so direct aromatisation would yield estradiol; its 17-oxidised metabolite boldione — a documented human metabolite — would yield estrone if aromatised. Estrone is genuinely weaker: estradiol is roughly 10× more potent in classic assays, with receptor-level estimates ranging 4–25× depending on assay — that range itself illustrates assay dependence.
• “It mainly makes estrone” — a community hypothesis, not a study. Its entire evidence base is a small number of self-reported blood panels (very high estrone, low sensitive estradiol on boldenone + testosterone) assembled by one community author — who himself concludes boldenone probably competes for aromatase and that metabolite-AI claims are “extrapolation from animal studies, in vitro models.” No controlled study measuring estrone and estradiol by LC-MS/MS in boldenone-treated men was found.
• The “AI metabolite” claim — unverifiable as a human in-vivo effect. The idea that a 1,4-dienedione metabolite inhibits aromatase has structural plausibility (related scaffolds include known steroidal AIs), but no study demonstrating inhibition at concentrations reached in human users was found.
• The one animal datapoint can't settle it: a rabbit study reported raised “estradiol” on boldenone — measured by ECLIA immunoassay, precisely the method that cross-reacts and reads high against mass spectrometry at male-range concentrations (see Lab traps, below).

Bottom line: boldenone is aromatizable in principle and is clinically much less estrogenic than testosterone at community doses — but neither the rate, nor the product, nor AI-metabolite activity has primary-source support in humans.

Sources: More Plates More Dates, boldenone aromatisation analysis (community) · ScienceDirect Topics, estrone · Endocrine Society position statement on estradiol measurement

5α-reduction to DHB — potent product, barely formed

Boldenone is a 5α-reductase substrate in principle; the product is 1-testosterone (dihydroboldenone, DHB). Formation from boldenone is said to be quantitatively very low — “affinity for this enzyme is extremely low” (Llewellyn, repeated in reference entries). No human study quantifying DHB formation after boldenone was found, so the pathway's relevance to an Equipoise user is presumed minor, not measured. DHB itself, given as a drug, is potent in rats: Friedel et al. 2006 found an equimolar dose matched testosterone propionate for levator ani, prostate and seminal vesicle growth (≈100:100 in that experiment) — which also conflicts with vendor tables listing DHB as “200:100.” Practical implication: the effects of DHB cypionate sold as a separate underground drug cannot be extrapolated back to boldenone undecylenate, because the conversion step is the bottleneck.

Pharmacokinetics — long-acting, but nobody has measured it in humans

• The hardest primary evidence is the horse label: 50 mg/mL in sesame oil, dosed every 3 weeks, with androgenic effects warned to persist 6–8 weeks after injection. The ester is genuinely long-acting — that much the label supports.
• “Half-life 14 days” — secondary citation, no human primary source. Repeated by Wikipedia, Llewellyn and virtually all profiles. No human IM boldenone-undecylenate PK study was found to support it.
• Measured animal PK: Soma et al. 2007, horses at the label dose (1.1 mg/kg IM), LC-MS/MS — median absorption half-life 8.5 h, median elimination half-life 123 h ≈ 5.1 days, Cmax ~1,128 pg/mL, plasma sampled to 64 days. A ~5-day elimination half-life and a “14-day biological half-life” are not the same quantity — but the discrepancy is never acknowledged in community sources.
• Time to peak / steady state — not measured in humans. By standard PK arithmetic only: if the effective half-life were ~14 days, weekly injections reach ~90% of steady state in ~6–7 weeks and ~97% in ~10 weeks, with an accumulation factor of ~3.4× the first-dose level; if the horse ~5-day figure applied, steady state would arrive in ~3 weeks. The truth for human IM undecylenate is unmeasured — which is why “judge it at week 6–8” is a heuristic, not a finding.

Sources: Equipoise US veterinary label (Zoetis, NADA 34-705), via Drugs.com · Soma et al., J Vet Pharmacol Ther 2007 · Gómez et al., J Steroid Biochem Mol Biol 2012 · WADA Prohibited List & TD2021IRMS

History and status — a horse drug with a brief human past

Ciba reportedly patented boldenone in 1949 and developed esters through the 1950s–60s (secondary chain, via Llewellyn). The undecylenate was introduced for human use as Parenabol in the late 1960s — retold as a lean-tissue-preserving anabolic for wasting and bone-mass retention in osteoporosis — and discontinued globally before the end of the 1970s. No published human efficacy trial from that era was located: at primary-source level, Parenabol's indications and its reason for discontinuation could not be verified. Do not repeat the common internet claim that it was “banned for safety” — no reason is documented in the sources found.

“An aid for treating debilitated horses when an improvement in weight, haircoat or general physical condition is desired” — debilitation following disease, overwork or overexertion; adjunctive only, “not a substitute for a well-balanced diet.” Dose: 0.5 mg/lb (≈1.1 mg/kg) IM, repeatable at 3-week intervals; most horses respond to 1–2 treatments.— Equipoise (boldenone undecylenate) US veterinary label, NADA 34-705

It was never labelled a performance / racing drug or a growth promoter for food animals, and there are no data for stallions or pregnant mares. Dose context: a 500 kg horse gets ~550 mg once every ≥3 weeks; a community human dose of 400 mg/week in an 80 kg man (5 mg/kg/week) is roughly 14× the labelled horse exposure rate on a mg/kg/time basis (arithmetic from the label dose). Current status: no human approval found anywhere; US Schedule III; WADA S1 at all times. The Ukrainian human product “Boldenol” in older reference entries could not be confirmed as a currently licensed human medicine.

Claimed effects — the evidence check

Each famous claim, with its best evidence and verdict

Claim	Best evidence	Verdict
Appetite — “EQ hunger”	Horse label trials: appetite improved in debilitated horses at label dose. No human appetite study exists.	Plausible; a boldenone-specific human effect is unproven. Debilitated-horse appetite restoration ≠ supraphysiologic hunger in fed athletes.
Erythropoiesis / RBC / hematocrit	Class mechanism (↑ erythropoietin, ↓ hepcidin, direct marrow effects). Rabbits at 5 mg/kg IM every 3 weeks: hematocrit 31.6% → 42.4%, hemoglobin 10.5 → 13.6 g/dL, RBC 5.12 → 6.42 ×10⁶/µL (small groups, n≈5). Broiler chicks: RBC, hemoglobin and hematocrit significantly increased.	A real class effect, confirmed for boldenone in animals. “Stronger than other AAS” is unverified — no head-to-head study in any species at matched exposure was found.
Vascularity	No study measures vascularity, endothelial function, or subcutaneous water with boldenone.	Community anecdote only.
Endurance / aerobic capacity	No boldenone endurance trial in humans or animals. In the best human class data, long-term AAS users showed reduced exercise capacity vs non-using strength athletes.	Unproven; class data partly contrary. The “↑RBC → ↑O₂ → endurance” chain ignores viscosity.
Collagen / joints / tendons	No boldenone-specific collagen, tendon, ligament or joint study found. Class animal data: AAS reduce tendon collagen biosynthesis, produce collagen dysplasia and increased tendon stiffness while muscle force rises faster.	Unsupported — and the direction of animal class data is opposite. The joint comfort lore belongs to nandrolone and does not transfer to a non-progestogenic compound.
Lean mass vs testosterone	Boldenone: no human data. Testosterone comparator (Bhasin et al., NEJM 1996): 600 mg/week testosterone enanthate × 10 weeks + standardised training → +6.1 kg fat-free mass.	“Slow, steady, dry gains” is anecdote that is pharmacokinetically coherent — and has never been measured against testosterone at matched dose / duration. There is no boldenone arm to stand on.

Sources: Equipoise veterinary label · Tousson et al., rabbit boldenone studies · Bhasin et al., NEJM 1996 · AAS cardiovascular-phenotype cohort, PMC9540672 · Tendon narrative review, PMC12579764

Dosing in practice — reported, not recommended

There is no medical dosing regimen for humans. Everything in this table is community / secondary-profile reporting, on nominal UGL label doses (§History note above).

What community sources converge on

Parameter	Older profile era (Llewellyn-derived)	Current guides / forums (2023–26)
Men, weekly dose	200–400 mg/week	300–600 mg/week most common; 400–600 the modal “serious” range; 800–1,000+ mg/week claimed at the advanced end (one forum repost claims up to 2,000 mg/week and that “1,000+ is where it becomes a good builder” — an outlier, low credibility)
Women (community)	50–100 mg/week	50–100 (some guides to 150) mg/week — with virilisation risk understated in these sources (see Risks)
Cycle length	8–12 weeks	12–16 weeks, commonly 16–20
Injection frequency	Weekly	1–2×/week, for volume / peak-smoothing reasons
First noticeable effects	—	Appetite / vascularity / endurance reports at 4–6 weeks; visible mass / strength judged at 6–8 weeks

Why cycles run long — the ester logic. The reasoning is internally consistent even though the half-life is unverified: a very long ester accumulates slowly, so an 8-week cycle spends most of its life below steady state and still has a long tail; extending to 14–20 weeks buys more time at plateau, at the cost of longer total exposure, longer suppression and more hematocrit / lipid burden. No study compares 8 vs 16 vs 20-week boldenone cycles on any outcome — the convention is arithmetic + tradition.

Front-loading. Community practice runs roughly double the maintenance dose for the first 1–2 weeks — the classic example is 600–800 mg/week × 2 weeks, then 300–400 mg/week. The rationale is sound PK arithmetic in principle (a loading dose can bring levels near plateau within the first interval instead of ~5 half-lives; at a 14-day half-life, instant plateau would imply a first dose ~3.4× the weekly dose, so 2× for 2 weeks is a partial version). Caveats: if the true human half-life is closer to the horse ~5 days, front-loading is mostly unnecessary; it front-loads peak-related harms and cannot shorten the tail; and no outcome data show front-loaded cycles gain more muscle. Mathematically coherent, clinically untested.

Sources: Muscle & Brawn, Equipoise cycle guide · Labgain, boldenone category · BodybuildBid, Equipoise profile (front-loading) · Evolutionary.org forum repost (high-dose outlier)

Risks — evidence-graded

• Hematocrit / viscosity / cardiovascular. Hematocrit rise is an established class effect, with boldenone animal data supportive. Viscosity rises exponentially with hematocrit; TRT guidance uses >54% as the stop / intervene threshold (see Monitoring) — a precautionary guideline extrapolation, not a proven cliff edge, and whether AAS-induced erythrocytosis causes thromboembolism at polycythemia rates is not established in RCTs. For cardiac structure, the human data are class / polypharmacy: long-term users show LV hypertrophy and remodelling, and a nationwide cohort (1,189 users vs 59,450 controls, ~11 years) found cardiomyopathy at adjusted HR 8.90 (95% CI 4.99–15.88). None of it isolates boldenone. Rabbit boldenone data add cardiac-muscle hypertrophy on histology. A published post-mortem case found blood boldenone 140 ng/mL in a man who had injected it — proof of exposure in a fatality context, not proof boldenone caused the death.
• Lipids. Boldenone-specific data are animal: rats at 5 mg/kg/week × 8 weeks showed ↑ total cholesterol, triglycerides, LDL-C and VLDL-C, ↓ HDL-C. Human class data associate AAS use with LDL ↑ >20% and HDL ↓ 20–70% across studies, worst with 17α-alkylated orals; non-alkylated injectables produce smaller but real HDL suppression (nandrolone 100 mg/week × 8 weeks: HDL −25–27%, reversible ~6 weeks after stopping). Expect boldenone in the injectable-class range — no boldenone-specific human lipid measurement exists.
• Blood pressure. AAS use is associated with hypertension in reviews / cohorts; there are no boldenone-specific human BP data. A long EQ cycle plausibly sustains any BP effect for months, including the tail.
• Mood / irritability / aggression. Class-level human data associate AAS use with hypomanic / aggressive symptoms in a subset of users and depressive symptoms on withdrawal; one cited comparison found 23% of AAS-using athletes met criteria for major mood disorders vs 4% of non-users. Boldenone-specific human psychiatric data: none. The Equipoise horse label itself notes androgenic “over-aggressiveness” in some animals persisting 6–8 weeks.
• Virilisation. Voice deepening and clitoromegaly can be irreversible; menstrual disruption, hirsutism, acne and scalp hair loss are dose / duration dependent. Boldenone's “mild, female-friendly” reputation rests on a rat assay rating and anecdote — and the long ester is a specific liability: once virilising effects appear, exposure continues for weeks after the last injection and cannot be switched off. No human female dose–response or safety data exist.
• Kidney. The community claim (“EQ is uniquely hard on the kidneys, because EPO production strains them”) splits three ways. The animal signal for boldenone itself is genuinely documented — rabbits: creatinine 1.25 → up to 2.30 mg/dL, urea up, glomeruli losing their typical shape, dose-duration dependent; rats: raised urea / creatinine / uric acid with renal histopathology. The human evidence is class / polypharmacy only: Herlitz et al. 2010 — 10 long-term AAS-using bodybuilders (mean BMI 34.7, high-protein diets, multiple agents) with proteinuria (mean 10.1 g/day) and renal insufficiency (mean creatinine 3.0 mg/dL), FSGS on biopsy in 9/10, improving on stopping in most, one progressing to ESRD, one relapsing on resuming. And creatinine alone misleads in this population — it rises with muscle mass, creatine, protein intake and hard training without GFR loss. Verdict: kidney injury in long-term AAS users is a real, case-series-grade class phenomenon; boldenone-specific human kidney injury is unproven, and the “EPO strain” mechanism is unverifiable.
• Hepatic. Not 17α-alkylated, so classic oral-AAS cholestasis / peliosis risk is not expected — nevertheless animal studies consistently show raised ALT / AST and hepatocellular vacuolation at 5 mg/kg-type doses. Human significance unknown.
• Gynecomastia is uncommon on EQ alone by reputation (low estrogenicity), possible at high doses or in stacks, and can persist after cessation per the ASIH literature.
• Long-term outcome data for boldenone: absent. No prospective cohort, no mortality data, no cancer-outcome data specific to this compound. Every long-term statement is an extrapolation from mixed-AAS cohorts with polypharmacy, training, diet and other-drug confounding. Anyone quoting a boldenone-specific long-term risk number is inventing it.

“EQ anxiety” — a real report pattern, an unproven mechanism

Anxiety / restlessness on boldenone is among the most consistently reported compound-specific effects in forums, often described as disproportionate to other AAS at similar doses. Animal work gives it partial support: in rats (Bueno et al.), higher anxiety measures appeared only in the high-dose / short protocol (5 mg/kg/week × 4 weeks), social interaction was altered in all protocols, and memory deficits appeared only in the low-dose / long protocol; Mowaad et al. 2023 (rats, 5 mg/kg/week × 2 months) found cortical oxidative stress and neuronal death — a neurotoxicity signal, though that paper's behavioural reporting is internally inconsistent.

The popular mechanism — a boldenone metabolite disrupts GABAergic neurosteroids (allopregnanolone signalling at GABA-A) — is unverifiable. No study measuring GABA, allopregnanolone or GABA-A function after boldenone was found, in any species; the story is told by analogy to class-level AAS effects on GABAergic transmission and to finasteride-type neurosteroid disruption, neither of which is boldenone data. Better-grounded competing explanations: low estradiol itself (E2 is neuroactive and anxiolytic-modulating), generic AAS CNS effects, and expectancy / nocebo in a community that primes users to expect “EQ anxiety.” None has been isolated experimentally for boldenone.

HPTA suppression — distinctive by duration, not by potency

Exogenous androgens suppress GnRH → LH / FSH → intratesticular testosterone and spermatogenesis; rabbit boldenone studies show testicular histology disturbance and suppressed testosterone / FSH / LH patterns, and rat studies show testicular oxidative damage. Duration is where boldenone stands apart: suppression cannot begin recovering until the depot clears, and the undecylenate tail plus a 12–20-week cycle means total suppression commonly spans 5–8+ months. Class recovery data: serum testosterone typically recovers over weeks–months; spermatogenesis over months — 4–12 months typical in case literature, up to 24–30 months in some; testicular volume can lag a year or more, and a minority have persistent hypogonadism symptoms years later.

Stack logic and interactions

• Why a testosterone base. Any suppressive AAS shuts down endogenous testosterone — and with it the body's main source of estradiol and DHT, which libido, erectile function, mood, bone and normal physiology require. Boldenone provides AR activation but unreliable E2 / DHT replacement, so testosterone is run to keep both in range — with the twist that EQ users often need more testosterone headroom because of the estrogen paradox.
• Ratios in practice — no consensus, no data. Examples found range from test : EQ ≈ 5:4 (500 mg test / 400 mg EQ, presented with the explicit rule “run test higher than EQ because EQ lowers estrogen”), through 1:1 (300–500 test / 400–600 EQ), to older EQ-dominant cycles; a minority run EQ higher than test and report needing exogenous estradiol — an escalation with no clinical literature at all. There is no evidence-based ratio; “test ≥ EQ” is a community response to the low-E2 phenomenon, not a studied rule.
• Aromatase inhibitors. If the paradox is even partly right, adding an AI by default — or dosed off a standard-E2 immunoassay — risks crashing estradiol outright, worsening exactly the symptoms users blame on EQ. Community harm-reduction practice has shifted to: no prophylactic AI on EQ cycles, AI only for clear high-E2 signs with a sensitive assay, and caution that exemestane (an irreversible inhibitor) overshoots longer. No controlled interaction study exists for any AI + boldenone combination. Conversely, gyno or water on an EQ stack should not be assumed “not estrogen” — the testosterone in the stack aromatises normally.
• The long ester shapes everything: late-emerging side effects (hematocrit peak, BP, lipids) may peak after the nominal cycle ends; PCT timing is contested (above); switching to shorter esters for the final weeks is a community workaround with no outcome data; and fertility-recovery clocks start from depot clearance, not from the last injection date.

The three “mild long-cycle” compounds, compared

Dimension	Boldenone undecylenate	Nandrolone decanoate	Methenolone (Primobolan) enanthate
Family	Testosterone (Δ¹)	19-nortestosterone	DHT-derived, 1-methyl-Δ¹ — a structural cousin of boldenone
A : A rating [Secondary]	100 : 50	~125 : 37 (other compilations differ)	88 : 44–57
Aromatisation	Yes — rate / product unresolved	Yes, lower (often quoted ~20% of test) [Secondary]; product estradiol	None — cannot aromatise
Progestogenic	No (PR <1%)	Yes — substantial PR activity	No
5α-reduction	Minimal → DHB (potent but barely formed)	To DHN, a weaker androgen	Already 5α-reduced framework; no relevant conversion
Human clinical history	Parenabol, discontinued 1970s; no trials found	Extensive human medical use / history and human studies exist	Human medical history (Schering) and some human literature exists

The honest summary: boldenone and methenolone are chosen for the same role — “mild, dry, long cycle” — but methenolone achieves low estrogenicity by non-aromatisation (certain), whereas boldenone's low estrogenicity is empirical and mechanistically unresolved. Nandrolone differs most sharply on progestogenic activity — and on having a real human literature.

Monitoring — framed by TRT / harm-reduction practice

There is no boldenone-specific monitoring trial; the thresholds below are extrapolations from TRT / AAS literature, adapted to this ester's kinetics. Information for discussion with a clinician — not a self-management protocol.

Markers that matter most for boldenone

Marker	Why for boldenone	Notes / assay
Hematocrit + hemoglobin (CBC) — priority 1	Signature risk; a long cycle means long accumulation	Same lab, hydrated, similar time of day; trend, don't react to one value
Blood pressure (home, validated cuff) — priority 1	Class effect, sustained by the long tail	Home averages beat clinic single readings
Lipids: HDL-C, LDL-C, triglycerides, ideally ApoB — priority 1	Rat boldenone data + class HDL suppression	Fasted; expect the HDL nadir mid / late cycle
Estradiol, sensitive LC-MS/MS — priority 1	Standard immunoassay can mislead precisely on this drug; estrone only if investigating the paradox (research-grade)	Interpret with symptoms, not to a forum target range
Total T (LC-MS/MS preferred), LH, FSH — priority 2	Confirms suppression depth; LH / FSH is the recovery marker after	On-cycle T reflects the test base, not EQ
Kidney: creatinine + cystatin-C eGFR, urinalysis + urine albumin / protein — priority 2	Animal hepatorenal signal + FSGS case series; creatinine alone is confounded	Proteinuria is the red-flag finding — not a modest creatinine bump in a large, creatine-using lifter
ALT / AST (± GGT, bilirubin) — priority 2	Animal liver signal; polypharmacy / oral add-ons are the bigger human risk	Marked rises in lifters also come from training itself — repeat rested
PSA (age / risk-appropriate, typically 40+ or 45+) — priority 2	Androgen exposure + TRT-guideline standard	Referral triggers below

Timing — built around the ester, not short-cycle calendars. Baseline before starting (without one, on-cycle numbers are uninterpretable). An early check at ~6–8 weeks — the first point levels plausibly approach plateau; a 3–4-week panel mostly measures the loading phase. A late-cycle panel at ~12–16 weeks / at the end, when hematocrit and lipids are expected to be at their worst. Post-cycle panels at ~6–8 weeks after the last injection (as the depot finally wanes) and ~3 months; semen analysis only ≥3 months after cessation if fertility is the question. On ongoing long cycles: hematocrit / BP at least every 8–12 weeks, lipids every 3–6 months.

• Hematocrit ≥54% — the Endocrine Society TRT threshold to stop / intervene (including therapeutic phlebotomy); >50% is a relative contraindication to starting. Many harm-reduction clinicians act at 52–54% with dose reduction / cessation and evaluate contributors (sleep apnea — which independently raises hematocrit — smoking, dehydration, altitude).
• Blood pressure: sustained home average ≥140/90 is hypertension-range requiring action; ≥160/100 or symptoms means prompt medical care.
• PSA: TRT-guideline urology-referral triggers — a confirmed rise >1.4 ng/mL within 12 months, or an absolute >4.0 ng/mL.
• Kidney: new proteinuria / albuminuria, a falling cystatin-C eGFR trend, or creatinine rising with falling cystatin-C eGFR is stop-and-evaluate territory — in the Herlitz series, cases improved when AAS stopped early.
• Lipids: HDL collapsing (e.g. <40 mg/dL) with rising LDL / ApoB mid-cycle is the expected harm pattern; persistence >3–6 months after cessation warrants a lipid workup rather than reassurance.
• Symptoms, any time — seek care promptly: chest pain / pressure, breathlessness, unilateral leg swelling, neurological symptoms, severe headache / visual change; severe or persistent anxiety, depression or suicidal ideation; painful gynecomastia lump, testicular pain, jaundice, dark urine / reduced urine output; in women, voice change or clitoral enlargement.

Claim check

Claims circulating about boldenone, checked against the research file:

• Not supported

  “It aromatises at 50% the rate of testosterone.”

  No in-vitro, animal or human measurement was located. The figure is traceable only to Llewellyn's Anabolics and its copies.

• Not supported

  “Boldenone aromatises to estrone, not estradiol — and its metabolite is an AI.”

  Both are hypotheses. The estrone story rests on a few self-reported panels with no LC-MS/MS study behind it; aromatase inhibition by a metabolite has never been demonstrated in human users.

• Not supported

  “The human half-life is 14 days.”

  No human PK study exists. The one measured figure — horses, LC-MS/MS — is an elimination half-life of ~123 hours (≈5.1 days); the two numbers are different quantities, and community sources never acknowledge the gap.

• Not supported

  “EQ raises red cells more than any other steroid.”

  Erythropoiesis is a class effect — Llewellyn himself says so — and no head-to-head comparison against testosterone or nandrolone was found in any species at matched exposure. Long cycles and high nominal doses may explain the impression.

• Not supported

  “It's good for joints and tendons — it boosts collagen.”

  No boldenone connective-tissue study exists, and the animal class data point the other way: reduced collagen biosynthesis, dysplasia, stiffer tendons. Flag this claim whenever it's quoted.

• Not supported

  “EQ anxiety is caused by GABA / allopregnanolone disruption.”

  No boldenone neurosteroid or GABA study exists in any species. Low estradiol, generic AAS CNS effects and expectancy are all better-grounded candidate contributors — none isolated experimentally.

• Needs the qualifier

  “It's detectable for up to 1.5 years.”

  A secondary repetition with no controlled human IM detection study behind it. ~4–5 months is plausible but likewise unmeasured; the window is dose-, duration-, matrix- and assay-dependent.

• Needs the qualifier

  “100:50 — as anabolic as testosterone, half as androgenic.”

  A rodent Hershberger-assay figure from a 1969 compilation, with no primary boldenone paper located. Directionally “moderately dissociated” is fair; the precision is not a human measurement.

• Not supported

  “EQ is uniquely nephrotoxic in humans because of EPO strain.”

  Human evidence is polypharmacy case-series only and the mechanism is speculation. Animal renal injury is documented — which is a different, narrower claim.

Open questions

Marked plainly, because they are open — every item below was searched for on 23 Oct 2026, with no primary source found:

• Any human dose–response for lean mass, strength, appetite or mood — no trial exists to verify against.
• Human pharmacokinetics of IM boldenone undecylenate — half-life, time to peak and steady state are all unmeasured in humans.
• Estrone vs estradiol — no LC-MS/MS study in boldenone-treated men; the aromatisation rate and the AI-metabolite claim are likewise unmeasured.
• Parenabol's record — exact human indications, doses and reason for discontinuation; and whether any currently licensed human boldenone product exists anywhere (including the older “Boldenol, Ukraine” listing).
• Assay cross-reactivity — boldenone-specific percentages in commercial testosterone / estradiol immunoassays were not found, though the estradiol / estrone interference concern is well founded.
• DHB's “200:100” ratio — a vendor / community figure inconsistent with Friedel 2006's equimolar, ~testosterone-like result in rats.

Sources

1. Kicman AT. Pharmacology of anabolic steroids. Br J Pharmacol 2008. pmc.ncbi.nlm.nih.gov/articles/PMC2439524
2. Equipoise (boldenone undecylenate) US veterinary label, Zoetis / Fort Dodge, NADA 34-705. drugs.com/vet/equipoise · Regulation text: govinfo.gov (21 CFR 522.204)
3. Soma LR et al. Pharmacokinetics of boldenone and stanozolol in horses. J Vet Pharmacol Ther 2007. reference.medscape.com/medline/abstract/17348894
4. Gómez C et al. New potential markers for the detection of boldenone misuse. J Steroid Biochem Mol Biol 2012. sciencedirect.com/science/article/abs/pii/S096007601200115X
5. Behairy A et al. Boldenone undecylenate-mediated hepatorenal impairment. Front Pharmacol 2021. pmc.ncbi.nlm.nih.gov/articles/PMC8111012
6. Mowaad NA et al. Behavioral and neurochemical changes induced by boldenone and/or tramadol in rats. 2023. pmc.ncbi.nlm.nih.gov/articles/PMC10066173
7. Tousson E et al. Rabbit boldenone studies (hematology, hepatic / renal structure-function). hero.epa.gov/reference/1538303 · researchgate.net (rabbit boldenone study)
8. Bhasin S et al. Effects of supraphysiologic testosterone on muscle size / strength. NEJM 1996. nejm.org/doi/abs/10.1056/NEJM199607043350101
9. Cardiac / metabolic effects of AAS (lipids, BP, LV) — review. pmc.ncbi.nlm.nih.gov/articles/PMC4111565 · AAS cardiomyopathy review (nationwide cohort HR). pmc.ncbi.nlm.nih.gov/articles/PMC12467473 · Cardiovascular phenotype of long-term AAS users. pmc.ncbi.nlm.nih.gov/articles/PMC9540672
10. AAS and the kidney — systematic review (incl. Herlitz FSGS series). pmc.ncbi.nlm.nih.gov/articles/PMC6545019 · Boldenone testicular oxidative damage in rats. pmc.ncbi.nlm.nih.gov/articles/PMC7694087
11. Recovery of spermatogenesis after TRT / AAS. pmc.ncbi.nlm.nih.gov/articles/PMC4854084 · ASIH recovery scoping review. pmc.ncbi.nlm.nih.gov/articles/PMC10620455
12. Erythrocytosis following testosterone therapy (Endocrine Society 54% threshold). pmc.ncbi.nlm.nih.gov/articles/PMC5690890 · Erythrocytosis / phlebotomy guideline comparison. pmc.ncbi.nlm.nih.gov/articles/PMC11466264
13. Endocrine Society position statement: challenges to measurement of estradiol. pmc.ncbi.nlm.nih.gov/articles/PMC3615207 · LC-MS/MS steroid assay utility. pmc.ncbi.nlm.nih.gov/articles/PMC9900367 · Estradiol immunoassay cross-reactivity. pmc.ncbi.nlm.nih.gov/articles/PMC8859944
14. Impact of anabolic steroids on tendons — narrative review. pmc.ncbi.nlm.nih.gov/articles/PMC12579764
15. WADA Prohibited List S1. wada-ama.org/en/prohibited-list · WADA TD2021IRMS (boldenone 2.5–30 ng/mL IRMS rule). wada-ama.org (TD2021IRMS PDF) · US Schedule III listing, 21 CFR 1308. ecfr.gov (21 CFR 1308)
16. Australian black-market AAS analysis (boldenone oils 101–206 mg/mL vs 200 label; substitution). researchgate.net (Australian AAS market analysis) · Phytosterol / boldenone human pilot (n=10, negative). ncbi.nlm.nih.gov/pubmed/17613052
17. Secondary / community, cited as that tier — Llewellyn Anabolics profile and reference entries. en.wikipedia.org/wiki/Boldenone_undecylenate · bpmedical.info (Equipoise profile) · wikibolics.com (Equipoise profile) · Relative binding affinities compilation. en.wikipedia.org (relative affinities template)
18. Secondary / community — aromatisation hypotheses (MPMD). moreplatesmoredates.com (boldenone aromatisation) · 1-Testosterone (Friedel 2006 summary). alchetron.com/1-Testosterone · Estrone potency overview. sciencedirect.com/topics (estrone)
19. Community dosing guides, cited as community tier. muscleandbrawn.com/anabolics/equipoise-cycle · labgain.com/product-category/injectable-steroids/boldenone · bodybuildbid.com (Equipoise profile) · evolutionary.org/forums/threads/this-is-gold.86774 · Primobolan comparison profiles. wikistero.com/primobolan · steroidal.com/steroid-profiles/primobolan

Catalogs decoded

2 — Tianjin Bio · High Purity

Listings

~240 SKUs (~110 + ~130)

Hard-stop listings

1 — insulin

Genuinely new names

1 — eloralintide

How to read these lists

Both documents are supplier price lists, not formularies. Products appear under short codes (EL5, SUR10, CS10 and so on), the same molecule can appear two or three times under different codes, and the Tianjin list is machine-translated — so spelling errors and loose labels are part of the document, not exceptions to it. Prices below are per 10-vial kit, USD, exactly as listed. A listing proves only that a name was offered for sale; it does not verify what is in a vial.

Sources: TIANJIN_BIO_PRODUCTS_LIST_1.pdf (Tina / Tianjin Bio) · High_Purity_Price_List__1___1___1.xlsx (High Purity) — both as filed 26 Oct 2026

The Khavinson bioregulator family — the bulk of what's “new”

Most of the unfamiliar names in both catalogs belong to one school: the Khavinson bioregulators — very short peptides of 2–4 amino acids, named for a target tissue, of the same school as Epitalon + Pinealon (the “Cogniblend” pairing covered elsewhere in this library). At Tianjin they are sold as 20 mg kits at ~$124–137; most also appear in High Purity. The evidence grade is the same across the whole family, and it travels with every name in the table: Russian institute literature, small and uncontrolled, with no independent RCTs.

Khavinson family in the catalogs · Sequences as listed in the research file

Name	Target / sequence	Decode note
Chonluten	Lung / bronchus · Glu-Asp-Gly (tripeptide)	Both lists, $124. Asked about 26 Oct 2026 — this decode is the answer.
Bronchogen	Lung / bronchial · Ala-Glu-Asp-Leu (tetrapeptide)	Yes — two different lung peptides in the same catalog: Chonluten (tripeptide) vs Bronchogen (tetrapeptide).
Crystagen	Immune system · Glu-Asp-Pro (tripeptide)	Family-grade evidence, as above.
Livagen	Liver · Lys-Glu-Asp-Ala	Family-grade evidence, as above.
Cartalax	Cartilage / joints · Ala-Glu-Asp	The one in this family that maps to a live issue in the documented protocol (shoulders) — on paper only; the evidence is the family grade above.
Pancragen	Pancreas · Lys-Glu-Asp-Trp	Marketed for glucose metabolism.
Cortagen	Cerebral cortex · Ala-Glu-Asp-Pro	Pinealon's sibling.
Prostamax	Prostate · Lys-Glu-Asp-Pro	Family-grade evidence, as above.
Testagen	Testes · Lys-Glu-Asp-Gly	Marketed as endogenous-testosterone support. Noted for literacy more than relevance: irrelevant on exogenous TRT, by the same category logic as enclomiphene.
Vilon	Immune / thymus · Lys-Glu (dipeptide)	The smallest Khavinson in the lists.
Thymalin	Thymus extract peptide complex (immune)	An older Soviet-era preparation. Distinct from Thymosin Alpha-1 (also listed, codes TA5 / TA10), which is the better-characterised immune peptide.

Prices and presence: Tianjin Bio list · High Purity list, as filed 26 Oct 2026

GLP-1 / metabolic frontier

The metabolic shelf is where the catalogs look most current — and where the decode matters most, because only one name here is actually new:

• Eloralintide (EL5 / EL10) — the genuinely new name. Lilly's long-acting amylin analog (the drug class of cagrilintide), designed as a weekly, in clinical development as an Ozempic-era combination partner. At $188–477 per kit it is the most expensive metabolic on either list. The decode's verdict: this is cagrilintide's next-generation competitor, not a new mechanism.
• Survodutide (SUR10) — Boehringer's dual GLP-1 / glucagon agonist, the same dual family tree as mazdutide (also listed, MDT5 / 10). A Phase 3 obesity / MASH drug.
• Retatrutide + tirzepatide fixed blend (High Purity, 20+40 mg) — two incretins stacked in one vial. The decode is blunt about why it exists: because catalogs sell combinations, not because the pharmacology asks for it.
• Cagrilintide + semaglutide blend (CS10) — the “CagriSema” combination, following Lilly / Novo trial logic, pre-mixed.
• Dulaglutide (the Trulicity molecule) and a plain “GLP-1 5 mg” (GP) — older or unspecified incretins. “GLP-1” with no molecule name is a labelling red flag for quality.
• Adipotide / FTPP (AP5 / AP10) — an experimental pro-apoptotic peptide that targets fat-tissue blood vessels. Primate data showed kidney toxicity signals. A famous research curiosity, stalled for a reason.
• AOD-9604 (“AOD”) — a modified HGH C-terminal fragment that failed its own obesity trials and is sold anyway. HGH Fragment 176-191, its parent concept, is also listed. Both are fragments riding HGH's reputation.
• Lipo-C / MIC / L-Carnitine (LC216, LC516, “MIC”) — injectable “lipotropic” blends (methionine / inositol / choline ± B12) from the weight-loss-clinic world. Clinic-margin products; in the decode's words, nothing here moves fat mass meaningfully.

Growth axis / myostatin cluster

• ACE-031 (AE1) — a soluble activin receptor decoy (a myostatin / activin trap). It reached real human trials (BMS / Acceleron) and was halted over nosebleed / gum-bleed and vascular signals. The follistatin idea with a worse risk sheet.
• Follistatin / FST-344 (FN1) — a myostatin-binding protein; FST-344 is the gene-therapy isoform name borrowed for the peptide listing. Rodent data and supplement-world legend, with essentially no human dosing literature.
• GDF-8 (F81) — literally myostatin itself, sold by the vial. Research-reagent logic (antibodies / standards) — not a builder, despite the name association. Readers of the Trevogrumab deep dive will recognise the target: here it is being sold, not blocked.
• IGF-DES (IGD) — a truncated IGF-1 analog, more potent and less binding-protein-limited than LR3 on paper, with even less human data than LR3 — which itself has none formal.
• HMG (G75) — human menopausal gonadotropin (FSH + LH from urine), a fertility drug: the other half of the HCG shelf.
• Gonadorelin (GND2) and triptorelin (TRI2) — GnRH and a GnRH superagonist. PCT folklore uses them for axis restart; triptorelin at dose suppresses the axis — chemical-castration pharmacology in prostate cancer. The margin between the folklore and the opposite effect is the whole story.
• Kisspeptin-10 (KS) — an upstream GnRH trigger; a legitimate research molecule for axis diagnostics.
• Teriparatide (TER10) — Forteo: recombinant PTH 1-34, a prescription osteoporosis drug. A real anabolic — for bone. Its presence shows these lists blur “peptide” into “any injectable biologic.”
• ACTH 1-39 (High Purity) — full-length corticotropin; drives adrenal cortisol output. A diagnostic / legacy drug, not a performance compound.

Repair / immune / miscellaneous biologics

• Ara-290 (RA10 / RA16) — cibinetide: a peptide spun out of EPO's tissue-protective domain, engineered to keep repair signalling without raising red cells — designed, in other words, to avoid exactly the EPO problem flagged above. It has a neuropathic-pain trial history.
• KPV (KPV5 / 10) — an alpha-MSH fragment tripeptide; gut / skin anti-inflammatory. Also the fourth ingredient that turns the GLOW blend into KLOW (see alias traps).
• LL-37 (cat. 375) — the human cathelicidin antimicrobial peptide; an immune / skin research staple.
• Cerebrolysin (CBL60) — a pig-brain-derived peptide mixture, used as a nootropic / stroke drug in some countries. An IV infusion product; the evidence is contested.
• VIP (VIP5 / 10) — vasoactive intestinal peptide; a potent vasodilator with a research / clinic niche (the CIRS world).
• Oxytocin (OT2 / OT5) — the hormone itself; social-bonding intranasal folklore, a short half-life, and a real uterotonic drug.
• Alprostadil (PRO20) — prostaglandin E1 (Caverject class), an intracavernosal erectile drug. Catalogs file it under peptides because it ships in vials.
• Melatonin 10 mg vials (MT10) — injectable melatonin; a curiosity next to the sleep file documented elsewhere (where the relevant use was oral and occasional).

Neuro / nootropic corner

• Semax (XA) and Selank (SK) — Russian nootropic peptides; anxiolytic / cognitive claims and an intranasal tradition, at family-grade evidence.
• Adamax (High Purity) — a Semax analog with an adamantane group bolted on for stability / potency; newer, and thinner data.
• Dihexa — an angiotensin-derived nootropic famous for the “million times more potent than BDNF” headline. Animal cognition data; a mechanism worry (growth signalling) follows it.
• P21 / P021 — a CNTF-derived neurogenesis peptide (the P021 research line); Alzheimer's-model animal data.
• PE-22-28 (PE) — a spadin analog: a TREK-1 channel blocker peptide studied as a fast antidepressant candidate in animals.
• Orexin A / Orexin B (High Purity; A is $550 per kit — the priciest item on either list) — the wakefulness peptides themselves. The mirror image of the Dayvigo file covered elsewhere in this project: Dayvigo (lemborexant) blocks orexin receptors, while orexin agonism is the insomnia-opposite pharmacology (narcolepsy research). Research reagents, not usable sleep / wake tools — but, in the decode's judgement, the single most relevant listing in either catalog to understand.

Cosmetic / clinic peptides

Snap-8, Matrixyl, AHK-Cu, GHK-Cu, hyaluronic acid, and the $31 “booster” line (PDRN, collagen, vitamin C, periocular, acne) — skincare / mesotherapy catalog filler. PDRN is salmon-DNA polynucleotide injections (the Rejuran class). Not the lane this library covers — but it explains why these catalogs run to 100+ SKUs in the first place: a large share of the count was never aimed at the performance or metabolic reader at all.

Alias traps & labelling notes

The practical core of the decode. Each of these is a place where reading a catalog at face value gives you the wrong product, the wrong count, or the wrong molecule:

Alias traps — as recorded in the research file

Listing	What it actually is
“CJC-1296 with DAC” (High Purity, CD5)	A typo for CJC-1295 with DAC. There is no CJC-1296.
“Epithalon”	Epitalon — a spelling variant of the Cogniblend component covered elsewhere in this library.
GLOW vs KLOW	The same BPC-157 + GHK-Cu + TB-500 blend; KLOW adds KPV. Tianjin's BBG70 is the GLOW formula under a code.
BT5 / BT20 vs TB5 / TB10	Both are TB-500 (Thymosin Beta-4) — two suppliers' codes inside the same lists. Watch for duplicate ordering.
Lipo-C	Appears three times (LC216, LC516, MIC) at three prices with near-identical descriptions.
“Samaglutide” (Tianjin header)	Semaglutide. Together with renderings like “L-Carn itine,” it marks a machine-translated list — expect label sloppiness generally.
B12 (Tianjin, “10ml 大瓶” = “big bottle”)	Injectable B12 at kit pricing — already covered in the documented protocol (methylcobalamin, alongside injectable nicotinamide).

Price spot-checks

Spot-checks only — twelve overlapping SKUs, not a price list. The pattern the file records: High Purity runs cheaper on most overlapping SKUs; Tianjin carries the broader Khavinson line. Tianjin shipping, as listed: $90 under 5 kits, free over $1,000; FedEx / UPS, 5–7 days to the US.

Price spot-checks · Per 10-vial kit, USD · As listed 26 Oct 2026

Item	Tianjin	High Purity
Retatrutide 10 mg	$85	$80
Tirzepatide 10 mg	$50	$45
Semaglutide 10 mg	$47	$40
MOTS-C 10 mg	$71	$55
Epitalon 10 mg	$71	$40
Tesamorelin 10 mg	$197	$160
SS-31 10 mg	$88	$80
BPC-157 5 mg	$44	$45
Cagrilintide 5 mg	$121	$125
5-Amino-1MQ 5 mg	$59	$65
Chonluten 20 mg	$124	— (not listed)
Eloralintide 10 mg	$477	$350

Source: Tianjin Bio list · High Purity list, price spot-checks as filed 26 Oct 2026. Two exceptions to the “High Purity is cheaper” pattern are in the table itself: BPC-157 and cagrilintide.

Claim check

Readings of these catalogs that the decode corrects, checked against the research file:

• Not supported

  “CJC-1296 with DAC is a distinct compound.”

  It is a typo for CJC-1295 with DAC. There is no CJC-1296.

• Not supported

  “GLOW and KLOW are the same blend.”

  Same BPC-157 + GHK-Cu + TB-500 base — but KLOW adds a fourth ingredient, KPV. The names differ because the formulas differ.

• Not supported

  “BT and TB codes are different products.”

  BT5 / BT20 and TB5 / TB10 are both TB-500 (Thymosin Beta-4) — two suppliers' codes in the same lists.

• Needs the qualifier

  “Eloralintide is a new mechanism.”

  It is the one genuinely new drug-class name in either catalog — a long-acting amylin analog, i.e. cagrilintide's next-generation competitor. New name, familiar mechanism family.

• Not supported

  “Lemon Bottle is a fat-loss compound.”

  It is a local cosmetic lipolysis injection, not systemic fat loss — and it can cause uneven necrosis if misused.

• Not supported

  “AOD-9604 works for obesity — it's the fat-burning fragment of HGH.”

  AOD-9604 failed its own obesity trials. It is sold anyway, alongside its parent concept, HGH Fragment 176-191.

• Needs the qualifier

  “The Khavinson peptides are proven tissue regenerators.”

  The family evidence is Russian institute literature — small, uncontrolled, with no independent RCTs. That grade applies to every name in the family table, Cartalax included.

• Not supported

  “Triptorelin restarts the axis.”

  At dose, triptorelin suppresses the axis — the pharmacology used for chemical castration in prostate cancer. The folklore use and the drug's actual effect point in opposite directions.

Open questions

Marked plainly, because they are open — the catalogs themselves cannot answer these, and the research file does not pretend otherwise:

• What is actually in any given vial. A price list is an offer, not an assay. The decode verifies names, codes and listed prices — not identity, purity or dose accuracy of any product.
• Unnamed “GLP-1 5 mg” (GP). With no molecule named, there is nothing to decode — which is itself the finding, and why the file flags it as a quality red flag.
• Whether the Khavinson family works at all, by independent standards. No independent RCTs exist for the family as filed; the open question is the evidence, not the catalog.
• Human dosing for the reagent-grade listings (follistatin / FST-344, GDF-8, orexin A / B, IGF-DES). The file records essentially no human dosing literature for these — they are listed as research reagents and curiosities, and this page keeps them there.

Sources

1. Tina / Tianjin Bio — product & price list, TIANJIN_BIO_PRODUCTS_LIST_1.pdf (~110 SKUs), as filed 26 Oct 2026. All Tianjin prices, codes (including XT100, BBG70, LC216 / LC516) and the shipping terms quoted above are taken from this document.
2. High Purity — price list, High_Purity_Price_List__1___1___1.xlsx (~130 SKUs), as filed 26 Oct 2026. All High Purity prices and codes (including ISU, CD5, EL5 / EL10, SUR10, CS10, GP) are taken from this document.
3. Chinese peptide catalog decode — research report, 26 Oct 2026. Cross-referenced against the documented inventory, prior adjudications and hard-check framework as of 26 Oct 2026. Every family grouping, evidence grade, alias correction and flag on this page follows that report, with its hedges carried over.

Construct

Engineered Follistatin-344 · ~40 kDa

Inventor

Michael Farber

Sold by

BioLongevity Labs · ~$299 / 10 mg

Status

Research chemical — not for human use

What it is — an engineered follistatin, per the seller

FLGR242 / FLGR-242 is a codename given by Jay Campbell to a follistatin-based protein sold by BioLongevity Labs — a company Campbell co-founded. It is sold as a research chemical (“not for human use”), at roughly $299 per 10 mg.

The construct, per the seller: a fragmented, modified Follistatin-344 whose base is the FS I-I follistatin variation — a myostatin-selective follistatin derivative described in the literature (Nakatani et al. 2008) — engineered to bind myostatin while largely sparing activin, and fused to an albumin-binding peptide. The seller has published that peptide's sequence (GGSGGSGGSGGRLIEDICLPRWGCLWEDD); its tail is SA21, an albumin binder published by Genentech scientists (Dennis et al. 2002). Molecular weight ~40 kDa.

The inventor is Michael Farber, and production is described as modified CHO (Chinese hamster ovary) cells — which is the correct platform for a glycoprotein of this class. The design, on paper, is serious protein engineering built from real published components. What follows is about the gap between the design and the evidence for the product.

Sources: BioLongevity Labs product page · Jay Campbell, jaycampbell.com · Tarzanoptimized, Farber / Kruder talk recap

The numbers are the seller's

Every quantitative property attached to FLGR-242 comes from the selling side, and none of it has been independently published — in any species:

Seller-asserted properties — and their verification status

Property	Seller-asserted figure	Verification status
Serum half-life	~19 days (vs ~1–2 hours for unmodified Follistatin-344)	Unverified — never published in any species.
Albumin affinity	<20 nM Kd	Unverified — binding data never published.
Myostatin selectivity	Binds myostatin while largely sparing activin	Design intent per the seller; the binding has never been published.

Source: Kalios, FLGR-242 compound page · Frontload, “Is FLGR-242 Legit?”

The headline claim — 10–12 lb

“5 mg per week for 4 months → 10–12 lb of lean muscle.”— Attributed to Michael Farber, the construct's inventor (26 Oct 2026)

Trace the number and it does not lead to a study. Campbell's own published writing promises “roughly 0.5 to 1 pound of true skeletal mass per week” over 12–16 weeks, alongside hormones, training and nutrition — the 10–12 lb figure is that marketing arithmetic restated. Both men are on the selling side of the product: Farber invented it, Campbell co-founded the company that sells it and gave it its name. A claim sourced to the people who profit from it is a conflict of interest, not a citation.

The supporting testimonials in circulation are all seller-affiliated and unverified: a BioLongevity affiliate's self-reported 5.2 lb DEXA gain in 8 weeks (via the Tarzanoptimized recap), and Campbell's claim that IFBB pro Alex Padron added 15 lb of lean mass in a 10-week contest prep. No methods, no independent measurement, no cohort.

Evidence state — the core problem

Zero published human data exists on FLGR-242 itself. Kalios's check (28 Sep 2026) found no record for FLGR242 / FLGR-242 on PubMed or ClinicalTrials.gov. The construct's full sequence, manufacturing method, glycosylation, myostatin / activin binding, and half-life in any species have never been published, and no laboratory independent of the seller has analysed the product.

What would change the verdict: publication of the construct's binding and pharmacokinetic data, independent lab analysis of the product, or any human cohort with methods attached — even a case series with DEXA methods stated. None of that exists today.

The follistatin in the literature is gene therapy — not this

The follistatin results that are in the scientific literature come from AAV gene therapy — a virus installed in muscle producing follistatin continuously — not from injecting a follistatin protein. That distinction is the whole ballgame: the gene-therapy experiments never had to solve the delivery, half-life or folding problems that an injected product lives or dies on.

Follistatin is a 300+ amino-acid glycoprotein, and correct folding and glycosylation are a precondition for activity. Which is why what is actually in any given vial cannot be assumed from the label — a point that applies before any question of dose or effect is even reached.

Source: Frontload, “Is FLGR-242 Legit?”

Class context — what myostatin blockade actually does in humans

FLGR-242 aims at the same target class as Deep Dive No. 01, and that class has a twenty-year controlled record: stamulumab, domagrozumab, ACE-031 (halted on bleeding signals), bimagrumab, trevogrumab. The pattern across all of it is lean-mass preservation or modest gains, with no proven functional (strength) benefit. In the COURAGE trial (n=975), even pharmaceutical-grade myostatin blockade on semaglutide did not build lean mass — every arm still lost it; the win was losing less of it.

Class adverse signals to hold alongside any follistatin pitch: bleeding / burst vessels, muscle cramps and spasms, LDL increases — and a theoretical concern about growth-pathway blockade and existing tumours.

Sources: Frontload FAQ · See Deep Dive No. 01 (Trevogrumab) for the COURAGE data in full

Claim check

Claims made for FLGR-242, checked against the research file:

• Needs the qualifier

  “FLGR-242 is follistatin.”

  It is an engineered follistatin-derived construct, per the seller's own description — not native follistatin.

• Not supported

  “It has a 19-day half-life.”

  Unverified. Seller-asserted, and unpublished in any species.

• Needs the qualifier

  “It's myostatin-selective, with no activin effects.”

  That is the design intent, per the seller. The binding has never been published.

• Not supported

  “0.5–1 lb of muscle per week.”

  A promoter claim. No human data exists for this molecule.

• Not supported

  “Myostatin inhibition builds large amounts of muscle in humans.”

  The class record shows preservation and modest effects, and no proven strength benefit.

• Not supported

  “The dramatic follistatin results in the literature apply.”

  Those were gene therapy — a virus producing follistatin in muscle — not injected protein.

• Not supported

  “It's being studied.”

  No registered trials, as of 28 Sep 2026.

• Checks out

  “Research chemical, not for human use.”

  That is literally how it is sold.

Open questions

Marked plainly, because they are open:

• What is actually in any given vial. No independent laboratory has analysed the product; identity, folding, glycosylation and bioactivity are unverified, and the construct's full sequence has never been published.
• Whether the binding and half-life data will ever be published. Every quantitative property — the ~19-day half-life, the <20 nM albumin affinity, the myostatin selectivity — currently rests on the seller's assertion alone.
• Any human cohort, with methods. None exists. A single case series with DEXA methods stated would move this product out of the testimonial tier; until then, the 10–12 lb figure stays where it started — with the inventor and the promoter.

Sources

1. Kalios — FLGR-242 compound page; registry check (28 Sep 2026): no record on PubMed or ClinicalTrials.gov. www.kalios.health/compounds/flgr-242.html
2. Frontload — “Is FLGR-242 Legit?” (gene-therapy distinction; class adverse signals via its FAQ). frontloadapp.com/blog/flgr-242
3. Jay Campbell — “Recombinant Follistatin: The Future Of True Muscle Growth,” including the published “0.5 to 1 pound of true skeletal mass per week” expectation and the seller-stated “normal use” regimens. jaycampbell.com/blog/recombinant-follistatin-the-future-of-true-muscle-growth/
4. Jay Campbell — “Follistatin vs Myostatin Inhibitors.” jaycampbell.com/blog/follistatin-vs-myostatin-inhibitors/
5. BioLongevity Labs — FLGR-242 product page (seller): construct description, albumin-binder sequence, price. biolongevitylabs.com/product/follistatin/
6. Tarzanoptimized — “Klotho & FLGR242 Explained” — recap of the Farber / Kruder talk: Farber named as inventor; the affiliate's self-reported 5.2 lb DEXA testimonial. tarzanoptimized.com/blog/klotho-flgr242-explained-mike-farber-trevor-kruder
7. Tarzanoptimized — “FLGR242 + Klotho + Cell Factors: The Anabolic Stack.” tarzanoptimized.com/blog/anabolic-stack-flgr242-klotho-cell-factors
8. LA Weekly — BioLongevity Labs founder piece on the new generation of muscle-sparing peptides helping GLP-1 users prevent muscle loss. www.laweekly.com/biolongevity-labs-founder-on-the-new-generation-of-muscle-sparing-peptides-helping-glp-1-users-prevent-muscle-loss/
9. Nakatani et al. 2008 — the FS I-I myostatin-selective follistatin derivative the construct's base is drawn from (as cited in the research file).
10. Dennis et al. 2002 — SA21, the albumin-binding peptide published by Genentech scientists, forming the tail of the seller-published binder sequence (as cited in the research file).
11. FLGR-242 (engineered follistatin) — profile & claim check, research report filed 26 Oct 2026. Every figure, attribution and hedge on this page follows that report.

Molecule

17β-hydroxyestra-4,9,11-trien-3-one

Class

19-nortestosterone · estrane (C18)

Human product

Parabolan (France) · 1980–1997

Human trials

None — controlled, of any kind

What it is — a cattle drug with a brief human chapter

Trenbolone is nandrolone with two additional double bonds: nandrolone already lacks the C19 angular methyl group (the “19-nor,” C18 estrane skeleton) and carries a Δ4 double bond; trenbolone adds Δ9 and Δ11, producing a conjugated 4,9,11-triene system. The Δ9-ene is credited with preventing aromatization; the triene raises androgen-receptor binding affinity, rigidifies the ring system, and makes the molecule not a substrate for 5α-reductase. In practice it is used only as C17β ester prodrugs — acetate, enanthate, hexahydrobenzylcarbonate — each hydrolysed to the same free 17β-trenbolone. US Schedule III / Canada Schedule IV.

• First synthesized in 1963 as a Roussel-Uclaf (France) development project; a primary paper trail exists by 1967–1975. Named inventors and the original patent number could not be verified from a primary source.
• The veterinary chapter is regulatory-grade. Trenbolone acetate was marketed in the early 1970s as the injectables Finajet and Finaject; in the late 1980s the injectable was discontinued and replaced by subcutaneous ear-implant pellets — Finaplix. The FDA approved trenbolone-acetate ear implants in 1987 (Finaplix-S for steers, Finaplix-H for heifers); the FDA Environmental Assessment describes an implant that “improves average daily gain and/or efficiency of feed conversion in feedlot cattle,” at 140 mg per steer and 200 mg per heifer, the ear being discarded at slaughter. Combination implants (trenbolone acetate + estradiol — Revalor, Component, Synovex Plus) followed from 1991. A vendor claim that trenbolone “was outlawed in the US” is false — FDA-approved implants remain marketed.
• The diversion history: in the 1980s–90s, Finaplix pellets were famously ground and solvent-extracted into injectable oil, or used transdermally with DMSO — the origin of trenbolone's underground supply before purpose-made underground-lab esters existed.

Parabolan — the one human product

The only trenbolone ester ever marketed for human medical use was trenbolone hexahydrobenzylcarbonate, brand names Parabolan and Hexabolan, made by Negma Laboratories, France — introduced in France in 1980 and voluntarily discontinued by its manufacturer in 1997. Presentation: a 1.5 mL ampoule of 76.5 mg of the ester, corresponding to 50 mg trenbolone base. The indications repeated in vendor and secondary sources (malnutrition / severe undernutrition, osteoporosis, cachexia) are repetition of a secondary claim: no original French label, regulatory record or primary clinical paper for Parabolan was retrieved, so the publishable formulation is “reported in secondary references to have been prescribed for wasting-type indications.” Why it ended could not be verified — the only sourced fact is voluntary discontinuation in 1997; vendor assertions that the reason was “commercial / regulatory, not safety” are unsourced and self-interested, and no safety-withdrawal notice was located. It was not, on the evidence found, a documented safety withdrawal — and that negative is itself unverified. No trenbolone product has ever been FDA-approved for human use, and none has received human approval anywhere since Parabolan.

Sources: FDA Environmental Assessment, trenbolone acetate implants (1987) · FDA FOI Summary, Revalor-S, NADA 140-897 (1991) · DailyMed, Revalor-200 label · Oklahoma State Extension, implants in beef cattle production

Esters and half-lives — the honest version

All common esters deliver the same 17β-trenbolone and differ only in the C17β ester and release rate:

The esters — and what each one actually is

Ester	Status
Acetate (C2)	Veterinary only — Finajet / Finaject injectable, then Finaplix / Revalor / Component implants.
Enanthate (C7)	Never a pharmaceutical product, human or veterinary. Its origin is underground: it first appeared commercially in 2004 as British Dragon's “Trenabol.” The romantic back-story (“synthesized by Roussel-Uclaf in 1967 and shelved”) appears only in underground retellings of Llewellyn's speculation and could not be verified from any primary source — folklore.
Hexahydrobenzylcarbonate	Human — Parabolan (Negma, France, 1980–1997).
Undecanoate	Listed in drug references; no marketed product history verified.

Half-lives: there are no human IM pharmacokinetic trials for any ester. This is the most misreported area of the whole trenbolone file. Circulating figures are secondary-source conventions and disagree with each other: one current reference table gives free trenbolone 48–72 h, acetate 1–2 days (also given as 3 days), enanthate 11 days, hexahydrobenzylcarbonate 8 days; popular / vendor consensus gives acetate ~3 days, enanthate 7–10 days, hexahydrobenzylcarbonate ~14 days — and a single vendor page gives the hex ester as both ~14 days and ~8 days in different tables. The “14-day Parabolan” figure may conflate duration of activity / dosing interval with a measured half-life; no retrieved source distinguishes them.

The only human pharmacokinetic data located are from a single-subject tracer study (Spranger & Metzler 1991): one human subject ingested tracer-labelled unesterified 17β-trenbolone at 0.04 mg/kg — a food-residue study design, not a therapy trial; 54% of administered radioactivity appeared in urine by 26 h and 63% by 72 h, and the authors concluded human disposition differs significantly from rats. That gives urinary-excretion timing for oral unesterified trenbolone in one person. It does not give an IM ester half-life, a plasma half-life, or steady-state data. Doping-control detection windows (urinary metabolites detectable for weeks) are detection, not half-life, and must not be conflated. The honest formulation: quote a range, name it as secondary-source convention rather than measured human pharmacokinetics, and flag the hex-ester discrepancy (8 vs 14 days) explicitly.

Source: Spranger & Metzler, J Chromatogr B 1991 (n=1 tracer study)

Pharmacology — what the assays actually say

Androgen-receptor binding — ~3× testosterone, DHT-like, NOT a clean “5×”

• Yarrow, McCoy & Borst 2010 (Steroids 75(6):377–389) — the anchor review — states 17β-trenbolone “binds to androgen receptors (ARs) with approximately three times the affinity of testosterone.”
• Bauer et al. 2000: in [³H]-DHT displacement from recombinant human AR, 17β-trenbolone's affinity was similar to DHT. Its affinity for human SHBG was much lower than DHT's — a high free fraction is a reasonable inference.
• Wilson et al. 2002 (US EPA group): trenbolone “induced AR-dependent gene expression with a potency equal to or greater than dihydrotestosterone,” with AR nuclear translocation at concentrations as low as 1 pM.
• Versus nandrolone: no single head-to-head human-AR binding study with a clean trenbolone : nandrolone ratio was retrieved. Both are 19-nors; nandrolone is the weaker AR agonist, aromatizes weakly, and is 5α-reduced to a weaker metabolite; trenbolone is the more potent AR agonist and is not 5α-reduced. A quantified “tren = X × nandrolone” figure could not be verified — do not publish one.
• The “5× testosterone / 500:500” claim is a debunk. “Five times testosterone” binding and the anabolic : androgenic ratio 500:500 (testosterone = 100:100) appear in every vendor profile, trace to Llewellyn's Anabolics, and contradict the primary assays: Yarrow says ~3× affinity; Bauer / Wilson say DHT-like. A secondary audit found the exact 500:500 figures in none of the primary literature it could reach; Wilson 2002's actual Hershberger finding was qualitative — trenbolone “was as potent as testosterone propionate in tissues that lack 5α-reductase but less effective at increasing weight of tissues with this enzyme.” The 500:500 numbers purport to be rodent Hershberger-assay ratios; presented as human potency data, they are folklore.

Aromatization — precisely zero for the parent compound, with two caveats

Trenbolone itself is not aromatized to estradiol. It is not an aromatase substrate — the anchor review uses testosterone → estradiol via aromatase as the explicit contrast case, and secondary sources credit the 19-nor triene system, the Δ9-ene specifically, with inhibiting aromatization. Two caveats. First, low estrogenic activity is not aromatization: in-vitro bioassays find 17β-trenbolone and its metabolites have very low estrogen-receptor binding and low estrogenic efficacy (~20% of estradiol's efficacy in those assays) — weak direct ER activity, not conversion to estradiol. Second, in any combined regimen, estradiol comes from the other compounds — and trenbolone's suppression of endogenous testosterone can lower endogenous estradiol when it is the only androgen present. Do not generalise “19-nor = no estrogen”: nandrolone aromatizes weakly, and trestolone (MENT) aromatizes to a potent methyl-estradiol. Among the true 19-nors, trenbolone is the non-aromatizable one.

5α-reduction — it does not happen the testosterone way

Yarrow 2010, in substance: 17β-trenbolone “is metabolized to less potent androgens in vivo; while testosterone undergoes tissue-specific biotransformation to more potent steroids, dihydrotestosterone and 17β-estradiol” — and trenbolone “appears to induce less growth in androgen-sensitive organs which highly express the 5α-reductase enzyme (e.g., prostate).” Yarrow et al. 2011 calls trenbolone explicitly a “non-5α-reducible androgen.” Consequences: there is no DHT-like amplification in prostate / skin / scalp, and finasteride / dutasteride have no trenbolone metabolite to block — that last clause is an inference from the primary premise, and is labelled inference. Actual major metabolism (human tracer study plus bovine / rodent literature) is 17-oxidation / epimerization — 17β-trenbolone ↔ trendione → 17α-trenbolone (epitrenbolone) — plus hydroxylated polar metabolites, excreted as glucuronides / sulfates; the metabolites retain <5% of the parent's AR affinity (Bauer 2000). Contrast with nandrolone, often garbled: nandrolone is 5α-reduced, to dihydronandrolone (DHN), a weaker AR agonist — which is why nandrolone is relatively prostate-sparing. Trenbolone shows a superficially similar tissue pattern by not being reduced at all, remaining at full parent potency in every tissue. Same Hershberger appearance, opposite mechanism. Older profile copy claiming trenbolone is “metabolized by 5α-reductase into DHT” is wrong and contradicts Yarrow / Wilson.

Progesterone receptor — strong binding, weak functional potency

This is where community lore — “binds the progesterone receptor harder than progesterone, therefore prolactin gyno” — outruns the data by two steps. Binding (in vitro): Bauer et al. 2000 found 17β-trenbolone showed “a slightly higher affinity to the bovine progestin receptor (bPR) than progesterone” in displacement from bovine uterine PR; metabolites were ~1–2% / <5% of that affinity. (Vendor profiles repeat a different figure — “~60% of progesterone's strength” — from different assays; the two numbers come from different assay systems and must not be conflated.) Functional activation (yeast bioassays): Death et al. 2004 put trenbolone's PR ED50 at ~40 nM vs progesterone's 4.5 nM — roughly tenfold more compound needed — and McRobb et al. 2008 put its relative PR potency near that of testosterone, i.e. very low; Neumann 1976 described trenbolone as non-progestogenic or only weakly progestogenic. What can be claimed: trenbolone is a progestogen by class and a high-affinity PR ligand in vitro (bovine PR). What cannot: that binding equals agonist efficacy in humans. No study retrieved measured PR activation, prolactin, or gynecomastia incidence in human trenbolone users — adjudicated in full under Prolactin and gynecomastia, below.

Glucocorticoid receptor — real signal, contested mechanism

Three primary findings that do not agree on mechanism — all three belong in an honest account:

• Wilson et al. 2002: trenbolone “displayed antiglucocorticoid activity in vitro, inhibiting dexamethasone-induced transcriptional activity, and reduced adrenal gland size in vivo.”
• Danhaive & Rousseau 1986/1988 (rat): the authors suggested anabolic steroids “can act via muscle glucocorticoid receptors, thereby antagonizing the catabolic activity of endogenous glucocorticoids, rather than via muscle androgen receptors.”
• Sharpe, Buttery & Haynes 1986 (sheep) — the direct contradiction on binding: skeletal-muscle glucocorticoid receptors had “virtually no affinity for trenbolone,” but trenbolone-acetate implantation reduced muscle GR binding capacity for cortisol and dexamethasone within 2 days — i.e. receptor down-regulation, not direct competitive binding.

Synthesis: anti-glucocorticoid / anti-catabolic activity is supported in animals and in vitro (functional antagonism, GR down-regulation, reduced protein degradation). The popular claim “trenbolone binds and blocks the glucocorticoid receptor like RU486” is not supported as a binding claim — the sheep data show virtually no direct GR affinity — and vendor copy stating trenbolone “binds the glucocorticoid receptor with high affinity” is contradicted by Sharpe 1986. Yarrow 2010's own wording stays hedged: trenbolone “may also exert anabolic effects by … inhibiting the action of glucocorticoids.” Do not upgrade the hedge.

IGF-1 and satellite cells — bovine cell culture, not human muscle

All primary evidence here is bovine or rodent. Kamanga-Sollo et al. 2008: a combined estradiol / trenbolone-acetate implant increased muscle IGF-I mRNA in steers, and estradiol and trenbolone each increased IGF-I mRNA in bovine satellite cell cultures in 10% fetal bovine serum — but in 1% IGFBP-3-free swine serum, the same treatments increased satellite-cell proliferation with no effect on IGF-I mRNA, and the authors concluded increased IGF-I expression “may not be responsible” for the proliferation; their conclusion, in substance, is that “IGF-I plays a role” in the proliferation “even in the absence of increased IGF-I expression.” Thornton et al. 2016 (fused bovine satellite-cell cultures) found 10 nM trenbolone acetate increased protein synthesis rate and decreased protein degradation rate vs control. Dalbo et al. 2016/2017 (rodents, in vivo) found testosterone and trenbolone enanthate increased skeletal-muscle hypertrophy and satellite cell numbers while also increasing mature myostatin protein expression — a complicating, not purely anabolic, finding. What this does not show: systemic IGF-1 elevation in humans, IGF-1 elevation in human muscle, or satellite-cell proliferation in human biopsies on trenbolone. No human IGF-1 or satellite-cell study was located. Vendor claims that “tren stimulates IGF-1 production in muscle,” without the words bovine, in culture, and inconsistently at the mRNA level, are distortions of Kamanga-Sollo.

Sources: Yarrow et al., Steroids 2010 · Yarrow et al., Am J Physiol Endocrinol Metab 2011 · Bauer et al., APMIS 2000 · Wilson et al., Toxicol Sci 2002 · Sharpe et al., Br J Nutr 1986 · Kamanga-Sollo et al., Domest Anim Endocrinol 2008

The human-data state

No randomized controlled trial, no controlled dose-response study, and no prospective safety trial in humans was located. Both anchor papers frame human use as a future question: Yarrow 2010 proposes trenbolone as a SARM-like candidate for future research into muscle / bone wasting, obesity, and androgen insensitivity — in 2010, human efficacy was a hypothesis, not a finding. The structured case-report review below states the same negative in its conclusion: “The absence of clinical studies on its safety, along with widespread unregulated use, hinders risk assessment.” Parabolan-era French use (1980–1997) was pre-modern-trial-era marketed use — no Parabolan RCT, cohort, or published case series was retrieved, and it must not be described as “clinically studied” without that qualifier. What human data exist instead, in descending tier order:

• 1 · Single-subject metabolism / tracer study (Spranger & Metzler 1991), plus human urinary-metabolite analytical studies written for doping control.
• 2 · In-vitro human-tissue data: Bauer 2000 (recombinant human AR, human SHBG); Wilson 2002 (human-AR-transfected cells). Human receptor data exist; human organism data do not.
• 3 · Case reports: a structured review (Viana et al. 2025) searching PubMed / EMBASE / CENTRAL to 2024 found 23 case reports meeting inclusion (human, documented trenbolone use, adverse effect). Most frequent systems: cardiovascular (heart failure, MI, ischemic stroke), hepatic (cholestatic injury, hepatitis), renal (nephropathy, AKI), neuropsychiatric (psychosis, aggression, delirium). Most cases were young males using unregulated-market product, with toxicological confirmation in only a minority — causality is weak and polypharmacy confounding pervasive.
• 4 · Surveys and cohorts of AAS users generally — trenbolone is one compound among stacks; no trenbolone-isolated outcomes. Class data borrowed for trenbolone must be labelled class inference.
• 5 · Environmental / occupational literature: a large primary literature treats 17β-trenbolone as a feedlot-effluent endocrine disruptor in fish and rodents. It is real trenbolone pharmacology; its doses and models do not transfer to human exposure.

The anchor review deserves its own precision. Yarrow, McCoy & Borst 2010 is a narrative review from the VA Gainesville group whose core thesis is trenbolone as a SARM-like compound — a potent AR agonist (~3× testosterone's binding affinity) shown to augment skeletal muscle mass and bone growth and reduce adiposity “in a variety of mammalian species,” with reduced androgenic / estrogenic activity because its metabolites are less potent than the parent (the opposite of testosterone's amplification to DHT and estradiol) and because it is neither aromatized nor 5α-amplified. It is a synthesis of animal and in-vitro work — not a human trial — and its optimism is explicitly prospective (“a basis for future research evaluating its safety and efficacy … in humans”). Its companion rodent study (Yarrow et al. 2011, rats, 29 days) found levator ani / bulbocavernosus mass +35–40% above shams on trenbolone, dose-dependent partial protection against orchiectomy-induced bone-density loss and visceral fat accumulation — and, at the highest dose, prostate mass +68% and raised hemoglobin: tissue selectivity is dose-limited, not absolute, and the erythropoietic effect is dose-dependent, not absent. The authors' conclusion supports “future preclinical studies” — again, explicitly not a human finding.

Efficacy claims, adjudicated

Each claim is tiered on what actually supports it. The recurring pattern: real animal data, zero human controlled data, and a community reputation built on top.

Each headline claim, with its evidence tier and verdict

Claim	Best evidence	Verdict
“It builds muscle / increases lean mass”	Animal: Yarrow 2011 rats, levator ani / bulbocavernosus +35–40% over 29 days. Donner et al. 2015 (normogonadic male rats, 2 mg/kg/day, 6 weeks): lean mass +11±4% on DXA — with sex-hormone suppression and prostate enlargement (+49%) confirmed as adverse effects in the same study. Cattle: average-daily-gain and feed-conversion improvements are the regulatory basis of the implants. Bovine satellite cells: increased protein synthesis, decreased protein degradation.	Supported in animals; unverified in humans. There is no human muscle, strength, or body-composition endpoint for trenbolone in the controlled literature.
“It preserves muscle in a calorie deficit” (the cutting claim)	Mechanistic, animal: functional glucocorticoid antagonism and GR down-regulation; reduced protein degradation in bovine cell culture; fat-mass protection in orchiectomized rats. In Donner 2015, trenbolone-treated rats lost fat while gaining lean mass without an imposed deficit — supportive of partitioning, not a deficit study.	Plausible mechanism, animal-supported; human claim unverified. No human deficit study, no nitrogen-balance study, nothing — the human form is entirely community anecdote.
“Recomp” — simultaneous fat loss and muscle gain	Animal: the one headline claim with a direct animal demonstration. Donner 2015 — over 6 weeks at essentially unchanged total body mass, control rats gained fat (+34±7%) while trenbolone rats lost 37±6% of fat mass and gained 11±4% lean mass. Yarrow 2011 similarly showed visceral-fat protection alongside muscle growth in rats.	Demonstrated in rats; in humans, anecdote only. Any specific human magnitude (“X lb of muscle while losing Y lb of fat”) is folklore.
Strength	Muscle mass and protein-turnover data exist in animals; direct strength endpoints are sparse even there.	Unverified in humans. Strength reports are inseparable from concurrent testosterone and other compounds, training, and expectancy.
“Nutrient partitioning” / feed efficiency	This claim is the veterinary original — FDA-approved implants improve feed conversion efficiency in feedlot cattle. In that species, at those doses, “partitioning” is a measured regulatory endpoint.	Supported in cattle; folklore in humans. The phrase migrated from the cattle literature into bodybuilding with no human measurement behind it.
Cosmetic claims — “dry, hard, grainy,” “tren face,” delt / trap growth	No human study has measured trenbolone's effect on body composition, water, vascularity, fat distribution, or facial appearance. Skin, at low quality: a 2024 review of trenbolone harms summarizes user / case literature reporting severe acne in roughly a third of users, plus gynecomastia and increased body hair — polypharmacy-confounded.	Community anecdote / folklore as measured claims.

Sources: Yarrow et al. 2011 · Donner et al., Steroids 2015 · FDA implant approvals (see History) · Thornton et al. 2016

The harm file

The framing that controls everything below: the strongest human harm data here are AAS-class and prospective (the HAARLEM cohort) — real, quantified, and not trenbolone-isolated. Trenbolone-specific human evidence is case reports, one survey, and qualitative interviews. Trenbolone-specific preclinical evidence is real but low-directness. Each item below states which of those three it is standing on.

HDL / lipid suppression — class-established, trenbolone magnitude unquantified

• [AAS-class] HAARLEM (Smit et al., Andrologia 2022): prospective cohort, 100 men on self-administered androgen cycles, measured before, at end of cycle, 3 months after, and 1 year after start. End-of-cycle: HDL-C −0.40 mmol/L (≈ −15.5 mg/dL), LDL-C +0.45 mmol/L, ApoB +18.2 mg/dL, ApoA −36.6 mg/dL, Lp(a) −37.6%. All measures returned to baseline at 3 months / 1 year. Cycles were user-chosen stacks — not trenbolone-isolable.
• [AAS-class mechanism]: AAS raise hepatic lipase activity, converting HDL2 to smaller, rapidly cleared HDL3; the extreme HDL reductions in the class literature (39–70%) are dominated by oral 17α-alkylated compounds — e.g. the stanozolol experiment in which hepatic triglyceride lipase rose +230% by day 3 and HDL-C fell 39% (HDL2 −71%) by day 7. That is stanozolol data, and it is the usual origin of “70% HDL suppression” claims misquoted as trenbolone data.
• [Trenbolone-specific, animal — and it cuts the other way]: Donner 2015 rats: triglycerides, HDL and LDL all fell (HDL −57%, LDL −78%). Rat lipoprotein metabolism is HDL-dominant and lacks CETP; it does not translate to human HDL / LDL risk in either direction.
• Verdict: HDL suppression on trenbolone-containing cycles is established at AAS-class level and consistent for trenbolone; the magnitude for trenbolone alone is unquantified. Any trenbolone-specific figure (“suppresses HDL 30–50%,” “worse per mg than testosterone”) appears only on vendor sites — unverifiable.

Cardiovascular — LVH, blood pressure, hematocrit

• LVH / cardiac function [AAS-class only]: HAARLEM echocardiography (n=31, 3D echo before, at end of cycle — median 16 weeks — and ~1 year after): LVEF −4.9%, LV mass +28.3 g, left atrial volume +9.2 mL; the LV-mass increase correlated with average weekly AAS dose, and all parameters returned to baseline at follow-up. Baggish et al. (Circulation 2017; 140 male weightlifters aged 34–54): LVEF 52±11% in ≥2-year AAS users vs 63±8% in non-users, current users worse than past users, and coronary plaque volume higher in users and associated with cumulative duration. A 2025 meta-analysis (35 studies, 2,000 men) found LVEF −2.25%, septal thickness +1.24 mm, LV mass +13.29 g vs non-users — cross-sectional dominance means confounding by training, body mass, duration, and polypharmacy. Counterpoint, animal: Donner 2015 and Donner et al. 2016 found no adverse cardiac effect in rats under their protocols — which does not exonerate human use, and also shows “trenbolone is uniquely cardiotoxic in rodents” is not demonstrated either. Verdict on “tren causes LVH”: LVH and impaired function are a real AAS-class human phenotype — dose / duration-associated, largely reversible after single short cycles in prospective data, less clearly reversible after years of use. No study isolates trenbolone as the agent.
• Blood pressure [AAS-class]: HAARLEM — mean +6.87 mmHg systolic / +3.17 diastolic on cycle, reversible at 3 months; a HAARLEM review notes 41% of participants were hypertensive (>140/90) during cycle. Any trenbolone-specific BP figure (“10–25 mmHg on tren”) is vendor-only — could not be verified.
• Hematocrit / erythrocytosis [class + animal]: HAARLEM — mean hematocrit +0.03 L/L on cycle, reversible. Yarrow 2011 (rats): low trenbolone doses held hemoglobin at sham levels; high doses raised it — dose-dependent, not absent.

Kidneys — the folk claim vs the evidence

The folk claim (“tren destroys kidneys,” inferred from dark / rust-coloured urine) is flagged as unproven even by low-grade secondary sources: WikiDoc states kidney toxicity “has not yet been proven,” attributing the belief to rust-coloured oxidised trenbolone metabolites excreted in urine being mistaken for blood. Honestly: that explanation is plausible and widely repeated, but no primary study measuring urine colour vs renal injury in trenbolone users was found — and absence of evidence is not proof of renal safety either. The real AAS kidney literature is FSGS in massive bodybuilders, not trenbolone [AAS-class]: Herlitz et al. 2010 — 10 bodybuilders, long-term AAS use (compounds not trenbolone-isolated), mean proteinuria 10.1 g/day, mean creatinine 3.0 mg/dL; biopsy showed FSGS in 9 and glomerulomegaly; one progressed to ESRD; seven who stopped AAS stabilised or improved; one who resumed relapsed. The authors' hypothesis is secondary FSGS from post-adaptive glomerular hyperfiltration driven by extreme lean mass, plus possible direct podocyte / androgen toxicity — the relative weights are unproven. Trenbolone-attributed renal cases are single cases: Lizardo et al. 2026 — a patient reporting trenbolone use with fatigue, hypercalcemia 15 mg/dL and AKI (creatinine 6.02 mg/dL); malignancy, hyperparathyroidism and vitamin-D toxicity ruled out; improved with fluids and pamidronate; mechanism proposed but not established; no biopsy; product unverified. A second pathway is indirect — severe AAS cholestasis can cause bile-cast nephropathy; in the published plasma-exchange case the regimen included trenbolone among other ingredients and the authors could not separate it from co-ingredients. A vendor-asserted “2025 case requiring simultaneous liver + kidney transplantation after 7 weeks of trenbolone enanthate” was found only on a steroid-sales page, with no PubMed / journal record — do not cite it as a case. Verdict: direct trenbolone nephrotoxicity is unproven. AAS-era kidney harm is real but runs through the extreme-mass hyperfiltration / FSGS phenotype and secondary routes — none trenbolone-isolated.

Liver — the non-17α-alkylated context

Chemistry first: trenbolone esters are 17β-esterified, not 17α-alkylated. The classic AAS cholestasis / peliosis / tumour literature belongs overwhelmingly to 17α-alkylated orals. Llewellyn's Anabolics summary: “Trenbolone is not c-17 alpha alkylated, and is generally not considered a hepatotoxic steroid; liver toxicity is unlikely… [but] has a strong level of resistance to hepatic breakdown, and significant liver toxicity has been noted in bodybuilders abusing trenbolone… cannot be completely excluded, especially with high doses.” Documented cases are rare, idiosyncratic, cholestatic, real [case reports]: Anand et al. 2006 — a 21-year-old bodybuilder who reported overdosing on “Parabolan”; pruritus and jaundice with cholestasis, bilirubin up to 65.5 mg/dL; treated with 5 MARS (albumin dialysis) cycles; bilirubin 2 mg/dL at 2-month follow-up. (Source inconsistency, flagged rather than smoothed over: the abstract calls the drug “trenbolone acetate” while the trade name Parabolan denotes the hexahydrobenzylcarbonate ester, and a later MARS review describes the same case as hexahydrobenzylcarbonate.) Boks et al. 2017 — a young amateur bodybuilder with jaundice and appetite loss; work-up including liver biopsy made trenbolone enanthate the likely cause of cholestatic hepatitis; the authors' point was that cholestatic hepatitis “may not be limited to oral AAS” — with the key limitation that the injected product could not be analysed to confirm it contained only trenbolone. Counterpoint, animal: Donner 2015 rats — liver enzymes and histology showed no adverse hepatic effect; small, short, rodent — it bounds but cannot exclude rare human idiosyncratic injury. Verdict: expected hepatotoxicity is low, categorically below 17α-alkylated orals; documented harm is rare idiosyncratic cholestatic hepatitis, case-report grade only, incidence unknowable, product adulteration / co-use never excluded. Two cautions: do not conflate trenbolone with metribolone (17α-methyltrenbolone), a different, genuinely hepatotoxic research compound; and in lifters, ALT / AST are muscle-confounded — GGT, bilirubin and ALP are the cholestasis signals in these cases.

Sleep — a class effect with a trenbolone reputation

Trenbolone-specific evidence is essentially none above qualitative: in the interview research below, insomnia was a recurring participant-attributed theme for trenbolone, and a 2021 forensic review lists insomnia among short-term trenbolone effects as a secondary listing with no primary sleep measurement. No trenbolone polysomnography, actigraphy, or trenbolone-only sleep cohort was found. Vendor statistics (“55–65% of tren users report moderate-to-severe insomnia”; “22% REM reduction on polysomnography”) appear on sales sites with no traceable study — unverifiable. The best available evidence is [AAS-class]: Klonteig, Scarth & Bjørnebekk 2024 — Norwegian men, long-term AAS users (n=68) vs non-using weightlifting controls (n=58). 66% of the AAS group reported sleep problems as a side effect; 38% had used sleep medication. Global PSQI 6.85 vs 4.34; PSQI >5 in 72% vs 34%. And sleep was significantly worse during withdrawal than during use — an important complication for any simple “on-tren insomnia” timeline; the authors interpret hormone deviation in both directions as the driver, partly mediated by psychological distress. Night sweats: consistently reported in user qualitative / anecdotal literature and listed alongside insomnia in reviews; no primary study measuring trenbolone, core temperature, or sweating was found — proposed mechanisms are anecdote / folklore. Verdict: sleep disruption in AAS users is substantial and class-supported; that trenbolone is worse than other AAS for sleep is qualitative reputation only.

Mood and aggression — evidence quality

• [AAS-class RCT meta-analysis sets the ceiling]: Chegeni et al. 2021 — 12 RCTs, 562 healthy males, AAS vs placebo. Self-reported aggression rose slightly (Hedges' g = 0.171) after excluding one outlier; the effect was not replicated for observer-reported aggression, nor when restricted to higher-dose or longer administration. The compounds were predominantly testosterone esters — there are no trenbolone RCTs.
• [Trenbolone-specific, survey]: Piatkowski et al. 2024 — online survey, 282 males using AAS (93 current trenbolone, 156 past, 33 never). After controlling for age and BMI, trenbolone dose was associated with verbal aggression only (p = 0.045); no significant association for physical aggression, anger, hostility, or psychological distress. Borderline p, multiple outcomes, self-selected, cross-sectional, and near-universal polypharmacy (testosterone in 97.9% of the sample).
• [Trenbolone-specific, qualitative]: Piatkowski, Neumann & Dunn 2023 — semi-structured interviews with 16 Australian strength athletes, only 50% of whom had used trenbolone. Trenbolone was perceived as the most harmful AAS; themes included aggression, violent behaviour, insomnia, impulsivity problems, cardiovascular concerns. This is reputation / perception data from 8 trenbolone-experienced interviewees — valuable for lived experience, zero incidence or causal weight.
• Confounders pervading all of it: expectancy (“tren rage” is culturally scripted), sleep loss, caloric deficit (most trenbolone use occurs in cutting phases), stimulant / other-drug co-use, pre-existing traits selecting into trenbolone use, and withdrawal-phase effects. The proposed serotonin mechanism is animal nandrolone data; “trenbolone decreases serotonin neurotransmission in humans” has never been measured in trenbolone users.
• Verdict: a small self-reported aggression signal exists for supraphysiologic AAS in RCTs [class]. The trenbolone-specific signal is one borderline verbal-aggression association plus a consistent qualitative reputation. “Tren rage” as a deterministic pharmacological syndrome is not established.

Prolactin and gynecomastia — adjudicating the community story

The community story runs: trenbolone binds the progesterone receptor → raises prolactin → causes gynecomastia and sexual dysfunction → requires a dopamine agonist. Traced link by link: Link 1 — PR binding: real, in vitro, bovine (Bauer 2000, §Pharmacology). Link 2 — PR activation: weak — functional progestogenic potency near testosterone's in yeast bioassays; binding affinity is not agonist efficacy. Link 3 — prolactin: no human data. No human study measuring prolactin in trenbolone users was found; the only direct trenbolone prolactin measurement located in secondary curation was in castrated sheep with trenbolone-acetate implants, prolactin unaltered — and even that primary full text was not retrieved, so it stands as unverified secondary. Vendor statements presenting trenbolone-prolactin elevation as measured fact have no primary citation behind them. Link 4 — gynecomastia: not described for trenbolone in the literature — the Yarrow correspondence states trenbolone gynecomastia “is not described in the literature” and an association “would be worthy of publication.” Meanwhile gynecomastia is documented at class level: in the HAARLEM cohort (testosterone-based stacks), observed gynecomastia ran ~19%, and prevalence returned to baseline by 1-year follow-up. The sophisticated forum position — that most “prolactin gyno” on testosterone + 19-nor regimens is actually estradiol-driven from the testosterone — is consistent with the evidence state, since estradiol is a measured, established gynecomastia driver and trenbolone's prolactin effect is unmeasured. Community ancillary practice, as actually reported [community anecdote]: vitamin B6 (P5P) is the reported first line for 19-nor prolactin concerns, and cabergoline is reported when bloodwork / symptoms are invoked, with “do not run caber preventatively” the modern stated forum norm — descriptions of community practice, not evidence and not recommendations. Verdict: PR binding — supported [in vitro]. Functional progestogenic potency — weak / contested. Trenbolone raises prolactin in humans — unverified; no human measurement exists. “Tren gyno is prolactin gyno” — folklore as a mechanism; gynecomastia on trenbolone-containing regimens is real at class level and is most parsimoniously attributed to the aromatizable components until trenbolone-specific data exist.

Sexual function

Established components: exogenous androgens suppress LH / FSH and endogenous testosterone [AAS-class; HAARLEM showed gonadal disruption with recovery in the majority after cessation; Donner's rats confirmed sex-hormone suppression with trenbolone]. Trenbolone provides no aromatization to estradiol, so sexual function on a trenbolone-containing regimen depends heavily on what aromatizable androgen is or isn't co-present, and on post-cycle hypogonadism — low estradiol itself impairs male libido and erectile function in human experimental literature, which is class physiology, not trenbolone data. User reports are bidirectional [community anecdote]: short-term increased libido is listed in review sources, while low libido and erectile dysfunction (“tren dick”) are staples of qualitative / community reports — consistent with phase-dependence (on-cycle androgen load vs low-estradiol states vs post-cycle suppression), none of it quantified in a trenbolone cohort. The prolactin route specifically: unproven (§above). The PR route: binding real, clinical relevance unresolved. Verdict: sexual-dysfunction risk via axis suppression plus non-aromatization is mechanistically sound and class-supported; via prolactin, unproven; any incidence figure, unverifiable.

Hair loss

No trenbolone-specific human alopecia study, incidence figure, or scalp study was found — everything here is class extrapolation. What is solid pharmacology: androgenic alopecia requires genetic follicle sensitivity plus AR signalling in scalp; trenbolone (a) binds human AR at DHT-like affinity in vitro and (b) is not a 5α-reductase substrate, so its androgenicity is intrinsic — a 5α-reductase inhibitor, whose action is blocking testosterone → DHT conversion, has no substrate-level route to block trenbolone itself. That inference is sound chemistry; it is still an inference about scalp outcomes. A correction of a common error: trenbolone is a 19-nortestosterone (estrane) derivative, not a “DHT-derived” steroid — vendor pages that classify it as DHT-derived and then reason about hair from that classification are chemically wrong. And because trenbolone is not 5α-reduced and not DHT-derived yet is itself a potent direct AR agonist, “no DHT conversion = hair-safe” is an unsupported inference. Verdict: “trenbolone can accelerate male-pattern hair loss in predisposed men” — plausible, class-consistent, trenbolone-unmeasured. Any percentage risk, timeline, or “worse than X for hair” ranking — unverifiable.

“Tren cough” — mechanism evidence

Headline finding: no peer-reviewed study, case series, or trenbolone-specific case report of “tren cough” could be located. Every mechanistic statement in circulation traces to vendor / community secondary content. The proposed mechanisms, graded: (1) Pulmonary oil microembolism (POME) — the only arm with real primary literature, and it is not trenbolone literature. The FDA label for Aveed (testosterone undecanoate) describes serious POME reactions — cough, urge to cough, dyspnea, sweating, throat tightening, chest pain, dizziness, syncope — during or immediately after IM injection, most lasting minutes; trial experience in labelling summaries: 9 POME events in 8 patients among 3,556 treated. A published case (Russell, Storck & Ainslie 2011): a 21-year-old bodybuilder after accidental intravascular injection of an oil solution of testosterone enanthate + boldenone undecylenate (not trenbolone) — transient dyspnea within 1 minute, bilateral diffuse opacities and ground-glass on CT, hypoxemia, improving over 48 h. That proves oil-based androgen injections can produce immediate cough / dyspnea and, rarely, lung injury — but the documented severe case is slower and more serious than the typical self-limited seconds-to-minutes “tren cough” description, and involved different compounds. (2) Prostaglandin / bradykinin bronchoconstriction — repeated verbatim across vendor sites; no primary source could be verified: no study measuring prostaglandins or bradykinin after trenbolone injection was found. Folklore — plausible physiology, zero direct evidence. (3) Solvent / vehicle reaction — a genuine differential for immediate post-injection cough / flushing with depot oils; again, not trenbolone-tested. Folklore checks: “tren cough is unique to trenbolone” — false (POME cough occurs with testosterone undecanoate and other oil injectables). “Acetate causes it more than enanthate” — unanimous community claim, no comparative study found; faster vascular uptake of a short ester is a rationalization, not data. “It is harmless” — usually brief in anecdote, but the POME literature includes hypoxemia and hospitalization. Verdict: the phenomenon is widely and consistently self-reported [community anecdote]; the mechanism is unresolved — POME is the best-anchored analogy, the prostaglandin / bradykinin story is unverified repetition, and trenbolone-specific evidence is absent.

Sources: Smit et al., HAARLEM, Andrologia 2022 & Front Reprod Health 2021 · Baggish et al., Circulation 2017 · Herlitz et al., J Am Soc Nephrol 2010 · Anand et al. 2006 · Boks et al. 2017 · Klonteig et al., BMC Psychiatry 2024 · Piatkowski et al. 2023 & 2024 · Russell et al., Can Respir J 2011 · FDA, Aveed approval package

The assay trap — trenbolone and estradiol immunoassays

The claim: trenbolone (or a metabolite) cross-reacts on standard estradiol immunoassays, producing falsely elevated E2 readings that provoke unnecessary aromatase-inhibitor use. It grades in layers, because the layers have very different strengths.

What is solid

• Trenbolone does not aromatize, so trenbolone itself should not raise true estradiol; true E2 on a trenbolone-containing regimen comes from co-administered aromatizable compounds and endogenous production.
• LC-MS/MS (“sensitive” / “ultrasensitive”) estradiol is not subject to antibody cross-reactivity — mass spectrometry physically separates molecules — and LabCorp itself directs men to LC-MS/MS over ECLIA as “the more appropriate choice” for these clinical situations. The standard ECLIA estradiol test is a poor test in men anyway, trenbolone or not — an important confounder for every anecdote below.
• The mechanism class is real for this assay family. Fulvestrant — a steroidal estradiol analog — cross-reacts on the Roche Elecsys Estradiol II and III assays: Roche's Field Safety Notice SBN-CPS-2016-005 documents serum spiked with 25,000 pg/mL fulvestrant reading apparent E2 increases of ~19–20 pg/mL (~0.076–0.079% cross-reactivity), and Roche instructs that the assay not be used to monitor estradiol in fulvestrant-treated patients, directing users to LC-MS instead; the FDA logged it as a Class 2 device correction. Abbott's ARCHITECT estradiol showed 0.23% fulvestrant cross-reactivity in its own 2016 field notice. Structural-analog cross-reactivity on estradiol immunoassays is therefore a regulator-acknowledged phenomenon — for fulvestrant.

What the manufacturers actually publish — trenbolone is absent

The Roche Elecsys Estradiol III method sheet (V6.0, 2020-05) publishes a cross-reactivity table of ~30 substances — the highest being 6α-hydroxyestradiol (74.1%) and estrone (0.757%), with testosterone, 5α-DHT, mesterolone and DHEA-S all “not detectable.” Trenbolone, 17α-trenbolone, trendione and nandrolone do not appear in the table at all. The sheet carries two relevant general warnings: “Due to the risk of cross reactivity, this assay should not be used when monitoring Estradiol levels in patients being treated with Fulvestrant,” and “Steroid drugs may interfere with this test.” Interpretation: Roche never spiked trenbolone-class compounds — unsurprising for a non-human drug. Absence from the list means untested, not tested-and-clean. The Beckman Coulter and Abbott estradiol tables likewise contain no trenbolone row, and Krasowski et al. 2014, experimentally testing cross-reactivity on the Roche Elecsys Estradiol II assay, concluded clinically significant cross-reactivity was “unlikely” for the compounds tested — trenbolone was not among the compounds tested. The only actual spiking numbers for trenbolone in an estradiol assay are veterinary, and they point in both directions: a commercial bovine estradiol ELISA lists trenbolone cross-reactivity of 1.0%; a different bovine anti-E2 antiserum used in heifer research listed trenbolone acetate 0.01%, trenbolone-17α 0.04%, trenbolone-17β 0.04%. None of these is a Roche clinical number.

The community evidence — consistent, old, and uncontrolled

The phenomenon claim originates in a 2013 eroids thread and has been repeated, always naming the same platform, ever since. Reported LabCorp / Roche ECLIA values in that thread: E2 481 pg/mL on trenbolone enanthate 200 mg/week; a collated range of 170–1,000 pg/mL on testosterone 100–500 mg/week plus trenbolone 200–700 mg/week, often barely moving despite high-dose aromatase inhibitors; E2 234 (later recounted in the same thread as 243 — internal inconsistency, no lab image) on trenbolone enanthate 200 mg/week with no testosterone at all, one week in. A counter-example at Quest: testosterone propionate + trenbolone acetate + drostanolone propionate with total estrogens 102 (ref <130) — no spike. Later repetitions include an AnabolicSteroidForums poster reporting E2 613 via LabCorp on trenbolone, a 2018 T-Nation post stating flatly that “if you are using tren and get lab draws at using labcorp your estrogen levels will be falsely high” and warning against dosing an AI off that number, and Professional Muscle threads repeating the claim and advising the LC-MS sensitive test. Confounders the anecdotes cannot separate: ordinary ECLIA inaccuracy at male E2 ranges, real aromatization from the testosterone in the stack, and unverified underground AI products.

Sources: Roche, Elecsys Estradiol III method sheet V6.0 (2020-05) · Roche Field Safety Notice SBN-CPS-2016-005 · Krasowski et al., BMC Clin Pathol 2014 · RIDASCREEN 17β-Estradiol (bovine) cross-reactivity table · eroids (2013–) · T-Nation (2018)

Community use patterns — reported practice only

What the sources report — with the source type attached

Source	Reported figures	Tier
Llewellyn, Anabolics	Trenbolone acetate 100–300 mg/week for 6–8 weeks; trenbolone enanthate 150–300 mg/week for 6–10 weeks. Both profiles state “never approved for use in humans; prescribing guidelines are unavailable.”	Secondary — reference book
steroid.com profile lineage (Anthony Roberts)	Trenbolone enanthate 200–600 mg/week — markedly higher than Llewellyn, illustrating the spread. Its cycle templates run from a “beginner” trenbolone acetate 300 mg/week × 8 weeks to an “advanced” trenbolone enanthate 800 mg/week × 12 weeks.	Secondary — profile lineage
Vendor / product pages	Typically list trenbolone enanthate 200–400 mg/week, cycles of 8–12 weeks, “some advanced users up to 600 mg/week.”	Commercial pages — lowest evidentiary weight
Compiled forum standards (“Steroid Wiki” tradition)	A “Silver Standard” basic cut framed as testosterone propionate 100–175 mg/week + trenbolone acetate 350 mg/week (50 mg/day), 8–12 weeks; some versions put a first trenbolone exposure at 200–300 mg/week. A “Gold Standard” adds drostanolone propionate.	Community anecdote — compiled
Piatkowski et al. 2024 survey	Respondents' mean self-reported trenbolone exposure was ~245 mg/week (SD 158.74) over cycles of ~8.7 weeks (SD 6.89) — the only sample-based (rather than prescriptive) central figure here, and it sits well below the older forum templates.	Clinical / human — descriptive survey

• The high-test vs low-test debate: an older orthodoxy paired low testosterone (“TRT-level,” 100–250 mg/week) with high trenbolone (400–700 mg/week in older logs; 200–400 in newer), arguing estradiol worsens trenbolone side effects and trenbolone “outcompetes” testosterone at the AR; the opposing camp runs testosterone ≥ trenbolone (an Evolutionary.org thread argues the receptor-competition framing is a misunderstanding). No human trial compares ratios; the current forum synthesis is goal- and total-dose-dependent.
• Ester preference, as the community frames it: acetate is the near-universal “first trenbolone” advice, explicitly as an exit strategy rather than a potency choice — short active life by community convention (see the half-life section for the evidence state), so intolerable side effects clear in days, at the cost of daily or every-other-day injections. Enanthate horror threads are a genre: the same slow clearance that suits experienced users' twice-weekly schedules means side effects, once established, persist. Forum ester-weight caveat: acetate is ~87% trenbolone by weight vs enanthate ~70%, so mg-for-mg comparisons across esters overstate enanthate — a community-carried but chemically sound point.
• The microdosing trend (2023–2026): a visible recent shift in forum practice toward trenbolone acetate 10 mg/day (~70 mg/week), 10–25 mg/day, or 25 mg every other day, often layered on TRT bases for 8–16+ weeks. The forums themselves debate whether “microdose tren is safe tren” — there are no human threshold data to settle that debate; lower reported exposure is a trend in practice, not a safety finding.
• The deterrent literature: forum trenbolone discourse has a named cautionary genre — trensomnia (2–4 h sleep), night sweats, cardio collapse, injection cough, anxiety / paranoia, jealousy / relationship threads, aggression — reported mostly at 300–700 mg/week in older logs, occasionally below 100 mg/week in self-described sensitive users — capped by the repeated sanction, “If you're not competing, why are you running tren?” Two-19-nor combinations (trenbolone + nandrolone) carry no pharmacological prohibition and no study; the forum heuristic against them is attribution — if something goes wrong you cannot tell which drug is responsible.

Sources: Llewellyn, Anabolics (mirrored profiles) · Steroid Wiki (compiled) · Piatkowski et al., Int J Drug Policy 2024 · Forum-cycle research filed 23 Oct 2026

Creator layer — opinion, labelled as opinion

Everything in this section is creator opinion and community content, kept separate from the primary literature above precisely so it is not mistaken for it. Where a creator cites the literature, the citation is checked at source; the creator's gloss is identified as gloss.

Dr. Alex Tatem — the trenbolone long-form

“Trenbolone: The Most Powerful Steroid Ever Created” (YouTube). Tatem is a board-certified urologist and men's-health specialist; his stated frame for the video is “informed consent,” not fear-mongering — “Not the memes. Not the gym lore… This isn't fear-mongering. It's informed consent.” Its chapters run: Introduction – What Is Trenbolone? · Biochemistry – The 19-Nor Family · Progesterone & Anti-Glucocorticoid Activity · Human Data (Or Lack Thereof) · The Benefits – Why Anyone Uses Tren · Side Effects – The Real Cost of Tren · Trenbolone in Popular Culture · The Honest Conclusion. His citation list, as printed, includes Yarrow et al. 2010 and 2011, Ma & Liu 2015, Piatkowski et al. 2023 and 2024, the 2025 structured review of 23 case reports, case reports (MI in a 23-year-old trenbolone user, 2015; trenbolone hepatotoxicity / cholestasis), the historical sources (Roussel-Uclaf 1963 synthesis; Parabolan / Negma 1980–1997), and Pope et al. on adverse consequences of performance-enhancing drugs. Checked at source: his two anchor citations say what the sections above say — Yarrow 2010 is a tissue-selectivity optimism review (not a toxicity paper), and Ma & Liu 2015 is rat + cell work showing Aβ42 changes, not “amyloid plaques in the human brain.” Where the presentation compresses the rodent findings into human-sounding risk language, that compression is creator gloss. One internal inconsistency worth flagging: in his separate tier-list video, Tatem says trenbolone was “briefly marketed as a potential treatment option for women with hormone-sensitive breast cancer in France in the 1970s” — a claim that could not be verified against the Parabolan record retrieved for this page — while also saying “we've never seen a pharma grade [trenbolone] in the United States,” the defensible clause; “never meant for human consumption” is not strictly accurate given his own deep-dive's Parabolan citation.

Dr. Alex Tatem — the tier list (a separate video)

“Trenbolone F-tier” comes from “Steroid Tier List: Doctor Explains Best & Worst Compounds,” not from the deep-dive video. His stated criterion: effectiveness for its purpose weighed against risk. On trenbolone: “There is nothing quite like Tren. Tren hits harder than almost any other injectable compound… incredible aggression… incredible amount of strength… dry, grainy look.” Then the verdict: “Everything else is a dumpster fire. This is an absolute poison. This is toxic… people are getting excited about microplastics and then they're injecting this.” His three toxicities, in substance: (i) nephrotoxicity — “Tren is one of the few injectables that is actually been linked with nephrotoxicity”; (ii) hepatotoxicity — “it is interestingly enough hepatoxic despite being injectable”; (iii) brain amyloid — “It causes amyloid plaques in the brain that have been linked to Alzheimer's… if you can think of a way that Tren can poison you, Tren can poison you.” Placement: “this is straight F-tier. This is trash. This is not something that anybody should ever put into their bodies.” Against this page's grading — creator opinion, labelled as such: the nephrotoxicity framing overstates the evidence (direct trenbolone nephrotoxicity is unproven); the hepatotoxicity framing omits the non-17α-alkylated context (real but rare and idiosyncratic); the amyloid sentence converts a rat Aβ42 marker study into a human plaque claim. His S-tier placements in the same video — testosterone cypionate, oxandrolone / Anavar, and methenolone / Primobolan — are creator opinion of the same status.

More Plates More Dates' “Does Trenbolone Cause Alzheimer's Disease?” write-up is the popularization layer of the neurotoxicity claim — see the trace in the Ma & Liu panel above. No substantive trenbolone material from the other creators in this library was located in the research file, and none is folded in without a source.

Sources: Tatem, “Trenbolone: The Most Powerful Steroid Ever Created” (YouTube) · Tatem, “Steroid Tier List” (transcript) · More Plates More Dates, “Does Trenbolone Cause Alzheimer's Disease?”

Claim check

Claims circulating about trenbolone, checked against the research file:

• Not supported

  “Tren is 5× testosterone — anabolic : androgenic 500:500.”

  Folklore. The figures trace to Llewellyn's Anabolics, not a study; Yarrow 2010 puts AR binding at ~3× testosterone and Bauer / Wilson at DHT-like, and the exact 500:500 figures appear in no primary source located.

• Checks out

  “Trenbolone does not aromatize.”

  Supported — it is not an aromatase substrate, the consensus of the anchor review and all pharmacology sources, for the parent compound. Estradiol on a trenbolone-containing regimen comes from the other compounds.

• Not supported

  “Tren converts to DHT — finasteride matters on tren.”

  Wrong. Trenbolone is a non-5α-reducible androgen (Yarrow 2011); its metabolites are weaker, and 5α-reductase inhibitors have no trenbolone metabolite to block.

• Needs the qualifier

  “Tren binds the progesterone receptor stronger than progesterone.”

  True as binding in one bovine-PR in-vitro assay (Bauer 2000). Functional progestogenic potency in bioassays is weak, near testosterone's — binding is not activation.

• Not supported

  “Tren raises prolactin — that's what tren gyno is.”

  Unverified as a claim, folklore as a mechanism. No human prolactin measurement in trenbolone users exists; trenbolone gynecomastia is not described in the literature; gyno on tren-containing regimens is documented at class level and most parsimoniously estradiol-driven.

• Needs the qualifier

  “Tren blocks the glucocorticoid receptor like RU486.”

  Overstated. Functional anti-glucocorticoid activity and GR down-regulation are animal-supported; direct GR binding was ~zero in sheep muscle (Sharpe 1986).

• Needs the qualifier

  “Tren raises IGF-1 in muscle.”

  Overstated. Bovine satellite-cell culture only, and the IGF-I mRNA effect is serum-condition-dependent (Kamanga-Sollo 2008); no human IGF-1 study exists.

• Needs the qualifier

  “Tren lets you gain muscle in a deficit / recomp.”

  Animal-supported, human-unverified — directly demonstrated in rats (Donner 2015: fat −37%, lean +11% at stable body mass); no human controlled data.

• Needs the qualifier

  “Nutrient partitioning.”

  Supported in cattle — feed-conversion efficiency is the FDA-approved veterinary endpoint. Folklore in humans: no human measurement exists.

• Needs the qualifier

  “Tren crushes HDL.”

  Supported at class level — HAARLEM: HDL-C −0.40 mmol/L on (stack) cycles. The trenbolone-specific magnitude is unverified, and the extreme percentages in circulation are stanozolol data.

• Needs the qualifier

  “Tren causes LVH.”

  Overstated. LVH / dysfunction is a real AAS-class, dose-associated, partially reversible phenotype (HAARLEM, Baggish); no study isolates trenbolone, and rat cardiac studies found no adverse effect.

• Not supported

  “Tren causes Alzheimer's / amyloid plaques in the brain.”

  Folklore in its human form. The source is Ma & Liu 2015 — single-dose rat Aβ42 changes + neuronal apoptosis in culture; no human Aβ / tau / cognition study exists, and Kaufman 2019 states this explicitly. Animal data only.

• Unverified

  “Tren destroys your kidneys.”

  No trenbolone-only renal study or biopsy series exists; the AAS kidney literature is FSGS of extreme lean mass, and dark urine is attributed to oxidised metabolites — itself unmeasured.

• Needs the qualifier

  “Tren is hepatotoxic like an oral.”

  Overstated. It is not 17α-alkylated and expected hepatotoxicity is low — but rare idiosyncratic cholestatic hepatitis is case-documented, including one biopsy-supported case.

• Needs the qualifier

  “Tren wrecks sleep / trensomnia.”

  Supported as a class effect plus a consistent reputation — AAS users: PSQI 6.85 vs 4.34, 66% reporting sleep problems (Klonteig 2024), and sleep was worse in withdrawal. The trenbolone-specific magnitude is unverified; no trenbolone sleep study exists.

• Needs the qualifier

  “Tren rage is a pharmacological syndrome.”

  Overstated. The class RCT meta-analysis shows a small self-report aggression signal (g = 0.171), not observer-confirmed; trenbolone-specific evidence is one borderline verbal-aggression association (p = 0.045) plus qualitative reputation.

• Not supported

  “Tren ruins your hair — or is hair-safe because it makes no DHT.”

  Both forms unverified. No trenbolone alopecia study exists; it is a potent direct AR agonist that is not 5α-reduced, so both the doom claim and the safety claim outrun the data.

• Not supported

  “Tren cough is caused by prostaglandins / bradykinin — and only happens with tren.”

  No primary source exists for the mechanism; the best-anchored mechanism is pulmonary oil microembolism, documented for oil injectables generally. And POME cough is on the FDA label of testosterone undecanoate — it is not trenbolone-unique.

• Needs the qualifier

  “Tren falsely raises E2 on standard (ECLIA) estradiol tests.”

  Supported as a reported phenomenon; mechanism unquantified. A decade of platform-specific paired lab reports, a veterinary ELISA listing 1.0% trenbolone cross-reactivity, and the fulvestrant precedent for this assay family — but Roche never tested trenbolone and no controlled study exists. Not proven fact.

• Not supported

  “Trenbolone was never used in humans.”

  Wrong as stated. Parabolan (trenbolone hexahydrobenzylcarbonate, Negma) was an approved human product in France, 1980–1997. What is true: never FDA-approved, and no human approval since 1997.

• Unverified

  “Parabolan was pulled for safety.”

  Only “voluntarily discontinued 1997” is sourced; no regulatory or manufacturer reason was located.

• Unverified

  “Trenbolone enanthate's half-life is 7–10 days (acetate ~3, hex ~14).”

  Unverified as measurement. No human IM pharmacokinetic trial exists for any ester; published figures are mutually inconsistent secondary conventions — the hex ester is given as both 8 and 14 days.

• Needs the qualifier

  “Two 19-nors must never be combined.”

  Folklore as pharmacology; sound as attribution hygiene. No pharmacologic contraindication and no study of the combination exists; the forum rationale is that side-effect attribution becomes impossible.

• Unverified

  “Microdose tren is safe tren.”

  A real 2023–2026 practice trend (10–25 mg/day acetate, reported) with zero human threshold data behind the safety inference.

Open questions

Marked plainly, because they are open — each item below was searched for in the research file, with no adequate source found:

• Why Negma discontinued Parabolan (1997) — and Parabolan's exact labelled indications. “Commercial / regulatory, not safety” is a vendor assertion with no primary source; the related claim that trenbolone was marketed in France in the 1970s for hormone-sensitive breast cancer in women could not be verified against the Parabolan record either.
• Any human IM half-life for trenbolone acetate, enanthate, or hexahydrobenzylcarbonate — no human pharmacokinetic trial exists in the sources found.
• Trenbolone raising prolactin in humans — and, with it, any prolactin-mediated gynecomastia or sexual-dysfunction mechanism. No human prolactin measurement in trenbolone users was located.
• Trenbolone raising IGF-1 in humans, systemically or in muscle — the evidence is bovine satellite-cell culture only. Likewise, a quantified trenbolone-vs-nandrolone AR binding ratio from a single head-to-head assay: only qualitative comparison is supportable.
• Any trenbolone-specific human figure for HDL suppression %, blood-pressure change, LVH incidence, hematocrit exceedance rate, sleep architecture change, hair-loss incidence, or dementia / cognitive outcomes. All such figures in circulation are vendor / forum numbers or AAS-class data relabelled.
• “Tren cough” mechanism and ester comparison — no primary tren-cough paper exists; the prostaglandin / bradykinin mechanism has no located source study; acetate-vs-enanthate frequency has no comparative study. Night-sweat mechanism and prevalence for trenbolone specifically are likewise unmeasured.
• Which molecule cross-reacts on the Roche estradiol assay, and at what percentage — no manufacturer listing, peer-reviewed spiking study, or controlled split-sample comparison exists; and whether Quest's non-LC-MS estradiol is truly unaffected rests on total-estrogens anecdotes, not a method comparison.
• Trenbolone enanthate's alleged 1967 Roussel-Uclaf origin, the named inventors / original 1963 patent for the first synthesis, and trenbolone undecanoate's marketing history — none verified from a primary source.

Sources

1. Yarrow JF, McCoy SC, Borst SE. Tissue selectivity and potential clinical applications of trenbolone. Steroids 2010;75(6):377-389. PMID 20138077. pubmed.ncbi.nlm.nih.gov/20138077
2. Yarrow JF et al. 17β-Hydroxyestra-4,9,11-trien-3-one (trenbolone) exhibits tissue selective anabolic activity. Am J Physiol Endocrinol Metab 2011;300(4):E650-E660. PMID 21266670. pubmed.ncbi.nlm.nih.gov/21266670
3. Bauer ERS et al. Receptor binding of trenbolone and its metabolites. APMIS 2000;108(12):838-846. PMID 11252818. hero.epa.gov/reference/6558070
4. Wilson VS et al. In vitro and in vivo effects of 17β-trenbolone. Toxicol Sci 2002;70(2):202-211. PMID 12441365.
5. Kamanga-Sollo E et al. IGF-I in bovine satellite cell cultures treated with estradiol / TBA. Domest Anim Endocrinol 2008;35(1):88-97. PMID 18403176. pubmed.ncbi.nlm.nih.gov/18403176
6. Donner DG et al. Improvements in body composition, cardiometabolic risk factors and insulin sensitivity with trenbolone in normogonadic rats. Steroids 2015. sigmaaldrich.com (Donner 2015 abstract) · Repository record: research-repository.griffith.edu.au
7. Sharpe PM, Buttery PJ, Haynes NB. Muscle glucocorticoid receptors in sheep. Br J Nutr 1986. resolve.cambridge.org (Sharpe 1986)
8. Danhaive PA, Rousseau GG. Steroid–glucocorticoid receptor interaction studies. dial.uclouvain.be/pr/boreal/object/boreal:21642
9. Spranger B, Metzler M. Disposition of 17β-trenbolone in humans. J Chromatogr B 1991;564:485-492. hero.epa.gov/search/?query=Trenbolone
10. FDA Environmental Assessment, trenbolone acetate implants (1987). animaldrugsatfda.fda.gov (EA 180)
11. FDA FOI Summary, Revalor-S, NADA 140-897 (1991). animaldrugsatfda.fda.gov (FOI 2486)
12. DailyMed, Revalor-200 label, NADA 140-992. dailymed.nlm.nih.gov (Revalor-200 label)
13. Oklahoma State Extension, implants in beef cattle production. extension.okstate.edu/fact-sheets/implants-and-their-use-in-beef-cattle-production
14. Smit D et al. HAARLEM cohort, androgen cycles. Andrologia 2022. PMID 35014715. medscape.com/medline/abstract/35014715
15. Smit D et al. HAARLEM echocardiography. Front Reprod Health 2021. PMID 36304014. pubmed.ncbi.nlm.nih.gov/36304014
16. Baggish AL et al. Long-term AAS use and cardiovascular dysfunction. Circulation 2017;135:1991-2002.
17. Meta-analysis of AAS cardiac effects. Int J Cardiol 2025. internationaljournalofcardiology.com/article/S0167-5273(25)00939-8/abstract
18. Ma F, Liu D. 17β-trenbolone contributes to neurodegeneration. Toxicol Appl Pharmacol 2015;282(1):68-76. PMID 25461682. pubmed.ncbi.nlm.nih.gov/25461682
19. Kaufman MJ et al. Supraphysiologic-dose AAS use: a risk factor for dementia? Neurosci Biobehav Rev 2019;100:180-207. PMID 30817935. pmc.ncbi.nlm.nih.gov/articles/PMC6451684
20. Zelleroth S et al. AAS neurotoxicity in rat cortical cultures. J Steroid Biochem Mol Biol 2021;210:105863. PMID 33677017. pubmed.ncbi.nlm.nih.gov/33677017
21. Zuo et al. 17β-trenbolone under chronic social defeat stress, mice. Mol Neurobiol 2025. PMID 40622522. pubmed.ncbi.nlm.nih.gov/40622522
22. Herlitz LC et al. FSGS in long-term AAS-using bodybuilders. J Am Soc Nephrol 2010;21:163-172. PMC2799287.
23. Lizardo et al. Hypercalcemia from trenbolone use: a case report. Cureus 2026;18(1):e102580. PMID 41769525. pubmed.ncbi.nlm.nih.gov/41769525
24. Anand JS et al. Parabolan overdose, cholestasis, MARS treatment. ASAIO J 2006. PMID 16436902. pubmed.ncbi.nlm.nih.gov/16436902
25. Boks M et al. Cholestatic hepatitis from trenbolone enanthate. J Sports Sci 2017;35:2262-2264. PMID 27937337. pubmed.ncbi.nlm.nih.gov/27937337
26. Klonteig S, Scarth M, Bjørnebekk A. Sleep in long-term AAS users. BMC Psychiatry 2024. PMID 38254047. pmc.ncbi.nlm.nih.gov/articles/PMC10804719
27. Chegeni R et al. AAS and aggression: RCT meta-analysis. Psychopharmacology 2021.
28. Piatkowski TM, Neumann DL, Dunn M. “My mind pretty much went to mush.” Drug Alcohol Rev 2023;42:1566-1576.
29. Piatkowski TM et al. Trenbolone, psychological distress and aggression survey. Int J Drug Policy 2024;134:104636. PMID 39486244. pubmed.ncbi.nlm.nih.gov/39486244
30. Viana et al. Structured review of trenbolone case reports (23 reports). World J Pharm Sci Res 2025. zenodo.org/records/15108706
31. Russell M, Storck A, Ainslie M. Pulmonary oil embolism after oil-based androgen injection. Can Respir J 2011;18(4):e59-e61. PMID 22059184. pubmed.ncbi.nlm.nih.gov/22059184
32. FDA, Aveed (testosterone undecanoate) approval package — POME labeling. accessdata.fda.gov (Aveed approval package)
33. Applebaum-Bowden D, Hazzard WR et al. Stanozolol and hepatic lipase / HDL. PMID 3657514. pubmed.ncbi.nlm.nih.gov/3657514
34. Roche, Elecsys Estradiol III method sheet V6.0 (2020-05). assets.roche.com (Elecsys Estradiol III method sheet)
35. Roche Field Safety Notice SBN-CPS-2016-005 — fulvestrant cross-reactivity, Elecsys Estradiol II/III. moh.gov.cy (Roche FSN SBN-CPS-2016-005)
36. FDA / ICIJ device record — Elecsys estradiol assay correction (fulvestrant), Z-1277-2016. medicaldevices.icij.org (Elecsys estradiol correction)
37. Krasowski MD et al. Cross-reactivity of steroid hormone immunoassays. BMC Clin Pathol 2014;14:33. bmcclinpathol.biomedcentral.com/articles/10.1186/1472-6890-14-33
38. RIDASCREEN 17β-Estradiol (bovine) — cross-reactivity table incl. trenbolone 1.0%. ridacom.com/en/products/view/193
39. Community and creator sources, labelled as such wherever used — eroids, “Tren Causing Bogus Elevated Estradiol Levels — Roche ECLIA methodology” (2013–). eroids.com (tren / ECLIA estradiol thread) · AnabolicSteroidForums, LabCorp ECLIA E2 report. anabolicsteroidforums.com · T-Nation, “An Important Note on Bloodwork” (2018). t-nation.com/t/an-important-note-on-bloodwork/239662 · Professional Muscle, estradiol-on-tren threads. professionalmuscle.com
40. Community / secondary, cited as that tier — Llewellyn, Anabolics, mirrored profiles. bpmedical.info (trenbolone acetate) · bpmedical.info (trenbolone enanthate) · Steroid Wiki (compiled). steroidsbible.s3.us-east-2.amazonaws.com (Steroid Wiki) · Evolutionary.org, tren-over-test thread. evolutionary.org/forums/threads/tren-over-test.8793 · Evolutionary.org, Deca-vs-tren prolactin thread (hosts the Yarrow letter). evolutionary.org/forums/threads/deca-prolactin-vs-tren-prolactin.4151 · WikiDoc, Trenbolone. wikidoc.org/index.php/Trenbolone · The Compound Coaching, trenbolone audit. thecompoundcoaching.com/library/trenbolone
41. Creator sources, cited as opinion — Tatem A. “Trenbolone: The Most Powerful Steroid Ever Created” (YouTube). youtube.com/watch?v=IQ9-P0ncR1A · Tatem A. “Steroid Tier List” — transcript. youtubetotranscript.com/transcript?v=Qn4N2RTkgKU · More Plates More Dates. “Does Trenbolone Cause Alzheimer's Disease?” moreplatesmoredates.com/does-trenbolone-cause-alzheimers-disease-what-the-data-shows
42. Trenbolone — profile & claim check, research report filed 26 Oct 2026. Every figure, attribution and hedge on this page follows that report.

Molecule

Androst-4-en-17β-ol-3-one

Class

Endogenous androgen · C19

Esters covered

Cypionate · propionate · acetate

US status

Cypionate FDA-approved · Schedule III

The health & TRT layer — who replacement is actually for

Testosterone is the principal endogenous androgen in men. More than 95% is synthesised in the Leydig cells of the testes under pituitary LH control; reported daily production is roughly 3–11 mg/day across physiology reviews (≈5–7 mg/day in the sources most often cited), secreted in pulses, with intratesticular concentrations ~100× circulating concentrations — the local concentration spermatogenesis depends on. Its ordinary jobs, per the FDA label's clinical-pharmacology section and standard physiology: development and maintenance of male sex organs and secondary sex characteristics, and — the parts a wellness reader actually feels — contributions to sexual desire and function, muscle mass and strength, bone density, red-blood-cell production, fat distribution and mood. Several of those jobs are not done by testosterone itself but by its two active metabolites: dihydrotestosterone (DHT) in prostate/skin-type tissues, and estradiol in fat, bone and brain (full pharmacology below).

When production fails — testicular failure (primary hypogonadism) or hypothalamic/pituitary failure (hypogonadotropic hypogonadism) — the result is hypogonadism: a syndrome of symptoms plus consistently low testosterone, not a single low number. Three positions agree on who replacement is for, and they are narrower than most marketing:

• Endocrine Society Clinical Practice Guideline (Bhasin et al., 2018): diagnose hypogonadism only in men with symptoms/signs of testosterone deficiency and unequivocally and consistently low serum testosterone; confirm with a second fasting morning total testosterone — a technical remark in the guideline notes ~30% of men with an initial value in the hypogonadal range are normal on repeat. Do not screen the general population. LH/FSH distinguish primary from secondary causes.
• American Urological Association guideline (Mulhall et al., 2018; validity confirmed 2024): total testosterone <300 ng/dL is a reasonable cut-off in support of a diagnosis — not a diagnosis by itself; two separate early-morning measurements plus symptoms/signs are required; validated symptom questionnaires are not recommended to select candidates.
• FDA label (Depo-Testosterone, testosterone cypionate): “Prior to initiating… confirm the diagnosis of hypogonadism by ensuring that serum testosterone concentrations have been measured in the morning on at least two separate days and that these serum testosterone concentrations are below the normal range.” Approved indications: replacement in (1) primary hypogonadism — testicular failure from cryptorchidism, bilateral torsion, orchitis, vanishing testis syndrome, or orchidectomy; and (2) hypogonadotropic hypogonadism — gonadotropin/LHRH deficiency or pituitary-hypothalamic injury from tumours, trauma or radiation. The same label version states safety and efficacy in “age-related hypogonadism” have not been established — with a live regulatory caveat, flagged in the Regulatory section below.

What replacement is not approved or guideline-endorsed for: age-related decline alone (in the label version read), low energy in men with normal testosterone, athletic performance, or use in women (no FDA-approved testosterone product for women in the US). Testosterone is a DEA Schedule III controlled substance (Anabolic Steroids Control Act, 1990).

Sources: Depo-Testosterone (testosterone cypionate), DailyMed FDA label · Bhasin et al., Endocrine Society guideline 2018 · Mulhall et al., AUA guideline 2018

What replacement demonstrably does — the Testosterone Trials

The TTrials are the anchor efficacy evidence: seven coordinated, double-blind, placebo-controlled trials, 790 men aged ≥65 years with two morning testosterone values averaging <275 ng/dL and symptoms, treated for 1 year with testosterone gel titrated to the mid-normal range for young men vs placebo gel (Snyder et al., NEJM 2016; ClinicalTrials.gov NCT00799617).

TTrials outcomes — one year of replacement in men ≥65

Outcome	Result
Sexual function (primary Sexual Function Trial)	Significantly increased sexual activity (P<0.001), sexual desire and erectile function — described by the investigators as moderate benefit.
Walking / physical function (primary Physical Function Trial)	The % achieving a ≥50 m increase in 6-minute walk distance did not differ significantly in the Physical Function Trial itself; it did differ when all trial participants were pooled (20.5% on testosterone) — a small effect.
Vitality / energy (primary Vitality Trial)	No benefit on the primary vitality outcome.
Mood / depressive symptoms	Slight improvement in mood and depressive-symptom severity.
Anaemia	Haemoglobin rise ≥1.0 g/dL in 54% vs 15% (unexplained anaemia) and 52% vs 12% (anaemia of known cause), testosterone vs placebo.
Bone	Spine trabecular volumetric BMD change +7.5 vs +0.8 mg/cm³ (testosterone vs placebo); increased volumetric BMD and estimated bone strength, spine > hip. Surrogate outcome — no fracture endpoint.
Cognition	No improvement in memory or any other cognitive measure (~500 men with age-associated memory impairment).
Coronary plaque (surrogate)	Non-calcified coronary plaque volume increased more on testosterone (mean change ≈54 vs ≈14 mm³); major adverse CV events were similar in number, but 788 men × 1 year is far too few to judge clinical CV risk — the question TRAVERSE was built to answer.
Prostate / blood (safety)	PSA rose more on testosterone, with more urology referrals; erythrocytosis occurred in some men.

Independent confirmation of size: Ponce et al. (2018), the systematic review commissioned around the Endocrine Society guideline — 4 low-risk-of-bias RCTs, 1,779 hypogonadal men — found small significant gains in libido (SMD 0.17, 95% CI 0.01–0.34), erectile function (SMD 0.16, 0.06–0.27) and sexual satisfaction (SMD 0.16, 0.01–0.31), and no significant effect on energy or mood. Two honest conclusions follow: (i) the sexual-function benefit is the most reproducible symptom benefit, and it is small-to-moderate, not transformative on average; (ii) “testosterone for energy” is the marketing claim with the weakest trial support.

Diabetes — a genuinely split result, and the split is instructive

• T4DM (Wittert et al., Lancet Diabetes Endocrinol 2021): 1,007 Australian men 50–74 y, waist ≥95 cm, impaired glucose tolerance or newly diagnosed type 2 diabetes, testosterone ≤14 nmol/L, all enrolled in a structured lifestyle programme; long-acting testosterone undecanoate injections vs placebo for 2 years. Type 2 diabetes at 2 years (OGTT): 12% (55/443) vs 21% (87/413) — about a 40% lower risk, on top of the lifestyle programme.
• TRAVERSE diabetes substudy (JAMA Intern Med 2024, as summarised in secondary reporting): in men with prediabetes (n=1,175) progression to diabetes did not differ significantly, and in men with diabetes (n=3,880) testosterone did not improve glycaemic control or remission — without a structured lifestyle programme.
• Reading: testosterone is not a diabetes drug. At most, one large trial shows an added preventive effect when combined with a lifestyle programme in a specific metabolic population that largely did not have pathological hypogonadism.

Sources: Snyder et al., NEJM 2016 · Roy et al., JAMA Intern Med 2017 · Budoff et al., JAMA 2017 · Resnick et al., JAMA 2017 · Ponce et al., J Clin Endocrinol Metab 2018 · Wittert et al., Lancet Diabetes Endocrinol 2021

TRAVERSE — the cardiovascular question, tested properly

Design. TRAVERSE (NCT03518034; Lincoff et al., NEJM 2023) was the FDA-required, randomised, double-blind, placebo-controlled non-inferiority cardiovascular-outcomes trial: 5,246 men aged 45–80 with symptoms, two fasting testosterone levels <300 ng/dL, and pre-existing cardiovascular disease or high cardiovascular risk, at 316 US sites. Intervention: daily 1.62% transdermal testosterone gel, dose-adjusted to 350–750 ng/dL, vs placebo gel. Mean treatment duration ≈21.8 months; mean follow-up ≈33 months. Safety population 5,204 (the 42-man difference from enrolment is the enrolled-vs-treated distinction that explains the two n's in circulation). Mean age ≈63; this was a metabolically sick cohort (≈70% diabetes, >90% hypertension in commentary summaries of the baseline table).

Primary result. First occurrence of cardiovascular death, non-fatal MI or non-fatal stroke: 182 men (7.0%) on testosterone vs 190 (7.3%) on placebo; HR 0.96, 95% CI 0.78–1.17; P<0.001 for non-inferiority. No component and no secondary composite (including coronary revascularisation, heart-failure hospitalisation, all-cause death) differed meaningfully. That is a non-inferiority result — not proof of cardiovascular benefit.

What else TRAVERSE found — the part selective quoting drops

Rates as carried into updated US product labels, reproduced in label-summary reporting:

TRAVERSE safety outcomes — testosterone vs placebo

Outcome	Testosterone	Placebo	Note
Non-fatal arrhythmias warranting intervention	5.2%	3.3%	Secondary/exploratory — the trial was powered only for the primary MACE endpoint.
Atrial fibrillation	3.5%	2.4%	91 vs 63 men in one secondary summary.
Acute kidney injury	2.3%	1.5%	Mechanism not established.
Venous thromboembolism (total)	1.7%	1.2%	Incl. pulmonary embolism 0.9% vs 0.5%; DVT 0.6% vs 0.5%.
Bone fracture	3.5%	2.5%	Fracture substudy (Snyder et al., NEJM 2024): adjudicated clinical fractures 3.50% vs 2.46%, HR 1.43 — testosterone did not reduce fractures despite raising BMD in the TTrials. A standing lesson in surrogate-vs-clinical outcomes.

The prostate substudy

Bhasin et al., J Clin Endocrinol Metab 2024: 5,204 men, mean age 63.3 y, mean follow-up 33.0 months, 14,304 person-years; men with PSA >3.0 ng/mL or severe BPH symptoms (IPSS >19) were excluded at entry, with a pre-specified urology-referral protocol. High-grade prostate cancer: 5/2,596 (0.19%) vs 3/2,602 (0.12%); HR 1.62, 95% CI 0.39–6.77, P=.51. Any prostate cancer: HR 1.07 (0.47–2.42). No significant difference in acute urinary retention, invasive BPH procedures, new BPH pharmacotherapy, or IPSS change. PSA rose slightly more on testosterone — between-group difference ≈0.11 ng/mL at 3 months and ≈0.15 ng/mL by 12 months — and the difference did not widen after month 12, the pattern (early small rise, then parallel) that monitoring guidance is built on.

Limits that must travel with the headline: gel only (not injections, pellets or oral products); ~33 months (not decades); protocol-titrated to a mid-normal target (not supratherapeutic exposure); a screened population (PSA ≤3.0, no prostate-cancer history); high discontinuation. TRAVERSE answers “does guideline-style replacement cause heart attacks in high-risk hypogonadal men over ~3 years?” — answer: no detectable increase. It does not answer any bodybuilding-exposure question.

Regulatory consequence. On 28 February 2025 the FDA required class-wide label changes: TRAVERSE results added to labels, the boxed-warning language about increased adverse cardiovascular outcomes removed, and a warning that testosterone can increase blood pressure added/strengthened, based on FDA-required ambulatory blood-pressure monitoring (ABPM) studies (e.g., 24-hour systolic BP +3.8 mmHg on an oral undecanoate product in 138 men in one ABPM study, as summarised in secondary reporting). A cardiovascular sentence left the boxed warning and a blood-pressure warning arrived in the same action; both halves are the story.

Sources: Lincoff et al., NEJM 2023 (TRAVERSE) · Bhasin et al., J Clin Endocrinol Metab 2024 (prostate substudy) · Snyder et al., NEJM 2024 (fracture substudy) · FDA February 2025 label changes (secondary reporting — see Regulatory section)

Blood markers and safety — with numbers

The safety profile, domain by domain, with its evidence tier

Domain	Effect (named source)	Tier
Haematocrit / haemoglobin	Hct +3.18 points (95% CI 1.35–5.01), Hb +0.80 g/dL (0.45–1.14) vs control — Fernández-Balsells et al., 2010 meta-analysis. Erythrocytosis: OR 3.69 (Calof et al., 2005); RR 8.14 (1.87–35.40), n=1,579 (Ponce et al., 2018). By formulation (network meta-analysis, 29 RCTs, 3,393 men — Nackeeran et al.): Hct rise vs placebo — IM enanthate/cypionate +4.0, gel +3.0, IM undecanoate +1.6, patch +1.4 points; only the IM-vs-patch contrast was clearly significant. Observational clinic data report Hct >50% in up to 66.7% of injectable users vs 12.8% of gel users, but the groups were not comparable — treat as hypothesis, not a rate to quote naked. Guidelines act at Hct >54% (Endocrine Society: withhold until normalised).	Human RCT / meta-analysis
Estradiol (aromatisation)	Estradiol rises with testosterone, dose-dependently, and more in older men. Direct PK illustration: a single studied cypionate exposure in hypogonadal men produced a 3-fold estradiol rise on days 2–7 (Nankin, 1987). Estradiol is not waste: in Finkelstein 2013, estrogen deficiency — not androgen deficiency — accounted for fat gain, and both contributed to sexual function.	Human RCT / PK study
Lipids	Pooled across formulations, HDL falls by only −0.49 mg/dL (−0.85 to −0.13), LDL not significant — Fernández-Balsells 2010. Restricted to intramuscular esters in hypogonadal men (272 men — Whitsel et al., Am J Med 2001): total cholesterol −14, LDL −5, HDL −4 mg/dL, triglycerides −1 (NS). In healthy men at graded supratherapeutic exposures, HDL fell in proportion to dose (Bhasin 2001) — the small replacement-dose lipid effect and the larger bodybuilding-dose lipid effect are the same dose–response at different points.	Meta-analysis / RCT
PSA & prostate monitoring	Average PSA rise is small and front-loaded (TRAVERSE substudy, above). Older pooled data found combined prostate events more frequent on testosterone (OR 1.78, Calof 2005) — a composite dominated by PSA rises, biopsies and BPH diagnoses rather than cancers, in the pre-TRAVERSE era. Monitoring position (Endocrine Society/AUA): baseline PSA/DRE by age and risk, repeat PSA in the first year, urology referral for confirmed rises above guideline thresholds (commonly cited: a rise >1.4 ng/mL in a year). The saturation model (Claim check) is the mechanistic frame; TRAVERSE is the best clinical-outcome data and it is ~3 years in screened men.	RCT + guideline
Fertility / HPTA suppression	Exogenous testosterone suppresses GnRH→LH/FSH; intratesticular testosterone collapses (a controlled study measured a ~94% fall; Coviello et al., 2005), and sperm output falls. WHO Task Force (Lancet 1990): 271 healthy fertile men on a weekly enanthate contraceptive-trial exposure — ~65% azoospermic by 6 months (mean time to azoospermia ~120 days), most of the remainder severely oligozoospermic; the 1996 efficacy phase reported contraceptive efficacy of 96.6–98.6% among men who suppressed — which is precisely why testosterone is not an approved or reliable contraceptive (a third did not fully suppress). Recovery after stopping contraceptive regimens (Liu et al. pooled analysis, as reported in secondary sources): ~67% back to normal sperm counts by 12 months, ~90% by 24 months; recovery is usual but not guaranteed, and is slower/less certain after longer, higher, multi-drug exposure — for which there is no controlled recovery trial.	Human RCT (contraceptive trials)
Blood pressure	Class effect confirmed by FDA-required ABPM studies; label warning added 2025 (see TRAVERSE section). In TRAVERSE itself the gel arm's systolic BP rose modestly more than placebo.	RCT / regulatory
Sleep apnoea	The honest summary is small, short-term, dose-related worsening — not a proven long-term effect. Hoyos et al. (2012): RCT, 67 obese men with severe OSA on long-acting undecanoate study exposures — oxygen desaturation index worsened by +10.3 events/h (0.8–19.8) vs placebo at 7 weeks, not significant at 18 weeks. An earlier crossover study in 17 older men on short-term high-dose mixed esters found ~1 h less sleep and ~7 events/h higher respiratory disturbance indices. A 2021 systematic review concluded testosterone “likely plays a small role in exacerbating” OSA that “may be time limited” (Payne et al.). Endocrine Society: recommend against starting therapy in untreated severe OSA — a precautionary guideline stance resting on thin trial data, and this page says so.	Small human RCTs + review
Acne / oily skin	Listed on androgen labels and in guideline adverse-effect lists as an effect with evidence of association (Endocrine Society evidence review: common events with evidence — erythrocytosis, acne/oiliness, PSA/prostate-volume increase, reduced sperm production). No reliable placebo-controlled incidence figure at replacement exposure exists — acne is among the most asked-about and least counted effects. Observational AAS-user surveys report far higher rates at supratherapeutic poly-drug exposure and cannot be transferred to replacement.	Label / guideline; incidence = evidence gap
Gynecomastia	Mechanistically aromatisation-driven and on the cypionate label's adverse-reaction list. The same evidence review classifies gynecomastia among uncommon events with weak evidence of association at replacement exposure — direction only, no pooled incidence (Fernández-Balsells 2010). Community incidence claims (percentages circulating online) have no RCT source.	Label / meta-analysis (weak)

Monitoring — what guidelines actually ask for (description, not advice)

Endocrine Society (2018) and AUA (2018) monitoring frameworks, as described in the guidelines: evaluate symptom response and adverse effects after starting; measure serum testosterone and haematocrit (baseline, early — commonly 3–6 months — then periodically); evaluate prostate cancer risk in the first year (PSA ± exam by age/risk); ask about / act on untreated severe sleep apnoea, fertility intent (both guidelines: do not start exogenous testosterone in men planning near-term fertility — AUA statement graded Strong Recommendation), blood pressure, and cardiovascular status. The FDA label independently requires the two-morning-test diagnostic confirmation (above). Fertility intent, haematocrit and prostate risk are the three conversations both guidelines treat as mandatory before, not after, starting.

Pharmacology — the molecule, the receptor, the metabolites

The molecule and where it comes from

Testosterone is a C19 steroid (androst-4-en-17β-ol-3-one), synthesised from cholesterol in Leydig cells via pregnenolone → DHEA/androstenedione intermediates. Unesterified testosterone given intravenously has a half-life measured in minutes (~10–100 min across physiology sources) — hepatic clearance and tissue uptake are that fast — which is the entire reason ester prodrugs exist. In plasma, ~98% is protein-bound: tightly to sex-hormone-binding globulin (SHBG) and loosely to albumin, with ~2% free — the label's own figures (98% bound to “a specific testosterone-estradiol binding globulin,” ~2% free; Depo-Testosterone label). The free plus albumin-bound fractions are generally treated as bioavailable; SHBG concentration therefore moves free testosterone independently of total testosterone (raised by ageing, hyperthyroidism, estrogens; lowered by obesity/insulin resistance — the reason guidelines ask for free testosterone when total is borderline or SHBG is likely abnormal).

Receptor pharmacology

Testosterone is an agonist of the androgen receptor (AR), a ligand-activated nuclear transcription factor. The canonical genomic pathway (Handelsman 2021): free testosterone diffuses across the cell membrane → binds cytosolic AR → heat-shock proteins dissociate → receptor phosphorylates, dimerises → the complex translocates to the nucleus, binds androgen-response elements (AREs) in target-gene promoters, recruits co-activators/co-repressors, and changes transcription. Downstream gene programmes are tissue-specific, which is why one hormone touches muscle protein balance, erythropoiesis (via erythropoietin sensitivity and direct marrow effects — the dose–response study found haemoglobin rose without a matching erythropoietin explanation; Coviello et al., 2008), sebaceous glands, larynx, bone (largely via estradiol, below), brain and prostate simultaneously. Non-genomic (rapid, membrane-initiated) androgen signalling is described in the basic-science literature; it has no established clinical dosing or outcome correlate in humans and is flagged here as mechanistic, not clinical, evidence.

There is only one androgen receptor. Every claim that an ester, brand or blend has a different kind of anabolic/androgenic effect at the receptor is therefore false by construction for testosterone esters: after hydrolysis the ligand is identical. What varies between regimens is the concentration–time curve the receptor sees.

Tissue amplification I — DHT and 5α-reductase

In tissues expressing steroid 5α-reductase, testosterone is irreversibly converted to dihydrotestosterone (DHT) — a local amplification step:

• Enzymes: type 1 5α-reductase — liver, skin, brain, kidney; type 2 — prostate (dominant), seminal vesicles, hair-follicle dermal papilla, some skin/liver (Handelsman 2021).
• Scale: ~4% of circulating testosterone is converted to DHT (some reviews give 4–9%/5–10% ranges); circulating DHT runs at ~10% of testosterone concentrations, because most DHT acts and is degraded locally.
• Potency: DHT binds AR with ~2-fold higher affinity (Kd ≈0.25–0.5 nM vs ≈0.4–1.0 nM for testosterone), dissociates ~5-fold more slowly, and shows 3–10× greater molar potency in transactivation assays (Grino et al., 1990, via pharmacology reviews). Prostate tissue DHT concentrations are ~20× tissue testosterone (meta-analysis of biopsy samples: DHT 11.71 ± 4.07 vs testosterone 0.54 ± 2.13 pmol/g, n=256; Rittmaster et al., 2008).
• Tissue logic: skeletal muscle and bone express negligible 5α-reductase and rapidly inactivate DHT — so muscle sees testosterone itself, while prostate, skin/sebaceous glands and hair follicles see amplified DHT. This single fact explains the side-effect geography of testosterone (prostate growth, acne/oiliness, androgenetic hair loss in the genetically susceptible) and disproves the folklore that “DHT builds the muscle”: in the dutasteride RCT (Bhasin group, JAMA 2012), blocking 5α-reduction did not reduce the fat-free-mass or strength response to graded testosterone doses.

Tissue amplification II — aromatisation to estradiol

Aromatase (CYP19A1 — adipose, bone, brain, Leydig cells and other tissues) irreversibly converts a small fraction of testosterone to estradiol: ~0.2–0.4% of testosterone per day across sources, yet >80% of circulating estradiol in men derives from aromatisation (Finkelstein et al., 2013). Estradiol acts at estrogen receptors α/β and is required male physiology, not a by-product:

• Finkelstein et al., NEJM 2013 is the decisive human experiment: 400 healthy men 20–50 y, endogenous steroids suppressed with goserelin, graded testosterone gel replacement for 16 weeks, half with anastrozole added to block aromatisation. Result: lean mass, muscle size and strength tracked testosterone; fat accumulation tracked estradiol deficiency; sexual desire and erectile function needed both. The testosterone level at which each function failed varied widely between men.
• Estradiol also carries most of testosterone's bone effect in men (aromatase-deficient men develop severe osteoporosis; the Endocrine Society/AUA positions caution against long-term aromatase-inhibitor use in men on this basis) and provides the dominant negative-feedback signal to the hypothalamus/pituitary (see HPTA, below).
• Practical corollary, graded in the Claim check: any regimen or adjunct whose purpose is to drive estradiol as low as possible is fighting a hormone with RCT-proven jobs. Gynecomastia is the opposite tail — aromatisation in breast tissue — and is dose/exposure related, which is why it appears at supratherapeutic exposure far more plausibly than at replacement, even though a clean replacement-dose incidence was never counted.

Pharmacodynamics — the HPTA off-switch

Exogenous testosterone (and the estradiol made from it) feeds back on hypothalamus and pituitary: GnRH pulses slow, LH and FSH fall, Leydig-cell production stops, intratesticular testosterone — the ~100× serum concentration spermatogenesis requires — collapses (~94% measured fall; Coviello 2005), testicular volume and sperm production fall. This is not an idiosyncratic side effect; it is the control loop working as designed, it begins with the first studied exposures (LH was already suppressed after a first cypionate injection in Nankin 1987), and it is dose- and duration-related in the graded-dose PK/PD modelling of depot cypionate (Bi et al., 2018). Every other pharmacodynamic effect on this page — erythrocytosis, lipid shift, PSA rise, acne, apnoea signal — sits downstream of the same exposure integral: area under the testosterone curve, peak height, and time, which is exactly what ester and regimen change (Ester chemistry, below).

Metabolism and excretion — and what “detection” actually measures

Freed testosterone is cleared mainly by the liver along two 17-ketosteroid pathways to androsterone and etiocholanolone, excreted predominantly as glucuronide and sulfate conjugates: the label states ~90% of a dose is excreted in urine as glucuronic/sulfuric conjugates of testosterone and metabolites and ~6% in faeces, mostly unconjugated (Depo-Testosterone label). Glucuronidation is chiefly by UGT2B17 (and UGT2B15).

Anti-doping detection exploits this (WADA framework, from the doping-detection reviews): the urinary testosterone/epitestosterone (T/E) ratio is normally ~1; exogenous testosterone raises testosterone glucuronide excretion while epitestosterone — not a testosterone metabolite — does not rise (and endogenous production is suppressed), so the ratio climbs. A T/E ≥4.0 triggers confirmatory isotope-ratio mass spectrometry (GC/C/IRMS), which distinguishes exogenous (plant-derived, 13C-depleted) steroid carbon from endogenous carbon. The older ≥6 threshold was lowered to 4 by WADA in 2004. Genetics break the simple test: the UGT2B17 deletion polymorphism (del/del genotype — ~66.7% prevalence in East Asian vs ~9.3% in Caucasian populations in Schulze et al., 2008) produces mean T/E ratios of ~0.14 vs 1.4 (one copy) and 2.9 (two copies), and drastically lower T/E-test sensitivity after a studied enanthate exposure — one documented reason modern anti-doping uses longitudinal athlete biological passports and IRMS rather than a single ratio. Ester-specific detection windows quoted on vendor/forum pages were not found in primary sources in this research: what is documented is that intact esters and their hydroxylated metabolites are themselves detectable in blood/urine by LC-MS/MS for periods governed by the same depot kinetics as the ester section below, and that window varies with exposure, assay and genotype. Any precise per-ester detection calendar is, on current sourcing, folklore-grade.

Sources: Depo-Testosterone label (binding, excretion) · Handelsman 2021 (AR signalling, DHT) · Finkelstein et al., NEJM 2013 · Coviello et al. 2005 & 2008 · Bi et al. 2018 · Schulze et al. 2008

Ester chemistry — how the ester rewrites the curve

Chemistry. Esterification attaches a fatty-acid chain to testosterone's 17β-hydroxyl. The ester is more lipophilic (oil-soluble) and less polar than free testosterone; injected in oil (cypionate's FDA product uses cottonseed oil with benzyl benzoate/benzyl alcohol; Depo-Testosterone label), it forms a depot in muscle. Release is governed by partitioning out of the oil and hydrolysis by non-specific esterases in blood/tissue — the ester itself has no androgenic activity and the fatty acid released is pharmacologically inert at these quantities. Longer/bulkier, more lipophilic ester → slower depot escape → longer apparent half-life. This is flip-flop kinetics: the measured “half-life” of an injectable ester is an absorption half-life (release is slower than elimination, so release is rate-limiting), formally discussed for depot cypionate in Bi et al., 2018. Consequences that follow from the physics, not from opinion: the same ester in a different oil, volume, site or depth can shift the curve (demonstrated rigorously for nandrolone esters by Minto et al., 1997 — ester, injection site and volume all mattered); and peak–trough swing is a function of interval relative to half-life, which is why interval claims and ester claims must never be conflated.

Ester weight — the quiet arithmetic

Only part of each milligram of ester is testosterone (testosterone MW 288.4):

Testosterone fraction by ester

Ester	Ester MW	Testosterone fraction
Acetate	~330.5 (C21H30O3)	~87%
Propionate	~344.5	~83–84%
Cypionate	412.61 (label)	69.9%
(Reference) Enanthate	~400.6	~72.0%

So equal milligrams of ester are not equal milligrams of testosterone — a ~17-point spread between propionate and cypionate — and, in the one head-to-head that matters, investigators matched on unesterified testosterone content (140 mg) when comparing cypionate and enanthate (Schulte-Beerbuhl & Nieschlag, 1980, below). The acetate fraction is arithmetic from its formula (supplier listing), not a trial figure.

The three esters, from primary PK sources — not folklore tables

Cypionate, propionate, acetate — regulatory reality and sourced kinetics

	Cypionate	Propionate	Acetate
Regulatory reality	FDA-approved (Depo-Testosterone + generics), cottonseed-oil IM product.	US human products discontinued (originally FDA-approved 1974, Watson; now veterinary/compounded/grey-market supply in practice).	No FDA-approved product located; no marketed human medicine found.
Half-life (sourced)	~8 days IM — FDA label figure.	Terminal t½ 0.8 days; MRT 1.5 days — Nieschlag et al., 1976 single-dose study (50 mg study exposure, 7 hypogonadal men), as reported in Nieschlag's Testosterone: Action, Deficiency, Substitution chapter.	No primary human figure exists in the literature located. Secondary compilations list <1 day; vendor pages claim 1–2 or 2–3 days — mutually inconsistent, none primary.
Peak (sourced)	Serum T peak days 2–5 after a studied single exposure, ~3-fold rise — Nankin, 1987 (n=11 hypogonadal men).	tmax ≈14 h; peak 40.2 nmol/L (supraphysiological) — Nieschlag, 1976.	No human data located.
Duration (sourced)	Steroid values back to basal by days 13–14 (Nankin, 1987); label: “can be given at intervals of two to four weeks” (a label statement about the product, describing a wide peak–trough pattern, not an outcome claim).	Below the normal range by ~57 h (day 2); textbook duration of action 2–4 days; the authors' own simulation concluded propionate is not suitable for substitution therapy because of wide fluctuations at any practical interval (Nieschlag chapter).	No human data located.
Head-to-head evidence	Schulte-Beerbuhl & Nieschlag, Fertil Steril 1980: crossover in 6 healthy men, cypionate vs enanthate at matched unesterified content — serum testosterone profiles were identical in maximal concentration and in duration of elevation. Cypionate has no PK advantage over enanthate; the US/Europe split between them is market history, not pharmacology.	No modern head-to-head vs cypionate at matched content located; its kinetic disadvantage (left) is the primary-source finding.	None.

What ester choice therefore does and does not change. It changes: time-to-peak, peak height for a given interval, trough depth, how long a tail persists after the last exposure, and the testosterone fraction per milligram. It does not change: the active molecule, the receptor, the metabolite menu (DHT/estradiol proportions are set by tissue enzymes and total exposure, not by which fatty acid delivered the testosterone), or — on any head-to-head human evidence located — muscle, symptom or safety outcomes at matched testosterone exposure. Claims of ester-specific “quality” of effect (drier, cleaner, more anabolic, less estrogenic per unit of testosterone delivered) have no primary human source; see the Claim check.

Sources: Depo-Testosterone label · Nankin 1987 · Nieschlag et al. 1976 / Nieschlag, Testosterone: Action, Deficiency, Substitution · Schulte-Beerbuhl & Nieschlag, Fertil Steril 1980 · Bi et al. 2018 · Minto et al. 1997

The bodybuilding layer — the graded dose–response

Study exposures are named because the evidence is uninterpretable without them; they are trial designs, not recommendations. Reported community practice is labelled as such.

Bhasin et al., NEJM 1996 — muscle without training

43 healthy normal men, standardised diet, randomised to placebo or a supraphysiological weekly enanthate study exposure (600 mg/week × 10 weeks), with or without supervised resistance training. Fat-free-mass change: placebo/no-exercise +0.8 kg; testosterone/no-exercise +3.2 kg; placebo/exercise +2.0 kg; testosterone/exercise +6.1 kg. Two facts bodybuilding retellings often separate: testosterone alone beat training alone on fat-free mass in this short trial — and testosterone plus training beat everything by a wide margin. Strength (bench/squat) rose in parallel.

Bhasin et al., 2001 — the graded dose–response

61 healthy men 18–35 y; endogenous production suppressed with a GnRH agonist; weekly enanthate at five graded study exposures (25 / 50 / 125 / 300 / 600 mg) for 20 weeks; diet standardised; no training programme — so the numbers isolate the drug. Mean nadir testosterone: 253 / 306 / 542 / 1,345 / 2,370 ng/dL respectively.

Graded study exposure vs fat-free-mass change, 20 weeks (Bhasin 2001)

Graded study exposure	Fat-free-mass change
Lowest graded exposure (25 mg/wk arm; nadir T 253 ng/dL — below replacement)	Negative — men lost lean mass vs their eugonadal baseline state.
125 mg/wk arm (nadir T 542 ng/dL)	+3.4 kg
300 mg/wk arm (nadir T 1,345 ng/dL)	+5.2 kg
600 mg/wk arm (nadir T 2,370 ng/dL)	+7.9 kg

Changes in fat-free mass correlated with log testosterone concentration (r = 0.73, P = 0.0001). Leg-press strength, leg power, thigh/quadriceps volume, haemoglobin and IGF-I rose with concentration; fat mass and HDL cholesterol fell with concentration. Sexual function, mood, visual-spatial cognition and PSA did not change significantly at any graded exposure over 20 weeks in these young men. A companion analysis in older men (Bhasin 2005, same graded design, 60–75 y) found a near-identical fat-free-mass gradient (+4.2 / +5.6 / +7.3 kg at the 125/300/600 arms in secondary tabulations) with, in Coviello 2008's companion paper, a steeper haemoglobin/haematocrit rise in older men at the same exposure.

What “log-linear” means for the “more is better” question: equal multiplications of exposure buy roughly equal additions of fat-free mass — so returns per added milligram diminish continuously, while several harm markers (haemoglobin/haematocrit up, HDL down) scale with the same curve. Within the graded range there was no plateau and no threshold; above the top studied exposure there is no trial at all.

The questions bodybuilders ask — graded

Bodybuilding questions, answered only from human evidence

Question	Best human evidence	Verdict
Does testosterone build muscle without training?	Bhasin 1996: yes, +3.2 kg FFM in 10 wk at the studied supraphysiological exposure, no exercise — less than half the drug+training gain.	Supported — Human RCT (short-term, single exposure level).
Is more linearly better?	Bhasin 2001: log-linear, not linear; no plateau up to the highest studied exposure; no data above it.	Partly supported, mis-stated — “more gives more, with diminishing per-mg returns, up to the studied ceiling; beyond that, unknown.”
Does the ester change the muscle result?	No ester-vs-ester body-composition RCT exists. Cypionate ≡ enanthate PK at matched content (Schulte-Beerbuhl 1980); freed molecule identical (Pharmacology, above).	Ester-superiority claims: Folklore (PK timing differences are real; outcome differences are unmeasured).
Is propionate “drier / better for cutting,” cypionate “for bulking”?	No human trial compares esters on fat, water or body composition at matched testosterone exposure. Water/fluid retention tracks total androgen/estradiol exposure and diet in the clinical literature, not ester identity.	Folklore.
Do blends (multi-ester products) give a flatter/better curve?	The blend rationale (fast ester fills the early window) is contradicted for replacement by the textbook PK analysis: short and intermediate esters both peak shortly after injection, so combining them adds a peak, not a plateau (Nieschlag chapter). No blend-vs-single-ester outcome RCT located.	Unsupported; mechanistic rationale partly contradicted.
“Receptors saturate / down-regulate, so you must keep raising exposure”	No primary human study demonstrating AR down-regulation limiting muscle response was located; the graded data show response still rising at the top studied exposure. The one documented saturation phenomenon is prostatic (Claim check), not muscular.	Folklore / unverified.
What are the long-term risks at bodybuilding exposures?	No RCT exists (such a trial would be unethical). Best data are observational on mixed AAS use: Danish register cohort — 1,189 men sanctioned for AAS doping vs 59,450 matched controls, ~11 y follow-up: all-cause mortality HR 2.81 (1.98–3.99), natural-death HR 2.24, unnatural-death HR 3.64 (JAMA 2024); companion Circulation 2025 analysis: adjusted HRs ≈3.0 acute MI, 2.4 VTE, 2.3 arrhythmia, 3.6 heart failure, 8.9 cardiomyopathy. Cross-sectional imaging/echo studies of long-term AAS users (mean ~11 cumulative years) show higher LV mass index (106 vs 80 g/m²), lower LVEF (49% vs 59%) vs non-using lifters. None of this isolates testosterone — poly-drug use, dose misreporting, and sanction-cohort selection are structural confounders; it bounds the risk conversation, it does not price testosterone alone.	Human observational — real signal, wrong granularity for testosterone-only claims.
Does stopping restore everything?	HPTA/sperm recovery is usual after contraceptive-regimen exposures (safety table, above) but there is no controlled recovery trial after years of supratherapeutic multi-drug use; cardiac-imaging abnormalities have been observed in former users years after quitting (cross-sectional) — reversibility is an open question (below), not a promise either way.	Open — observational only.

Sources: Bhasin et al., NEJM 1996 · Bhasin et al., Am J Physiol Endocrinol Metab 2001 · Bhasin et al. 2005 · Coviello et al. 2008 · Windfeld-Mathiasen et al., JAMA 2024 (Danish AAS cohort)

Creator layer — opinion, graded claim by claim

Everything in this section is creator and community content, kept separate from the primary literature above precisely so it is not mistaken for it. Creator claims are never evidence; they are claims being graded. Where a creator's content exists mainly as video/podcast audio and only episode descriptions, transcripts or reposts were indexed, the claim rests on indexed text and is graded on that basis. Doses named inside creator claims are the claim's content, not recommendations.

Vigorous Steve (coach/creator)

• “Full HRT” beats “cookie-cutter TRT”: over time neurosteroid production and estrogen balance deteriorate on testosterone alone; he advocates lower-dose testosterone via daily subcutaneous micro-administrations of longer esters (enanthate/cypionate) to reduce conversion to estradiol and DHT, plus DIM for estrogen-metabolite balance, supplemental DHEA and pregnenolone to sustain neurosteroid levels, and hCG to preserve fertility [interview repost, MuscleChemistry]. Verdict — partly supported in pieces, unproven as a package. (i) Frequent small administrations do flatten peak–trough mathematically and by PK principle — but no RCT shows daily SC administration of a long ester improves symptoms, estradiol-related outcomes or safety vs less frequent administration at matched exposure. (ii) hCG maintaining intratesticular testosterone/fertility has mechanistic and small-study support and is guideline-discussed; the specific recombinant product preference is not trial-tested. (iii) DHEA/pregnenolone “backfill” and DIM: no RCT in men on testosterone located — mechanistic folklore tier. (iv) “Neurosteroid deterioration on TRT” as a clinical syndrome: not an outcome measured in TTrials/TRAVERSE — unverified.
• Start low; increase only while side effects are tolerable and progress continues; many users take far more than needed — gains attributed to very high total exposures could be matched at lower exposure with better nutrition/training [Q&A transcript]. Verdict — directionally supported. Bhasin 2001's log-linear curve is exactly a diminishing-per-mg-returns curve, and harm markers scale with exposure — “lowest exposure that produces the goal” is the only reading of the dose–response data that isn't contradicted by it. The specific community numbers in the transcript are reported practice, not evidence.
• Daily administration with MCT oil (any ester) smooths levels [as discussed, ThinkSteroids forum thread on his video]. Verdict — partly supported / carrier-oil part unverified. Frequency→smoothness: supported in principle. Carrier oil changing absorption: demonstrated for nandrolone esters (Minto 1997 — oil/site/volume matter), not demonstrated for testosterone cypionate in MCT vs cottonseed in humans. The claim borrows a real cross-compound finding.

Big Paul — Paul Barnett (Anabolic Bodybuilding podcast)

• Bloodwork and health markers should dictate decisions (blast/cruise transitions, contest-prep entry); going into prep “already red-lined” leads to worse results; prioritise health to maximise growth [Podbean episode descriptions; Clipzag episode summary]. Verdict — supported as a stance, untested as a protocol. Every marker he centres — haematocrit/haemoglobin, blood pressure, estradiol, kidney markers, sleep apnoea — is a marker the trials/guidelines independently flag. No trial tests his (or any) marker-guided community protocol as a whole; the stance is concordant, the thresholds used in community practice are not trial-derived.
• Slower, more measured escalation beats using the strongest tools early; don't reflexively add another drug to fix a side effect [Podbean episode description]. Verdict — consistent with the evidence structure. Dose–response harms scale with exposure; several community “fixes” (e.g., routine aromatase inhibition) carry their own RCT-documented costs. As general harm-reduction logic: supported. As specific advice: outside trial evidence.
• High testosterone/estrogen makes other compounds (his example: trenbolone) harder to manage; an estradiol–prolactin interaction explains some side effects [Podbean episode description]. Verdict — unverified for testosterone itself. Plausible-sounding endocrine interaction talk with no primary human study located; trenbolone has essentially no human trial literature at all, so any claim stacking on it is folklore-tier by construction.
• Equipoise (boldenone) crushed estradiol in clients, verified on bloodwork [reported via forum summary of his content]. Verdict — out of scope for testosterone, flagged for method: a bloodwork-verified observation in clients is still an uncontrolled case series — and it is a claim about boldenone, not testosterone. It illustrates the layer's recurring pattern: real measurements, uncontrolled design.

Dr. Alex Tatem (urologist; YouTube / clinic content)

• “Does testosterone cause prostate cancer? No.” — an old myth; research has consistently shown therapy does not cause prostate cancer or make it worse; PSA still monitored as a precaution [Testosterone 101 page attributed to Tatem, via ExcelMale]. Verdict — partly supported; the flatness overstates it. “No increased incidence in RCTs including TRAVERSE's prostate substudy, in screened men, over ~3 years” is supported. “Does not… make it worse” is not established: men with prostate cancer were excluded from the trials, the label still contra-indicates known/suspected prostate carcinoma, and androgen deprivation treating prostate cancer is the very observation (Huggins) the myth grew from. Correct claim, wrong certainty.
• Newer research including TRAVERSE showed “improved cardiovascular outcomes” in men with low testosterone who receive treatment [same page]. Verdict — contradicted as stated. TRAVERSE showed non-inferiority (HR 0.96, CI 0.78–1.17) — compatible with anything from a 22% benefit to a 17% harm on the primary endpoint, designed and reported as a safety result, with several adverse secondary signals. No testosterone RCT has shown improved cardiovascular outcomes. This is the single clearest creator-layer misstatement found in this research, because it converts a safety trial into an efficacy claim.
• Daily microdosed testosterone vs traditional weekly protocols presented as a cutting-edge strategy worth comparing [Asprey podcast episode description of Tatem interview]. Verdict — unproven superiority, legitimate question. Same grading as Vigorous Steve, above: PK smoothing is real, outcome superiority is untested in any RCT located.
• Customise formulation/regimen by body composition, aromatisation risk, lifestyle and fertility goals; estradiol views are evolving; LH/FSH/prolactin can change management in younger/fertility-concerned men [Urology Times, Pearls & Perspectives interview]. Verdict — supported. Concordant with Endocrine Society/AUA positions (fertility intent changes management categorically; estradiol is physiologically necessary) and the most guideline-aligned creator content reviewed.
• Natural boosters (ashwagandha, tongkat ali, boron, zinc/vitamin-D-rich foods, sleep, weight loss) can boost testosterone [Testosterone 101 page]. Verdict — mixed, mostly low-tier. Sleep, weight loss and resistance training raising endogenous testosterone in deficient states: supported by physiology/observational and intervention literature for lifestyle. The named supplements: small, short, heterogeneous trials at best, none approaching replacement effects in hypogonadal men — folklore-to-weak-RCT tier, and presenting them alongside therapy without that grading misleads.

Hunter Williams (podcast / coaching content)

• “Estrogen is not the enemy”: routine aromatase inhibition for the average man on TRT is almost never necessary; the “crush estrogen” mindset is bodybuilding folklore; low estrogen in men causes more body fat, worse bone density, lower libido, mood problems and worse cardiovascular health [podcast episode description, “Estrogen Is Not the Enemy”]. Verdict — mostly supported; the best-aligned creator claim in this review. Finkelstein 2013 (RCT) directly supports fat (estradiol-deficiency-driven) and sexual function (both hormones). Bone: supported by aromatase-deficiency and AI literature/guideline caution. Libido/mood: sexual function supported; mood specifically is weaker (Ponce 2018 found no mood effect of testosterone itself). Cardiovascular harm of low estradiol in men: plausible/observational, not RCT-proven — that one clause outruns the evidence. “Almost never necessary” is a practice stance, not a trial result, but it points the same way the guidelines point.
• Pregnenolone for men on TRT: TRT suppresses LH/FSH, lowering testicular upstream-steroid production; pregnenolone (as neurosteroid precursor to allopregnanolone) often improves sleep, stress tolerance and mood [podcast description via Castbox]. Verdict — mechanism real, intervention unproven. LH/FSH suppression and intratesticular collapse are documented. That supplemental pregnenolone improves any outcome in men on testosterone: no RCT located — folklore/mechanistic tier, same grading as Vigorous Steve's backfill claim. Two creators converging on a claim is community consensus, not evidence.

TRT Dad Unscripted (host Matt Jimenez; podcast)

• You can look great on TRT and still be metabolically unhealthy; testosterone is not a substitute for fixing nutrition, training, sleep and lifestyle; TRT shouldn't be used as a metabolism fix [Ep. 84 description]. Verdict — supported. TTrials (no vitality benefit; modest effects), and the T4DM-vs-TRAVERSE-diabetes contrast (benefit appeared with a lifestyle programme, not without) — this framing is what the trials actually say.
• Low testosterone as a root-cause / ignored factor in chronic disease (obesity, diabetes, heart disease, depression, metabolic dysfunction) [Ep. 63 description]. Verdict — partly supported, direction-of-causation warning. Low testosterone is associated with all of these observationally; obesity and illness also lower testosterone (reverse causation is documented physiology — acute illness, opioids, obesity suppress the axis). Replacement reversing chronic disease is not shown, except the narrow T4DM prevention result. Association → “root cause” is the layer's standard overreach.
• Sleep apnoea as the hidden problem sabotaging men whose testosterone “looks great” [Ep. 87 description]. Verdict — partly supported. OSA lowering testosterone and TRT mildly/transiently worsening OSA are both real; the episode's instinct to look at sleep before escalating hormone exposure is concordant with guideline cautions. Magnitude in either direction is modest in trials.

Dan the Bodybuilder (Dan the Bodybuilder in Thailand; host, The Steroids Podcast)

• A gram-level weekly testosterone exposure is “the threshold dosage” where “the full power of testosterone” is experienced and serious bodybuilding size is made; a lower range is described as insufficient for those goals, with “strong diminishing return” only beyond his stated band [testosterone-dosage page, bodybuilderinthailand.com — reported claim, reproduced only to grade it; not a recommendation]. Verdict — contradicted in its specifics, folklore in its structure. (i) Bhasin 2001 found no threshold: response rose log-linearly from sub-replacement to the highest studied exposure — a threshold at any specific community number is incompatible with the only graded human data. (ii) No human trial at or above his claimed threshold exists, so “full effects begin there” is untestable-by-design folklore. (iii) The diminishing-returns direction is the one part the data support — but in log-linear form diminishing returns operate at every doubling, not only above a magic band. (iv) The harm side of his framing (“increased risk and responsibility,” gynecomastia, water retention, blood pressure, prostate symptoms at that exposure unless ancillaries are used) is directionally consistent with dose-scaling harms — at exposures where, unlike the trials, no monitoring or exclusion criteria exist.
• Cypionate vs propionate injection-site reactions / “testosterone flu” as a discussion topic distinguishing esters [Steroids Podcast Ep. 4 topic list]. Verdict — anecdote tier. Injection-site reactions are label-listed for androgens generally; an ester-comparative incidence from controlled data was not located. A topic list is not a finding.
• “You will never look like a bodybuilder without hardcore supplementation” [Get Big page]. Verdict — untestable as stated / rhetorical. Competitive-bodybuilder physiques at the elite level are universally associated with PED use in the observational and historical record, but the absolute claim about any individual's ceiling is not a scientific proposition and no trial could test it. Included because it is the premise the threshold claim rests on.

Creator-layer synthesis. Ranked by alignment with primary evidence on the claims actually indexed: Hunter Williams' estradiol position and TRT Dad's lifestyle framing sit closest to the trials; Tatem is guideline-concordant on management and fertility but produced the clearest factual overstatement (TRAVERSE as cardiovascular benefit) and an over-flat prostate answer; Vigorous Steve and Big Paul are directionally harm-reductionist (start low / let bloodwork rule) while packaging untested protocol specifics (daily micro-administration as superior, neurosteroid backfill) with the same confidence as the supported parts; Dan the Bodybuilder's signature quantitative claim is contradicted by the only graded human dose–response data and is structurally unfalsifiable above the trial ceiling. The recurring failure mode across the layer is not inventing effects — it is assigning trial-level certainty to mechanism-level or anecdote-level claims.

Sources: Vigorous Steve (MuscleChemistry interview repost; Q&A transcript; ThinkSteroids thread) · Anabolic Bodybuilding / Big Paul (Podbean; Clipzag) · Tatem (ExcelMale, Testosterone 101; Urology Times; Asprey podcast) · Hunter Williams Podcast (podcast.app; Castbox) · TRT Dad Unscripted (episode descriptions, Eps. 63/84/87) · Dan the Bodybuilder in Thailand (bodybuilderinthailand.com; Steroids Podcast Ep. 4)

Claim check

Popular claims about testosterone, traced to their origin and graded:

• Not supported

  “Testosterone causes / fuels prostate cancer.”

  Myth as stated — human RCT + observational evidence against it. Origin: Huggins & Hodges, Cancer Res 1941 — castration/estrogen caused metastatic prostate cancer to regress (real, Nobel-winning work), and testosterone administration was said to activate it, on results from essentially one non-castrated patient (Morgentaler, Eur Urol 2006). The logical error — “lowering T shrinks cancer, therefore raising T must grow it” — is resolved by the saturation model (Morgentaler & Traish, Eur Urol 2009): maximal prostatic androgen stimulation occurs at low T concentrations (receptor saturation, in the castrate-to-low-normal range); above it, added T adds little. TRAVERSE's prostate substudy found no significant difference in high-grade or any prostate cancer over ~3 y in screened men. Residual truth: androgen deprivation genuinely treats prostate cancer; men with known/suspected prostate cancer remain a label contraindication; unscreened/high-risk and long-horizon data do not exist.

• Not supported

  “Bioidentical means risk-free — natural means safer.”

  Folklore / marketing — contradicted. “Bioidentical” is a marketing term, not a pharmacological category: all approved testosterone is bioidentical — the molecule is identical to endogenous testosterone. The term is most often used to sell compounded preparations, which are not FDA-approved: FDA does not review compounded drugs for safety, efficacy or quality, and the National Academies' 2020 report on compounded bioidentical hormone therapy found the evidence did not support broad clinical use. Risk follows dose, route and exposure — every harm in the safety table above is a harm of the bioidentical molecule itself.

• Not supported

  Ester superiority — “cypionate is smoother/stronger than enanthate”; “propionate is drier”; “acetate is best for X.”

  Folklore. Forum/vendor tables assigning personalities to esters ultimately confuse curve shape with drug identity. Cypionate ≡ enanthate head-to-head (Schulte-Beerbuhl 1980); no ester-vs-ester outcome trial exists for any pair; acetate has no human PK data at all.

• Needs the qualifier

  “More is linearly better.”

  Mis-statement of human RCT data. The Bhasin dose–response is log-linear to the studied ceiling — diminishing per-mg returns throughout, harms scaling on the same curve, and a hard evidence stop at the ceiling.

• Needs the qualifier

  “Testosterone causes heart attacks.”

  Tested and not supported at replacement exposure — human RCT. Origin of the 2010s form: small, flawed observational studies and the TOM trial's early stop (209 older men with mobility limitation and heavy disease burden; CV-related events 23 vs 5 — a real trial signal in a frail population, on a broad event definition); FDA added CV warnings in 2015 and required TRAVERSE. Result: HR 0.96 for MACE, and FDA removed the boxed CV language in 2025 — while adding the BP warning, and while TRAVERSE's AF/AKI/VTE/fracture signals stand. “Proven heart-safe for everyone at any exposure” would be the mirror-image myth: supratherapeutic AAS cohorts show the opposite association.

• Not supported

  “TRT is an anti-ageing / vitality / cognition treatment.”

  Contradicted — human RCT. Clinic marketing built on symptom overlap with ageing. TTrials: no vitality benefit, no cognition benefit, no primary walking benefit; Ponce 2018: no energy/mood effect. The benefits are specific (sexual function, anaemia, bone density, mood slightly) and in deficient men.

• Not supported

  “Estradiol must be blocked/crushed on testosterone.”

  Contradicted — human RCT. Bodybuilding practice imported into TRT clinics (routine aromatase-inhibitor co-prescription). Finkelstein 2013: estradiol deficiency caused fat gain and contributed to sexual dysfunction; bone harm from AI use is the guideline concern. Gynecomastia risk is real but is an argument about exposure, not for universal suppression.

• Needs the qualifier

  “Testosterone is reliable contraception.”

  Partly supported, dangerously incomplete. 96.6–98.6% efficacy among men who suppressed to azoo-/severe oligozoospermia in the WHO trials — but ~35% did not reach azoospermia at 6 months. Not approved, not reliable for an individual.

• Needs the qualifier

  Fixed half-life tables for every ester (incl. acetate).

  Mixed: two rows sourced, the rest folklore. Vendor/forum charts copying each other; the cypionate “~8 days” and propionate “~0.8–2 days” figures do trace to a label and a 1976 study respectively — the acetate and several other rows trace to nothing. Any table that presents all rows with equal confidence is misrepresenting the record.

• Not supported

  “PSA rising on TRT means cancer is developing.”

  Contradicted in pattern — human RCT. Conflation of the expected small PSA rise with carcinogenesis. TRAVERSE: small early PSA rise, no widening after 12 months, no cancer excess. PSA monitoring exists to catch the uncommon true signal against this expected background, per guideline thresholds.

Regulatory status

• United States: Testosterone cypionate — FDA-approved prescription product (Depo-Testosterone and generics), Schedule III controlled substance, IM use per label; indications and diagnostic-confirmation requirement as in the Health & TRT layer. Testosterone propionate — historically approved (1974), US human products discontinued. Testosterone acetate — no FDA-approved product located. Other approved testosterone forms (enanthate, undecanoate IM/oral, gels, pellets, nasal, buccal historically) are outside this page's ester scope but share the class labelling.
• February 2025 class labelling: TRAVERSE results added; boxed CV-risk language removed; class-wide blood-pressure warning added/strengthened.
• Sport: testosterone in all forms is prohibited at all times in WADA-governed sport (S1 anabolic agents); detection per the Pharmacology section.
• Women: no FDA-approved testosterone product for women in the US; any such use is off-label/compounded — relevant because “bioidentical” marketing (Claim check) is aimed largely at this market.

Open questions

Marked plainly, because they are open — where the evidence genuinely runs out:

• Long horizons. The longest large placebo-controlled safety record is ~33 months (TRAVERSE) at replacement exposure. Lifetime replacement — decades — has no RCT; prostate, cardiovascular and renal outcomes beyond ~4 years are observational only.
• Everything above the Bhasin ceiling. No controlled human data on single-agent exposures above the highest graded study exposure, on durations beyond 20 weeks at supratherapeutic exposure, or on any multi-compound combination. Dose–response, harm rates and reversibility in actual bodybuilding practice are unknowable from current designs — only bounded by poly-drug observational cohorts.
• Acetate (and propionate, modernly). No human PK study of testosterone acetate was located at all; propionate's human PK rests on small 1970s studies in 6–11 men, as does much of cypionate's primary PK (Nankin n=11; Schulte-Beerbuhl n=6). Modern, adequately sized, head-to-head ester PK — let alone outcome — trials do not exist and are unlikely to be funded.
• Fractures. Testosterone raises BMD (TTrials) yet was associated with more clinical fractures in TRAVERSE (HR 1.43). Mechanism unknown (falls/behaviour? bone quality vs density? chance?) — a live contradiction between a surrogate and an outcome.
• TRAVERSE's secondary signals. Atrial fibrillation, acute kidney injury, VTE/PE and arrhythmia excesses were not primary endpoints, are unreplicated, and have no established mechanism (haematocrit-mediated viscosity is hypothesised, not shown). Whether they are real, and whether they apply to injections (higher peaks, higher haematocrit rises) more than to the gel studied, is unknown — arguably the most important near-term safety question, since injectables are the dominant real-world formulation.
• Reversibility after long supratherapeutic use. Sperm/HPTA recovery probabilities after years of multi-drug exposure, and whether AAS-associated cardiac remodelling normalises after cessation (cross-sectional data show abnormalities in former users), have no prospective controlled answer.
• Who benefits most? TTrials/Ponce effect sizes are averages over broadly symptomatic older men; no validated predictor (baseline level, symptom cluster, SHBG/free T) reliably identifies responders. Finkelstein's wide inter-individual thresholds suggest “deficiency” is partly individual — guidelines still operate on population cut-offs.
• Estradiol management. That estradiol is necessary is RCT-settled; what (if any) estradiol level or symptom pattern justifies intervention in a man on testosterone, and with what, has never been RCT-tested. Both “always block” and “never measure” exceed the evidence.
• Frequent-administration superiority. Daily/micro-fractionated and subcutaneous regimens are widely advocated by creators on PK-smoothing logic. No RCT compares them with conventional intervals on symptoms, haematocrit, estradiol or any clinical outcome at matched weekly exposure.
• Younger men, and men without classic hypogonadism. The efficacy trials are in older, unequivocally low men; the dose–response trials are in healthy young men at mostly supratherapeutic exposure. The large middle — young/midlife men with borderline values and non-specific symptoms, the demographic most marketed to — is the population with the least direct RCT evidence for benefit.

Sources

1. Depo-Testosterone (testosterone cypionate) — DailyMed FDA label (indications, ~8-day half-life, 98% binding, ~90% urinary excretion, Schedule III, diagnostic confirmation). dailymed.nlm.nih.gov (Depo-Testosterone label)
2. Lincoff AM et al. Cardiovascular safety of testosterone-replacement therapy (TRAVERSE). NEJM 2023. PMID 37326322. pubmed.ncbi.nlm.nih.gov/37326322
3. Bhasin S et al. TRAVERSE prostate substudy. J Clin Endocrinol Metab 2024. academic.oup.com (JCEM 2024) · TRAVERSE label-carried adverse-event rates (arrhythmia 5.2/3.3, AF 3.5/2.4, AKI 2.3/1.5, fracture 3.5/2.5, VTE 1.7/1.2). prnewswire.com (Tlando approval release)
4. Snyder PJ et al. Effects of testosterone treatment in older men (Testosterone Trials). NEJM 2016. PMID 26886521. wikijournalclub.org (Testosterone Trials, 2016) · TTrials lessons/review (Snyder; Matsumoto 2019). pmc.ncbi.nlm.nih.gov/articles/PMC6287281
5. TTrials 2017 companion results — anaemia 54% vs 15% / 52% vs 12%, bone, plaque, cognition (NIH/EurekAlert release). eurekalert.org/news-releases/676913 · ClinicalTrials.gov TTrials record/results, NCT00799617 (bone vBMD 7.5 vs 0.8 mg/cm³; plaque volume). cdek.pharmacy.purdue.edu/trial/NCT00799617
6. Fernández-Balsells MM et al. (2010) & Calof OM et al. (2005), adverse-effect meta-analyses (Hct +3.18 points, Hb +0.80 g/dL, HDL −0.49 mg/dL; erythrocytosis OR 3.69); Ponce OJ et al. (2018) figures as reported therein and in the research file. pmc.ncbi.nlm.nih.gov/articles/PMC5080989
7. Whitsel EA et al. Intramuscular testosterone esters and plasma lipids in hypogonadal men: a meta-analysis. Am J Med 2001. PMID 11566455. pubmed.ncbi.nlm.nih.gov/11566455 · Nackeeran et al., network meta-analysis — haematocrit by formulation. thebh.us (haematocrit on TRT)
8. Wittert G et al. T4DM, Lancet Diabetes Endocrinol 2021 (12% vs 21%). fcmhw.org.au (T4DM) · TRAVERSE diabetes substudy contrast (secondary summary). veedma.com (low-testosterone triangle)
9. WHO Task Force male contraceptive trials (1990/1996 — ~65% azoospermia, efficacy among suppressors; recovery). pmc.ncbi.nlm.nih.gov/articles/PMC2664382
10. Hoyos CM et al. (2012) sleep-apnoea RCT (ODI +10.3/h at 7 wk, NS at 18 wk) & Payne et al. review (secondary summaries). researchgate.net (OSA and testosterone therapy) · trtcatalog.com (TRT & sleep apnea)
11. Morgentaler A. Testosterone and prostate cancer: an historical perspective on a modern myth. Eur Urol 2006 (Huggins 1941 single-patient trace; saturation model). reference.medscape.com/medline/abstract/16875775
12. Nankin HR. Hormone kinetics after intramuscular testosterone cypionate. 1987. PMID 3595893. pubmed.ncbi.nlm.nih.gov/3595893
13. Schulte-Beerbuhl M, Nieschlag E. Comparison of testosterone, dihydrotestosterone, luteinizing hormone, and follicle-stimulating hormone in serum after injection of testosterone enanthate or testosterone cypionate. Fertil Steril 1980 — via Nieschlag textbook chapter summary. scribd.com (pharmacology of testosterone) · Propionate PK (Nieschlag et al., 1976 — tmax 14 h, t½ 0.8 d, MRT 1.5 d), textbook excerpt. excelmale.com (testosterone propionate PK thread)
14. Bi Y et al. Population pharmacokinetic/pharmacodynamic modelling of depot testosterone cypionate (flip-flop kinetics; dose-related HPTA suppression). 2018. pmc.ncbi.nlm.nih.gov/articles/PMC5915615
15. Handelsman DJ (2021) androgen pharmacology — AR signalling, DHT amplification, 5α-reductase isoforms (Tulane pharmacology wiki summary). tmedweb.tulane.edu (anabolic androgens — pharmacology) · DHT metabolism/amplification review excerpt (Handelsman; ~4% conversion, DHT ~10% of T). excelmale.com (low DHT thread)
16. Testosterone physiology (production 3–11 mg/day, >95% testicular, binding, 10–100 min half-life). sciencedirect.com/topics (testosterone) · Ester chemistry / fractions / cypionate-vs-enanthate comparison table. trtcatalog.com (cypionate vs enanthate)
17. Anti-doping detection — T/E ratio, UGT2B17, IRMS. pmc.ncbi.nlm.nih.gov/articles/PMC3944571 · pmc.ncbi.nlm.nih.gov/articles/PMC4760435 · pmc.ncbi.nlm.nih.gov/articles/PMC2684592
18. Testosterone propionate US status (discontinued human products). en.wikipedia.org (androgens available in the US) · Testosterone acetate — research-chemical status / no clinical product (supplier listing, formula C21H30O3). akumaenterprises.in (testosterone acetate) · Ester list noting acetate has no commercial/clinical availability. testosteroneesters.org (all testosterone esters compared)
19. FDA February 2025 label changes — secondary reporting. andromd.com (FDA testosterone labeling changes) · kymatahealth.com (TRT risks, labs, FDA label) · urologytimes.com (testosterone label updates) · FDA abuse/dependence warning & Schedule III history (RAPS summary). raps.org (FDA abuse/dependence warning) · Compounded bioidentical hormones / National Academies 2020 position (pharmacy summary). bayviewrx.com (hormone replacement therapy)
20. Bhasin S et al. Testosterone dose-response relationships in healthy young men. Am J Physiol Endocrinol Metab 2001. PMID 11701431 (FFM +3.4/+5.2/+7.9 kg; log r=0.73). nutrition-evidence.com (Bhasin 2001) · Bhasin S et al. (1996), NEJM supraphysiological trial (FFM +3.2 no-exercise / +6.1 with exercise) — secondary summary. doctor-muscle.com (Bhasin 1996 summary)
21. Finkelstein JS et al. Gonadal steroids and body composition, strength, and sexual function in men. NEJM 2013. PMID 24024838 (abstract). sigmaaldrich.com (Finkelstein 2013 abstract)
22. Windfeld-Mathiasen J et al. Danish AAS mortality cohort. JAMA 2024 (HR 2.81; n=1,189 vs 59,450). jamanetwork.com (JAMA 2024) · AAS cardiovascular status, JAMA Network Open 2025. jamanetwork.com (JAMA Network Open 2025) · Circulation 2025 HRs via review. rpthjournal.org (Circulation 2025 review)
23. Creator sources, cited as claims graded in the Creator layer, not evidence — Vigorous Steve: “full HRT”/daily micro-administration/DHEA-pregnenolone/DIM (interview repost). musclechemistry.com (Vigorous Steve interview) · Start-low Q&A transcript. youtubetotranscript.com (U05ugUedLQM) · Daily/MCT discussion. thinksteroids.com (carrier oil thread)
24. Creator sources — Big Paul / Anabolic Bodybuilding, episode descriptions. anabolicbodybuilding.podbean.com · Cruise/health-markers episode. clipzag.com (Anabolic Bodybuilding episode)
25. Creator sources — Tatem A. Testosterone 101 (attributed). excelmale.com (Testosterone 101) · Urology Times, Pearls & Perspectives interview. urologytimes.com (Tatem interview) · Daily-microdosing interview description. daveasprey.com/1424-alex-tatem
26. Creator sources — Hunter Williams. “Estrogen Is Not the Enemy.” podcast.app (Estrogen Is Not the Enemy) · Pregnenolone episode description. castbox.fm (Hunter Williams Podcast) · TRT Dad Unscripted, episode descriptions (Eps. 63/84/87). ie.radio.net (TRT Dad) · audible.com (TRT DAD Unscripted)
27. Creator sources — Dan the Bodybuilder. “Threshold dosage” claim page. bodybuilderinthailand.com/testosterone-dosage · Get Big page. bodybuilderinthailand.com (how to get bigger) · Steroids Podcast Ep. 4 topics. podm8.com (Steroids Podcast Ep. 4)
28. Testosterone — profile & claim check, research report filed 27 Oct 2026. Every figure, attribution and hedge on this page follows that report.

Molecule

191-amino-acid protein

Class

Recombinant human GH · somatropin

US status

Rx-only · § 333(e) controlled distribution

Sport

WADA S2 · prohibited at all times

The health layer — deficiency, and what replacement demonstrably does

Adult GH deficiency is almost never a stand-alone ageing phenomenon. In the guidelines and reviews it is a consequence of hypothalamic–pituitary disease — pituitary tumours and their surgery or radiotherapy, other sellar lesions, traumatic brain injury, subarachnoid haemorrhage, or childhood-onset deficiency that persists into adult life — usually alongside other pituitary hormone deficiencies. Diagnosis requires a high pre-test probability plus biochemical confirmation, normally a GH stimulation test (the insulin tolerance test is the reference test in both the Endocrine Society guideline, as summarised in the Korean Endocrine Society position statement, and the AACE/ACE 2019 guideline). A random GH or a single IGF-1 is not sufficient, and a normal IGF-1 does not exclude deficiency. The one widely accepted shortcut is a low IGF-1 in a patient with three or more other pituitary hormone deficiencies and typical clinical features, where stimulation testing may be omitted.

The adult GH-deficiency syndrome

Findings in untreated adult GH deficiency (Cuneo / Carroll, summarised in Kargi & Merriam 2013)

Domain	Finding
Body composition	Increased fat mass, especially central/visceral adiposity; decreased lean body mass.
Muscle / function	Decreased muscle strength, decreased exercise performance, decreased cardiac capacity.
Bone	Decreased bone mineral density, increased fracture risk.
Lipids / cardiovascular risk	Atherogenic profile — raised total and LDL cholesterol, raised triglycerides, low HDL.
Skin	Thin, dry skin.
Psychological	Psychosocial problems and decreased quality of life, fatigue, depression, anxiety, impaired sleep, social isolation.

These features overlap with normal ageing — which is precisely why the guidelines insist on stimulation testing in a plausible clinical context rather than treating an age-related IGF-1 decline. Elderly patients with true pituitary deficiency can be distinguished from healthy elderly people, whose lower GH secretion is physiological.

What replacement demonstrably improves

Replacement outcomes in confirmed adult GH deficiency, with evidence tier

Outcome	Effect (named source)	Grade
Lean body mass	Meta-analysis of 22 placebo-controlled RCTs, 591 GH-treated / 562 placebo, 6 months (Li et al., Endocrine 2015): +2.61 kg on GH vs +0.04 kg on placebo (p < 0.0001); dose-related in that analysis.	Human RCT meta-analysis
Fat mass	Same meta-analysis: −2.19 kg on GH vs +0.31 kg on placebo (p = 0.0002).	Human RCT meta-analysis
Lipids	Same meta-analysis: total cholesterol −0.38 mmol/L vs +0.01; LDL cholesterol −0.42 vs −0.10 mmol/L. A separate meta-analysis of 37 blinded RCTs found consistent total/LDL benefit but inconsistent HDL and triglyceride effects.	Human RCT meta-analysis
Body composition, long term	39 adults with adult-onset deficiency followed 33 months after an initial RCT: fat mass −2.18 (±4.87) kg and lean mass +2.01 (±3.25) kg sustained; LDL −0.6 mmol/L, HDL +0.2 mmol/L; fasting insulin rose 110 → 159 pmol/L (p < 0.02).	Human RCT + observational follow-up
Bone mineral density	Retrospective study, 68 adults on individualised replacement for 18 months: BMD first fell to a nadir at ~6 months (early resorption), then rose by month 18; lumbar spine gain exceeded hip; gain was larger in patients with low baseline Z-scores.	Human observational (retrospective)
Quality of life	Same 18-month study (36 patients with QoL data): QoL improved significantly (p < 0.001). Across the wider literature QoL instruments and results are heterogeneous — improvement is the common finding, not a universal one.	Mixed RCT / observational
Strength / exercise capacity in deficiency	Exercise capacity and protein-synthesis increases are reported in deficient adults — unlike in healthy adults (Bodybuilding layer, below) — but effects are smaller and less consistent than the body-composition effects, and at least one 166-patient, 12-month RCT found no significant effect on strength/endurance, QoL or BMD.	Mixed

What replacement has not been shown to do in deficiency: there is no randomised trial with a mortality or fracture endpoint demonstrating that GH replacement reduces death or fractures in adults with GH deficiency. Those remain extrapolations from surrogate outcomes (see Open questions).

Monitoring — IGF-1 is the control variable

Because GH itself is pulsatile (Pharmacology, below), day-to-day management of replacement uses serum IGF-1, interpreted against the age-adjusted reference range, plus clinical response and side effects. The guideline pattern (Endocrine Society guideline as summarised in the Korean Endocrine Society position statement; AACE/ACE 2019): titrate against IGF-1 kept within the age-adjusted reference range — AACE/ACE phrasing is the minimal exposure that normalises IGF-1 without side effects, stated here as the guideline's position, not as advice to any reader. Measure IGF-1 monthly or bimonthly during titration, then about twice a year on maintenance, alongside clinical response and adverse effects. Monitor fasting glucose/HbA1c and lipids at least annually, repeat DXA every ~1.5–2 years where baseline BMD was abnormal, and track waist circumference and quality of life. Thyroid and adrenal replacement needs can change after GH is started, and are reassessed.

Guideline positions — who replacement is actually for

• For: adults (and, under separate paediatric indications, children) with confirmed deficiency — a plausible hypothalamic–pituitary cause plus failed stimulation testing, or the low-IGF-1-plus-multiple-deficits pattern above. Childhood-onset patients are retested at final height unless the cause is known to be irreversible. FDA-labelled somatropin indications are of this character — deficiency, specific paediatric growth conditions, and, for particular brands, HIV-associated wasting and short bowel syndrome.
• Not for: ageing itself, obesity without deficiency, athletic performance, bodybuilding, or “low IGF-1 for age” in an otherwise healthy adult. Liu et al. 2007's conclusion — GH “cannot be recommended as an anti-aging therapy” — is the position the guideline literature has not moved from.
• Contraindicated (FDA labels, all somatropins): active malignancy; active proliferative or severe non-proliferative diabetic retinopathy; acute critical illness after major surgery/trauma or with acute respiratory failure; in children, closed epiphyses and the high-risk Prader-Willi pattern; hypersensitivity to the product.

Sources: Li et al., Endocrine 2015 · Kargi & Merriam, Front Endocrinol 2013 · Korean Endocrine Society position statement (Endocrine Society guidance) · AACE/ACE guideline 2019 · FDA somatropin labels

Pharmacology — the axis, the pulses, the two routes of action

The molecule and the axis

Human GH (somatotropin) is a single-chain 191-amino-acid protein secreted by somatotroph cells of the anterior pituitary. Somatropin is its recombinant form, identical in sequence. Control is a three-level axis with negative feedback. In the hypothalamus, GH-releasing hormone (GHRH) stimulates GH synthesis and release while somatostatin inhibits it; ghrelin — from the stomach and hypothalamus — is a further potent secretagogue acting at the GH secretagogue receptor (GHS-R). The pituitary integrates those signals and releases GH in pulses. In the periphery, GH binds its receptor — a cytokine-receptor family member signalling principally through JAK2–STAT — in liver and other tissues; hepatic receptor activation is the main source of circulating IGF-1. IGF-1 then feeds back to suppress GH, stimulating somatostatin release and suppressing GH transcription, and GH itself short-loop feeds back on GHRH.

Pulsatile secretion — the defining feature

Endogenous GH is not secreted steadily. In healthy adults, pulses occur roughly every 1–3 hours — on the order of ~45 µg per pulse, ~400–500 µg per 24 hours — with the major surge at onset of slow-wave (deep) sleep and smaller pulses after meals. Exercise, fasting and stress stimulate release; raised glucose and free fatty acids suppress it. Secretion falls with age and with adiposity — visceral fat in particular suppresses GH — which is the physiological background against which both the “ageing = deficiency” claim and obesity-related low GH readings must be judged: low output in those settings is common and is not, by itself, deficiency. The consequence for measurement is direct: a single blood GH value is near-uninterpretable, which is why diagnosis uses stimulation tests and management uses IGF-1, stable across the day because most circulating IGF-1 is carried bound to IGF-binding proteins. Exogenous daily somatropin does not reproduce this pattern — a subcutaneous injection produces one absorption-driven peak lasting hours (see the PK table below). Long-acting weekly somatropin preparations exist as separate FDA-approved products for deficiency; they are replacement products for deficient patients, not evidence about non-deficient use.

IGF-1-mediated vs direct effects

Two routes, and what each one carries

Route	What it mediates	Evidence anchor
IGF-1-mediated	Most growth effects: linear growth at the growth plate, much of the anabolic/protein-accretion signalling, collagen expression in tendon/muscle in the Doessing study (tissue IGF-I mRNA correlated with collagen I mRNA). IGF-1 is produced in the liver for circulation and locally in tissues.	Liver-specific GH-receptor deletion lowers circulating IGF-1 drastically yet compromises growth less than whole-body deletion — local IGF-1 contributes independently.
Direct GH effects	Lipolysis, insulin antagonism, and sodium/water retention (antinatriuretic effect, including increased epithelial sodium-channel activity described in acromegaly).	Acromegaly and GH-physiology reviews list the anti-insulin, lipolytic and antinatriuretic actions as GH effects “not mediated by IGF-I”. The FDA Humatrope label shows a single somatropin exposure raising median IGF-1 to a peak only after a median 14 hours (women) / 21 hours (men), while metabolic GH effects are faster.

The practical implication: raising IGF-1 is a marker of GH exposure, not proof that a desired direct effect — or a desired growth effect — has occurred in a particular tissue.

Lipolysis — mobilised is not the same as oxidised

GH is a fasting-adapted, fat-mobilising hormone. It stimulates adipose lipolysis — increasing hormone-sensitive lipase–driven triglyceride breakdown, releasing free fatty acids (FFA) and glycerol — and shifts substrate use toward lipid oxidation while decreasing glucose oxidation. In the Hermansen et al. 2017 meta-analysis of placebo-controlled trials in healthy young adults, GH significantly raised exercising glycerol (p = 0.01) and FFA (p < 0.01) levels, without changing respiratory quotient during exercise (p = 0.30) — more fat was mobilised, which is not the same measurement as more fat being oxidised. Net fat-mass loss in trials is real but modest (Bodybuilding layer), systemic, and not site-selective (Claim check).

The insulin-antagonist action

GH opposes insulin in liver, muscle and fat: it decreases insulin-stimulated glucose uptake and glucose oxidation, increases hepatic glucose production (gluconeogenesis), and — via lipolysis — floods the circulation with FFA that compete with glucose as an oxidative substrate and further impair insulin signalling. Luger et al. (Diabetes Care 2012, KIMS) summarise it directly: GH “decreases glucose oxidation and glucose uptake by muscle and increases gluconeogenesis… stimulates lipolysis, lipid oxidation, and protein synthesis”. This is physiology, not a side-effect surprise — it is why fasting GH protects against hypoglycaemia, and why exogenous GH predictably worsens glucose tolerance in susceptible people (Safety, below). IGF-1, by contrast, has insulin-like glucose-lowering actions; the net glucose effect of GH treatment is the balance, and in trials the GH (diabetogenic) side usually wins early in treatment.

Pharmacokinetics, metabolism and clearance — from the FDA labels

Somatropin PK — Humatrope and Genotropin labels compared

Parameter	Humatrope label	Genotropin label
Absolute bioavailability, subcutaneous	75%	~80%
Tmax, subcutaneous	Median 4.0 h (range 3–8) after 0.1 mg/kg in healthy subjects.	Mean ~5.9 h in adult GH-deficient patients.
Half-life, subcutaneous	3.8 (±1.4) h	3.0 h in GH-deficient adults.
Half-life, intravenous	— (label notes the longer SC value reflects slow absorption from the injection site).	0.4 h — the true elimination half-life; the SC figure is absorption-limited (“flip-flop”).
Volume of distribution	0.96 (±0.3) L/kg	1.3 (±0.8) L/kg
Clearance, subcutaneous	0.18 (±0.03) L/h/kg	0.3 (±0.11) L/h/kg
Metabolism / excretion	“Classical protein catabolism in both the liver and kidneys”; some breakdown products from renal cells return to circulation; intact somatropin in urine not measured (small amounts detected in urine of treated children).	Same statement of liver/kidney protein catabolism.

Two consequences follow. First, endogenous GH's minute-to-minute pulses and injected somatropin's hours-long tail are pharmacokinetically different exposures even before exposure size is considered. Second, somatropin is a protein that is catabolised to amino acids and peptides — it is not converted into other hormones, a point that matters for one creator claim graded below.

Sources: HUMATROPE (somatropin) US prescribing information, DailyMed · GENOTROPIN (somatropin) FDA label, 2024 · GH and metabolic homeostasis, EMJ · GH central regulation review, MDPI

The bodybuilding layer — the evidence in non-deficient adults

Body composition: real, small, and substantially water

Body composition in non-deficient adults — the anchor trials and meta-analyses

Source	Finding	Interpretation in the source
Liu et al. 2007	Healthy elderly, mean age 69; 18 RCT populations, 220 GH completers, mean 27 weeks. Fat mass −2.1 kg (95% CI −2.8 to −1.35); lean body mass +2.1 kg (1.3 to 2.9); total weight unchanged (+0.1 kg, ns).	“Small changes in body composition.”
Liu et al. 2008	Healthy fit young adults, 13–45 y; 27 study samples, 303 GH-treated. Lean mass +2.1 kg (meta-analysis of 9 studies, mean treatment ~4 weeks in the Birzniece/Ho summary).	No accompanying strength gain (below).
Hermansen et al. 2017	Healthy young adults; 11 trials, 254 subjects. Fat mass significantly decreased (p < 0.01); lean mass significantly increased (p < 0.01).	Composition changes without performance changes.
Sergeeva et al. 2019	Well-trained healthy young subjects; 11 RCTs. Fat mass −0.67 kg (CI −0.93 to −0.40); lean mass +2.72 kg (2.02 to 3.38) — with total body water +1.10 kg and extracellular water +1.78 kg.	Authors' conclusion: the lean-mass gain is due to extracellular-water accumulation; no hypertrophy of muscle fibres.
Meinhardt et al. 2010	96 recreational athletes, 8 weeks (trial exposure 2 mg/day — a study design, not guidance). Fat mass significantly reduced; lean mass increased through an increase in extracellular water; body cell mass increased only in men also receiving testosterone.	Fluid accounts for most of the lean-mass signal.
Birzniece / Ho group study	Healthy adults, 8 weeks, double-blind. DXA lean mass +~2.5 kg with a concomitant +~2 L expansion of extracellular water (bromide dilution).	“Strong evidence that fluid retention accounts for most of the increase of LBM induced by GH.”

Read together: the fat loss — roughly 0.7–2.1 kg depending on population and duration — is the most consistent finding in the entire non-deficient literature. The lean-mass figure that dominates marketing is, on the studies that measured water compartments directly, mostly extracellular fluid — which also explains the oedema, joint symptoms and carpal tunnel symptoms in the Safety section: they are the same phenomenon.

Strength and athletic performance

• Strength: no effect. Hermansen 2017: muscle strength p = 0.36 (not significant). Sergeeva 2019: strength effect −0.02 (CI −0.05 to 0.02, p = 0.36) and muscle protein synthesis rate 0.00 (p = 0.70). Liu 2008: lean mass rose while “strength and exercise capacity did not improve”; exercise lactate was significantly higher on GH in 2 of 3 studies that measured it, and Liu's stated conclusion was that “claims that growth hormone enhances physical performance are not supported by the scientific literature.”
• Aerobic capacity: no effect. Hermansen 2017: VO₂max p = 0.89 (not significant). Blackman 2002 (healthy elderly): VO₂max did not improve significantly on GH alone in women or men.
• The exception — sprint capacity. Meinhardt 2010: Wingate sprint capacity +3.9% on GH (men and women combined; 0.71 kJ, CI 0.1–1.3) and +8.3% in men receiving GH plus testosterone; no significant change in endurance, strength or jump height; the sprint effect was not maintained 6 weeks after discontinuation. Hermansen's meta-analysis notes anaerobic capacity was significantly increased in “the only study which investigated this” — this single trial is the whole of that evidence base.
• Resistance-training interaction: no augmentation. In trials combining GH with resistance exercise in young and elderly men (Yarasheski et al.; Lange et al. 2002; Taaffe et al. 1994, cited in the reviews), GH did not augment training-induced muscle hypertrophy or strength beyond training alone.
• Standing caveat in every review: trial exposures and durations (weeks, single-agent) may not match reported community practice, which typically involves longer use and stacking with androgens/insulin. That caveat cuts both ways — higher and longer exposure is unstudied for benefit, while the harms below are known to be exposure-related. It is not evidence that untested regimens work.

Collagen synthesis and tendon — the most interesting, and most over-extended, data

Doessing et al., Journal of Physiology 2010 — double-blind crossover, 10 healthy young men, 14-day trial exposure: tendon collagen I mRNA +3.9-fold and tendon collagen protein synthesis +1.3-fold (p = 0.02); muscle collagen I mRNA +2.3-fold and muscle collagen protein synthesis +5.8-fold (p = 0.06); myofibrillar protein synthesis unaffected. The authors' conclusion: GH “is more important in strengthening the matrix tissue than for muscle cell hypertrophy in adult human musculotendinous tissue.”

Immobilisation and rehabilitation studies (young men, n = 20; elderly men, n = 12) found GH increased collagen/IGF-1 gene expression during immobilisation and retraining and, in the young-men study, tendon stiffness fell during immobilisation only in the placebo group and rose during rehabilitation only in the GH group — but muscle cross-sectional area and maximal voluntary contraction declined and recovered similarly in both groups, and tendon fibril morphology did not differ. These are small, surrogate-endpoint studies in uninjured tissue.

What is missing: any human RCT of GH for an actual tendon, ligament or muscle injury with a healing, re-rupture, return-to-function or pain endpoint. Direct laboratory evidence points the other way: human tendon/ligament fibroblasts exposed to HGH in vitro showed no improvement in proliferation or differentiation and no dose-response for collagen production (“this study does not support the use of HGH for accelerating recovery from injury”), and in a rat rotator-cuff repair model HGH produced no mechanical benefit and significantly worse force-to-failure in one protocol.

Rudman 1990 — what it did and did not show

Rudman et al., New England Journal of Medicine 1990;323:1–6, is the paper the anti-aging industry was built on. Read closely:

Rudman 1990, as published

Element	What Rudman actually was
Design	21 healthy men aged 61–81 selected for low IGF-I (<350 U/L); 12 treated, 9 untreated controls — no placebo, no blinding; 6 months.
Trial exposure	~0.03 mg/kg subcutaneously three times a week — a trial-design exposure, stated to identify the study, not as guidance.
Results	Lean body mass +8.8%, adipose mass −14.4%, lumbar vertebral bone density +1.6% (all p < 0.05); skin thickness +7.1% (p = 0.07, not significant); no significant change in radius or proximal femur bone density.
What it did not measure	Strength, physical function, cognition, quality of life, cardiovascular outcomes, mortality — no functional outcome of any kind.
What happened next	Later controlled trials at similar exposures reported substantial side effects (oedema, carpal tunnel, glucose intolerance — Safety, below) and had to reduce exposures; a longer Rudman-group follow-up reported 9 cases of carpal tunnel syndrome and 4 of gynaecomastia during treatment. The famous sentence — that 6 months of GH reversed “10 to 20 years of aging” in body composition — compares a percentage change in two DXA compartments to ageing, and was never a finding about ageing itself.

The accurate summary is Liu 2007's, 17 years later, pooling the whole literature Rudman started: small composition changes, increased adverse events, cannot be recommended as an anti-aging therapy.

Sources: Liu et al., Ann Intern Med 2007 & 2008 · Hermansen et al. 2017 · Sergeeva et al. 2019 · Meinhardt et al., Ann Intern Med 2010 · Doessing et al., J Physiol 2010 · Rudman et al., NEJM 1990

Safety — with numbers

Common adverse effects in controlled trials (non-deficient adults)

Adverse effects in healthy-volunteer GH trials

Effect	Number and source
Any of oedema / arthralgia / carpal tunnel symptoms	24–46% of GH-treated participants — Blackman et al., JAMA 2002 (131 healthy adults 65–88 y, 26 weeks).
Oedema	Women on GH 39% vs 0% on placebo; GH + hormone therapy 38% vs 0% (Blackman 2002). Liu 2007: soft-tissue oedema significantly more likely on GH.
Arthralgia	Men on GH 41% vs 0% (Blackman 2002, p < 0.01 in the trial report); significantly more likely on GH in Liu 2007. A secondary summary of Liu 2007 put arthralgia at ~21% of GH-treated participants across trials.
Carpal tunnel syndrome / symptoms	Men on GH + testosterone 32% vs 0% (Blackman 2002); significantly increased in Liu 2007's pooled analysis (a secondary summary put it at ~19%); 9 cases in the Rudman-group longer follow-up. In replacement practice in deficient adults, a guideline-era review puts carpal tunnel at ~2% and fluid retention at 5–18% — the contrast with the figures above is the exposure/context difference between replacement and the exposures used in healthy-volunteer trials.
Gynaecomastia	Significantly increased in Liu 2007; 4 cases in the Rudman-group follow-up.
Fatigue	More frequent on GH than control in Liu 2008's athletic-performance review, alongside soft-tissue oedema.
Exposure relationship	In Blackman 2002, the number of oedema/arthralgia/carpal-tunnel events per participant correlated directly with mean IGF-I during treatment (r = 0.41–0.53, p < 0.001) — these are exposure effects, not idiosyncrasies.

Insulin resistance and diabetes

Glucose outcomes across settings

Setting	Number and source
Healthy elderly RCT	Diabetes or glucose intolerance in 18 GH-treated men vs 7 men not receiving GH (p = 0.006); in women the glucose-intolerance increase was non-significant and no diabetes occurred (Blackman 2002).
Healthy elderly meta-analysis	GH-treated participants were “somewhat more likely” to develop diabetes mellitus and impaired fasting glucose — direction consistent, pooled estimate not statistically significant in that synthesis (Liu 2007). A secondary summary put impaired fasting glucose / impaired glucose tolerance / diabetes at ~22% of GH-treated participants.
Deficient adults on replacement — KIMS registry	n = 5,143; 20,106 patient-years; mean follow-up 3.9 y. Diabetes incidence 2.6 per 100 patient-years; 523 patients (10.2%) developed diabetes. Observed/expected 6.02 against a Swedish reference population and 2.11–5.22 against four other reference populations. Incidence was highest in year 1 (4.1/100 patient-years) and fell to 1.0 after >8 years. Those who developed diabetes were older and had higher baseline BMI, waist, triglycerides and blood pressure. No association with GH dose (p = 0.74) or IGF-I SDS (p = 0.47) at 1 year in that analysis (Luger et al. 2012).
Glucose drift on replacement	In the non-diabetic subset: fasting glucose 84.4 → 89.5 mg/dL and HbA1c 4.74% → 5.09% over 6 years of replacement (Luger 2012).
Interpretation	Deficiency itself, adiposity and surveillance all confound the registry comparison, and the authors' own conclusion is limited to incidence appearing increased in GH-deficient patients on replacement with an adverse risk profile at baseline. The RCT data in healthy adults — where confounding by deficiency is absent — show the same direction of effect. There is no setting in which GH improved glucose tolerance.

Acromegaly — the natural model of chronic excess

Acromegaly — usually a GH-secreting pituitary adenoma; gigantism if onset precedes epiphyseal closure — is what uncontrolled GH/IGF-1 excess does over years, and it is the correct frame for thinking about long-term pharmacological exposure:

• Heart: biventricular hypertrophy / acromegalic cardiomyopathy, diastolic then systolic dysfunction, arrhythmia, hypertension — cardiovascular disease is the leading cause of death.
• Metabolic: insulin resistance and type 2 diabetes; dyslipidaemia.
• Joints and soft tissue: cartilage and soft-tissue overgrowth causing destructive, largely irreversible arthropathy/osteoarthritis; carpal tunnel syndrome; sleep apnoea; enlargement of hands, feet, jaw, tongue and internal organs (heart, liver, spleen, kidneys, thyroid).
• Mortality: uncontrolled acromegaly carries at least twice the mortality of the general population (Endotext); older series report standardised mortality ratios of 1.2–3.3, roughly 10 years of life lost if untreated, with deaths attributed ~60% cardiovascular, ~25% respiratory, ~15% neoplasia. Mortality returns toward the general population's when GH/IGF-1 are biochemically controlled — which underlines that the risk tracks the hormone excess itself.
• Relevance and limit: acromegaly involves years of continuous, often very high GH/IGF-1, with a pituitary tumour as confounder. It does not prove that any particular shorter exposure causes these outcomes; it proves the direction of travel of chronic excess — and every element of the acromegaly phenotype (fluid retention, insulin resistance, carpal tunnel, joint disease, cardiac hypertrophy) already appears in miniature in the short GH trials above.

Cancer risk — what the cohorts do and do not cover

Cancer evidence from GH-treated cohorts

Evidence	Finding
SAGhE cohort	23,984 patients treated with rhGH in childhood, mean cancer follow-up 14.8 years/patient. Overall cancer risk not increased in patients treated for growth failure only (isolated deficiency, idiopathic short stature, small-for-gestational-age). Risk was increased in patients with a previous cancer, and in some non-growth-failure diagnoses signals appeared for bone cancer (SIR 2.8) and bladder cancer (SIR 16.3) on very small absolute numbers. Cancer risk was unrelated to duration/cumulative dose overall, but in the previous-cancer group cancer mortality rose with daily dose.
Swedish population cohort 2024	3,408 childhood rhGH patients, 15 matched controls each, median follow-up 19.8 years, up to 35 years. Neoplastic events overall HR 1.28 (95% CI 1.12–1.46), driven by non-malignant events and significant in males only; malignant neoplasms HR 0.91 (0.66–1.26) — no increase. Longer treatment duration was associated with neoplastic events; mean/cumulative dose was not.
Registry review	Pharmaceutical GH registries + SAGhE, 396,344 person-years: overall risk of primary cancer not increased in GH-treated patients without previous risk factors; risk of a second neoplasm is increased in childhood cancer survivors, particularly after CNS radiotherapy — a group in whom the radiotherapy is itself the dominant risk.
Acromegaly cohorts	Active acromegaly is associated with increased colon polyps and, in several cohorts, increased incidence of some cancers (digestive and others reported); neoplasia accounts for ~15% of deaths in older acromegaly series. IGF-1 is mitogenic — the mechanistic basis for concern and for the active-malignancy contraindication.
The gap	No cohort of comparable size or duration exists for healthy adults who used GH for bodybuilding, sport or anti-aging. “No increased cancer in children treated for deficiency, followed to a mean age in their twenties” is a true statement that answers a different question from “is years of adult pharmacological GH use cancer-safe.”

Other labelled warnings (FDA somatropin labels) include intracranial hypertension, fluid retention, hypothyroidism unmasking, pancreatitis (rare), scoliosis progression in children, and lipoatrophy at repeatedly used injection sites — hence the label instruction to rotate sites (see Claim check, on “spot reduction”).

Sources: Blackman et al., JAMA 2002 · Liu et al. 2007 & 2008 · Luger et al., Diabetes Care 2012 (KIMS) · SAGhE cohort · Swedish cohort, Front Endocrinol 2024 · Acromegaly, Endotext · FDA somatropin labels

Creator layer — opinion, graded claim by claim

Everything in this section is creator and community content, kept separate from the primary literature above precisely so it is not mistaken for it. Claims are included only where they are attributable to the named creator in a retrieved source — a transcript, the creator's own video description, or their channel's own episode description. Where only an episode title or description could be retrieved and not the audio argument itself, the entry says so and the verdict grades the claim as framed, with that limitation stated. Verdicts: supported / partly supported / contradicted / unproven.

Vigorous Steve

Source: Mark Bell's Power Project Ep. 1010, transcript (Podscripts) — full transcript retrieved.

• GH is “a very expensive fat loss aid because it does cause lipolysis, fat loss from adipose tissue through hormone-sensitive lipase” (24:16). Verdict — supported on mechanism and direction; the magnitude is modest. GH-driven lipolysis is established physiology (Pharmacology, above); pooled fat loss in non-deficient adults is ~0.7–2.1 kg (Bodybuilding layer) — which is what “expensive fat loss aid” implies.
• GH “can help with skin rejuvenation… fills in your wrinkles, makes you look more youthful, your nails grow better, your skin grows better, your hair grows better” (24:16, 26:52). Verdict — partly supported. Skin thickness rose 7.1% in Rudman 1990 but at p = 0.07 (non-significant), and skin changes are documented in deficiency replacement. No human RCT measures wrinkle appearance, hair or nails as outcomes in non-deficient adults.
• “People think it's an anti-aging drug. In reality, probably it's not because it also raises IGF-1” — with raised IGF-1 proposed to age you faster via cell division/telomere shortening (24:16–25:26). Verdict — partly supported. The conclusion matches the evidence (Liu 2007: cannot be recommended as anti-aging). The telomere-shortening mechanism is a plausible-sounding extrapolation with no human trial showing GH accelerates ageing by that route — that step is unproven.
• “It doesn't seem to have a huge performance enhancing benefit. Maybe through the liberation of fat stores” before training (27:20). Verdict — supported (first half); unproven (second half). Strength and aerobic capacity do not improve in meta-analyses; the single contrary signal is Meinhardt's 3.9% sprint finding. No trial tests pre-workout fat liberation as a performance mechanism.
• Doubt that GH regenerated his thymus — “otherwise all the bodybuilders would have giant thymuses” (30:12). Verdict — partly supported (as scepticism). Thymus-regeneration claims rest on one very small, multi-drug exploratory study (GH combined with other agents, epigenetic-clock endpoints); there is no evidence of thymic regrowth as a bodybuilding outcome. His stated reason is anecdote, but the sceptical conclusion is where the evidence sits.

Overall: the most evidence-concordant of the creators graded here on the claims retrieved — his headline positions (fat loss yes, anti-aging no, performance unimpressive) are the trial literature's positions.

Big Paul — Paul Barnett (Anabolic Bodybuilding)

Sources: his channel's own episode descriptions (Podbean channel page; Listen Notes episode listings) — descriptions retrieved, full audio arguments not retrieved.

• HGH affects blood glucose and insulin resistance, and can worsen hand numbness and carpal tunnel symptoms — topics he presents as real effects to manage. Verdict — supported (as framed). Both are documented, exposure-related effects: diabetes/glucose intolerance 18 vs 7 in Blackman 2002; carpal tunnel significantly increased in Liu 2007 and 32% vs 0% for GH + testosterone in Blackman (Safety, above).
• “The GH Paradox: why growth hormone makes you look 20 years younger on the outside but might be accelerating the clock on the inside” (episode title/framing). Verdict — partly supported. “Look younger” overstates a non-significant skin-thickness finding (Rudman, p = 0.07) into a 20-year claim no trial supports; “accelerating the clock inside” is unproven as stated, though chronic excess (acromegaly) does shorten life. The paradox framing is directionally reasonable; the numbers in it are not evidence.
• What “really causes acromegaly and ‘bubble gut’ and whether HGH or insulin are to blame” (episode topic). Verdict — unproven (as a bodybuilding claim). Acromegaly's cause is established — a GH-secreting pituitary adenoma in the overwhelming majority — it is not caused by taking GH, though exogenous excess can produce acromegaly-like features. “Bubble gut” (distended abdomen in bodybuilders) has no controlled-study literature attributing it to GH, insulin, food volume, or any combination; any confident attribution exceeds the evidence. The substantive answer given in the audio was not retrieved, so only the framing is graded.
• A comparison of the pros and cons of two different community GH dose levels (episode topic; the levels are claim content and are not reproduced here as guidance). Verdict — unproven. No human RCT compares community dose levels of GH in bodybuilders for any benefit or harm outcome; dose-response for harms exists only across the replacement-to-trial range covered on this page.

Overall: his harm framing (glucose, carpal tunnel, water retention) is evidence-concordant. Efficacy and “bubble gut” content could not be verified beyond episode descriptions.

Dr. Alex Tatem

Sources: his own YouTube video “Sermorelin: The ‘Safer HGH’ Has One Huge Problem” (video description, in his own words); “Peptide Expert” interview transcript; a secondary summary of his peptide tier-ranking — flagged as secondary wherever used.

• “Sermorelin isn't growth hormone. It's a fragment of growth hormone-releasing hormone (GHRH) that signals your pituitary to release your own GH” (own video description). Verdict — supported. Correct pharmacology: sermorelin is a GHRH analogue/secretagogue, not somatropin; it acts upstream and depends on a functioning pituitary (Pharmacology, above).
• “If you're expecting a nightly injection to pack on muscle like pharmaceutical growth hormone, the research tells a very different story” — with human trials producing body-composition changes using GHRH doses in milligrams while many compounded sermorelin protocols are in hundreds of micrograms (own video description; the dose figures are the content of his claim). Verdict — supported. Even pharmaceutical GH itself does not increase contractile muscle protein synthesis in non-deficient adults (Doessing 2010; Sergeeva 2019), so a secretagogue producing a smaller, physiological GH rise cannot be expected to do more. The dose-gap observation is his; it is consistent with the trial literature's exposures.
• “The evidence for improved sleep is especially interesting” for sermorelin/GHRH (own video description). Verdict — partly supported. GH's major pulse is tied to slow-wave sleep and GHRH-axis manipulation affects sleep architecture in small human studies; there is no large RCT showing a clinically meaningful sleep outcome for compounded sermorelin in healthy adults.
• MK-677 (ibutamoren) “binds to [the] ghrelin [receptor] and actually stimulates the release of significant growth hormone” and “would actually stimulate hunger a profound amount” (Peptide Expert interview, 0:14:33). Verdict — supported. Correct: ibutamoren is an oral ghrelin-mimetic GH secretagogue (GHS-R agonist) that raises GH/IGF-1 and appetite in human trials. In the best-known such trial in older adults, fat-free mass rose ~1.1 kg vs placebo without gains in strength or function — the pattern of the Bodybuilding layer.
• GH itself is the “S-tier” benchmark, “proven for muscle, fat loss, collagen, and tissue repair” (as reported by the secondary tier-list summary). Verdict — partly supported. Fat loss: supported (modest). Collagen synthesis: supported as a surrogate (Doessing 2010). “Muscle”: contradicted if it means contractile muscle — myofibrillar synthesis and strength do not increase. “Tissue repair”: unproven — no human injury-healing RCT. Because this attribution comes from a secondary summary rather than a retrieved primary transcript, it should be re-checked against the original video before republication.

Overall: on the directly retrieved material (sermorelin), Tatem is notably sceptical and evidence-concordant. The tier-list claim, if accurately reported, overstates GH on muscle and repair.

Hunter Williams

Source: “HGH vs Peptides: Should You Combine Both for Maximum Results?”, The Hunter Williams Podcast, 15 July 2025, full transcript (Podscan) — retrieved in full.

• On GH/peptide cancer fears: “they're not going to cause cancer by the way” (00:39). Verdict — contradicted (as an absolute). The accurate statement is narrower: no increased malignant neoplasia was found in childhood-treatment cohorts without prior risk factors, followed into early/mid-adulthood. Active malignancy is a labelled contraindication, acromegaly cohorts show cancer signals, and there are no long-term data in adult non-deficient users. A blanket “will not cause cancer” is not a claim the evidence can carry.
• GH is “the gold standard… you can control the exact dose… results seem to be better than growth hormone peptides”, and using GH and peptides together at different times of year “can be beneficial” (00:39–01:09). Verdict — partly supported / unproven. It is true that exogenous somatropin bypasses pituitary control and raises IGF-1 reliably, and secretagogue outcome data are weaker still. But “better results” for muscle/performance is unproven for both, and no RCT tests GH-plus-secretagogue combinations at all — the combination claim is unproven.
• Exogenous HGH produces a sustained “square wave” that “produces stronger anabolic signals” than pulsatile peptide-driven GH (03:52). Verdict — partly supported. The pharmacokinetic contrast is real (Pharmacology, above: hours-long absorption tail vs minutes-long pulses). “Stronger anabolic signals” does not translate into the outcome that matters: myofibrillar protein synthesis and strength do not increase on GH in non-deficient adults.
• Pituitary suppression from exogenous GH is “more or less overblown” and endogenous production returns “in as quick as three days” after stopping (04:17). Verdict — unproven. Negative feedback of GH/IGF-1 on the axis is established physiology; recovery of endogenous secretion after exogenous GH is plausible, and GH suppression is generally considered shorter-lived than testosterone suppression — but no source retrieved in this research supports a specific “three days” figure.
• “Both HGH itself and GHRP/GHRH therapies can increase lean body mass by stimulating protein synthesis and muscle cell growth… HGH injections reliably increase IGF-1, which directly promotes muscle hypertrophy… [secretagogues] support new muscle cell creation… CJC and ipamorelin combo accelerates muscle fiber development” (08:14). Verdict — contradicted (for contractile muscle) / unproven (for the peptide stack). Directly refuted for GH by Doessing 2010 (myofibrillar synthesis unaffected) and Sergeeva 2019 (muscle protein synthesis effect 0.00); DXA lean mass rises via extracellular water. For CJC-1295 + ipamorelin there is no controlled human trial showing any body-composition, strength or muscle-fibre outcome — only short pharmacokinetic studies showing GH pulses.

Overall: the least evidence-concordant creator graded here on the claims retrieved — the central muscle/hypertrophy framing is contradicted by the primary literature, and the cancer reassurance is stated more absolutely than the evidence allows.

TRT Dad Unscripted

No HGH-specific claim attributable to TRT Dad was found in the indexed episode descriptions or transcripts retrieved (podcast feeds and episode listings on Listen Notes, TopPodcast and Podscan, read 27 October 2026). The retrieved catalogue covers TRT itself, estrogen management, libido/fertility, body composition on TRT (Ep. 78), and a general discussion of an FDA meeting on peptides (Ep. 77) whose description makes no specific HGH claim. Grading a claim that could not be retrieved would require inventing one, so none is graded. Verdict — could not verify. This is a gap in this page's research, flagged again in Open questions.

Dan the Bodybuilder — Steroids Podcast

Sources: episode listings with timestamped contents (Ivy.fm podcast page; Spotify/creator listing carrying Episode 55 contents) — listings retrieved, full audio transcripts not retrieved.

• “Growth Hormone fat burning and recomposition effects” and “How HGH burns fat — mechanism” (Ep. 55, segments at 46:47 and 52:00). Verdict — partly supported (as framed). Fat burning via lipolysis: supported (Pharmacology and Bodybuilding layer), with modest pooled fat loss. “Recomposition” in the sense of simultaneous contractile-muscle gain: contradicted — lean-mass gains are predominantly extracellular water and strength does not increase. The mechanism as argued in the audio was not retrieved.
• “Growth Hormone Degrading into Prolactin type hormones and sexual dysfunction from generic growth hormone in bodybuilding” (Ep. 26, segment at 43:05). Verdict — contradicted (as titled). Somatropin is catabolised by classical protein catabolism in liver and kidneys to peptides/amino acids (FDA labels). It does not degrade or convert into prolactin or “prolactin-type hormones”. GH and prolactin are evolutionarily related hormones with some receptor cross-talk at high concentrations, which is not a degradation pathway — and product-quality problems in unregulated “generic” GH, whatever they are, cannot change the molecule's catabolic fate. No human evidence retrieved links somatropin itself to sexual dysfunction via such a mechanism.
• “Growth hormone use and effects in Strongman and Powerlifting” (Ep. 26, segment at 20:40). Verdict — unproven (as a performance framing). No RCT shows GH improves maximal strength in any population of non-deficient adults; strongman/powerlifting use is reported practice, not demonstrated efficacy.

Overall: the fat-loss framing is broadly concordant; the prolactin-degradation claim as titled is contradicted by established pharmacology.

Creator summary

The creator layer at a glance — best-supported and weakest claims retrieved

Creator	Best-supported claim retrieved	Weakest claim retrieved	Basis
Vigorous Steve	GH is an expensive fat-loss aid, not an anti-aging drug, with unimpressive performance effects.	Skin/wrinkle rejuvenation as a stated benefit.	Full transcript
Big Paul	GH worsens glucose/insulin resistance and carpal tunnel (harm framing).	“Look 20 years younger” framing; “bubble gut” attribution.	Episode descriptions only
Dr. Alex Tatem	Sermorelin is not GH and should not be expected to build muscle like GH.	GH as “S-tier… proven for muscle… and tissue repair” (secondary attribution).	Own video description + interview transcript; tier list secondary
Hunter Williams	Exogenous GH bypasses pituitary control and reliably raises IGF-1.	GH/secretagogues build muscle via protein synthesis and new muscle cells; “not going to cause cancer”.	Full transcript
TRT Dad Unscripted	— none retrieved —	— none retrieved —	Could not verify
Dan the Bodybuilder	HGH burns fat (mechanism segment).	GH “degrading into prolactin-type hormones”.	Episode listings only

Sources: Vigorous Steve (Power Project Ep. 1010 transcript) · Big Paul (Podbean; Listen Notes) · Tatem (YouTube description; HoldsUp interview; unfinishedman.com secondary summary) · Hunter Williams Podcast (Podscan transcript) · TRT Dad Unscripted (Listen Notes; TopPodcast) · Dan the Bodybuilder / Steroids Podcast (Ivy.fm; Spotify creator listing)

Claim check

Popular claims about HGH, traced and graded at the tier their evidence actually sits at:

• Not supported

  “HGH builds contractile muscle.”

  Contradicted by human RCT evidence. DXA lean mass rises ~2.1–2.7 kg, but compartment studies show the rise is extracellular water (+1.78 kg ECW in Sergeeva 2019). Myofibrillar protein synthesis: unchanged in Doessing 2010 (direct biopsy measurement) and a 0.00 pooled effect in Sergeeva 2019. Muscle protein synthesis, fibre hypertrophy and strength are the contractile-muscle endpoints, and all are null in non-deficient adults. In deficient adults, replacement restores lean mass toward normal — a different claim.

• Not supported

  “Spot fat reduction” — inject where you want the fat gone.

  Folklore, contradicted as a cosmetic method, with one real adverse-effect kernel. GH lipolysis is systemic and hormonal; pooled trials show whole-body fat-mass change with no site-selective measurement supporting targeting. The kernel: repeatedly injecting the same site can cause local lipoatrophy — which is why FDA labels instruct site rotation — and two children in an early treatment series developed markedly diminished fat where they repeatedly injected one arm. That is an unwanted, uncontrolled local adverse effect, not a controllable cosmetic technique, and no trial supports using it as one.

• Unproven

  “HGH heals injuries.”

  Unproven in humans; animal/in-vitro evidence mixed-to-negative. For: GH raises collagen synthesis in uninjured human tendon/muscle (Doessing 2010) and collagen gene expression during immobilisation/rehab. Against: no human RCT on any actual injury; human tendon/ligament cells in vitro did not proliferate or differentiate better on HGH; a rat rotator-cuff repair model showed no benefit and worse force-to-failure in one protocol. Collagen synthesis in healthy tissue ≠ faster or stronger healing of injured tissue.

• Not supported

  “Anti-aging in a syringe.”

  Contradicted by human RCT evidence. Liu 2007's pooled verdict on the healthy-elderly RCTs: small composition changes, more adverse events, “cannot be recommended as an anti-aging therapy”. No trial shows improvement in strength, function, cognition, cardiovascular outcomes or mortality from GH in ageing adults. Rudman 1990 is the origin, not the proof. Distribution for this use is also specifically illegal in the US (Regulatory, below).

• Needs the qualifier

  “HGH increases strength / athletic performance.”

  Contradicted for strength and endurance; one narrow human RCT signal for sprint. Strength p = 0.36 and VO₂max p = 0.89 in Hermansen 2017; Liu 2008's conclusion is quoted in the Bodybuilding layer. Sole exception: Meinhardt 2010 Wingate sprint +3.9% (GH alone) / +8.3% (GH + testosterone, men), lost after washout.

• Not supported

  “HGH makes adults taller / regrows height.”

  Contradicted. Somatropin is contraindicated for growth promotion once epiphyses are closed (FDA labels); in adults, excess GH grows acral and soft tissue — the acromegaly pattern — not stature.

• Unproven

  “‘GH gut’ / ‘bubble gut’ is caused by HGH.”

  Folklore / unproven. No controlled literature isolates GH (vs insulin, androgens, food volume, visceral adiposity, organ growth) as the cause of abdominal distension in bodybuilders. Chronic GH excess does enlarge internal organs in acromegaly, which makes the claim biologically narratable but not demonstrated for this phenotype.

• Unproven

  “Raising GH with secretagogues gives GH's benefits without GH's risks.”

  Unproven. Secretagogues raise endogenous GH/IGF-1 (pharmacology supported), but there are no long-term outcome RCTs for body composition, healing, sleep or safety for the commonly marketed stacks in healthy adults — and any effect they have is still mediated by the same GH/IGF-1 axis, with the same theoretical risk profile.

• Not supported

  “GH is safe because it's ‘natural’ / bioidentical.”

  Folklore reasoning, contradicted by the harm data. Somatropin is sequence-identical to endogenous GH. Identity does not produce safety at non-physiological exposure: the endogenous excess state (acromegaly) is a disease, and the trial harms in the Safety section occurred with the identical molecule.

Regulatory status

United States — a controlled-distribution regime stricter than ordinary prescription status, and distinct from Controlled Substances Act scheduling.

• Somatropin products are prescription-only drugs, FDA-approved for defined indications: GH deficiency in adults and children, specific paediatric growth conditions (e.g. Turner syndrome, Prader-Willi syndrome, small-for-gestational-age without catch-up growth, idiopathic short stature, chronic renal insufficiency — varying by brand), and, for particular brands, HIV-associated wasting/cachexia and short bowel syndrome. Ageing, bodybuilding, fat loss in non-deficient people, and athletic performance are not approved indications for any somatropin product.
• 21 U.S.C. § 333(e) — “Prohibited distribution of human growth hormone.” Whoever knowingly distributes, or possesses with intent to distribute, HGH “for any use in humans other than the treatment of a disease or other recognized medical condition, where such use has been authorized by the Secretary of Health and Human Services… and pursuant to the order of a physician” commits an offence punishable by up to 5 years' imprisonment, or up to 10 years if the offence involves an individual under 18. A conviction is treated as a felony violation of the Controlled Substances Act for forfeiture purposes, and the DEA is authorised to investigate offences under the subsection. “Human growth hormone” is defined as somatrem, somatropin, or an analogue of either. Perls et al. (JAMA 2005) document the FDA's position that distribution for anti-aging is illegal under this provision — GH is, in effect, the one prescription drug with its own statutory ban on off-label distribution.
• Precision point: HGH is frequently described in gym and media sources as a “controlled substance”. Federally, it is not scheduled under the Controlled Substances Act in the way anabolic steroids are; its control operates through § 333(e) above (with CSA forfeiture consequences) plus ordinary prescription-drug law. Some individual US states do schedule or additionally restrict HGH under state law — state-by-state status was not verified in this research (see Open questions).
• Sport: GH is on the WADA Prohibited List, class S2 (peptide hormones, growth factors, related substances and mimetics), prohibited at all times, in and out of competition — the WADA athlete guide to the 2026 Prohibited List lists S2 with human growth hormone as its exemplar. GH secretagogues and GHRH analogues fall in the same class.
• Elsewhere (brief, not exhaustively verified): somatropin is a prescription-only medicine in the UK, EU, Canada and Australia, with supply for non-medical purposes an offence in several of those jurisdictions; jurisdiction-specific penalties were not verified in this research and should not be stated without a separate check.

Open questions

Marked plainly, because they are open — where the evidence genuinely runs out:

• Long-term safety of adult non-deficient use. No RCT or cohort follows healthy adults using GH for bodybuilding/anti-aging purposes for the years-to-decades over which acromegaly and cancer outcomes develop. Cancer reassurance comes from childhood replacement cohorts followed to a mean age in the mid-twenties — a different population, exposure and horizon.
• Whether GH replacement in deficiency reduces the outcomes that matter. Mortality and fracture benefits in adult deficiency are inferred from surrogates (lipids, BMD, body composition); no adequately powered RCT with those endpoints exists.
• Untested regimens. Community practice — longer durations, combinations with androgens and insulin — has never been studied as a regimen for benefit or harm. The trial caveat that study exposures “may not reflect real-world doses and regimens” (Liu 2008) remains unresolved in both directions.
• The sprint signal. Meinhardt 2010's 3.9% Wingate finding is a single trial; it has not been independently replicated as a primary endpoint, and its mechanism (it did not track body-cell-mass change except with testosterone co-administration) is unexplained.
• The collagen → clinical outcome gap. Whether GH-induced collagen synthesis in tendon/muscle ever translates into stronger tendon, fewer injuries, or faster rehabilitation in humans is untested in an injury population.
• Bone in non-deficient older adults. Liu 2007 found no bone-density change in healthy elderly trials; whether longer exposure changes fracture risk (in either direction) is unknown.
• Thymus / epigenetic-ageing claims. Based on very small, multi-agent, exploratory work; neither confirmed nor refuted by an adequately powered trial.
• Long-acting (weekly) GH preparations. Efficacy/safety equivalence to daily GH is supported for deficiency replacement over trial horizons; very long-term comparative safety is still accruing.

Sources

1. Liu H et al. Systematic review: the safety and efficacy of growth hormone in the healthy elderly. Ann Intern Med 2007;146:104–115. healthpolicy.fsi.stanford.edu (Liu 2007) · Liu H et al. Systematic review: the effects of growth hormone on athletic performance. Ann Intern Med 2008;148:747–758 (summary and conclusions via Stanford/AHRQ reporting and the Garvan review).
2. Hermansen K et al. Impact of GH administration on athletic performance in healthy young adults: systematic review and meta-analysis of placebo-controlled trials. Growth Horm IGF Res 2017. PMID 28514721. pubmed.ncbi.nlm.nih.gov/28514721
3. Sergeeva KV et al. Effect of growth hormone administration on the mass and strength of muscles in healthy young adults: systematic review and meta-analysis. Hum Physiol 2019 (record via EPA HERO). hero.epa.gov/reference/7434477
4. Meinhardt U et al. The effects of growth hormone on body composition and physical performance in recreational athletes: a randomized trial. Ann Intern Med 2010;152:568–577. drhigashi.com.br (Meinhardt 2010, full text PDF)
5. Rudman D et al. Effects of human growth hormone in men over 60 years old. N Engl J Med 1990;323:1–6 (abstract). worldhealth.net (Rudman 1990)
6. Blackman MR et al. Growth hormone and sex steroid administration in healthy aged women and men: a randomized controlled trial. JAMA 2002;288:2282–2292. jamanetwork.com (Blackman 2002)
7. Doessing S et al. Growth hormone stimulates the collagen synthesis in human tendon and skeletal muscle without affecting myofibrillar protein synthesis. J Physiol 2010;588:341–351. pmc.ncbi.nlm.nih.gov/articles/PMC2821728 · GH, tendon and skeletal muscle during immobilisation and rehabilitation (Boesen et al.; Langberg/Kjaer group). pubmed.ncbi.nlm.nih.gov/24081158 · pmc.ncbi.nlm.nih.gov/articles/PMC3872768 · Can human growth hormone accelerate tendon and ligament injury recovery? (human fibroblast study), 2024. pmc.ncbi.nlm.nih.gov/articles/PMC11571130
8. Li S et al. Effects of low-dose versus high-dose human growth hormone in adults with GH deficiency: meta-analysis of placebo-controlled RCTs (22 trials). Endocrine 2015. springermedizin.de (Li 2015) · Adult GH deficiency reviews and meta-analyses (37 blinded RCTs; GHD syndrome). pmc.ncbi.nlm.nih.gov/articles/PMC3183535 · frontiersin.org (Kargi & Merriam 2013) · Effects of 18 months of GH replacement on BMD in adult GHD (n = 68), 2023. pmc.ncbi.nlm.nih.gov/articles/PMC10081901
9. Luger A et al. Incidence of diabetes mellitus and evolution of glucose parameters in GH-deficient subjects during GH replacement therapy (KIMS, n = 5,143). Diabetes Care 2012. pmc.ncbi.nlm.nih.gov/articles/PMC3241307
10. Swerdlow AJ et al. SAGhE cohort cancer outcomes (n = 23,984) — summarised in: Long-term safety of childhood growth hormone treatment. springermedizin.de (SAGhE summary) · pmc.ncbi.nlm.nih.gov/articles/PMC8416866 · Swedish population cohort: long-term risk of neoplastic events after childhood GH treatment (n = 3,408). Front Endocrinol 2024. frontiersin.org (Swedish cohort 2024)
11. Acromegaly: Endotext chapter (mortality at least 2×; causes of death). ncbi.nlm.nih.gov (Endotext, acromegaly) · Acromegaly prognosis review (SMR 1.2–3.3; ~10 years' life lost untreated). pmc.ncbi.nlm.nih.gov/articles/PMC2459162 · Cardiovascular effects of excess GH. pmc.ncbi.nlm.nih.gov/articles/PMC11273026
12. HUMATROPE (somatropin) US prescribing information, DailyMed (PK §12.3; contraindications §4; acute-critical-illness mortality §5.1). dailymed.nlm.nih.gov (Humatrope label) · GENOTROPIN (somatropin) FDA label, 2024 (PK: IV half-life 0.4 h, SC half-life 3.0 h). accessdata.fda.gov (Genotropin label 2024)
13. Diagnosis and treatment of growth hormone deficiency: position statement, Korean Endocrine Society / Korean Society of Pediatric Endocrinology (summarising Endocrine Society guidance; IGF-1 monitoring). ncbi.nlm.nih.gov/pmc/articles/PMC7386113 · AACE/ACE Growth Hormone Deficiency Clinical Practice Guidelines 2019 (guideline summary). reference.medscape.com (AACE/ACE 2019 summary)
14. Growth hormone and metabolic homeostasis (pulsatility, GHRH/somatostatin, IGF-1 feedback). EMJ. emjreviews.com (GH and metabolic homeostasis) · Growth hormone: central regulation of metabolism (191-aa structure, GHR–JAK/STAT, ghrelin/GHS-R, hepatic vs local IGF-1). Encyclopedia / MDPI. encyclopedia.pub/entry/15782
15. 21 U.S.C. § 333(e), Prohibited distribution of human growth hormone (statute text). law.justia.com (21 U.S.C. § 333(e)) · Perls TT et al. Provision or distribution of growth hormone for “antiaging”: clinical and legal issues. JAMA 2005 (summary). natap.org (Perls 2005) · WADA, Athlete and Athlete Support Personnel Guide to the 2026 Prohibited List (S2 includes human growth hormone, prohibited at all times). wada-ama.org (2026 Prohibited List guide)
16. Creator sources, cited as claims graded in the Creator layer, not evidence — Vigorous Steve on Mark Bell's Power Project Ep. 1010 (transcript). podscripts.co (Power Project Ep. 1010) · Big Paul / Anabolic Bodybuilding — channel and episode descriptions. anabolicbodybuilding.podbean.com · listennotes.com (Anabolic Bodybuilding episode)
17. Creator sources — Dr. Alex Tatem, “Sermorelin: The ‘Safer HGH’ Has One Huge Problem” (own video description). youtube.com (Tatem, sermorelin) · “Peptide Expert” interview transcript. holdsup.app (Peptide Expert interview) · Tier-ranking as reported (secondary). unfinishedman.com (peptides, evidence ranked) · Hunter Williams, “HGH vs Peptides: Should You Combine Both for Maximum Results?” transcript, 15 July 2025. podscan.fm (Hunter Williams, HGH vs Peptides)
18. Creator sources — TRT Dad Unscripted, podcast feeds/listings searched, no HGH-specific claim retrieved. listennotes.com (TRT Dad Unscripted) · toppodcast.com (TRT Dad Unscripted) · Dan the Bodybuilder / Steroids Podcast, episode listings with timestamped contents. ivy.fm (Steroids Podcast) · creators.spotify.com (Steroids Podcast Ep. 55)
19. HGH (somatropin) — evidence deep-dive, research report filed 28 Oct 2026. Every figure, attribution and hedge on this page follows that report.

Molecule

39-amino-acid synthetic peptide

Class

Dual GIP / GLP-1 receptor agonist

Brands

Mounjaro · Zepbound

US status

Rx-only · boxed warning

The health layer — what tirzepatide is for in legitimate medicine

These are chronic-disease treatments, not cosmetic or performance products. Every pivotal trial paired the drug with lifestyle intervention — the Zepbound label trials used a ~500 kcal/day deficit diet, a recommended minimum of 150 min/week of physical activity, and behavioural counselling. Results quoted without that context overstate what the drug alone was tested doing.

The legitimate jobs — indications in the FDA labels

Indication	Brand / population	What “success” means in the trials
Type 2 diabetes — improve glycemic control	Mounjaro, as adjunct to diet and exercise; the label as of late 2026 also covers pediatric patients aged 10+ (per Lilly's August 2026 release wording)	HbA1c reduction; proportions reaching HbA1c <7% and <5.7% (Efficacy, below)
Chronic weight management	Zepbound, with reduced-calorie diet and increased physical activity, in adults with obesity, or overweight with ≥1 weight-related comorbid condition	Mean % weight change and proportions reaching ≥5/10/15/20% loss at 72 weeks
Moderate-to-severe obstructive sleep apnea in adults with obesity	Zepbound, with reduced-calorie diet and increased physical activity	Reduction in apnea–hypopnea index (AHI) at 52 weeks
Reduce risk of major adverse cardiovascular events	Mounjaro, added 28 Aug 2026, in adults with type 2 diabetes at high risk, based on SURPASS-CVOT	Noninferiority to dulaglutide for CV death / nonfatal MI / nonfatal stroke

Sources: FDA Zepbound label §1, DailyMed · Lilly press release, 28 Aug 2026

Glycemic control and diabetes prevention — what it means day to day

• In SURPASS-2, 82–86% of tirzepatide participants reached HbA1c <7% vs 79% on semaglutide 1 mg, and 27–46% reached HbA1c <5.7% — a non-diabetic range — vs 19%. (Frías et al., NEJM 2021)
• Hypoglycemia on tirzepatide alone is uncommon because insulin secretion is glucose-dependent (Pharmacology, below). In the label's diabetes weight trial, glucose <54 mg/dL occurred in 4.2% on Zepbound vs 1.3% on placebo, rising to 10.3% in those also taking a sulfonylurea vs 2.1% without. (Zepbound label §5.7)
• Prevention: in the 3-year SURMOUNT-1 prediabetes cohort (n=1,032 with prediabetes at randomisation; 176 weeks), progression to type 2 diabetes was reduced by 94% (HR 0.06); 98.7% of tirzepatide participants remained diabetes-free vs 86.7% on placebo, with sustained weight loss of −22.9% at the highest efficacy estimand. A post-hoc mediation analysis attributed roughly half the prevention effect to weight loss. This is not a separate FDA prevention indication — it is trial evidence within the weight/diabetes story. (Jastreboff et al., NEJM 2024)

Sleep apnea — approval and data

FDA approved Zepbound for moderate-to-severe OSA in adults with obesity on 20 December 2024 — the first drug approval for OSA. Basis: SURMOUNT-OSA, two 52-week trials totalling n=469 adults with obesity and moderate-to-severe OSA, one without positive airway pressure (PAP) therapy and one with it. (Malhotra et al., NEJM 2024)

SURMOUNT-OSA at 52 weeks

Outcome	No-PAP trial	PAP trial
AHI change, tirzepatide (maximum tolerated trial exposure)	≈ −25 events/h (≈ −50–55%)	≈ −29 events/h (≈ −59%)
AHI change, placebo	≈ −5 events/h	≈ −5.5 events/h
Weight change, tirzepatide vs placebo (label)	−17.7% vs −1.6%	−19.6% vs −2.3%
Remission or mild non-symptomatic OSA	42% vs 16% placebo	50% vs 14% placebo

In plain language: breathing interruptions roughly halved on average, alongside ~18–20% weight loss. The trials do not establish tirzepatide as a replacement for PAP — one trial was explicitly conducted in people continuing PAP.

Cardiovascular risk markers — not the same as outcomes

In SURMOUNT-1 / label Study 1, tirzepatide improved, vs placebo: waist circumference (−14.0 to −18.5 cm vs −4.0 cm), triglycerides (−21.2% to −29.1% vs −5.6%), HDL cholesterol (+6.9% to +9.2% vs −0.7%), non-HDL cholesterol and blood pressure. Marker improvements are consistent and large, but markers are surrogates — hard-outcome status is dealt with separately in the Efficacy section.

Sources: FDA Zepbound label §§5.7, 14.1–14.2, DailyMed · Frías et al., NEJM 2021 · Malhotra et al., NEJM 2024 · Jastreboff et al., NEJM 2024

Pharmacology — one molecule, two incretin receptors

Tirzepatide is a 39-amino-acid synthetic peptide based on the native GIP sequence, engineered to activate both the glucose-dependent insulinotropic polypeptide receptor (GIPR) and the glucagon-like peptide-1 receptor (GLP-1R). Native GIP (from intestinal K cells) and GLP-1 (from L cells) are incretins — gut hormones released after eating that potentiate glucose-stimulated insulin secretion. (Coskun et al., Cell Metab 2018; Willard et al., JCI Insight 2020)

Label pharmacodynamics (§12.2): tirzepatide (i) decreases calorie intake, likely via appetite effects; (ii) lowers body weight with greater fat-mass than lean-mass loss; (iii) stimulates insulin secretion in a glucose-dependent manner and reduces glucagon secretion; (iv) increases insulin sensitivity (demonstrated in a hyperinsulinemic-euglycemic clamp study in type 2 diabetes at 28 weeks); and (v) delays gastric emptying, with the delay largest after the first dose and diminishing over time.

Why GIP co-agonism was pursued — and why it was not obvious

• The historical problem: in type 2 diabetes, GIP's insulinotropic effect is markedly attenuated, and GIP alone looked like a poor drug target; early candidates targeting GIPR alone showed much lower efficacy than GLP-1R agonists. GIP was also suspected of promoting adipose storage.
• The bet: preclinical work with the dual agonist (development code LY3298176) showed combined GIPR + GLP-1R agonism produced greater glucose lowering and weight loss than GLP-1R agonism alone, with complementary tissue distribution — GIPR in pancreatic beta cells, adipose tissue and central appetite circuits; GLP-1R in beta cells, gut, and brain. Proposed complementary mechanisms in the literature include GIPR-mediated effects on adipose insulin sensitivity and lipid handling, central food-intake circuits, and attenuation of GLP-1–related nausea/emesis in animal models.
• The paradox that remains open: GIPR antagonism also reduces body weight in preclinical models and is the basis of a separate drug programme (the antibody–peptide conjugate maridebart cafraglutide). Both agonism and antagonism at the same receptor producing weight loss is unresolved — proposed explanations include agonist-induced receptor desensitisation/down-regulation functionally mimicking antagonism. The honest pharmacological statement is: dual agonism works clinically; the precise share of the effect attributable to the GIP arm in humans has never been isolated, because no trial has compared tirzepatide against an exposure-matched GLP-1-only version of itself.

Receptor pharmacology of both arms — “imbalanced and biased”

Willard et al., JCI Insight 2020 — Lilly's own characterisation, albumin-free in-vitro assays

Property	GIP receptor	GLP-1 receptor
Binding affinity vs native ligand	Comparable to native GIP	~5-fold lower than native GLP-1
cAMP potency vs native ligand	Equipotent, full agonist	~20-fold lower potency (13–20× across assays)
Signalling bias	Similar to native GIP, including β-arrestin recruitment	Biased: favours cAMP generation over β-arrestin recruitment; weak/partial β-arrestin recruitment, hence less receptor internalisation in cell systems

The interpretation offered by the authors: imbalance toward GIPR plus biased signalling at GLP-1R (sustained cAMP signalling with reduced desensitisation) may together account for the clinical efficacy. That is a mechanistic hypothesis supported by cell pharmacology, not a human proof.

The fatty-acid / albumin-binding design, pharmacokinetics, metabolism, excretion

Design: a C20 fatty diacid moiety is conjugated (via a linker/spacer) to a lysine residue of the peptide. The fatty acid binds reversibly and with high affinity to serum albumin, shielding the peptide from rapid degradation/clearance and creating a circulating reservoir. Amino-acid substitutions in the backbone (including α-aminoisobutyric acid) confer resistance to DPP-4 cleavage. (Coskun et al. 2018; Willard et al. 2020)

Label pharmacokinetics — Zepbound / Mounjaro §12.3

Parameter	Label value
Half-life	Approximately 5 days (Mounjaro wording) / 5–6 days in patients with overweight/obesity and in OSA with obesity (Zepbound §12.3)
Tmax (subcutaneous)	Median 24 h (range 8–72 h)
Absolute bioavailability	80%
Steady state	After 4 weeks of once-weekly administration; exposure dose-proportional
Distribution	Apparent volume ~9.7 L (obesity) / 11.8 L (OSA); 99% bound to plasma albumin
Clearance	Apparent population mean ~0.06 L/h
Metabolism	Proteolytic cleavage of the peptide backbone, beta-oxidation of the C20 fatty diacid, and amide hydrolysis; low potential to inhibit/induce CYP enzymes
Excretion	Metabolites mainly via urine and feces; intact tirzepatide is not observed in urine or feces
Special populations	Renal and hepatic impairment do not have a clinically relevant impact on PK (label §12.3, single-dose studies)

Pharmacodynamic interaction consequence: delayed gastric emptying alters oral drug absorption — e.g., after a first 5 mg trial dose, acetaminophen Cmax fell 55% and tmax was delayed 1 h, with no meaningful effect by week 6 at the 15 mg exposure; a single dose reduced oral contraceptive steroid Cmax by 55–66% and AUC by ~20–23%. (Zepbound label §12.3) This is label information, not use guidance.

Sources: FDA Zepbound label §12, DailyMed · Willard et al., JCI Insight 2020 · Coskun et al., Cell Metab 2018 · “Clinical Potential of GIP in Type 2 Diabetes and Obesity,” Diabetes Care 2026

The bodybuilding / fitness layer — body composition, strength, and the gaps

Tirzepatide has no approval, and no pivotal trial, for physique, performance, or use in lean trained people. Trial populations were people with obesity/overweight, mostly untrained, eating in a prescribed deficit. Fitness-community use is described in this section only as reported practice/claims.

Body composition — the DXA evidence

SURMOUNT-1 DXA substudy (Look et al., Diabetes Obes Metab 2025): 255 participants enrolled in the substudy; n=160 had baseline and week-72 DXA and were analysed (pooled tirzepatide trial exposures n=124; placebo n=36; 73% female; mean weight 102.5 kg, BMI 38.0).

SURMOUNT-1 DXA substudy — change at 72 weeks

Measure	Tirzepatide (pooled)	Placebo
Body weight	−21.3%	−5.3%
Fat mass	−33.9%	−8.2%
Lean mass	−10.9%	−2.6%
Share of weight lost	~75% fat / ~25% lean	~75% fat / ~25% lean

Proportions were broadly consistent across sex, age (<50, 50–<65, ≥65) and weight-loss tertile subgroups. The key interpretation: the proportion of lean loss was not worse than placebo weight loss — the absolute lean loss was larger only because total loss was ~4× larger.

Comparison with semaglutide's DXA numbers — STEP 1 substudy (Wilding et al., J Endocr Soc 2021): n=140 (semaglutide n=95, placebo n=45), 68 weeks, semaglutide 2.4 mg trial exposure: total fat mass −19.3%, regional visceral fat −27.4%, total lean body mass −9.7%, with lean mass proportion of total mass rising +3.0 percentage points. Secondary analyses of STEP 1 estimate lean mass accounted for ~39–40% of total weight lost on semaglutide, vs ~25–26% for tirzepatide in SURMOUNT-1.

Muscle quality imaging: in the SURPASS-3 MRI substudy post-hoc analysis (n=246, type 2 diabetes, 52 weeks), thigh muscle volume fell 0.64 L — not significantly different from the volume loss predicted for the observed weight loss from UK Biobank population data (difference −0.04 L, p=0.22) — while muscle fat infiltration fell 0.36 percentage points, more than predicted. Interpretation: smaller muscles, but less fatty muscles, in line with weight loss itself.

What “lean mass” is not: DXA lean mass includes water, glycogen, organs, gut contents and connective tissue as well as skeletal muscle. Early “lean” loss partly reflects glycogen/water shifts. No DXA substudy can state how many kilograms of contractile muscle were lost.

Strength and function — the thin part of the evidence

• No SURMOUNT trial reported handgrip, leg-press, or other direct strength endpoints. That is a genuine evidence gap, flagged in reviews: body-composition studies of tirzepatide generally lacked functional measures, so whether lean loss translated into strength loss in trial populations is unknown from the pivotal programme.
• The same review literature notes short-to-mid-term incretin trials (mostly semaglutide/liraglutide) generally showed preserved handgrip despite lean loss; one small study of tirzepatide combined with resistance + aerobic training in young men found no strength benefit beyond exercise alone; and longer observational data in older adults with diabetes raise a possible strength-decline signal with prolonged semaglutide use. None of this is a tirzepatide RCT strength result.
• Functional capacity did improve where it was a disease endpoint: in SUMMIT (HFpEF + obesity), 6-minute walk distance and KCCQ health-status scores improved more on tirzepatide than placebo at 52 weeks — that is cardiac functional capacity in sick patients, not gym performance.
• Animal data (diet-induced obese mice) showing maintained grip strength on tirzepatide exist but are animal evidence only and are not graded as human evidence here.

Sources: Look et al., Diabetes Obes Metab 2025 (PMID 39996356) · Wilding et al., J Endocr Soc 2021 (STEP 1 DXA) · SURPASS-3 MRI substudy, Lancet Diabetes Endocrinol · Muscle strength review, Br J Pharmacol 2026 (PMID 41577337)

Efficacy — with numbers

SURMOUNT-1 — weight (n=2,539; 72 weeks; obesity/overweight, no diabetes; lifestyle counselling in all arms)

Two estimands are published and are often conflated. Both are shown, labelled:

SURMOUNT-1 — mean weight change by estimand (*trial-design exposures, not recommendations)

Estimand	5 mg*	10 mg*	15 mg*	Placebo
Treatment-regimen (label §14.1 / NEJM primary) — %	−15.0%	−19.5%	−20.9%	−3.1%
Treatment-regimen — placebo contrast	−11.9 pts	−16.4 pts	−17.8 pts	—
Efficacy (on-treatment) — %	−16.0%	−21.4%	−22.5%	−2.4%
Efficacy — kg (lb)	−16 kg (35 lb)	−22 kg (49 lb)	−24 kg (52 lb)	−2 kg (5 lb)

Responders (regimen estimand): ≥5% loss in 85% / 89% / 91% vs 35% placebo; ≥20% loss in 30% / 50% / 57% vs 3%. In participants with type 2 diabetes (SURMOUNT-2 / label Study 2, n=938, 72 weeks), weight loss was smaller, as is typical: −12.8% (10 mg*) and −14.7% (15 mg*) vs −3.2% placebo.

SURPASS — glycemic results (type 2 diabetes)

SURPASS programme — HbA1c change (*trial-design exposures)

Trial	Design	HbA1c change (tirzepatide)	Comparator
SURPASS-1 (n=478, 40 wk)	Monotherapy vs placebo	−1.87 / −1.89 / −2.07 pts (5/10/15 mg*)	+0.04 pts placebo; weight −7.0 to −9.5 kg
SURPASS-2 (n=1,879, 40 wk)	Head-to-head vs semaglutide 1 mg*	−2.01 / −2.24 / −2.30 pts	−1.86 pts semaglutide; weight −7.6 / −9.3 / −11.2 kg vs −5.7 kg
SURPASS-3 (n=1,444, 52 wk)	vs insulin degludec	up to −2.37 pts (15 mg*)	Weight −12.9 kg vs gain on insulin
SURPASS-5 (40 wk)	Added to insulin glargine	up to −2.59 pts (15 mg*)	—

All tirzepatide exposures in SURPASS-2 were noninferior and superior to semaglutide 1 mg for HbA1c (p<0.001 for weight differences). Note the comparator caveat: SURPASS-2 used the 1 mg diabetes exposure of semaglutide, not the 2.4 mg weight-management exposure — SURMOUNT-5 is the fair weight head-to-head. The SURPASS-1/-3/-5 figures above are taken from the respective papers as summarised in secondary trial reviews — index-level pending primary-text verification, unlike the SURPASS-2 row.

SURMOUNT-5 — head-to-head vs semaglutide (n=751; 72 weeks; obesity, no diabetes; open-label; maximum tolerated trial exposures)

SURMOUNT-5 at 72 weeks (Aronne et al., NEJM 2025)

Outcome	Tirzepatide	Semaglutide
Mean weight change	−20.2% (−22.8 kg)	−13.7% (−15.0 kg)
Waist circumference	−18.4 cm	−13.0 cm
Discontinuation for GI adverse events	2.7%	5.6%

≥25% weight loss was also higher on tirzepatide (per NEJM secondary endpoints). Sponsor: Eli Lilly (tirzepatide's maker) — disclosed as a design context, with open-label administration the other main caveat.

Cardiovascular outcomes — status as of late 2026

Cardiovascular outcome trials

Trial	Population / design	Result
SURPASS-CVOT (NEJM 2025)	n=13,299, type 2 diabetes + established ASCVD, vs dulaglutide 1.5 mg* (active comparator with known CV benefit), median ~4 years	3-point MACE 12.2% vs 13.1%; HR 0.92 (95.3% CI 0.83–1.01) — noninferiority met (p=0.003), superiority not met (p=0.09). All-cause death HR 0.84 (95% CI 0.75–0.94) as a secondary endpoint. HbA1c −1.66 vs −0.88 pts; weight −11.6% vs −4.8%
SUMMIT (NEJM 2025)	n=731, HFpEF (EF ≥50%) + obesity, vs placebo, median ~2 years	CV death or worsening HF event 9.9% vs 15.3%; HR 0.62 (95% CI 0.41–0.95; p=0.026), driven by worsening-HF events; CV death alone did not differ significantly (small numbers). KCCQ and 6-min walk improved; weight difference −11.6%

Regulatory consequence: on 28 August 2026 the FDA approved Mounjaro to reduce the risk of MACE (CV death, nonfatal MI, nonfatal stroke) in adults with type 2 diabetes at high risk — Lilly described it as the first dual GIP/GLP-1 agonist with such an indication, explicitly on a noninferiority basis vs dulaglutide. What remains unproven as of late 2026: placebo-controlled MACE superiority for tirzepatide, and any dedicated CV outcomes trial in obesity without diabetes (the analogue of semaglutide's SELECT does not exist for tirzepatide).

Sources: Jastreboff et al., NEJM 2022 (SURMOUNT-1) · FDA Zepbound label Table 2 · Frías et al., NEJM 2021 (SURPASS-2) · Aronne et al., NEJM 2025 (SURMOUNT-5) · Nicholls et al., NEJM 2025 (SURPASS-CVOT) · Packer et al., NEJM 2025 (SUMMIT) · Lilly, 28 Aug 2026

Safety — with numbers

Primary source for this section: the Zepbound FDA label §§5–6 (DailyMed), pooled weight-reduction trials Studies 1+2 (n=2,519 tirzepatide-treated; placebo n=958 in the adverse-reaction table), supplemented by SURMOUNT-1 NEJM reporting where labelled.

Gastrointestinal events and discontinuations

Label Table 1 — adverse reactions, pooled Studies 1+2 (*label exposures)

Adverse reaction	Placebo	5 mg*	10 mg*	15 mg*
Nausea	8%	25%	29%	28%
Diarrhea	8%	19%	21%	23%
Vomiting	2%	8%	11%	13%
Constipation	5%	17%	14%	11%
Any GI adverse reaction (pooled text)	30%	56%	56%	56%
Severe GI adverse reactions (§5.2)	1.0%	1.7%	2.5%	3.1%
Discontinued for GI adverse reactions	0.5%	1.9%	3.3%	4.3%
Permanently discontinued for any adverse reaction	3.4%	4.8%	6.3%	6.7%

SURMOUNT-1 alone (NEJM) reported nausea 24.6/33.3/31.0%, diarrhea 18.7/21.2/23.0%, vomiting 8.3/10.7/12.2% and discontinuation for adverse events 4.3/7.1/6.2% vs 2.6% placebo — consistent with the pooled label. Most nausea/vomiting/diarrhea occurred during dose escalation and decreased over time (label §6.1).

Gastroparesis / ileus signals

• Label §5.2: Zepbound is not recommended in patients with severe gastroparesis; severe stomach problems have been reported.
• Label §6.2 (postmarketing): ileus is listed among reported post-approval GI reactions. Postmarketing reporting cannot establish frequency or causality (the label's own wording).
• Lilly medical information (placebo-controlled SURMOUNT-1/-2/-3): intestinal obstruction or ileus in 4 of 2,806 tirzepatide-treated participants and 2 of 1,250 on placebo — rare, and numerically similar in proportion, in the controlled data.
• Observational/case literature reports gastroparesis and bowel-obstruction cases with GLP-1–class drugs; a 2023 cohort/JAMA analysis of GLP-1 agonists for weight loss found higher risks of gastroparesis, bowel obstruction and pancreatitis vs a comparator weight-loss drug, but tirzepatide-specific controlled incidence remains the label/Lilly figures above. Grade: signal present (postmarketing + class observational), causal frequency unquantified.

Gallbladder, pancreatitis, heart rate and other label signals

Further label signals — Studies 1+2 pool unless stated

Signal	Number and source
Gallbladder	Cholelithiasis 1.1% vs 1.0% placebo; cholecystitis 0.7% vs 0.2%; cholecystectomy 0.2% vs 0% (label §§5.4/6.1). The label notes acute gallbladder events were associated with weight reduction — a known consequence of rapid weight loss by any method. SURMOUNT-1 reported cholecystitis at ≤0.6% in tirzepatide arms.
Pancreatitis	Adjudication-confirmed acute pancreatitis in the weight trials was 0.2% on Zepbound (0.14 per 100 patient-years) vs 0.2% on placebo (0.15 per 100 patient-years) — no excess in the controlled pool. In trials for a different (diabetes) indication: 0.23 vs 0.11 per 100 years (tirzepatide vs comparator). In the OSA trials the exposure-adjusted rate was 0.84 per 100 years vs 0 on placebo (small n=467). Label §5.5 nonetheless warns that acute pancreatitis, including fatal/necrotising cases, has been observed with GLP-1–class drugs and tirzepatide, and mean amylase rose 20–25% and lipase 28–35% on treatment (clinical significance unknown absent symptoms). Grade: warning justified; controlled weight-trial data show no measurable excess.
Heart rate	Mean increase of 1 to 3 beats/min vs no increase on placebo (label §6.1).
Acute kidney injury	0.5% vs 0.2% placebo, mostly in the setting of GI fluid loss/dehydration (label §§5.3/6.1, plus postmarketing reports including cases requiring hemodialysis).
Hypotension	1.6% vs 0.1%, more frequent with concomitant antihypertensives.
Hair loss	5/4/5% across exposures vs 1% placebo (7.1% of females vs 0.5% of males on drug), associated with weight reduction (label §6.1).
Pulmonary aspiration	Rare postmarketing reports during anesthesia/deep sedation with GLP-1–class drugs in patients with residual gastric contents despite fasting (label §5.10) — a direct consequence of delayed gastric emptying (Pharmacology, above).

Sources: FDA Zepbound label §§5–6, 13.1, DailyMed · Lilly Medical — intestinal obstruction/ileus incidence with Zepbound

Creator layer — opinion, graded claim by claim

Everything in this section is creator and community content, kept separate from the primary literature above precisely so it is not mistaken for it. Only claims attributable to the named creator in a findable source are graded. Verdicts: supported / partly supported / contradicted / unproven. Grading a claim is not endorsing the practice it describes.

Vigorous Steve

No findable attributable claim. Three separate searches (general web, YouTube, and his own site vigoroussteve.com) returned no specific statement by Vigorous Steve about tirzepatide that could be quoted and graded. His indexed GLP-1–adjacent content did not surface a tirzepatide claim in the sources available to this review. No verdict is offered, and none should be invented.

Big Paul — Paul Barnett (Anabolic Bodybuilding)

Source: Anabolic Bodybuilding episode description, 9 Oct 2026 (Podbean).

• Tirzepatide and other GLP-1 medications may interfere with a gaining (bulking) phase. Verdict — partly supported. The direction is consistent with the label pharmacodynamics — decreased calorie intake via appetite effects is the drug's primary weight mechanism (label §12.2) — so a drug that suppresses intake plausibly opposes a deliberate energy surplus. But there is no RCT in bodybuilders or in a gaining phase; the claim as applied to trained people deliberately overfeeding is untested. The hedged wording (“may”) matches the evidence level.

No other specific, quotable tirzepatide claim by Big Paul was findable in indexed episode descriptions/transcripts for this review.

Dr. Alex Tatem

Source: “Doctor Reveals BRUTAL Truth About Peptides” — Diary of a CEO episode with Dr Alex Tatem, YouTube (claims as summarised in the episode's own description).

• A patient who was morbidly obese with a sperm count too low to conceive lost 100 lb on tirzepatide, after which his sperm count increased tenfold into the normal range; treating metabolic dysfunction, not the symptom (low testosterone), was the fix. Verdict — unproven (as a drug claim). Single-patient anecdote (n=1), reported in a podcast description. A 100 lb loss is within the absolute range seen in SURMOUNT-1 for people starting at high body weights (−24 kg ≈ 52 lb was the mean at the highest efficacy estimand; larger absolute losses occur at higher starting weights), so the weight part is plausible. There is no tirzepatide RCT with semen/fertility endpoints, and weight loss by any method is associated with improved semen parameters in observational literature — the anecdote cannot attribute a 10× rise to tirzepatide specifically.
• Compounded tirzepatide allows flexible “microdosing” throughout the week instead of one large dose, which many patients tolerate better with fewer side effects. Verdict — unproven. No RCT has tested split/more-frequent tirzepatide administration for tolerability or efficacy; the entire pivotal programme and the FDA label are built on once-weekly administration, and compounded products are not FDA-approved products. “Many patients tolerate better” is clinical impression, not trial evidence. This is reported practice, graded here only as a claim.

Hunter Williams

Sources: his own newsletter (Hunter's Health Hacks / Beehiiv) — primary creator texts.

• Tirzepatide and retatrutide have “PERMANENTLY reduced the amount of food I can consume”; GLP-1s “fundamentally rewire the brain”. Verdict — contradicted (as a general claim). SURMOUNT-4 randomised withdrawal: switching to placebo produced +14.0% regain in a year and only 16.6% kept ≥80% of their loss (Claim check, below) — appetite/weight effects are predominantly on-treatment effects in the controlled data. A personal report of lasting change is anecdote; “permanently” as a class claim conflicts with the withdrawal RCT.
• “Why EVERYONE Can (And Probably Should) Use GLP-1 Agonists” — including people already lean; GLP-1s change you permanently (rewire brain chemistry, change attitude to food, reduce smoking/drinking/gambling, remodel the gut) and are “the single greatest health discovery since penicillin”. Verdict — contradicted / unproven. No approval or RCT supports use in lean people for “optimisation”; pivotal trials enrolled people with obesity/overweight or diabetes. Permanence is contradicted by SURMOUNT-4. Addiction/vice and gut-remodelling claims have no tirzepatide RCT endpoints. The penicillin comparison is rhetoric, not a gradeable evidence claim.
• Tirzepatide “SLAYS” semaglutide: better appetite suppression, fewer side effects, and gut remodeling giving him a permanently smaller appetite. Verdict — partly supported. Greater weight loss than semaglutide is supported by SURMOUNT-5 (−20.2% vs −13.7%), and GI discontinuations were lower in that trial (2.7% vs 5.6%). “Fewer side effects” overall overstates it — GI events were the most common events in both arms and overall AE profiles were similar. “Gut remodeling / permanently smaller appetite” is unproven/contradicted per SURMOUNT-4.
• Comparison chart: tirzepatide ~22% maximum weight loss (15 mg trial exposure, 72 weeks). Verdict — supported (with estimand context). Matches SURMOUNT-1 efficacy estimand (−22.5% at 15 mg*) and the 3-year prediabetes efficacy figure (−22.9%); the regimen estimand is −20.9%.

TRT Dad Unscripted

Sources: episode #83 show notes (11 Sep 2026); GLP-1 discussion transcript.

• Ep. 83 framing: for obese men with low testosterone, the debate is TRT vs GLP-1s — obesity itself can suppress testosterone, significant weight loss could bring testosterone back naturally, and combining fat loss with TRT + resistance training + nutrition is the alternative path (presented largely as questions, not a declared winner). Verdict — partly supported. The underlying physiology is consistent with the wider literature (obesity-associated hypogonadism improves with substantial weight loss), and the show's refusal to declare a winner matches the evidence. But there is no tirzepatide RCT powered on testosterone recovery in this population, and the episode's framing is a discussion question — it should not be cited as a finding.
• Descriptive claim in the GLP-1 discussion: semaglutide acts through one pathway (GLP-1) whereas tirzepatide has GLP-1 and GIP — “two slightly different medications,” prescriber should be consulted before switching. Verdict — supported. Accurate receptor pharmacology (Pharmacology, above) and an appropriately cautious framing.

Dan the Bodybuilder — Steroids Podcast

No specific attributable tirzepatide claim was extractable for grading. The show unambiguously covers tirzepatide comparatively (Ep. 77's stated subject is a semaglutide/tirzepatide/retatrutide comparison; a Dawson Weiss guest-episode summary states Dan and his guest “discussed … the effects of different compounds like Retatrutide and Tirzepatide on body composition and performance” and the guest's personal experiences). However, the accessible transcript text for Ep. 77 did not contain a quotable, specific tirzepatide claim by Dan himself, and the guest-episode summary contains no quantitative claim. Grading an unquoted claim would require inventing one, so none is graded. This is a “could not verify” item (Open questions, below), not evidence the show has made no claims elsewhere.

Peptide of the Week — JD Denham & William T. Haas

Source: podcast episode notes as aggregated on Podtail (the GLP-1 breakdown below appears in the show's episode notes, e.g. Peptide Q&A #25). Attribution caveat: Podtail is an aggregator and repeats this GLP-1 breakdown across multiple episode pages; the claims are graded as the show's published episode-note content, not as verbatim spoken quotes.

• Tirzepatide = GLP-1 + GIP; ~20–22% average weight loss; still suppresses appetite heavily. Verdict — supported. Dual agonism (Pharmacology, above); −20.9% regimen / −22.5% efficacy in SURMOUNT-1; −20.2% in SURMOUNT-5.
• Tirzepatide has “much less nausea than semaglutide”. Verdict — partly supported / overstated. SURMOUNT-5: GI discontinuations were lower on tirzepatide (2.7% vs 5.6%), but nausea/GI events were common in both arms and no data support “much less” as an incidence claim. Cross-trial nausea rates (SURMOUNT-1 25–33% vs STEP 1 semaglutide ~44% in its own trial) point the same direction but are not a head-to-head incidence comparison.
• “Less muscle loss” on tirzepatide (vs semaglutide's “fat + muscle loss (indiscriminate)”). Verdict — partly supported. Cross-trial DXA shares (~25% lean for tirzepatide vs ~39–40% for semaglutide) favour tirzepatide, but there is no head-to-head DXA trial, and ~25% lean loss is not “muscle-sparing” in absolute terms.
• GIP “improves insulin efficiency, reduces nausea, enhances fat usage”; better insulin function → “better nutrient partitioning”. Verdict — unproven (mechanistic). GIP-arm contributions in humans have never been isolated (Pharmacology, above); anti-emetic and adipose effects are preclinical hypotheses. “Nutrient partitioning” has no clinical-trial endpoint.
• Hair loss, fatigue etc. “are not from the drug — they come from not eating (malnourishment)”. Verdict — partly supported / overstated. Rapid weight loss and under-nutrition are recognised drivers of telogen effluvium and fatigue, and the label itself associates hair loss with weight reduction. But the label lists hair loss as an adverse reaction at 4–5% vs 1% on placebo — “not from the drug” erases a documented, placebo-contrasted association and asserts a sole mechanism that has not been proven.

Creator summary

The creator layer at a glance — best-supported and weakest claims retrieved

Creator	Best-supported claim retrieved	Weakest claim retrieved	Basis
Vigorous Steve	— none retrieved —	— none retrieved —	No findable attributable claim
Big Paul	GLP-1s may interfere with a bulking phase (hedged, mechanism-consistent).	— no other quotable claim retrieved —	Episode description
Dr. Alex Tatem	The 100 lb weight loss itself is within SURMOUNT-1's absolute range at high starting weights.	Sperm count ×10 attributed to tirzepatide (n=1 anecdote); compounded “microdosing” tolerability.	Episode description (Diary of a CEO)
Hunter Williams	~22% maximum weight loss figure (efficacy estimand).	“Permanently rewires the brain”; “everyone should use GLP-1s”.	Own newsletter
TRT Dad Unscripted	Tirzepatide = GLP-1 + GIP, vs semaglutide's single pathway.	— framing presented as questions, not findings —	Show notes + transcript
Dan the Bodybuilder	— none extractable —	— none extractable —	Could not verify
Peptide of the Week	Dual agonism and ~20–22% average weight loss.	“Much less nausea”; hair loss “not from the drug”.	Aggregator episode notes (Podtail), not verbatim transcripts

Sources: Big Paul / Anabolic Bodybuilding (Podbean) · Tatem — Diary of a CEO (YouTube) · Hunter Williams (Beehiiv) · TRT Dad Unscripted (TopPodcast; YouTube-to-transcript) · Dan the Bodybuilder / Steroids Podcast (Podscan; Podtail) · Peptide of the Week (Podtail) · Vigorous Steve — no source: no attributable tirzepatide claim found

Claim check

Popular claims about tirzepatide, traced and graded at the tier their evidence actually sits at:

• Not supported

  “Dual agonist = double the loss.”

  Folklore extrapolation from a true mechanism, contradicted by the head-to-head human RCT. SURMOUNT-5: −20.2% vs −13.7% = 1.47× the percentage loss (6.5 points more), not 2×. Receptor count does not scale linearly with outcome; tirzepatide is also a weaker, biased GLP-1R agonist per molecule than native GLP-1 (Pharmacology, above).

• Not supported

  “Muscle loss is worse than other GLP-1s.”

  Contradicted by the human RCT substudies, with the cross-trial caveat stated. SURMOUNT-1 DXA lean share ~25% (n=160) — identical proportion to its own placebo arm and, cross-trial, lower than STEP 1 semaglutide's ~39–40% lean share (n=140). Absolute lean kg lost is larger because total loss is larger. No head-to-head DXA exists, so “better than semaglutide” is also not proven — but “worse” is the opposite of the available direction.

• Not supported

  “It permanently resets your set point.”

  Folklore; contradicted by human RCT withdrawal data. No trial measured a “set point,” and SURMOUNT-4 shows the defended weight reasserts on withdrawal (+14.0% in a year off drug). Continued treatment maintains loss; that is pharmacological maintenance, not a permanent reset. Hunter Williams' permanence claims (Creator layer, above) are the creator version of this myth.

• Needs the qualifier

  “Weight always fully returns after stopping.”

  Partly supported — direction supported; “always fully” contradicted at 1 year. SURMOUNT-4 (n=783 lead-in; 670 randomised): after 36 weeks (−20.9% mean), the switch-to-placebo group regained +14.0% over 52 weeks but still ended −9.9% from original baseline at week 88, vs −25.3% continued. Only 16.6% of the withdrawal group kept ≥80% of their loss (vs 89.5% continued), and a 2025 post-hoc analysis found ≥25% regain of lost weight in most withdrawers within a year, with reversal of cardiometabolic gains. So: substantial regain is the rule; complete return for everyone within a year is not what the trial showed. Longer-term (multi-year) off-treatment trajectories are not established by this 52-week withdrawal window.

• Unproven

  “Tirzepatide builds / directly burns fat and muscle via GIP ‘nutrient partitioning’.”

  Folklore built on preclinical mechanism. Label pharmacodynamics attribute weight loss to reduced calorie intake/appetite; no human trial shows anabolism or a weight-independent fat-oxidation effect of clinical size. GIP adipose effects are preclinical (Pharmacology, above).

• Unproven

  “The thyroid warning means it causes thyroid cancer in people.”

  Animal evidence misread as human. Rat 2-year study: dose/duration-dependent C-cell tumours at clinically relevant exposures; 6-month transgenic mouse study negative; the label states human relevance not determined; no human causal evidence. The contraindication (MTC/MEN 2) is precautionary. Unproven in humans; supported in rats.

Regulatory status

All dates are US FDA dates. The current Zepbound label on DailyMed is the version revised 01/2026.

Approval history

Date	Action
13 May 2022	Mounjaro (tirzepatide) approved — adjunct to diet and exercise to improve glycemic control in adults with type 2 diabetes. First dual GIP/GLP-1 agonist approved anywhere.
8 Nov 2023	Zepbound approved — chronic weight management in adults with obesity, or overweight with ≥1 weight-related comorbid condition, with reduced-calorie diet and increased physical activity.
20 Dec 2024	Zepbound label expanded — moderate-to-severe obstructive sleep apnea in adults with obesity (first drug approved for OSA). Based on SURMOUNT-OSA.
19 Dec 2024	FDA declared the tirzepatide injection shortage resolved (context for the end of shortage-based compounding discretion in early 2025 — regulatory context, not an approval).
Dec 2025	SURPASS-CVOT published in NEJM — not itself a regulatory action; the evidentiary basis for the 2026 CV indication. (The pediatric expansion of Mounjaro to ages 10+ is reflected in Lilly's 2026 wording — “adults and children 10 years of age and older” — the exact expansion date was not independently verified in this review; see Open questions.)
28 Aug 2026	Mounjaro label expanded — reduce risk of major adverse cardiovascular events (CV death, nonfatal MI, nonfatal stroke) in adults with type 2 diabetes at high risk, based on SURPASS-CVOT noninferiority vs dulaglutide.

Status as of late October 2026: prescription-only; both brands carry the thyroid C-cell boxed warning; Zepbound is not approved for children (label: safety/effectiveness in children not established for Zepbound); tirzepatide is not approved for physique/performance use, prediabetes prevention as a standalone indication, HFpEF, or any “optimisation” use in lean people — the SUMMIT and prediabetes data are trial evidence without corresponding indications at the time of writing.

Open questions

Marked plainly, because they are open — where the evidence genuinely runs out:

• GIP's share of the effect is unquantified in humans. No trial isolates the GIP arm; the agonism-vs-antagonism paradox (Pharmacology, above) is unresolved. Any percentage split of tirzepatide's effect between receptors, from any source, exceeds the evidence.
• Strength/function with tirzepatide: no direct strength endpoints in the pivotal programme. Whether DXA lean loss costs strength in older or trained people on tirzepatide is an open RCT question.
• Head-to-head body composition: no DXA/MRI comparison of tirzepatide vs semaglutide exists; the widely quoted 25% vs ~40% lean-share contrast is cross-trial.
• Cardiovascular outcomes in obesity without diabetes: no SELECT-analogue for tirzepatide; MACE superiority vs placebo is unproven in any population (SURPASS-CVOT was active-comparator noninferiority).
• Long-term safety beyond ~4 years: thyroid C-cell human relevance, pancreatitis/gastroparesis causal frequencies, and effects of decade-long use are not established; postmarketing signals (ileus, aspiration, kidney injury) cannot yield rates.
• Off-treatment durability beyond 1 year after withdrawal, optimal maintenance strategies, and outcomes in lean/athletic populations are all untested in RCTs — claims in any direction on these points should be graded unproven by default.

Sources

1. FDA Zepbound (tirzepatide) prescribing information, DailyMed (current version, rev. 01/2026) — verified live 28 Oct 2026. dailymed.nlm.nih.gov (Zepbound label) · Lilly — FDA approves Mounjaro to reduce cardiovascular risk in adults with type 2 diabetes, 28 Aug 2026. lilly.mediaroom.com (Mounjaro CV indication) · Lilly Medical — incidence of intestinal obstruction/ileus with Zepbound. medical.lilly.com (ileus incidence)
2. Jastreboff AM et al. SURMOUNT-1, N Engl J Med 2022;387:205–216 (DOI 10.1056/NEJMoa2206038). investor.lilly.com (SURMOUNT-1 NEJM release) · hormoneinsight.com (SURMOUNT-1) · Jastreboff et al. SURMOUNT-1 3-year prediabetes, NEJM 2024 — Lilly release. nasdaq.com (prediabetes release)
3. Look M et al. SURMOUNT-1 DXA substudy, Diabetes Obes Metab 2025 (PMID 39996356). pubmed.ncbi.nlm.nih.gov/39996356 · pmc.ncbi.nlm.nih.gov/articles/PMC11965027 · Wilding JPH et al. STEP 1 DXA substudy, J Endocr Soc 2021. pmc.ncbi.nlm.nih.gov/articles/PMC8089287
4. Frías JP et al. SURPASS-2, N Engl J Med 2021;385:503–515 — summary. gpnotebook.com (SURPASS-2) · Aronne LJ et al. SURMOUNT-5, NEJM 2025 — results summary. rx.com (SURMOUNT-5 summary)
5. Aronne LJ et al. SURMOUNT-4, JAMA 2024;331:38–48 (NCT04660643) — summary. microdose-glp.com (SURMOUNT-4 summary) · Post-hoc cardiometabolic analysis, JAMA Intern Med 2025/2026. jamanetwork.com (SURMOUNT-4 post-hoc) · Malhotra A et al. SURMOUNT-OSA, NEJM 2024 — summary. thrivesleepusa.com (SURMOUNT-OSA)
6. Nicholls SJ et al. SURPASS-CVOT, N Engl J Med 2025;393:2409–2420 (DOI 10.1056/NEJMoa2505928) — reports. tctmd.com (SURPASS-CVOT) · Packer M et al. SUMMIT, NEJM 2025 — ACC summary. acc.org (SUMMIT)
7. Willard FS et al. Tirzepatide is an imbalanced and biased dual GIP and GLP-1 receptor agonist, JCI Insight 2020. pmc.ncbi.nlm.nih.gov/articles/PMC7526454 · Clinical Potential of GIP in Type 2 Diabetes and Obesity, Diabetes Care 2026. diabetesjournals.org (GIP review)
8. Skeletal muscle / incretin review, Adv Ther 2026 (SURPASS-3 MRI and SEMALEAN summaries). link.springer.com (Adv Ther 2026) · Muscle strength review, Br J Pharmacol 2026 (PMID 41577337). pubmed.ncbi.nlm.nih.gov/41577337
9. Creator sources, cited as claims graded in the Creator layer, not evidence — Big Paul / Anabolic Bodybuilding (episode description, 9 Oct 2026). anabolicbodybuilding.podbean.com · Dr Alex Tatem — Diary of a CEO episode (YouTube description). youtube.com (Diary of a CEO, Tatem)
10. Creator sources — Hunter Williams (Hunter's Health Hacks / Beehiiv). huntershealthhacks.beehiiv.com (GLP-1s vs Men in Black) · huntershealthhacks.beehiiv.com (Everyone Can Use GLP-1s) · huntershealthhacks.beehiiv.com (Tirzepatide Slays Semaglutide) · huntershealthhacks.beehiiv.com (Next Wave of GLP-1s)
11. Creator sources — TRT Dad Unscripted, Ep. 83 notes (11 Sep 2026). toppodcast.com (TRT Dad Unscripted) · GLP-1 discussion transcript. youtubetotranscript.com (TRT Dad GLP-1 discussion) · Dan the Bodybuilder / Steroids Podcast, Ep. 77. podscan.fm (Steroids Podcast Ep. 77) · Dawson Weiss episode. podtail.com (Steroids Podcast, Dawson Weiss) · Peptide of the Week, episode notes via Podtail. podtail.com (Peptide of the Week Q&A #25) · Vigorous Steve — no source: no attributable tirzepatide claim found.
12. Tirzepatide — evidence deep-dive, research report filed 28 Oct 2026. Every figure, attribution and hedge on this page follows that report.

Molecule

Single peptide · C20 fatty-diacid conjugate

Class

Triple GIP / GLP-1 / glucagon agonist

Code

LY3437943 (Eli Lilly)

Status

Investigational — approved nowhere

The health layer — what this could mean for ordinary health, if it is ever approved

Status first, because everything else is conditional: retatrutide is investigational and approved nowhere. Nothing in this section describes a treatment anyone can be prescribed today. It describes what the trial evidence would mean if regulators eventually approve it and if longer-term safety holds up.

The context. Obesity is described in the Phase 2 NEJM paper as “a chronic, treatable, neurometabolic disease that is projected to affect nearly a quarter of the world population by 2035” (Jastreboff et al., NEJM 2023). The World Obesity Atlas projects adults living with obesity rising from 0.81 billion in 2020 to 1.53 billion in 2035 (World Obesity Federation, Atlas 2024). For ordinary health, the relevance is not cosmetic: excess adiposity drives type 2 diabetes, fatty liver disease, knee osteoarthritis, obstructive sleep apnea, hypertension and dyslipidaemia — exactly the conditions the retatrutide programme measured.

What the Phase 2 outcomes would translate to clinically

• Weight of a magnitude previously associated mainly with surgery. A mean −24.2% at 48 weeks (Phase 2, highest exposure) means an average participant starting at ~108 kg lost ~26 kg. At that magnitude, downstream measures moved with it: 72% of participants with prediabetes at baseline reverted to normoglycaemia vs 22% on placebo, and 41% (combined 8 mg group) and 30% (12 mg group) were able to stop at least one blood-pressure medicine during the trial. (Jastreboff 2023)
• Diabetes control into the non-diabetic range for many. In Phase 2 type 2 diabetes, HbA1c fell up to −2.02 points, and secondary reporting of the Lancet paper notes 67% of the highest-exposure group reached HbA1c <5.7% — below the diabetes/prediabetes threshold — a result that, if durable, is the territory of diabetes-remission discussions rather than incremental control. (Rosenstock 2023)
• Liver fat. In the MASLD substudy, liver fat fell by >80% relatively at the higher exposures and normalised (<5%) in 86–93% of higher-exposure participants vs 0% on placebo (Sanyal et al., Nature Medicine 2024). Fatty liver is common, usually silent, and has few treatments — arguably the most “ordinary health” result in the programme. But it is a 98-person substudy, not an outcomes trial.
• Joints and sleep (Phase 3). TRIUMPH-1's peer-reviewed report includes pre-specified baskets: knee-osteoarthritis pain (WOMAC) fell −3.2 to −3.6 points vs −1.9 on placebo (n=574), and apnoea–hypopnoea index fell −22.9 to −34.3 events/hour vs −9.9 on placebo (n=243) (NEJM 2026, via concordant secondary reporting). TRIUMPH-4 (sponsor topline) reported 12.0–14.1% of retatrutide participants completely free of knee pain at 68 weeks vs 4.2% on placebo. For daily life — walking, stairs, sleeping — those are the outcomes people feel, not percentages on a scale.

Pharmacology — one molecule, three receptors

Retatrutide is a single peptide conjugated to a fatty-diacid moiety, based on a GIP peptide backbone, engineered to activate three class-B G-protein-coupled receptors simultaneously (Jastreboff 2023, citing Coskun et al., Cell Metabolism 2022). The fatty-diacid (C20) conjugation enables albumin binding and underlies its long half-life.

Activity at each receptor — as published

Two complementary published framings exist, and both are reported here because secondary sources often conflate them:

Retatrutide at the three human receptors (Coskun et al., Cell Metab 2022, as reported in Jastreboff 2023 and Sanyal 2024)

Measure	GIP receptor	GLP-1 receptor	Glucagon receptor
Potency vs the endogenous ligand	8.9× more potent than native GIP	0.4× — less potent than native GLP-1	0.3× — less potent than native glucagon
In-vitro EC₅₀ (human receptors)	0.0643 nM	0.775 nM	5.79 nM

The interpretation: retatrutide is GIP-dominant by design, with deliberately sub-native activity at the GLP-1 and glucagon receptors — the engineering premise being that incretin (GIP+GLP-1) insulinotropic and appetite effects can counterbalance glucagon's glucose-raising effect while retaining its other actions.

What each arm does physiologically

• GLP-1 receptor (best-established arm; human RCT evidence across the drug class). Glucose-dependent insulin secretion, suppression of inappropriate glucagon secretion, delayed gastric emptying, and central appetite/satiety signalling. This arm is the shared basis of semaglutide-type drugs and explains most of the GI adverse-effect profile and a large part of the reduced food intake.
• GIP receptor (human evidence from the dual-agonist tirzepatide, plus retatrutide trials). A second incretin: glucose-dependent insulin secretion, plus effects on adipose-tissue lipid handling and central appetite pathways. Whether GIP agonism's contribution is mainly insulinotropic, adipose, or central — and how it interacts with GLP-1 — remains mechanistically debated even for tirzepatide; retatrutide data do not resolve it.
• Glucagon receptor (the novel arm). Endogenous glucagon raises hepatic glucose output, which is why adding it to a diabetes drug is counter-intuitive. The design hypothesis is that its other actions can be retained — increased energy expenditure, hepatic fatty-acid oxidation, reduced hepatic lipogenesis, amino-acid catabolism, and reduced meal size — while GIP/GLP-1 activity prevents hyperglycaemia. The NEJM 2023 discussion explicitly frames this as speculation: glucagon activation “may augment effects on energy intake, substrate utilization, and energy expenditure.”

The glucagon hypotheses — graded by evidence tier

What actually exists behind each glucagon claim

Hypothesis	Evidence tier	What actually exists
Glucagon arm increases energy expenditure	Preclinical (animal) + hypothesis in humans	In diet-induced-obese mice, the glucagon component of LY3437943 increased energy expenditure while the GIP/GLP-1 components reduced intake (Coskun 2022). A later DIO-MASH mouse study found no significant energy-expenditure difference vs vehicle despite weight loss (Briand et al., Obesity 2026) — illustrating model dependence. No published retatrutide human trial reports a direct energy-expenditure endpoint. Human EE increase: unproven.
Glucagon arm increases hepatic fat oxidation / reduces liver fat	Human RCT for the outcome; preclinical/indirect for the mechanism	Liver fat fell −81.4% / −82.4% relatively at 24 weeks at the two higher Phase 2 exposures (Sanyal 2024) — a real human RCT outcome. But the substudy reports liver-fat reduction was strongly associated with weight loss and visceral-fat change, so the data do not isolate a direct glucagon-mediated hepatic effect from losing ~20%+ of body weight. Dose-dependent rises in β-hydroxybutyrate (a fat-oxidation/ketogenesis marker) in Phase 1b are consistent with, but do not prove, increased fat oxidation.
Glucagon arm reduces food intake / meal size independently	Preclinical / physiological literature	Cited by Jastreboff 2023 from glucagon physiology literature; not separately measurable in the human triple-agonist trials, where all three arms act together.
Glucagon arm explains the extra weight loss vs dual agonists	Unproven in humans	No trial compares retatrutide against a matched GIP/GLP-1-only control. The decisive test — TRIUMPH-5 vs tirzepatide — had not read out as of late Oct 2026. Cross-trial comparisons cannot attribute causality to the glucagon arm.

Pharmacokinetics — as published

• Half-life ≈ 6 days; pharmacokinetics dose-proportional; this enables the once-weekly subcutaneous administration used in all trials (Jastreboff 2023, citing Urva et al. 2022).
• First-in-human single-dose study: median Tmax 12–72 hours; geometric mean terminal half-life 134–165 hours across cohorts (Phase 1 data as summarised in peer-reviewed reviews).
• Phase 1b (12 weeks, type 2 diabetes, n≈72 dosed): placebo-adjusted weight reduction of 8.96 kg (~10%) at 12 weeks in the highest-exposure group (Urva 2022); adverse events 63% on retatrutide vs 54% on placebo, mostly GI.
• Gastric-emptying delay was greatest after the first dose and attenuated with subsequent doses despite escalation (Urva 2022, as summarised in reviews) — partial tachyphylaxis, as with other GLP-1-based agents.
• Not published in the sources reviewed: dedicated renal/hepatic-impairment PK studies, a formal exposure–response model, or drug-interaction characterisation sufficient for prescribing. These are pre-approval gaps, not findings.

Sources: Jastreboff et al., NEJM 2023 · Coskun et al., Cell Metab 2022 · Sanyal et al., Nat Med 2024 · Urva et al., Lancet 2022 · Briand et al., Obesity 2026

The bodybuilding / fitness layer — body composition, the expenditure question, and the gaps

No retatrutide trial has enrolled a bodybuilding population or measured strength or performance outcomes. Trial populations were people with obesity and/or type 2 diabetes — not lean or trained athletes. Fitness-community use is described in this section only as reported practice/claims.

Body composition — the DXA substudy (read carefully: it is not the obesity trial)

The only published retatrutide DXA substudy is Coskun et al., Lancet Diabetes & Endocrinology 2025 — a substudy of the Phase 2 type 2 diabetes trial (36 weeks), frequently misattributed online to the NEJM obesity trial. Of 189 enrolled in the substudy, 155 had baseline DXA and only 103 had paired baseline + week-36 scans — attrition of ~45% from enrolment to paired scans is a major interpretive limit.

DXA substudy, type 2 diabetes — change at 36 weeks, least-squares mean % (*trial exposures)

Measure	Placebo	Dulaglutide 1.5 mg*	4 mg*	8 mg*	12 mg*
Total fat mass	−4.5%	−2.6%	−15.2%	−26.1%	−23.2%
Total lean mass	—	—	−6.0%	−12.5%	−11.7%

The lowest trial exposure (not tabulated) showed −4.9% fat mass. The non-monotonic 8 mg > 12 mg fat-mass ordering most plausibly reflects small-sample noise. In absolute terms, lean-mass loss reached up to −6.5 kg alongside up to −10.9 kg total fat-mass loss and up to −0.6 kg android visceral fat mass. Fat accounted for roughly 62–69% of the weight lost across retatrutide exposures. The substudy authors' own interpretation: “The proportion of lean mass loss to weight loss was similar to other obesity treatments” and the findings “could provide reassurance that a greater proportion of lean mass is not lost with retatrutide despite the overall increased weight loss.”

What DXA lean mass is not: DXA “lean mass” includes water, glycogen and connective tissue, not just skeletal muscle; the substudy measured no strength, function or performance endpoint. Claims of “muscle preservation” or “muscle loss” in either direction go beyond the measurement. And there is no published DXA substudy of the Phase 2 obesity population, and none in Phase 3 populations — lean-mass data in non-diabetic obesity, the population closest to fitness-community interest, are a genuine data gap (Open questions, below).

The energy-expenditure question

Glucagon has a thermogenic reputation from physiology and animal work. For retatrutide specifically: in animals, an energy-expenditure increase was attributed to the glucagon component in DIO mice in the discovery paper (Coskun 2022), while a 2026 DIO-MASH mouse study found weight loss driven by reduced intake with unchanged energy expenditure (Briand 2026) — even the preclinical record is not uniform. In humans, no published direct measurement (indirect calorimetry, metabolic chamber, doubly-labelled water) demonstrating increased energy expenditure on retatrutide was found; the human evidence is indirect — large weight and liver-fat losses, β-hydroxybutyrate rises in Phase 1b, and a heart-rate increase consistent with, but not proof of, sympatho-metabolic activation.

Sources: Coskun et al., Lancet Diabetes Endocrinol 2025 · Coskun et al., Cell Metab 2022 · Briand et al., Obesity 2026

Efficacy — with numbers

Phase 2 obesity — Jastreboff et al., NEJM 2023 (NCT04881760; n=338; no type 2 diabetes; 48 weeks) — peer-reviewed

Least-squares mean % change in body weight, efficacy estimand (*trial exposures; combined escalation arms where applicable)

Arm	24 weeks (primary)	48 weeks
Placebo	−1.6%	−2.1%
Retatrutide 1 mg*	−7.2%	−8.7%
Retatrutide 4 mg* (combined)	−12.9%	−17.1%
Retatrutide 8 mg* (combined)	−17.3%	−22.8%
Retatrutide 12 mg*	−17.5%	−24.2%

By individual escalation arm at 48 weeks: 4 mg −16.3% (started 2 mg) / −17.8% (started 4 mg); 8 mg −21.7% / −23.9%; 12 mg −24.2% (−26.2 kg). At 48 weeks, ≥5% / ≥10% / ≥15% loss was achieved by 92% / 75% / 60% (4 mg), 100% / 91% / 75% (8 mg), 100% / 93% / 83% (12 mg) vs 27% / 9% / 2% (placebo); 26% of the 12 mg group lost ≥30%. Weight curves had not plateaued at 48 weeks. Greater % loss in women than men and in baseline BMI ≥35 than <35 (prespecified exploratory).

Phase 2 type 2 diabetes — Rosenstock et al., Lancet 2023 (NCT04867785; n=281; 36 weeks; dulaglutide 1.5 mg active comparator) — peer-reviewed

HbA1c change at 24 weeks · body-weight change at 36 weeks (*trial exposures)

Arm	HbA1c, 24 wk	Weight, 36 wk
Placebo	−0.01%	−3.00%
Dulaglutide 1.5 mg*	−1.41%	−2.02%
Retatrutide 0.5 mg*	−0.43%	−3.19%
Retatrutide 4 mg* (escalation / none)	−1.39% / −1.30%	−7.92% / −10.37%
Retatrutide 8 mg* (slow / fast escalation)	−1.99% / −1.88%	−16.81% / −16.34%
Retatrutide 12 mg* (escalation)	−2.02%	−16.94%

HbA1c reductions were significantly greater than placebo at all but the 0.5 mg exposure, and greater than dulaglutide at the 8 mg slow-escalation and 12 mg exposures; results were consistent at 36 weeks.

Liver-fat substudy — Sanyal et al., Nature Medicine 2024 (substudy of NCT04881760; n=98 with MASLD and ≥10% liver fat by MRI-PDFF) — peer-reviewed

Relative liver-fat change (*trial exposures)

Arm	24 weeks	48 weeks	Liver fat <5% at 24 wk / 48 wk
Placebo	+0.3%	−4.6%†	0% / —
Retatrutide 1 mg*	−42.9%	−51.3%	27% / —
Retatrutide 4 mg*	−57.0%	−59.0%	52% / —
Retatrutide 8 mg*	−81.4%	−81.7%	79% / 89%
Retatrutide 12 mg*	−82.4%	−86.0%	86% / 93%

† Secondary summaries differ on the sign of the small 48-week placebo value (+4.6% vs −4.6%); the retatrutide-arm values are consistent across sources. Evaluable MRI data fell to ~44% of substudy participants by week 48 (discontinuation / out-of-window scans), a further limit. Liver-fat change correlated with weight and visceral-fat change.

Phase 3 efficacy — estimand and tier flagged for every row

Phase 3 headline results — sponsor release vs peer-reviewed, efficacy vs treatment-regimen estimand

Trial	Population / n / duration	Highest-exposure result	Estimand & tier
TRIUMPH-4	Obesity/overweight + knee OA, no diabetes; n=445; 68 wk	−28.7% (12 mg*) vs −2.1% placebo; treatment-regimen: −23.7% vs −4.6%	Efficacy estimand; sponsor release, 11 Dec 2025
TRANSCEND-T2D-1	T2D, diet/exercise only; n=537; 40 wk	HbA1c −1.94% and weight −15.3% (12 mg*, treatment-regimen) vs −0.81% / −2.6% placebo; sponsor efficacy estimand: HbA1c up to −2.0%, weight −16.8%	Peer-reviewed, Lancet Jun 2026 + sponsor release 19 Mar 2026
TRIUMPH-1	Obesity/overweight, no diabetes; n=2,339; 80 wk	−25.0% (12 mg*) vs −3.9% placebo (treatment-regimen); sponsor efficacy estimand −28.3% vs −2.2%; 104-wk extension (n=532, BMI ≥35): −30.3%	Peer-reviewed, NEJM 29 Sep 2026 + sponsor release 21 May 2026
TRIUMPH-2	Obesity/overweight + T2D; n=1,152; 80 wk	−18.8% (12 mg*) vs −5.1% placebo (treatment-regimen); sponsor efficacy estimand −20.8% vs −4.0%; A1C −1.5% (12 mg*) / −1.6% (9 mg*) vs −0.2%	Peer-reviewed, Lancet Sep 2026 + sponsor release 23 Jul 2026
TRIUMPH-3	BMI ≥35 + established CVD; n=1,949; 80 wk	−22.6% (12 mg*) vs −3.2% placebo	Efficacy estimand; sponsor release, 23 Jul 2026; peer review pending

Phase 3 did not simply repeat Phase 2: the same drug reads −28.3% or −25.0% depending on estimand in TRIUMPH-1, and −20.8% / −18.8% in obesity with type 2 diabetes in TRIUMPH-2. Any single-number summary that doesn't name its population and estimand is incomplete.

Sources: Jastreboff et al., NEJM 2023 · Rosenstock et al., Lancet 2023 · Sanyal et al., Nat Med 2024 · Lilly releases, 11 Dec 2025 – 23 Jul 2026 · TRIUMPH-1, NEJM 2026 · TRIUMPH-2, Lancet 2026 · TRANSCEND-T2D-1, Lancet 2026

Safety — with numbers

Phase 2 obesity — Jastreboff 2023, Table 3 (safety population n=337) — peer-reviewed

Adverse events by arm (*trial exposures; 4 mg and 8 mg split by starting exposure)

Event	Placebo (n=70)	1 mg*	4 mg* (start 2 / start 4)	8 mg* (start 2 / start 4)	12 mg* (n=62)
Any adverse event	70%	84%	73% / 85%	80% / 94%	92%
Nausea	11%	14%	18% / 36%	17% / 60%	45%
Diarrhoea	11%	9%	12% / 12%	20% / 20%	15%
Vomiting	1%	3%	12% / 12%	6% / 26%	19%
Constipation	3%	7%	15% / 6%	11% / 11%	16%
Decreased appetite	9%	13%	18% / 24%	11% / 31%	29%
Cardiac arrhythmia category†	3%	4%	0% / 6%	0% / 14%	11%
Skin hyperesthesia / related	1%	1%	6% / 6%	3% / 14%	13%
Discontinued due to adverse event	0%	7%	6% / 9%	14% / 6%	16%

† MedDRA cluster including supraventricular arrhythmias and conduction disorders. GI events were dose-related, mostly mild–moderate, concentrated in escalation, and partially mitigated by a 2 mg vs 4 mg starting exposure. Serious AEs: 4% in both retatrutide (pooled) and placebo groups; one death (drowning, 4 mg start-4 arm) assessed as unrelated.

Heart rate — magnitude and time course

NEJM 2023: heart rate “increased in a dose-dependent manner with retatrutide up to 24 weeks and then declined thereafter” (36 and 48 weeks); the authors compared the increases to those reported for GLP-1 receptor agonists.

Arrhythmia, blood pressure, pancreas, hypoglycaemia

• Arrhythmia (Phase 2): events in the MedDRA arrhythmia/conduction category were mostly mild–moderate; one severe event — prolonged QT syndrome — occurred in a participant also treated with ondansetron (Jastreboff 2023). TRIUMPH-1: severe/serious arrhythmia or conduction disorders in 0.2–0.9% on retatrutide vs 0.9% on placebo. TRANSCEND-T2D-1: arrhythmia/conduction events 1–3% on retatrutide vs 0% on placebo. No trial was powered for arrhythmia outcomes.
• Blood pressure: improved, not worsened, on average. Phase 2: systolic/diastolic BP improved at 24 and 48 weeks and antihypertensive medicines were stopped in a substantial minority. Phase 3 (sponsor): systolic BP −14.0 mmHg at 12 mg in TRIUMPH-4 and −9.3 mmHg in TRIUMPH-3 (efficacy estimands). Hypotension events were nonetheless more common on retatrutide in TRIUMPH-2's published report (1–6% vs <1% placebo).
• Pancreatic: Phase 2 obesity: one confirmed acute pancreatitis (12 mg group, <1% overall); amylase/lipase increases otherwise asymptomatic; transient ALT rises >3× ULN in 1% on retatrutide, with mean ALT/AST unchanged or lower by week 48. No medullary thyroid cancer or C-cell hyperplasia cases in Phase 2.
• Hypoglycaemia in non-diabetic participants: no cases of clinically significant hypoglycaemia (level 2, <54 mg/dL, or level 3/severe) in the Phase 2 obesity trial. In the diabetes trials: no severe hypoglycaemia in Phase 2 T2D, TRANSCEND-T2D-1, or TRIUMPH-2's published report. Caveat: these populations were not on insulin/sulfonylurea regimens that drive hypoglycaemia risk in real-world diabetes care; TRANSCEND-T2D-3 (kidney impairment, on basal insulin) had not reported as of late Oct 2026.

Phase 3 safety headlines — sponsor releases; peer-reviewed where noted

Event rates at the 12 mg trial exposure, placebo in parentheses

Signal	TRIUMPH-4 (68 wk)	TRIUMPH-1 (80 wk)	TRIUMPH-2 (80 wk)	TRIUMPH-3 (80 wk)	TRANSCEND-T2D-1 (40 wk)
Nausea	43.2% (10.7%)	42.4% (14.8%)	28.0% (8.0%)	22.4% (5.8%)	26.5% (3.7%)
Diarrhoea	33.1% (13.4%)	32.0% (13.5%)	33.6% (13.2%)	24.4% (8.7%)	~22.8–26.3% across doses (4.5%)
Vomiting	20.9% (0.0%)	25.3% (4.8%)	15.7% (4.2%)	—	17.6% (2.2%)
Dysesthesia	20.9% (0.7%)	12.5% (0.9%)	7.3% (0.7%)	6.4% (1.3%)	4.4–4.5% (0%)
Discontinued due to AE	18.2% (4.0%)	11.3% (4.9%)	7.7% (4.9%)	13.5% (4.8%)	5.1% (0% in published report)

TRIUMPH-2's published AE rates (diarrhoea 27/34/34%, nausea 14/21/28%, vomiting 6/10/16% across 4/9/12 mg; discontinuation 4/12/8% vs 5% placebo; dysesthesia 4/6/7% vs 1%) differ slightly from the sponsor release because the paper reports the treatment-regimen framework — both are shown in this review's sources, and the difference is estimand/population handling, not a contradiction to resolve by picking one silently. Dysesthesia events were generally mild–moderate, mostly resolved during treatment, and rarely caused discontinuation (Lilly releases). In TRIUMPH-4, Lilly notes discontinuation-for-AE rates were “highly correlated with baseline BMI and included discontinuations for perceived excessive weight loss” (12.1% at 12 mg among baseline BMI ≥35).

Deaths in Phase 3 (for completeness): TRIUMPH-2's published report records 7 deaths (2 on 4 mg, 3 on 9 mg, 1 on 12 mg, 1 on placebo), all judged unrelated to study intervention. TRIUMPH-1 records 4 deaths on retatrutide and 2 on placebo, with causes including motor-vehicle accidents and one cardiovascular death 189 days after stopping treatment. Event counts are far too small for causal inference in either direction.

Sources: Jastreboff et al., NEJM 2023 · Lilly releases, 11 Dec 2025 – 23 Jul 2026 · TRIUMPH-1, NEJM 2026 · TRIUMPH-2, Lancet 2026 · TRANSCEND-T2D-1, Lancet 2026

Development status — as of late October 2026

The programme has moved from “Phase 2 promise” to “pre-filing.” Five Phase 3 trials have reported; Lilly states it now has “the clinical data package to support global submissions for retatrutide as a potential treatment for obesity, knee osteoarthritis pain, and obstructive sleep apnea” and plans to submit a BLA to the FDA in Q1 2027, after completing the Chemistry, Manufacturing and Controls package (Lilly, 23 Jul 2026). An earlier ambition to file in 2026 slipped to Q1 2027 for CMC reasons, per sponsor-linked reporting. No approval, price, or FDA decision date exists. A limited Lilly expanded-access programme for patients who cannot enrol in trials was confirmed in August 2026 — it is not commercial availability.

Read out (reported)

Phase 3 trials that have reported

Trial	Registry	Readout	Publication status (late Oct 2026)
TRIUMPH-4 (knee OA)	NCT05931367 per most listings — see flag below	Topline 11 Dec 2025 (sponsor)	No peer-reviewed publication found in sources reviewed
TRANSCEND-T2D-1 (T2D)	NCT06354660	Topline 19 Mar 2026 (sponsor)	Published, Lancet, Jun 2026
TRIUMPH-1 (obesity; OA + OSA baskets)	NCT05929066	Topline 21 May 2026 (sponsor); ADA presentation Jun 2026	Published, NEJM, 29 Sep 2026
TRIUMPH-2 (obesity + T2D)	NCT05929079	Topline 23 Jul 2026 (sponsor); EASD Sep 2026	Published, Lancet, Sep 2026
TRIUMPH-3 (severe obesity + CVD)	NCT05882045	Topline 23 Jul 2026 (sponsor)	Peer-reviewed publication pending in sources reviewed

Pending (registry-based; completion dates are estimates and move)

Phase 3 trials still to report

Trial	Question	Registry expectation
TRANSCEND-T2D-2 (NCT06260722)	Head-to-head vs semaglutide in T2D (n≈1,250)	Primary completion Aug 2026 (reported as actual); no topline as of early Oct 2026
TRANSCEND-T2D-3	T2D with moderate/severe kidney impairment, on basal insulin (n≈320)	Primary completion est. Oct 2026
TRIUMPH-5 (NCT06662383)	Head-to-head vs tirzepatide in obesity (n≈800)	Primary completion est. Nov 2026; study completion est. Dec 2026
TRIUMPH-7 (NCT07035093)	Obesity + chronic low back pain (n=586)	Est. Sep 2027; active, not recruiting
TRIUMPH-6 (NCT06859268)	Weight-loss maintenance (n=643)	Est. Apr 2028
TRIUMPH-8 / TRIUMPH-9	Additional obesity populations (n≈250 / 600)	Est. Jul 2027 / Oct 2028
TRIUMPH-Outcomes (NCT06383390)	Cardiovascular + kidney outcomes (n=10,000; age ≥45, BMI ≥27, ASCVD and/or CKD)	Primary completion est. Feb 2029
MASLD Phase 3 master protocol (NCT07165028)	Liver disease (multi-agent programme including retatrutide)	Primary completion est. Aug 2030

Timeline arithmetic (labelled as inference, not Lilly guidance): a Q1 2027 BLA submission plus a standard ~10-month review points to a possible decision around late 2027–Q1 2028; priority review would shorten that. No review track has been assigned, because no application has been filed yet. Cardiovascular outcomes remain unproven in the meantime: TRIUMPH-3 was not powered for events (MACE-3 HR 1.12, 95% CI 0.64–1.96) and the dedicated outcomes trial does not complete until ~Feb 2029 per registry.

Creator layer — opinion, graded claim by claim

Everything in this section is creator and community content, kept separate from the primary literature above precisely so it is not mistaken for it. Only claims attributable to the named creator in a findable source are graded. Verdicts: supported / partly supported / contradicted / unproven. Doses inside creator claims are the content of the claim being graded — several are trial-unlike regimens, and they are not recommendations. Grading a claim is not endorsing the practice it describes.

Vigorous Steve

Source: “Safest Dose Of Retatrutide (NEXT-LEVEL Appetite Suppression)… Deep-Dive”, YouTube, 31 Mar 2025 (transcript read in full).

• Retatrutide is a triple GLP-1/GIP/glucagon agonist, investigational, “not FDA approved yet… only way to get it [outside trials] are through the Chinese generic peptide resellers.” Verdict — supported (status/mechanism). Matches NEJM 2023 and Lilly's statements; the non-trial supply channel he describes is unregulated — a risk he himself flags only lightly.
• The combination “promotes lipolysis through the glucagon and GIP receptor… makes retatrutide almost the ideal fat loss agent,” and stimulant fat-burners (clenbuterol etc.) “are probably not required now” / are less healthy alternatives. Verdict — partly supported / partly unproven. Appetite suppression and weight loss: supported by RCTs. Direct human lipolysis/energy-expenditure attribution to the glucagon arm: unproven (Pharmacology, above). Comparative safety vs clenbuterol/DNP-type agents: no comparative trial exists; plausible rhetoric, not evidence.
• Community users “get a much more pronounced effect for the same amount of appetite suppression” vs earlier GLP-1 drugs, and retatrutide is “the superior compound in this class… until NA-931 hits.” Verdict — unproven. No head-to-head RCT vs tirzepatide/semaglutide had reported at the time (or by Oct 2026). Cross-trial % comparisons cannot establish superiority.
• After stopping incretin drugs, people regain and “end up fatter than they ever were before” (300 lb → 200 lb → 350 lb example). Verdict — contradicted as stated. Withdrawal data for predecessor drugs show substantial regain toward baseline (e.g. STEP-1 extension ~two-thirds regained), not systematic overshoot beyond baseline. No retatrutide withdrawal data exist. The regain direction is supported; the overshoot claim is not.
• Personal framing: he had not used retatrutide at the time of the deep-dive (“not quite on retatrutide yet”), and described a hypothetical split, thrice-weekly low-exposure regimen he would consider. Verdict — unproven (reported practice claim). All retatrutide RCTs used once-weekly administration with a ~6-day half-life; no trial supports split/thrice-weekly regimens. Noted only as claim content.
• (Podcast appearance) Retatrutide's “real effects on dopamine, motivation, and relationships.” Verdict — unproven. No retatrutide trial reports dopamine, motivation or relationship endpoints. SF-36 quality-of-life domains in Phase 2 are not dopamine measures.

Big Paul — Paul Barnett (Anabolic Bodybuilding)

Sources: Anabolic Bodybuilding podcast listings / episode descriptions (ListenNotes). Graded from episode descriptions only — see flags above.

• A dedicated video (18:13) in which “Big Paul breaks down the new Phase 3 TRIUMPH-4 topline results in adults with obesity + knee osteoarthritis (without diabetes),” framing retatrutide as “Eli Lilly's investigational triple agonist (GLP-1 + GIP + glucagon).” Verdict — supported (as a description of the trial). TRIUMPH-4 is exactly that population/design; Lilly topline 11 Dec 2025: −28.7% at 12 mg, WOMAC pain −4.4 to −4.5 points (sponsor). The framing “investigational” is correct.

Any more specific falsifiable claim could not be verified. No transcript or detailed show notes for Big Paul's retatrutide episodes were accessible in the sources searched — only episode descriptions/titles. Rather than invent claims, this review grades only the description-level claim above and flags the rest as unextractable. The episode title itself (“Does Retatrutide Actually Work Or Is It Overhyped?”) is a question, not a claim.

Dr. Alex Tatem

Sources: Diary of a CEO peptide episode, claim-checked transcript page; “Retatrutide Dosing Explained (Doctor Breaks Down the Data)”, YouTube description; “Retatrutide vs Diabetes: The Results Are INSANE”, YouTube Short description.

• “Clinical trials of retatrutide show a 20% to 25% loss of total body weight within a relatively short period of time” (Diary of a CEO). Verdict — partly supported. 22.8–24.2% at 48 weeks is correct for the Phase 2 8/12 mg exposures (NEJM 2023). “Relatively short” understates that this required 48 weeks at the highest exposures; lower exposures gave 8.7–17.1%.
• “With average weight loss approaching 30%… Retatrutide is outperforming many existing GLP-1 medications, including Semaglutide… and potentially even Tirzepatide” (TRIUMPH-1 video description). Verdict — partly supported. “Approaching 30%” matches the sponsor efficacy estimand (−28.3% at 80 wk; −30.3% in the 104-wk extension subgroup). Outperforming semaglutide/tirzepatide is not established: no head-to-head had reported; TRIUMPH-5 (vs tirzepatide) pending. The description does correctly state retatrutide “remains investigational and is not FDA-approved.”
• Diabetes-trial short: patients with diabetes lost ~19% vs ~25% in previous non-diabetes trials; ~one-third achieved normal-range HbA1c; risks include diarrhoea, dizziness from low blood pressure, increased heart rate, discontinuation. Verdict — supported (with estimand note). ~19% matches TRIUMPH-2's 9 mg efficacy result / 12 mg treatment-regimen result (−18.8%); ~25% matches TRIUMPH-1 treatment-regimen −25.0%. HbA1c <5.7% was reached by 39.3% at 12 mg (efficacy estimand, sponsor) — “approximately one-third” is if anything conservative. Listed risks all appear in the trial reports.

Hunter Williams

Sources: his own newsletter (Hunter's Health Hacks / Beehiiv) — “A Retatrutide Masterclass”; “The Maximum Effective Dose of Retatrutide” (8 Aug 2025); “Landmark New Retatrutide Study” (DXA); “The Peptide Stack That Reverses Liver Fat” (podcast).

• “8 mg is the sweet spot for most people… pushing to 12 mg is rarely worth the side effect burden”; “for many, 8 mg/week may be the maximum effective dose… The Lancet data stands as proof that more is not always better.” Verdict — contradicted as a general claim. The kernel comes from the DXA substudy's non-monotonic fat-mass result (−26.1% at 8 mg vs −23.2% at 12 mg, n=103 paired scans). But weight loss was dose-ordered in every larger trial (Phase 2 obesity: −22.8% combined 8 mg vs −24.2% 12 mg; Phase 3 programmes dose-ordered to 12 mg). A small substudy's secondary pattern does not “prove” a maximum effective dose. Stated here only as claim content, not as dosing guidance in either direction.
• The DXA study “proved that you can lose more fat than ever before, without losing more muscle than you'd expect”; lean loss “approximately 20–40% of total weight lost… 60–80% was actual fat”; retatrutide produced “the largest absolute fat loss ever recorded in an obesity/diabetes drug trial.” Verdict — partly supported. Fat share 62–69% and “proportion similar to other obesity treatments” match the substudy authors. “20–40% lean” brackets the published ~31–38% lean share loosely. “Largest absolute fat loss ever recorded” is an unverifiable superlative — no such cross-trial record is established, and the substudy was in T2D, n=103 paired scans. DXA lean ≠ muscle.
• “76% of people regain the weight after stopping. Here's how to not be one of them” (cycling-off claims in the Masterclass). Verdict — unproven. No retatrutide discontinuation/withdrawal trial has published at any dose. Any % regain figure for retatrutide is extrapolated from other drugs' data and cannot be stated as a retatrutide finding.
• Stacking BAM15 and SLU-PP-332 with retatrutide gives “extra calorie burn without muscle loss” / lean-mass protection; retatrutide is “the most powerful metabolic compound we currently have available. Full stop… nothing else is touching it.” Verdict — unproven (stack); partly supported (superlative). Neither BAM15 nor SLU-PP-332 has human RCT evidence for lean-mass preservation stacked with retatrutide; both are preclinical-stage compounds. The superlative rests on cross-trial comparisons with no head-to-head readout.
• Skin sensitivity and heart-rate effects are real, dose-related phenomena users should expect (Masterclass topics: “the skin sensitivity issue nobody talks about”, “the heart rate question… when it peaks”). Verdict — supported (existence/pattern). Dysesthesia: 7% vs 1% (Phase 2, as skin sensitivity) rising to 12.5–20.9% at 12 mg in Phase 3. Heart rate: dose-dependent, peak ~24 weeks, decline thereafter (NEJM 2023). His suggested personal remedies for these effects are anecdote, ungraded as efficacy claims.
• GLP-1-class drugs are “the best longevity medications… pound for pound” and retatrutide's liver-fat effects are “incredible” (liver-fat podcast). Verdict — unproven (longevity); supported outcome, unproven attribution (liver). No longevity/outcomes RCT exists for retatrutide (TRIUMPH-Outcomes due ~2029). Liver-fat reductions are real RCT outcomes; attributing them specifically to the glucagon arm is not established.

TRT Dad Unscripted

Sources searched: podcast listings / show notes (TopPodcast; Podscan; Podtail).

No findable attributable claim specifically about retatrutide. The show's searchable episode descriptions cover GLP-1 medications as a class — e.g. episode #83 (11 Sep 2026), “Do You Need TRT… or Do You Just Need to Lose Weight?”, debating “TRT vs GLP-1 medications for obese men with low testosterone”, and a summer fat-loss episode discussing “GLPs… pros and cons” — but no episode description, show note or indexed transcript segment naming retatrutide was found. No claim is graded, because none could be attributed without inventing one. A class-level GLP-1 discussion is not a retatrutide claim. This is a search-limitation finding (audio content not fully indexed), not proof the word was never spoken on the show.

Dan the Bodybuilder — Steroids Podcast

Sources: “Dawson Weiss Guest Episode on The Steroids Podcast” (12 Dec 2025), transcript read via Podscan; “Retatrutide Results — GLP1 Agonists Semaglutide Tirzepatide and Retatrutide Compared”, Ep. 77 (26 Oct 2025), listing only. Attribution: the substantive claims below are guest Dawson Weiss's, aired on Dan's show with the host framing/endorsing.

• Dawson Weiss: retatrutide “helped me get very, very lean because I stopped eating as much… staying around that nine to 10 percent [body fat] range was pretty effortless for me.” Verdict — unproven (anecdote consistent with mechanism). Reduced intake is the trials' core mechanism, but trials enrolled people with obesity/T2D, not athletes maintaining 9–10% body fat. A single self-report with unverified product cannot be graded higher.
• Dawson Weiss: retatrutide “flushes away a little bit of water retention… it has to do with sodium… I'd always be pissing on it”; and it “burns through carbs, glycogen a lot faster… a lot easier to go flat.” Verdict — unproven. No trial endpoint measures diuresis, sodium excretion, glycogen kinetics or physique “flatness”. Plausible-sounding extrapolations from eating less / weight loss, presented as drug effects.
• Dawson Weiss (speculating): prolonged glucagon activation “can kind of raise cortisol or stress you out… raise heart rate… make me feel kind of run down, a bit achy”; side effects (mainly sleep) became noticeable at his higher self-administered exposures. Verdict — partly supported / partly unproven. Heart-rate increase: supported (Safety, above). Cortisol elevation, achiness, sleep disturbance as retatrutide effects: no trial evidence found; he himself frames the cortisol point as guesswork (“I'm not a scientist”). Reported exposures in the conversation are reported practice only, not trial regimens for this population and not recommendations.
• Dan (host): an acquaintance's hair falling out and others' iron/platelet problems in the context of GLP-1 use at very high self-administered exposure. Verdict — unproven (second-hand anecdote). Hair loss and nutrient deficiency are biologically plausible with rapid weight loss/reduced intake generally, but these are unattributed second-hand reports with confounders (the guest himself attributes his own low ferritin to another compound, primobolan, not the GLP-1). No causal grade is possible.

Peptide of the Week — JD Denham & William T. Haas

Sources: “Peptide Q&A #26 – Top 5 Peptides…” episode description (Podtail; Castbox mirror); Spotify show page. The episode descriptions contain explicit dosing regimens and stacking protocols — not reproduced here; only falsifiable factual claims are graded.

• Retatrutide is “quickly becoming the most powerful fat loss compound in the peptide world” / “the undisputed #1… Nothing comes close.” Verdict — partly supported. Largest mean weight losses in RCTs to date, cross-trial; no head-to-head proof (TRIUMPH-5 pending). “Undisputed” overstates an open question.
• “Minimal to no nausea compared to semaglutide or tirzepatide.” Verdict — contradicted. Nausea on retatrutide: 45% at 12 mg in Phase 2 (vs 11% placebo) and 42.4% at 12 mg in TRIUMPH-1 — at the top of, not below, the class range. This is the clearest factual error among all creator claims reviewed.
• “Boosts metabolic rate” / “increases insulin sensitivity.” Verdict — unproven (metabolic rate); partly supported (insulin sensitivity). No human energy-expenditure measurement published (Bodybuilding layer, above). Glycaemic/insulin measures did improve in trials, inseparable from weight loss.
• “Torches visceral fat, especially in the midsection.” Verdict — partly supported. Visceral fat mass fell in the DXA substudy (up to −0.6 kg android visceral fat; visceral % reductions reported) and waist circumference fell up to −19.6 cm (Phase 2, 12 mg). “Spot-torching” language implies selective regional action the trials do not demonstrate — fat loss was systemic and tracked total weight loss.
• “Major brain and focus benefits — no more ‘hanger,’ no mental fog”; “keeps brain sharp while fasted.” Verdict — unproven. No cognition/focus endpoint in any retatrutide trial. Appetite/“food noise” reduction is trial-consistent; cognitive benefit is not measured.
• “Adaptable for both weight loss and lean bulking — it's all about dosage” (with bulking stacks). Verdict — unproven, and contrary to the trial direction of effect. Every RCT shows dose-dependent weight loss with appetite suppression in obesity/T2D populations. No trial, at any exposure, shows lean bulking or muscle gain. This is a practice claim with no supporting evidence base. (The stacking protocols themselves are not reproduced — see flags above.)
• Personal report: “JD's experience: [a split low-exposure regimen] got him ultra-lean in just 2 weeks.” Verdict — unproven (anecdote / folklore tier). A two-week “ultra-lean” transformation is incompatible with trial time courses (meaningful fat loss accrued over months) and is an unverifiable self-report using non-trial product. Noted only as claim content; the regimen itself is not reproduced.

Creator summary

The creator layer at a glance — best-supported and weakest claims retrieved

Creator	Best-supported claim retrieved	Weakest claim retrieved	Basis
Vigorous Steve	Triple agonist, investigational, not FDA approved.	Regain leaves people “fatter than they ever were before” (contradicted as stated).	YouTube deep-dive transcript, read in full
Big Paul	TRIUMPH-4 breakdown framing — investigational triple agonist in obesity + knee OA.	— nothing further gradable —	Episode descriptions only; no transcript accessible
Dr. Alex Tatem	Diabetes short: ~19% vs ~25%, ~one-third to normal-range HbA1c, listed risks.	“Outperforming… potentially even Tirzepatide” (no head-to-head).	Diary of a CEO claim-check page; YouTube descriptions
Hunter Williams	Skin sensitivity and heart-rate effects are real and dose-related.	“8 mg is the maximum effective dose… the Lancet data stands as proof” (contradicted).	Own newsletter (Beehiiv)
TRT Dad Unscripted	— none retrieved —	— none retrieved —	No findable retatrutide-specific claim
Dan the Bodybuilder	Heart-rate increase (within guest Dawson Weiss's speculation).	Water “flushing” / glycogen “flatness” presented as drug effects (unproven).	Guest-episode transcript (claims by Dawson Weiss, aired on Dan's show)
Peptide of the Week	Visceral fat fell in the DXA substudy (partly supported).	“Minimal to no nausea” (contradicted); lean-bulking adaptability (unproven).	Episode descriptions; protocols not reproduced

Sources: Vigorous Steve (YouTube transcript; Mark Bell's Power Project) · Big Paul / Anabolic Bodybuilding (ListenNotes listings) · Tatem (HoldsUp claim-check; YouTube) · Hunter Williams (Beehiiv) · TRT Dad Unscripted (TopPodcast; Podscan) · Dan the Bodybuilder / Steroids Podcast (Podscan; ListenNotes) · Peptide of the Week (Podtail; Castbox)

Claim check

Popular claims about retatrutide, traced and graded at the tier their evidence actually sits at:

• Needs the qualifier

  “24% loss means it beats everything.”

  Human RCT number + folklore inference — partly supported; the inference is unproven. −24.2% at 48 weeks is a real Phase 2 RCT result (NEJM 2023) and was the largest mean loss reported for an obesity drug at that duration at the time. “Beats everything” requires head-to-head trials: none vs tirzepatide or semaglutide had reported by Oct 2026 (TRIUMPH-5 and TRANSCEND-T2D-2 pending). Cross-trial comparisons differ in population, duration, estimand and era — and Phase 3 itself shows the number is population-dependent (18.8–25.0% treatment-regimen at 12 mg).

• Unproven

  “Glucagon burns fat directly in humans.”

  Animal/preclinical, popularised as human fact — unproven in humans. Mouse discovery data attribute expenditure/oxidation effects to the glucagon arm (Coskun 2022); a 2026 mouse study found no energy-expenditure effect. In humans: liver fat and weight fall (RCT), β-hydroxybutyrate rises (Phase 1b) — but no direct energy-expenditure or fat-oxidation measurement isolating glucagon's contribution, and liver-fat loss tracks total weight loss. Correct statement: glucagon agonism is hypothesised to add expenditure/hepatic oxidation on top of reduced intake; human proof is indirect.

• Not supported

  “Phase 2 results always hold in Phase 3.”

  Folklore — contradicted as a general rule; instructive here. Retatrutide is actually a favourable case — Phase 3 confirmed large effects — yet even here the numbers moved: longer duration (80 vs 48 weeks), different populations (T2D: −18.8% treatment-regimen), and estimand choice (efficacy −28.3% vs treatment-regimen −25.0% in the same TRIUMPH-1 trial) each shift the headline by several points, and a new safety signal (dysesthesia prominence) only became clear at Phase 3 scale/duration. In general, most Phase 2 programmes in medicine fail or shrink in Phase 3; “always holds” is false, and retatrutide's relative consistency is the exception that still required Phase 3 to define.

• Not supported

  “Retatrutide has minimal nausea — gentler than Ozempic/Mounjaro.”

  Folklore / creator claim — contradicted. Nausea 45% (Phase 2, 12 mg) and 42.4% (TRIUMPH-1, 12 mg); vomiting up to 25.3%; discontinuation-for-AE up to 18.2% (TRIUMPH-4, 12 mg). Squarely within/above the class range.

• Needs the qualifier

  “It spares muscle / loses almost only fat, unlike other GLP-1s.”

  Human RCT substudy, over-read — partly supported, commonly overstated. DXA substudy (T2D, n=103 paired): ~62–69% of loss was fat; the authors say the lean:weight proportion is similar to other obesity treatments — i.e. typical, not uniquely sparing — with up to −6.5 kg lean mass lost. No obesity-population or Phase 3 DXA data; no function measures.

• Needs the qualifier

  “28.7% / 30.3% is the Phase 3 result.”

  Sponsor claim, accurate within its estimand — partly supported; estimand cherry-picking risk. 28.7% is TRIUMPH-4's efficacy estimand in knee-OA patients; 30.3% is a 104-week extension subgroup (BMI ≥35 completers who tolerated treatment). The corresponding treatment-regimen figures are 23.7% (TRIUMPH-4) and 25.0% at 80 weeks (TRIUMPH-1). Quoting only the largest efficacy-estimand subgroup number misleads.

• Needs the qualifier

  “Heart-rate rise is only ~6.7 bpm and settles, so CV risk is settled.”

  Human RCT (supplementary data) + folklore reassurance — partly supported on magnitude/time course; CV safety unproven. Ambulatory data as reported: +9.2 bpm at week 24 (12 mg) declining to +6.7 at week 36 — the pattern (peak ~24 weeks, partial decline) is consistent across trials. But TRIUMPH-3's MACE analyses were non-significant with wide CIs (MACE-3 HR 1.12, 0.64–1.96) in a trial not powered for events; the powered outcomes trial completes ~2029. Risk-factor improvements (BP, lipids, hsCRP) and the HR signal pull in opposite directions, and only TRIUMPH-Outcomes resolves the net.

• Not supported

  “If it's sold online as ‘reta’, it's the same drug as in the trials.”

  Folklore / grey-market claim — contradicted. Lilly: retatrutide “cannot be legally sold or marketed for human use” and is legally available only to trial participants (plus the narrow 2026 expanded-access route). Non-trial product has no verified identity, purity, dose or sterility; trial safety/efficacy numbers cannot be transferred to it.

Open questions

Marked plainly, because they are open — where the evidence genuinely runs out:

• Cardiovascular safety given the heart-rate signal. Risk factors improve (SBP −9 to −14 mmHg, triglycerides −37%, hsCRP −51% in Phase 3 sponsor data) while heart rate rises dose-dependently and an arrhythmia-category signal appeared in Phase 2. TRIUMPH-3 was not powered for events (MACE-3 HR 1.12, CI crossing 1 widely). TRIUMPH-Outcomes (n=10,000, ~Feb 2029) is the resolving trial. Until then, any claim of cardiovascular benefit or harm is unproven.
• Durability and discontinuation. Longest peer-reviewed exposure is 80 weeks (104-week extension data are sponsor/subgroup). No published withdrawal, regain, or maintenance data for retatrutide; TRIUMPH-6 (maintenance) is due ~2028. Whether weight, glycaemic and liver benefits persist, and what happens on stopping, are unknown.
• Lean-mass data gaps. DXA data exist only in type 2 diabetes (Phase 2 substudy, 103 paired scans). No DXA in the obesity population, none in Phase 3, no strength/function endpoints, nothing in older/sarcopenia-risk or athletic populations. Hepatic amino-acid catabolism from chronic glucagon agonism is a mechanistic reason this gap matters, not a demonstrated harm.
• Dysesthesia mechanism and course. The signal is consistent (up to 20.9%) and mostly mild/resolving, but its mechanism, predictors, and very-long-term course are unexplained in published sources.
• Head-to-head standing. Vs tirzepatide (TRIUMPH-5, ~late 2026) and vs semaglutide in T2D (TRANSCEND-T2D-2, primary completion Aug 2026, unreported as of early Oct 2026). All “best-in-class” claims are provisional until these read out.
• Rare and long-latency harms. A few thousand trial participants cannot detect events at 1-in-several-thousand frequency (pancreatitis occurred once in Phase 2 obesity). Thyroid C-cell, gallbladder, gastroparesis/aspiration and psychiatric signals are class watch-items without retatrutide-specific resolution.
• Populations never studied: adolescents, pregnancy, lean/normal-weight people, athletes, severe renal/hepatic impairment (partly in progress), and people on insulin/sulfonylureas (hypoglycaemia risk untested in that context).

Sources

1. Jastreboff AM et al. Triple–Hormone-Receptor Agonist Retatrutide for Obesity — A Phase 2 Trial, N Engl J Med 2023 (DOI 10.1056/NEJMoa2301972) — verified. nejm.org (Phase 2 obesity) · Rosenstock J et al. Retatrutide for people with type 2 diabetes: phase 2 trial, Lancet 2023;402:529–544. thelancet.com (Phase 2 T2D)
2. Sanyal AJ et al. Triple hormone receptor agonist retatrutide for metabolic dysfunction-associated steatotic liver disease: a randomized phase 2a trial, Nat Med 2024. ncbi.nlm.nih.gov (MASLD substudy) · Coskun T et al. Effects of retatrutide on body composition in people with type 2 diabetes: a substudy of a phase 2 trial, Lancet Diabetes Endocrinol 2025. thelancet.com (DXA substudy)
3. Coskun T et al. LY3437943, a novel triple glucagon, GIP, and GLP-1 receptor agonist: from discovery to clinical proof of concept, Cell Metab 2022;34:1234–1247.e9 (PMID 35985340) — cited within sources 1–2 · Urva S et al. LY3437943 in people with type 2 diabetes: a phase 1b trial, Lancet 2022;400:1869–1881 — cited within source 1 and reviews · Briand F et al. Retatrutide in diet-induced obese MASH mouse and hamster models, Obesity 2026. onlinelibrary.wiley.com (Briand 2026)
4. Phase 3 publications, via concordant reports — TRIUMPH-1, N Engl J Med, 29 Sep 2026. hcplive.com (TRIUMPH-1) · alert.psychnews.org (TRIUMPH-1) · TRIUMPH-2, Lancet, Sep 2026. medscape.com (TRIUMPH-2) · TRANSCEND-T2D-1, Lancet, Jun 2026 — trial summaries. sites.google.com (retatrutide trials)
5. Sponsor (Lilly) releases — company-reported — TRIUMPH-4 topline, 11 Dec 2025 (verified). lilly.mediaroom.com (TRIUMPH-4) · TRANSCEND-T2D-1 topline, 19 Mar 2026. lilly.mediaroom.com (TRANSCEND-T2D-1)
6. Sponsor (Lilly) releases — TRIUMPH-1 topline, 21 May 2026 (verified). lilly.mediaroom.com (TRIUMPH-1) · TRIUMPH-2 & TRIUMPH-3 topline + Q1 2027 BLA plan, 23 Jul 2026 (verified). lilly.mediaroom.com (TRIUMPH-2 & -3)
7. Registry / context — ClinicalTrials.gov records via registry summaries: TRIUMPH-1 NCT05929066; TRIUMPH-2 NCT05929079; TRIUMPH-3 NCT05882045; TRIUMPH-Outcomes NCT06383390. inclinicaltrials.com (TRIUMPH-Outcomes) · TRIUMPH-5 NCT06662383; TRANSCEND-T2D-2 NCT06260722; TRIUMPH-6 NCT06859268 · World Obesity Federation, World Obesity Atlas 2024. worldobesity.org (Atlas 2024)
8. Creator sources, cited as claims graded in the Creator layer, not evidence — Vigorous Steve, retatrutide deep-dive transcript (YouTube, 31 Mar 2025), verified via transcript mirror. youtubetotranscript.com (Vigorous Steve) · Vigorous Steve on Mark Bell's Power Project. podscripts.co (Power Project) · Big Paul / Anabolic Bodybuilding, ListenNotes listing (episode descriptions). listennotes.com (Anabolic Bodybuilding)
9. Creator sources — Dr Alex Tatem, Diary of a CEO claim-check page. holdsup.app (Tatem claim-check) · TRIUMPH-1 video. youtube.com (Tatem, TRIUMPH-1) · Diabetes short. youtube.com (Tatem, diabetes short) · Hunter Williams (Beehiiv) — Masterclass. huntershealthhacks.beehiiv.com (Masterclass) · Maximum Effective Dose. huntershealthhacks.beehiiv.com (Maximum Effective Dose) · DXA study post. huntershealthhacks.beehiiv.com (DXA study)
10. Creator sources — TRT Dad Unscripted, show listings (no retatrutide-specific claim found). toppodcast.com (TRT Dad Unscripted) · podscan.fm (TRT Dad Unscripted) · Dan the Bodybuilder / Steroids Podcast, Dawson Weiss episode transcript (12 Dec 2025), verified via Podscan. podscan.fm (Dawson Weiss episode) · Ep. 77 listing. listennotes.com (Steroids Podcast Ep. 77) · Peptide of the Week, episode descriptions. podtail.com (Peptide of the Week Q&A #26) · castbox.fm (Peptide of the Week)
11. Retatrutide — evidence deep-dive, research report filed 28 Oct 2026. Every figure, attribution and hedge on this page follows that report.

Molecule

Human GLP-1(7–37) analogue · 31 amino acids

Class

GLP-1 receptor agonist

Brands

Ozempic · Rybelsus · Wegovy

US status

Rx-only · boxed warning

The health layer — semaglutide in legitimate medicine

This layer leads: what the drug does for ordinary health, in the populations actually studied.

Type 2 diabetes

Semaglutide was developed and first approved (Ozempic, December 2017; oral Rybelsus, September 2019) as a glucose-lowering treatment for type 2 diabetes, as an adjunct to diet and exercise (FDA labels; approval history). Its glucose effect is glucose-dependent insulin secretion plus glucagon suppression (labels, §12), which is why hypoglycaemia risk is low unless it is combined with insulin or a sulfonylurea (labels). Across the SUSTAIN phase-3 programme, the 1.0 mg weekly trial exposure reduced HbA1c by about 1.5–1.8 percentage points after 30–56 weeks, and the 0.5 mg exposure by 1.0–1.5 points, with weight reductions of 3.6–6.5 kg at the 1.0 mg exposure — consistently greater than the sitagliptin, exenatide-ER, insulin-glargine, dulaglutide and canagliflozin comparators used in individual trials (SUSTAIN programme summaries).

What this means for ordinary health: for a person with type 2 diabetes, this is a large glycaemic effect by historical standards, achieved with weight loss rather than the weight gain typical of insulin or sulfonylureas. It does not mean the drug “reverses” diabetes in the trial record: trials measured HbA1c reduction on treatment, and glycaemic improvements moved back toward baseline after withdrawal in the STEP 1 extension (Efficacy, below).

Chronic weight management

Wegovy was approved in June 2021 for chronic weight management in adults with obesity, or overweight with at least one weight-related condition, alongside reduced-calorie diet and increased physical activity; approval extended to adolescents aged ≥12 in December 2022 (FDA / approval history). The pivotal evidence is STEP 1 (Efficacy, below): an average −14.9% of body weight over 68 weeks, with 86.4% achieving ≥5% loss vs 31.5% on placebo.

What this means for ordinary health: the trials frame obesity as a chronic, relapsing condition managed long-term, not a short course with a permanent result. The withdrawal data are part of the health story, not a footnote: most of the benefit attenuates when treatment stops, in the same way blood-pressure control attenuates when an antihypertensive stops.

SELECT — cardiovascular outcomes, and what it does and does not mean

SELECT (NEJM 2023) randomised 17,604 adults aged ≥45, BMI ≥27, with established cardiovascular disease (76.3% prior myocardial infarction, 23.3% prior stroke in enrolment reports) and no diabetes, to semaglutide 2.4 mg weekly (trial exposure) or placebo, mean follow-up 39.8 months.

SELECT primary outcome — Lincoff et al., NEJM 2023

Outcome	Semaglutide	Placebo	Effect
3-point MACE (CV death, non-fatal MI, non-fatal stroke)	6.5%	8.0%	HR 0.80 (95% CI 0.72–0.90) — 20% relative reduction; 1.5 pp absolute reduction

Interpretation, graded:

• Supported (Human RCT): in secondary prevention — people who have already had cardiovascular disease — semaglutide reduced first MACE events on top of contemporary care (~90% on statins in secondary reporting).
• Not tested: primary prevention. SELECT says nothing direct about a person with no cardiovascular disease, no diabetes and no obesity. Extrapolating the 20% relative reduction to low-risk healthy users is unproven; the absolute benefit in a low-risk population would be far smaller even if the relative effect held.
• Mechanism unsettled: a 2025 mediation analysis (Colhoun et al., European Heart Journal, reported 2026) estimated that measured mediators jointly explained only ~31.4% of the MACE effect, with wide confidence intervals — “it is all just the weight loss” is not established, and neither is any single alternative mechanism.
• Regulatory consequence: in March 2024 FDA approved Wegovy to reduce cardiovascular death, heart attack and stroke in adults with established CVD and overweight/obesity — the first such indication for a weight-management drug (FDA news release, 8 March 2024).

FLOW — kidney outcomes

FLOW (Perkovic et al., NEJM 2024) was the first dedicated kidney-outcomes trial of a GLP-1 receptor agonist: 3,533 people with type 2 diabetes and albuminuric CKD, semaglutide 1.0 mg weekly (trial exposure) vs placebo, median follow-up 3.4 years, stopped early after an independent monitoring board judged a pre-specified efficacy boundary crossed.

FLOW — Perkovic et al., NEJM 2024

Outcome	Semaglutide	Placebo	Effect
Primary composite (kidney failure, sustained ≥50% eGFR decline, kidney or CV death)	18.7% (331 first events)	23.2% (410 first events)	HR 0.76 (0.66–0.88) — 24% relative reduction
eGFR slope	1.16 mL/min/1.73 m²/year slower decline (P<0.001)		—
MACE	HR 0.82 (18% relative reduction)		—
All-cause death	HR 0.80 (20% relative reduction)		—
Serious adverse events	49.6%	53.8%	Fewer with semaglutide

In January 2025 FDA expanded Ozempic's labelling to reduce risk of kidney-disease worsening, kidney failure and cardiovascular death in type 2 diabetes with CKD (approval-history sources). Boundary: FLOW applies to albuminuric diabetic kidney disease, not to kidney protection in healthy people, for which there is no trial.

Other approved territory (for completeness)

• MASH: in August 2025 Wegovy received accelerated approval for non-cirrhotic metabolic dysfunction-associated steatohepatitis (MASH) with moderate-to-advanced fibrosis, based on ESSENCE part 1: steatohepatitis resolution without worsening fibrosis 62.9% vs 34.3% on placebo (trial reportage). This is an accelerated approval; confirmatory outcomes data are pending.
• Oral weight-management formulation: Wegovy tablets were approved 22 December 2025 (approval history; FDA label §12.3). The oral formulation's pharmacology is covered in the Pharmacology section; its existence does not change the evidence grading of the injection trials.

Sources: FDA Ozempic / Wegovy / Rybelsus labels · Lincoff et al., NEJM 2023 (SELECT) · Perkovic et al., NEJM 2024 (FLOW) · SUSTAIN programme summaries · FDA news release, 8 Mar 2024

Pharmacology — GLP-1 receptor agonism, engineered for a week

All statements from FDA prescribing information unless another source is named. Doses named here are label/trial exposures, not recommendations.

Class and mechanism

Semaglutide is a glucagon-like peptide-1 (GLP-1) receptor agonist (FDA labels, §12.1–12.2). Native GLP-1 is an incretin hormone released from the gut after eating. Receptor activation:

• stimulates insulin secretion and suppresses glucagon secretion in a glucose-dependent manner (pancreas);
• delays early postprandial gastric emptying (labels, §12.2);
• acts on GLP-1 receptors in brain regions involved in appetite regulation — the labels state semaglutide lowers body weight through decreased calorie intake, with effects “likely mediated by affecting appetite,” and animal studies show distribution to and activation of neurons in food-intake regulation regions (Wegovy label, §12.2).

Molecular design — a human GLP-1 analogue, not a lizard peptide

This distinction is frequently garbled in popular content, so it is stated precisely:

Exenatide vs semaglutide — comparative pharmacology summaries; semaglutide structure per FDA labels

Feature	Exenatide	Semaglutide
Origin	Synthetic version of exendin-4, a peptide isolated from Gila monster (Heloderma suspectum) venom/saliva	Analogue of human GLP-1(7–37)
Sequence identity to human GLP-1	~53% (39 amino acids)	~94% (31 amino acids)
Key engineering	Exendin-4's intrinsic resistance to DPP-4	Three modifications (below)
Half-life	~2.4 hours (immediate-release form)	~1 week

Exenatide (Byetta, approved 2005) was the proof-of-concept that a DPP-4-resistant GLP-1 mimic could work as a drug; semaglutide's scaffold is human, and only the strategy descends from the exendin-4 story. Semaglutide's three modifications (labels / medicinal-chemistry literature):

• Position 8: alanine replaced by α-aminoisobutyric acid (Aib) — confers resistance to cleavage by dipeptidyl peptidase-4 (DPP-4), the enzyme that destroys native GLP-1 within ~2 minutes.
• Position 26 (Lys): acylation with a C-18 fatty di-acid chain via a linker (γ-Glu / oligoethylene-glycol spacer in the chemistry literature) — this is the albumin-binding moiety.
• Position 34: lysine replaced by arginine — prevents fatty-acid attachment at the wrong site, directing acylation to position 26.

Molecular formula C₁₈₇H₂₉₁N₄₅O₅₉; molecular weight ≈4,113.6 Da (chemistry summaries).

Albumin binding and the ~1-week half-life

• Semaglutide is >99% bound to plasma albumin (labels, §12.3). The fatty-acid chain binds albumin reversibly; the labels state this “results in decreased renal clearance and protection from degradation.”
• Elimination half-life ≈1 week; apparent clearance ≈0.05 L/h (subcutaneous, weight-management population); mean volume of distribution ≈12.5 L (labels, §12.3).
• Consequence stated on the label: semaglutide “will be present in the circulation for about 5 to 7 weeks after the last dose” of the 2.4 mg trial/label exposure (Wegovy label).
• Subcutaneous absolute bioavailability 89%; maximum concentration 1–3 days post-dose; steady state after 4–5 weeks (labels).
• Contrast: native GLP-1 half-life ~2 minutes. The entire pharmacokinetic advantage is engineered — DPP-4 resistance plus albumin binding — not a property of the GLP-1 sequence itself.

The oral formulation — SNAC and why bioavailability is ~1%

Peptides are normally digested before absorption. Rybelsus (and the newer oral formulations) co-formulate semaglutide with SNAC — salcaprozate sodium, sodium N-[8-(2-hydroxybenzoyl)amino]caprylate (labels: “coformulated with salcaprozate sodium which facilitates the absorption of semaglutide”). Mechanism, from the label and the peer-reviewed oral-semaglutide literature:

• absorption occurs predominantly in the stomach, not the intestine;
• SNAC creates a local, transient rise in pH around the tablet, protecting semaglutide from acidic/peptic degradation;
• SNAC promotes monomerisation of semaglutide and transiently increases gastric epithelial membrane fluidity, enabling transcellular passage across the stomach wall.

Route comparison — labels, §12.3

Parameter	Subcutaneous	Oral (Rybelsus, original formulation)	Oral (newer formulations on current labels)
Absolute bioavailability	89%	≈0.4–1%	≈1–2%
Time to peak concentration	1–3 days	~1 hour	~1 hour
Steady-state concentration (named label exposures)	~75 nmol/L (2.4 mg weekly, obesity population)	6.7 nmol/L (7 mg daily); 14.6 nmol/L (14 mg daily) in T2D	~77 nmol/L (25 mg daily Wegovy tablet, obesity population) — predicted comparable exposure to the 2.4 mg injection, with higher variability

The ~100-fold bioavailability gap is why oral trial exposures are in milligrams per day and injectable exposures in milligrams per week, and why the labels state the routes are not substitutable milligram-for-milligram. This is a pharmacokinetic fact, not a dosing instruction.

Metabolism and excretion

• Metabolism: “proteolytic cleavage of the peptide backbone and sequential beta-oxidation of the fatty acid sidechain” (labels, §12.3). No single organ is the major route; semaglutide is not primarily dependent on CYP enzymes.
• Excretion: semaglutide-related material is excreted mainly in urine and faeces; approximately 3% of the absorbed dose is excreted in urine as intact semaglutide (labels).
• Cardiac electrophysiology: no clinically relevant QTc prolongation in thorough-QT testing at exposures above the maximum labelled exposure (labels, §12.2).

Sources: FDA Wegovy / Ozempic / Rybelsus labels, §§12.1–12.3 (DailyMed) · Oral-semaglutide / SNAC mechanism reviews (PIONEER programme)

The bodybuilding / fitness layer — body composition, strength, and the gaps

No trial has enrolled bodybuilders, tested contest preparation, or measured physique outcomes. What follows is (a) the best body-composition data from obesity trials, (b) the thin strength/function record, and (c) how fitness communities discuss the drug — labelled as reported practice/claims, not evidence of effect.

The STEP 1 DXA substudy — the central dataset

Exploratory analysis of STEP 1 (Wilding et al., presented ENDO 2021; Endocrine Society abstract): substudy n = 140 (semaglutide n=95; placebo n=45), drawn from 9 sites, participants with BMI ≤40; mean weight 98.4 kg, BMI 34.8, 76% female. DXA at screening and week 68. Body-weight change in the substudy: −15.0% (semaglutide) vs −3.6% (placebo).

STEP 1 DXA substudy — change from baseline, semaglutide

DXA measure	Result
Total fat mass	−19.3%
Regional visceral fat mass	−27.4%
Total lean body mass	−9.7%
Lean mass as proportion of total body mass	+3.0 percentage points
Fat mass as proportion of total body mass	−3.5 percentage points
Lean:fat mass ratio	1.34 at baseline → +0.23 (95% CI 0.14–0.32) at week 68; greater improvement with ≥15% weight loss (+0.41) than <15% (+0.03)

Derived partition (the underlying review's arithmetic from the substudy's published percentages and baseline composition — fat ≈43.4% and lean ≈53.9% of body mass): of the tissue lost, roughly three-fifths was fat and two-fifths was lean (≈38–40% lean). Two cautions are mandatory:

• Lean mass ≠ muscle. DXA lean mass includes water, organs, connective tissue and gut contents. The skeletal-muscle fraction of the lean loss is unknown from DXA.
• This partition is not unusual for large weight loss by any method. The tirzepatide SURMOUNT-1 DXA substudy (n=160) found ~75% fat / ~25% lean, with a similar split in its placebo group — i.e., the drug groups do not obviously partition loss worse than the weight loss itself does.

Strength and physical function — what exists

There is no adequately powered RCT of semaglutide with strength as a primary outcome. The record, in descending order of quality:

• STEP UP thigh-imaging secondary analysis (semaglutide weight-management trial programme; Diabetes, Obesity and Metabolism): MRI thigh skeletal-muscle volume fell alongside large fat-volume reductions, muscle fat infiltration decreased, and sit-to-stand repetitions were unchanged from baseline to end of trial — “no deterioration in physical function” in a small imaging cohort.
• SEMALEAN (Alissou et al., Diabetes Obes Metab 2026; prospective, uncontrolled, n=106 with obesity): weight −12.7% at 12 months; lean mass −3.0 kg by month 7, then stable; handgrip strength increased (~+3.7 kg at 7 months, ~+4.1–4.5 kg at 12 months in reports of the study); sarcopenic-obesity prevalence fell from 49% to 33%. Uncontrolled design means training, diet and survivor effects cannot be separated from drug effect.
• Systematic review of handgrip strength on semaglutide (2026, PROSPERO-registered): 10 studies / 660 treated participants pooled for within-group change: −0.16 kgf (95% CI −1.98 to +1.65), non-significant, I² = 90.8%, certainty rated very low. The authors' conclusion — evidence is insufficient to determine whether semaglutide changes handgrip strength — is the honest summary of this literature.
• STEP 1's main paper reported greater improvement in participant-reported physical functioning vs placebo (NEJM 2021) — a questionnaire outcome in a population with obesity, not a strength test.

Fitness-layer translation: “it destroys strength” is unproven; “strength is proven safe” is also unproven. The defensible statement is that function measures have not deteriorated in the small datasets that measured them, while absolute lean mass falls.

How fitness communities discuss it — reported practice / claims (Folklore tier)

Described, not endorsed, and with no use instructions reproduced:

• Contest-prep / cutting appetite tool. Bodybuilding podcasts and channels discuss GLP-1 agonists as appetite-suppression tools for dieting phases, and debate whether they “interfere with a gaining phase” (Big Paul, Creator layer). This is reported practice discussion; there is no trial in lean athletes, and the trial populations (BMI ≥27–30 with comorbidity) do not resemble competitors.
• “Microdosing” discourse. Multiple creators discuss smaller, more frequent administrations than the labelled regimens in the content of their claims. No RCT has tested such regimens for weight, composition or safety outcomes: unproven as an efficacy claim, and not a labelled regimen.
• Preference migration to tirzepatide / retatrutide. Fitness content increasingly treats semaglutide as the “first-generation” option and discusses dual/triple agonists as superior for fat loss with less muscle loss. Head-to-head body-composition trials supporting the muscle claim do not exist (Creator layer).
• Muscle-preservation stacking discourse. Community discussion pairs GLP-1s with resistance training, high protein, and (in enhanced circles) anabolic agents, on the theory that these offset lean loss. Resistance training + adequate protein during weight loss is supported by general weight-loss literature; the specific claim that any stack has been shown to preserve lean mass on semaglutide in an RCT is unproven — no such trial was found.
• Regulatory status in sport: semaglutide is not on the WADA Prohibited List; it appears on WADA's Monitoring Program (secondary regulatory summaries). It is a prescription medicine, not a supplement, in all approved markets.

Sources: Wilding et al., STEP 1 DXA substudy, ENDO 2021 · STEP UP thigh-imaging analysis, Diabetes Obes Metab · Alissou et al., Diabetes Obes Metab 2026 (SEMALEAN) · Handgrip-strength systematic review, 2026

Efficacy — with numbers

Weight — STEP 1 (pivotal)

Wilding et al., NEJM 2021 (STEP 1; NCT03548935). n=1,961 adults without diabetes, BMI ≥30 or ≥27 with ≥1 weight-related condition; semaglutide 2.4 mg weekly (trial exposure) vs placebo, both + lifestyle intervention, 68 weeks. Evidence grade: Human RCT, large, double-blind. Funding: Novo Nordisk (declared in the paper).

STEP 1 — outcomes at week 68

Outcome	Semaglutide	Placebo	Difference
Body weight, mean change	−14.9%	−2.4%	−12.4 pp (95% CI −13.4 to −11.5), P<0.001
Body weight, mean change	−15.3 kg	−2.6 kg	−12.7 kg (−13.7 to −11.7)
≥5% loss	86.4%	31.5%	P<0.001
≥10% loss	69.1%	12.0%	P<0.001
≥15% loss	50.5%	4.9%	P<0.001
≥20% loss	32.0%	1.7%	(secondary reporting)

Glycaemic efficacy — SUSTAIN summary

SUSTAIN programme — pooled post-hoc summaries

Programme	Population / design	Headline result
SUSTAIN 1–10 (phase 3, T2D)	vs placebo and active comparators, 30–56 weeks	HbA1c −1.0 to −1.5 pp (0.5 mg exposure); −1.3 to −2.0 pp (1.0 mg exposure); weight −2.3 to −4.7 kg (0.5 mg) and −3.6 to −6.5 kg (1.0 mg) across trials/subgroups
SUSTAIN 7 (head-to-head vs dulaglutide)	T2D on metformin, 40 weeks, baseline HbA1c 8.2%	HbA1c −1.5% (0.5 mg) vs −1.1% (dulaglutide 0.75 mg); −1.8% (1.0 mg) vs −1.4% (dulaglutide 1.5 mg); HbA1c ≤7% reached by 69%/79% on semaglutide exposures
SUSTAIN-6 (CV outcomes)	See Safety	MACE HR 0.74; also the retinopathy signal (Safety, below)

Oral vs injectable — PIONEER, and the limits of the comparison

PIONEER 1 (Aroda et al., Diabetes Care 2019): n=703, T2D managed by diet/exercise alone, 26 weeks, oral semaglutide 3 / 7 / 14 mg daily (trial exposures) vs placebo, baseline HbA1c 8.0%.

PIONEER 1 — week 26

Outcome	Oral 14 mg	Placebo	Placebo-adjusted difference
HbA1c change (treatment-policy estimand)	—	—	−1.1 pp (P<0.001)
HbA1c change (trial-product estimand)	−1.4% (from-baseline reports −1.4 to −1.5%)	~−0.1 to −0.3%	−1.4 pp
Body-weight change	−3.7 to −4.1 kg from baseline	−1.4 to −1.5 kg	−2.3 kg (policy) / −2.6 kg (trial-product)
Discontinuation for AEs	7.4% (14 mg arm; 2.3% at 3 mg)	2.2%	GI events predominant

Across the PIONEER programme, oral 14 mg HbA1c reductions ranged −1.0 to −1.5 pp depending on background therapy (PIONEER subgroup analysis), and in PIONEER 4 oral semaglutide was non-inferior to injectable liraglutide 1.8 mg for HbA1c (−1.2% vs −1.1%, secondary reporting).

Cardiovascular — SELECT primary result

See the Health layer: MACE 6.5% vs 8.0%, HR 0.80 (0.72–0.90), n=17,604, mean follow-up 39.8 months (NEJM 2023). Absolute difference 1.5 pp over ~3.3 years in a secondary-prevention population ≈ number-needed-to-treat ~67 for that period (arithmetic from the published percentages).

What happens when treatment stops — withdrawal data

• STEP 1 extension (Wilding et al., Diabetes Obes Metab 2022): after 68 weeks on treatment (extension subset mean −17.3%), participants regained 11.6 percentage points in 1 year off treatment ≈ two-thirds of the loss, ending −5.6% vs baseline (placebo −0.1%); the share maintaining ≥5% loss fell from 86.4% to 48.2%. Cardiometabolic improvements largely reverted toward baseline.
• STEP 4 (withdrawal RCT): switching to placebo regained +6.9% over 48 weeks while continuers lost more.

The same dataset contradicts both popular extremes: the loss is not permanent off-drug, and it does not “all come back, and then some” — the group mean one year after stopping was still 5.6% below where it started.

Sources: Wilding et al., NEJM 2021 (STEP 1) · SUSTAIN programme summaries · Aroda et al., Diabetes Care 2019 (PIONEER 1) · Wilding et al., Diabetes Obes Metab 2022 (STEP 1 extension)

Safety — with numbers

Gastrointestinal events and discontinuation (Human RCT)

Pooled Wegovy weight-management trials (label data, as summarised from the prescribing information):

Pooled Wegovy trial data — label

Adverse reaction	Semaglutide 2.4 mg exposure	Placebo
Nausea	44%	16%
Diarrhoea	30%	16%
Vomiting	24%	6%
Constipation	24%	11%
Abdominal pain	20%	10%
Any GI adverse reaction	73%	47%
Severe GI reactions	4.1%	0.9%
Discontinuation due to any adverse reaction	6.8%	3.2%
Discontinuation due to GI events	4.3%	0.7%

STEP 1 specifically: GI-event discontinuation 4.5% vs 0.8% (NEJM 2021). Events were mostly transient and mild-to-moderate, clustering around exposure escalation in the trials. Oral PIONEER 1 showed the same pattern at lower absolute rates (nausea 5–16% across oral exposures vs 6% placebo in the sponsor's trial report; discontinuation 2–7.4%).

Gastroparesis

Pharmacology first: delayed gastric emptying is an intended, labelled effect (Pharmacology, above). The disputed question is persistent, pathological gastroparesis.

• Sodhi et al., JAMA 2023 (claims database; Human observational): GLP-1 agonists for weight loss vs bupropion-naltrexone — gastroparesis HR 3.67 (95% CI 1.15–11.90); incidence on semaglutide 9.1 per 1,000 person-years. Limitations that must travel with the number: the semaglutide group was only n=613, the comparator is another drug (not non-use), diabetes was excluded by design, and gastroparesis was defined by diagnostic code or promotility-agent use, not gastric-emptying testing.
• Labels state the drug is not recommended in severe gastroparesis (label summaries), and severe GI disease was an exclusion in the pivotal trials — so RCTs cannot estimate this risk.

Grade: signal, Human observational, low absolute incidence (~1% per year order of magnitude in that dataset), causality and persistence after discontinuation unresolved. Litigation and case reports exist but are not evidence of incidence.

Gallbladder events

• He et al., JAMA Internal Medicine 2022 (meta-analysis of 76 RCTs): GLP-1 receptor agonists associated with ~37% higher relative risk of gallbladder/biliary disease, with risk higher at weight-loss exposures and longer duration.
• STEP pooled analysis (4 placebo-controlled trials; Diabetes Obes Metab): semaglutide 2.4 mg exposure ≈2-fold gallbladder-event risk vs placebo, absolute incidence ~2 events per 100 patient-years, mostly cholelithiasis; cholecystitis and cholecystectomy rare.
• Label trial rates commonly cited: cholelithiasis ~1.6% vs 0.7% (label summaries). Mechanism is doubly plausible — reduced gallbladder emptying plus rapid weight loss itself (a known lithogenic state after bariatric surgery) — so the drug-specific vs weight-loss-mediated split is unresolved.

Pancreatitis

• Labels carry an acute-pancreatitis warning and trials excluded people with prior pancreatitis.
• RCT/meta-analytic data: no statistically significant pancreatitis excess for semaglutide vs control in the obesity GI-safety meta-analysis and in class meta-analyses — absolute rates in STEP 1 reporting are ~0.2% order.
• Observational counterpoint: Sodhi 2023 found pancreatitis HR 9.09 (1.25–66.00) vs bupropion-naltrexone, incidence 4.6 per 1,000 person-years on semaglutide — a very wide confidence interval from a small semaglutide sample.

Grade: Human RCT — no clear excess demonstrated; Human observational — elevated but imprecise signal. Both statements are true of their datasets; neither licenses “causes pancreatitis commonly” or “zero risk.”

Thyroid C-cell tumours — the boxed warning's rodent basis

The substance of the FDA boxed warning (all three brands): “In rodents, semaglutide causes dose-dependent and treatment-duration-dependent thyroid C-cell tumors at clinically relevant exposures. It is unknown whether [semaglutide] causes thyroid C-cell tumors, including medullary thyroid carcinoma (MTC), in humans as human relevance of semaglutide-induced rodent thyroid C-cell tumors has not been determined.” (DailyMed label.)

• Evidence tier for the hazard: Animal only. No human trial or, in the sources read, large human cohort has demonstrated a causal increase in MTC; human relevance is explicitly not determined on the label.
• The label contraindication — personal/family history of MTC or MEN 2 — is a precaution attached to that animal finding, not a finding of human cases.
• Popular phrasing “it causes thyroid cancer” converts a rodent finding into a human claim: contradicted as stated (i.e., not supported by human evidence), while “the risk in humans is proven zero” would equally exceed the evidence.

Retinopathy — SUSTAIN-6

Marso et al., NEJM 2016 (SUSTAIN-6, n=3,297, T2D at high CV risk, 104 weeks):

SUSTAIN-6 — Marso et al., NEJM 2016

Outcome	Semaglutide	Placebo	Effect
Primary MACE	6.6%	8.9%	HR 0.74 (0.58–0.95)
Retinopathy complications (vitreous haemorrhage, blindness, or need for intravitreal agent/photocoagulation)	3.0% (50 patients)	1.8% (29 patients)	HR 1.76 (1.11–2.78), P=0.02
New/worsening nephropathy	3.8%	6.1%	HR 0.64

The authors and subsequent literature note the association, seen with rapid glucose lowering by other means (including insulin in type 1 diabetes), between abrupt HbA1c falls in people with pre-existing retinopathy and early worsening; a direct drug effect could not be ruled out. SUSTAIN-6 was a non-inferiority safety trial, retinopathy was not a primary endpoint, and later GLP-1 retinopathy analyses (e.g., large database target-trial emulations reported 2025–26) have not reproduced an increased risk of treatment-requiring retinopathy — grade: Human RCT signal in one trial, mechanism plausibly glycaemic-speed-related, not confirmed as a class effect.

NAION — the 2024–2025 signal and the regulator split

NAION = non-arteritic anterior ischaemic optic neuropathy: sudden, usually painless, monocular vision loss; generally irreversible; no effective treatment (WHO/EMA summaries).

NAION — the evidence, graded source by source

Source	Design	Finding
Hathaway et al., JAMA Ophthalmology 2024 (Mass Eye and Ear)	Retrospective, single neuro-ophthalmology referral centre, n=16,827 records	NAION HR 4.28 (T2D cohort) and 7.64 (overweight/obesity cohort) for semaglutide prescriptions vs non-GLP-1 comparators; cumulative incidences in that referred population were far above population background — referral/selection bias is the standing critique
Cai et al., JAMA Ophthalmology 2025	Multicentre cohort, 37.1 million adults with T2D (≈810,390 new semaglutide users in secondary reports)	Much weaker association; some models/comparators not statistically significant
Danish registry study (cited in creator content and regulator reviews)	Nationwide observational, >424,000 people with T2D in the creator's description	~2-fold NAION risk; absolute excess ~1 additional case per ~7,000–10,000 treated per year in that description
EMA PRAC, June 2025	Regulator review of trials, literature, post-marketing data	Concluded NAION is a very rare side effect of semaglutide (up to 1 in 10,000 users); ~2-fold risk in T2D ≈ 1 additional case per 10,000 person-years; EU product information to be updated; stop treatment if NAION confirmed
WHO, 27 June 2025	Global alert	Echoed EMA: risk-management plan should include NAION as a potential risk
FDA (US labels)	—	In the sources read (to October 2026), NAION was not listed on the US label; FDA had not adopted the EMA wording. This is a live regulatory divergence, flagged in Open questions

Grade: Human observational only — no RCT can power for an event this rare. The defensible synthesis is EMA's: a possible ~2-fold relative increase on a very small background risk (≈1 in 10,000 per year order), causality not proven by any single study, with the largest early estimate (Hathaway) the least generalisable.

Other labelled risks (no numbers invented)

Acute kidney injury secondary to dehydration from GI losses, hypersensitivity, ileus/bowel-obstruction warnings in post-marketing labelling, increased resting heart rate (small, ~1–3 bpm in trial reporting), and hypoglycaemia risk when combined with insulin/secretagogues all appear in current labelling (label summaries). Sodhi 2023's bowel-obstruction HR was 4.22 (1.02–17.40) for the GLP-1 class vs bupropion-naltrexone, with zero bowel-obstruction cases in its small semaglutide arm — a fact routinely omitted when the class HR is quoted as a semaglutide figure.

Sources: FDA Wegovy label (pooled trial data) · Sodhi et al., JAMA 2023 · He et al., JAMA Intern Med 2022 · Marso et al., NEJM 2016 (SUSTAIN-6) · Hathaway et al. 2024 / Cai et al. 2025, JAMA Ophthalmology · EMA PRAC, Jun 2025 · WHO, 27 Jun 2025

Regulatory / approval history

All dates United States (FDA) unless noted. Sources: approval-history records, FDA news releases; label versions via DailyMed.

Semaglutide — approval history

Date	Action
Apr 2005 (context)	Exenatide (Byetta) — first GLP-1 receptor agonist, exendin-4-based — approved for T2D. Establishes the class semaglutide later joins with a human-analogue design (Pharmacology, above).
5 Dec 2017	Ozempic (semaglutide injection) approved for glycaemic control in type 2 diabetes — first semaglutide approval. Initial US approval year (2017) is carried on all subsequent semaglutide labels.
20 Sep 2019	Rybelsus (oral semaglutide + SNAC) approved for T2D — first oral GLP-1 receptor agonist.
Jan 2020	Ozempic labelling expanded: reduction of major adverse cardiovascular events in T2D with established CVD (basis: SUSTAIN-6).
4 Jun 2021	Wegovy (semaglutide injection, weight-management brand) approved for chronic weight management in adults — first drug approved for general chronic weight management since 2014 (basis: STEP programme).
Mar 2022	Higher Ozempic exposure added to labelling for additional glycaemic effect (approval-history sources).
Dec 2022	Wegovy approved for adolescents aged ≥12 with obesity (basis: STEP TEENS programme).
2022–Feb 2025	Semaglutide injection on the FDA drug-shortage list; scaled 503A/503B compounding of copies permitted under shortage rules. FDA repeatedly warned that compounded GLP-1 products are not FDA-approved and flagged salt-form ingredients and dosing-error reports.
8 Mar 2024	Wegovy approved to reduce risk of CV death, MI and stroke in adults with established CVD and overweight/obesity (basis: SELECT) — FDA news release.
Jan 2025	Ozempic labelling expanded for kidney protection: reduce risk of kidney-disease worsening, kidney failure and CV death in T2D with CKD (basis: FLOW).
21 Feb 2025	FDA declares the semaglutide-injection shortage resolved; transition windows for compounders followed in spring 2025 (503A ~Apr 2025, 503B ~May 2025, per FDA policy summaries).
6 Jun 2025	EMA PRAC concludes NAION is a very rare side effect of semaglutide medicines and recommends EU product-information updates (Safety, above). WHO issues a concordant global alert 27 Jun 2025. No equivalent US label change in sources read.
Aug 2025	Wegovy receives accelerated approval for non-cirrhotic MASH with moderate-to-advanced fibrosis (basis: ESSENCE part 1).
22 Dec 2025	Wegovy tablets (oral semaglutide for weight management) approved (approval-history / label sources).
2026 (in sources read)	Regulatory pressure on compounded / mass-marketed GLP-1 products continued, including FDA proposals regarding the 503B bulks list and warning letters over misleading compounded-GLP-1 claims; final outcomes of those proposals were not verified in this research (Open questions, below).

EU context (secondary sources): Ozempic authorised Feb 2018, Rybelsus Apr 2020, Wegovy Jan 2022.

Creator layer — opinion, graded claim by claim

Everything in this section is creator and community content, kept separate from the primary literature above precisely so it is not mistaken for it. Only claims attributable to the named creator in a findable source (video title/description/transcript, show notes, or attributed press quote) are graded. Verdicts: supported / partly supported / contradicted / unproven, judged against the sections above. Where the accessible source is a show description rather than a transcript, that limitation is stated. No creator dosing protocol is reproduced; claims are paraphrased to their testable content only. Creators are quoted and attributed — they are never treated as evidence.

Vigorous Steve

Source: “Safest Dose Of GLP (NEXT-LEVEL Appetite Suppression) Insulin Sensitivity, Fat Loss Deep-Dive”, YouTube (verified live; full transcript read) — a retatrutide-centred deep-dive comparing GLP-1 agonists, with chapters including “GLP-1 Medications Serious Side-Effects” and “Body Weight Reduction Vs. Body Composition.”

• When liraglutide / dulaglutide / semaglutide / tirzepatide are discontinued, “the weight loss comes right back with a vengeance and these people end up fatter than they ever were before… they went from 300 lb to 200 lb and then to 350 lb, for example.” Verdict — contradicted (as a general claim). STEP 1 extension (Human RCT follow-up): regain ≈ two-thirds at 1 year, net −5.6% vs baseline — overshoot above starting weight was not the group outcome (Efficacy, above). Individual overshoot anecdotes cannot be excluded, but the claim as stated describes the opposite of the trial mean.
• Title-level framing: GLP agonists provide “next-level appetite suppression” with insulin-sensitivity / fat-loss benefits. Verdict — supported (directionally). Appetite reduction and fat loss are the labelled pharmacology and STEP outcomes; “insulin sensitivity” per se was not a STEP primary outcome, hence directional only.
• (Channel-level, from clip chapters verified live) “We don't know the long-term effects” of GLP-1 agonists. Verdict — partly supported. Outcome data now extend to ~4 years (SELECT) and decades-long safety is genuinely unknown; stated without that boundary it understates the existing RCT record.

Big Paul — Paul Barnett (Anabolic Bodybuilding)

Sources: Anabolic Bodybuilding podcast feed (verified live); episode description for the GLP-1/insulin Q&A via Listen Notes. Graded from episode descriptions only — see flags above.

• “Tirzepatide and other GLP-1 medications may interfere with a gaining phase” (show description of a Q&A episode). Verdict — partly supported. Appetite suppression and reduced calorie intake are established pharmacology (Pharmacology, above), so interference with a deliberate calorie surplus is mechanistically coherent; there is no trial in people attempting a gaining phase — the claim is extrapolation, not evidence.
• GLP-1 medications' “potential effects on motivation and enjoyment of life (anhedonia)” (topic of a feed episode). Verdict — unproven. Reward-pathway effects are biologically discussed and addiction-related outcomes are under study, but no RCT or robust cohort in the sources read establishes GLP-1-induced anhedonia. Presented on the feed as a discussion topic, not a demonstrated finding — graded as a claim only insofar as it is implied.
• Segment claim (episode listing): a dedicated segment on “How do GLP-1s impact insulin use?” Not gradeable from accessible text. The listing proves the segment exists; its spoken content was not available in transcript in this research. No claim is invented for it.

Dr. Alex Tatem (board-certified urologist, men's-health / peptide creator)

Sources: YouTube video/short “Ozempic Doubles Your Risk of Blindness?!” (description read in full); Fox News attributed interview.

• Semaglutide is associated with ~twice the NAION risk, but absolute risk is very small (~1 extra case per 7,000–10,000 treated annually); EMA classified NAION as very rare in June 2025; SELECT found ~20% fewer major CV events — presented as a risk–benefit balance. Verdict — supported. Matches EMA PRAC 2025 and SELECT closely, including the relative-vs-absolute distinction most coverage omits. This is the most evidence-concordant creator item found in this research.
• GLP-1s are “truly miracle compounds… incredibly effective… benefiting metabolic health, cardiovascular health [and] neurologic health” (Fox News quote). Verdict — partly supported. Metabolic and (in defined populations) cardiovascular benefit: supported (Health layer, above). Neurologic health benefit: unproven — no completed RCT demonstrates a neurological outcome benefit for semaglutide; trials in neurodegeneration are investigational. “Miracle” is rhetoric, not a gradeable claim.
• Compounded semaglutide/tirzepatide versions exist because of the national shortage and offer dosing flexibility (smaller, more frequent administrations) that commercial pens don't (Fox News quote, paraphrased). Verdict — partly supported (existence/history) / unproven (benefit). Shortage-era compounding history is correct, but: the shortage was declared resolved Feb 2025; compounded products are not FDA-approved and some used salt forms FDA rejects (Claim check, below); and no RCT shows altered-frequency regimens improve outcomes. The flexibility claim is a practice claim, not an evidence claim.

Hunter Williams

Sources (opened and read): “GLP-1s Slow Aging?” (newsletter); “Why You're Stuck: GLP-1 Plateaus And How to Break Through” (full transcript).

• A randomised trial (Corley et al., Nat Commun 2026; n=84 with HIV-associated lipohypertrophy, 32 weeks) found semaglutide slowed epigenetic (blood-biomarker) aging clocks, while “under-the-hood” clocks and functional measures did not move; Williams concludes we now have tools to “potentially delay aging directly,” while noting the study is small with real limitations. Verdict — partly supported. His description of the study is unusually careful and matches his own caveats (blood-based clocks moved; structural/functional clocks did not; muscle clock unmoved). The extrapolation — that this establishes direct delay of aging in general users — is unproven: one small RCT, surrogate clocks, a specific population with accelerated aging, no functional or clinical aging outcome.
• “Drop a quarter of your bodyweight and let a third of it come from muscle, and you have traded one aging problem for another” (as a risk framing). Verdict — partly supported. The lean fraction of loss (~one-third to two-fifths, Bodybuilding layer) is in the right range, but it equates lean mass with muscle, which DXA does not support — the proportion of skeletal muscle lost is smaller and unmeasured in that framing.
• GLP-1 weight loss plateaus around ~60–68 weeks on semaglutide / ~72 weeks on tirzepatide, via adaptive thermogenesis, leptin fall, lean-mass loss and reduced NEAT; if the drug is stopped, regain is common. Verdict — supported (in substance). Consistent with STEP/SURMOUNT trajectories and the withdrawal literature. Individual mechanistic details (e.g., specific % metabolic-rate falls) are presented in the transcript without primary citations and were not individually verified.
• Tirzepatide “beats” semaglutide on visceral-fat reduction and retatrutide “beats” tirzepatide (aging post, as a hierarchy). Verdict — unproven (as stated). No head-to-head RCT of semaglutide vs tirzepatide on visceral fat, and retatrutide remains investigational with no completed head-to-head vs either drug in the sources read. Cross-trial comparisons cannot carry a “beats” verdict.

TRT Dad Unscripted (@TRTDAD)

Source: show notes, episode #83 “Do You Need TRT… or Do You Just Need to Lose Weight?” (11 Sep 2026). Only show-note framing was accessible; no transcript was read — claims below are the episode's stated premises/questions, graded as such.

• Obesity itself can “crush” testosterone, and significant (GLP-1-assisted) weight loss could bring testosterone back naturally — so the real question is treating the cause of low T, not only the number. Verdict — partly supported. Obesity-associated (functional) hypogonadism and testosterone rises with substantial weight loss are supported by the general obesity literature; however, no semaglutide RCT with testosterone restoration as an endpoint was found in this research, and the notes themselves pose this as a debate, including the counterpoint that severely symptomatic men may not wait months for weight loss. The GLP-1-specific version of the claim is unproven at RCT level.
• Combining “aggressive fat loss with TRT, resistance training and proper nutrition” (posed as a question). Verdict — unproven (as a combined-protocol claim). No trial of that combination was found; it is presented as a question in the source, and is graded only to prevent it being read as a finding.

No semaglutide-specific quantitative claim attributable to TRT Dad was found beyond the episode framing above.

Dan the Bodybuilder / Steroids Podcast (Dan, “Bodybuilder in Thailand”)

Sources: “Retatrutide Results — GLP1 Agonists Semaglutide Tirzepatide and Retatrutide Compared — Bodybuilding Podcast Ep. 77” (transcript page opened; GLP-1 segment timestamped 49:50 in the episode listing); “Muscle Gain Formula” episode transcript (opened).

• The liver produces glucose essentially all the time “unless you're taking… a GLP-1 agonist like semaglutide” (or metformin/vinegar), i.e., semaglutide switches off hepatic glucose production (Muscle Gain Formula transcript). Verdict — partly supported / overstated. Semaglutide suppresses (glucose-dependent) glucagon and thereby reduces hepatic glucose output — directionally correct (Pharmacology, above). It does not abolish gluconeogenesis, and the effect is glucose-dependent, not an on/off switch; the metformin/vinegar equivalence is folklore-tier.
• Ep. 77 performs a results comparison of semaglutide vs tirzepatide vs retatrutide for bodybuilders. Not gradeable in detail. The episode and its GLP-1 segment are verified to exist (listing + partial transcript read), but the segment's spoken comparison was beyond the accessible portion of the transcript in this research, and it was not graded. No specific comparative claim is attributed or invented. Any “retatrutide preserves muscle better” version of this comparison, wherever stated, would be unproven — no head-to-head body-composition RCT exists.

Peptide of the Week (podcast)

Source: episode descriptions syndicated on Podtail. Caution: the same GLP-1 breakdown text is repeated verbatim across multiple Q&A episode descriptions on that feed, suggesting a boilerplate show description rather than episode-specific content; it is graded as the show's published position, with that caveat.

• Semaglutide: GLP-1 only; strong appetite suppression; high nausea for many users; ~15–17% average weight loss. Verdict — supported. STEP 1: −14.9% (2.4 mg exposure); STEP programme range brackets 15–17% in some trials; nausea 44% in pooled label data (Efficacy / Safety, above).
• Semaglutide causes “fat + muscle loss (indiscriminate).” Verdict — partly supported. Absolute lean loss is real, but “indiscriminate” contradicts the DXA partition — fat loss was ~2× lean loss in % terms and lean proportion rose (Bodybuilding layer, above).
• Tirzepatide: “much less nausea than semaglutide; less muscle loss; ~20–22% weight loss.” Verdict — partly supported (weight figure) / unproven (comparative tolerability & muscle claims). ~20–21% matches SURMOUNT-1 reporting for tirzepatide; there is no head-to-head RCT vs semaglutide establishing less nausea or less muscle loss.
• Retatrutide: “minimal to no nausea; preserves muscle much better; prevents metabolic slowdown (plateau killer); ~24%+ weight loss.” Verdict — unproven / contradicted in part. Retatrutide is investigational; published trials report GI adverse events including nausea (so “minimal to no nausea” as a class claim is contradicted by trial reporting), and no head-to-head muscle-preservation or plateau-prevention RCT exists. The ~24% figure reflects early-phase trial reporting, not an approved-product outcome.
• “Hair loss, fatigue, etc. are not from the drug — they come from not eating (malnourishment).” Verdict — partly supported. Rapid weight loss / under-nutrition is a recognised contributor to telogen effluvium and fatigue, and labels list fatigue (11% vs 4% placebo in pooled data) as an adverse reaction — so “not from the drug” as an absolute is overstated; the nutritional mechanism is plausible but not proven as the sole explanation.
• “Appetite suppression is actually a side effect, not the goal” of GLP-1 drugs. Verdict — contradicted (framing). On the labels, decreased calorie intake via appetite effects is the stated weight-loss mechanism (Pharmacology, above), not an off-target side effect.

Creator summary

The creator layer at a glance — best-supported and weakest claims retrieved

Creator	Best-supported claim retrieved	Weakest claim retrieved	Basis
Vigorous Steve	Title-level framing: appetite suppression with fat-loss benefits (supported, directionally).	Discontinuation leaves people “fatter than they ever were before” (contradicted).	YouTube deep-dive transcript, read in full
Big Paul	GLP-1s “may interfere with a gaining phase” (partly supported, mechanistic).	Insulin-use segment — not gradeable from accessible text; nothing invented for it.	Episode descriptions only; no transcript accessible
Dr. Alex Tatem	NAION video: ~2× relative risk, tiny absolute risk, EMA “very rare” (supported — most concordant creator item found).	“Truly miracle compounds” incl. neurologic benefit (partly supported; neurologic unproven).	YouTube short description; Fox News interview
Hunter Williams	Plateau timing and regain-on-stopping (supported in substance).	“A third of it come from muscle” (partly — lean ≠ muscle); cross-drug “beats” hierarchy (unproven).	Own newsletter + plateau transcript, opened and read
TRT Dad Unscripted	Obesity “crushes” testosterone premise (partly supported).	GLP-1-specific testosterone restoration (unproven at RCT level).	Show notes only (Ep. 83); no transcript read
Dan the Bodybuilder	Liver-glucose direction (partly supported / overstated).	Ep. 77 comparison segment — beyond the accessible transcript, not graded; nothing invented.	Muscle Gain Formula transcript; Ep. 77 listing + partial transcript
Peptide of the Week	~15–17% weight loss + high nausea (supported).	“Appetite suppression is a side effect, not the goal” (contradicted by the label mechanism).	Show description repeated across episodes (Podtail)

The recurring failure modes across the layer: (1) cross-trial “beats / preserves muscle better” hierarchies with no head-to-head trials; (2) lean mass quoted as muscle; (3) post-discontinuation overshoot folklore (Steve) contradicted by the one RCT extension that measured it; and (4) investigational drugs (retatrutide) discussed with approved-drug certainty (Peptide of the Week, and the comparative framing in several shows). The most evidence-concordant items found were a risk-communication video (Tatem on NAION) and the descriptive parts of Williams' trial summary.

Sources: Vigorous Steve (YouTube; transcript) · Big Paul / Anabolic Bodybuilding (Podbean feed; Listen Notes) · Tatem (YouTube; Fox News) · Hunter Williams (Beehiiv; transcript library) · TRT Dad Unscripted (Audible; TopPodcast) · Dan the Bodybuilder / Steroids Podcast (Podscan) · Peptide of the Week (Podtail)

Claim check

Popular claims about semaglutide, traced and graded at the tier their evidence actually sits at:

• Needs the qualifier

  “Ozempic face” — the drug uniquely ages / hollows the face.

  Observational / expert commentary; no RCT — partly supported as “rapid weight loss changes faces”; contradicted as a unique drug effect. No medical condition by this name exists in the trial or label record, and no study shows a face-specific effect. Rapid, large fat loss from any cause (including bariatric surgery) reduces facial subcutaneous fat and can make skin laxity/ageing more visible; dermatology commentary (e.g., JAAD commentary 2023, via secondary reports) describes exactly this non-specific mechanism. Attributing individual celebrities' faces to the drug is speculation.

• Not supported

  “It melts muscle.”

  Human RCT substudy — contradicted as stated; the defensible claim is “absolute lean mass falls as part of large weight loss.” STEP 1 DXA substudy (n=140): fat −19.3%, lean −9.7%; lean share of body weight rose 3.0 pp. ~Two-fifths of tissue lost was lean, but lean includes water/organs, and function measures did not deteriorate in small datasets. No evidence of a direct catabolic action on muscle; loss tracks the calorie deficit/weight loss.

• Not supported

  “Compounded versions are the same drug.”

  Regulator (FDA) statements — contradicted. Compounded products are not FDA-approved, are not reviewed for safety/effectiveness/quality, and are not generics. FDA states some compounded products used semaglutide sodium / semaglutide acetate — different active ingredients from approved semaglutide base, with no lawful basis for their use in compounding known to the agency; FDA has also reported dosing errors with compounded injectable semaglutide, some requiring hospitalisation. Legitimate 503A/503B compounding using the correct base ingredient existed under shortage rules — which is still not “the same drug” in regulatory or quality terms. Shortage declared resolved 21 Feb 2025.

• Needs the qualifier

  “Weight always fully returns when you stop.”

  Human RCT (extension + withdrawal design) — partly supported: “most of it returns within a year without continued treatment” is supported; “always fully” is contradicted by the same data. STEP 1 extension: after 68 weeks on treatment (extension subset mean −17.3%), participants regained 11.6 percentage points in 1 year off treatment ≈ two-thirds of the loss, ending −5.6% vs baseline (placebo −0.1%); the share maintaining ≥5% loss fell from 86.4% to 48.2%. STEP 4 (withdrawal RCT): switching to placebo regained +6.9% over 48 weeks while continuers lost more. Cardiometabolic improvements largely reverted toward baseline.

• Not supported

  “It causes thyroid cancer” (common social-media form).

  Animal tier — unproven in humans; contradicted as a stated human fact. Rodent C-cell tumours only; the label states human relevance not determined (Safety, above).

• Needs the qualifier

  “It causes stomach paralysis in everyone / commonly.”

  Human observational — partly supported as a rare risk; contradicted as a common/inevitable outcome. Delayed gastric emptying is pharmacology; diagnosed gastroparesis signal is ~9.1/1,000 person-years in one claims study with a small semaglutide arm (Safety, above).

• Unproven

  “Healthy / lean people get the SELECT heart benefit too.”

  Human RCT (by its eligibility criteria) — unproven. SELECT enrolled established CVD, BMI ≥27, age ≥45, no diabetes. No primary-prevention or lean-population outcomes trial exists.

• Unproven

  “Retatrutide / tirzepatide are proven to spare muscle better than semaglutide.”

  Absence of RCT; cross-trial folklore — unproven. No head-to-head body-composition RCT among these drugs was found; SURMOUNT-1's ~75/25 fat/lean split resembles its own placebo group's split.

Open questions

Flagged plainly, in descending order of importance — where the evidence genuinely runs out:

• Long-term (decadal) safety and durability. The longest large outcome exposure is ~4 years (SELECT). Cancer, neurodegenerative, bone and very-long-term GI outcomes cannot be graded from RCTs; pharmacovigilance and registries are the only current instruments.
• NAION causality and the FDA position. Observational estimates range from null (some Cai 2025 models) to HR 4–7 (single-centre Hathaway 2024), with EMA settling on “very rare, ~2-fold in T2D.” Whether FDA will adopt NAION labelling was not verified — US labels read for this review did not list it, and litigation reporting on the topic is not evidence.
• Strength/function. No powered RCT with strength as a primary endpoint exists. The pooled handgrip estimate is non-significant with very low certainty; SEMALEAN and STEP UP are small and (for SEMALEAN) uncontrolled. Older adults and people with sarcopenia are the least studied and plausibly the most at risk — an evidence gap, not a reassurance.
• What DXA lean loss is made of. The skeletal-muscle fraction of STEP 1's −9.7% lean-mass change is unknown; imaging (MRI muscle volume + fat infiltration) data are only beginning to appear (STEP UP analysis).
• Oral vs injectable, head-to-head. No phase-3 direct comparison exists (Efficacy, above). All oral-vs-injectable efficacy ratios in popular content are cross-trial constructions.
• Pancreatitis and gallbladder mechanism split. RCT meta-analyses and claims studies disagree in precision, not just size (Safety, above); how much gallbladder risk is drug effect vs rapid-weight-loss effect is unresolved.
• Creator-layer gaps (stated, not filled): Big Paul — no transcript of the insulin-use segment was accessible. Dan the Bodybuilder — the Ep. 77 comparison segment was beyond the accessible transcript portion. TRT Dad Unscripted — show notes only, no transcript. Vigorous Steve — one clip verified by title/chapter text, no stable URL captured. Peptide of the Week — the graded text is a repeated show description; verbatim spoken delivery unverified.
• Post-2025 regulatory items not fully verified: the final disposition of FDA's 2026 proposal on the 503B bulks list for semaglutide, and any US label action on NAION, were reported in secondary sources but not confirmed on a primary FDA page in this research.
• MASH confirmatory outcomes for the August 2025 accelerated approval (ESSENCE part 2) and oral-formulation long-term outcomes were beyond the verified scope of this review and are not quantified here.
• Populations with no evidence at all: lean athletes, adolescents outside the obesity indication, pregnancy (labels direct discontinuation before planned pregnancy — a labelling fact, not advice), and people with prior pancreatitis or severe gastroparesis (excluded from pivotal trials). Any claim — positive or negative — specific to these groups is, on current evidence, unproven.

Sources

1. FDA / DailyMed — WEGOVY (semaglutide) prescribing information (§§12.1–12.3 pharmacology/PK, boxed warning). dailymed.nlm.nih.gov (Wegovy label) · FDA — WEGOVY label PDF (2023 version; PK, albumin binding, half-life, excretion). accessdata.fda.gov (Wegovy label PDF) · FDA / DailyMed — RYBELSUS / OZEMPIC tablets prescribing information (SNAC, 0.4–1% and 1–2% bioavailability, steady-state concentrations). dailymed.nlm.nih.gov (Rybelsus/Ozempic tablets)
2. Wilding JPH et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity (STEP 1). N Engl J Med 2021;384:989–1002. doi:10.1056/NEJMoa2032183; NCT03548935. s4me.info (STEP 1) · Wilding JPH et al. STEP 1 DXA body-composition exploratory analysis (ENDO 2021 abstract, n=140). newswise.com (STEP 1 DXA abstract) · Wilding JPH et al. Weight regain and cardiometabolic effects after withdrawal of semaglutide: the STEP 1 trial extension. Diabetes Obes Metab 2022. balancedbodiescobb.com (STEP 1 extension summary)
3. Lincoff AM et al. Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes (SELECT). N Engl J Med 2023;389:2221–2232. Via secondary summaries of the published paper. peptideai.co (SELECT) · Perkovic V et al. Effects of Semaglutide on Chronic Kidney Disease in Patients with Type 2 Diabetes (FLOW). N Engl J Med 2024. doi:10.1056/NEJMoa2403347; NCT03819153. healio.com (FLOW) · gpnotebook.com (FLOW)
4. Marso SP et al. Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes (SUSTAIN-6). N Engl J Med 2016; NCT01720446. reference.medscape.com (SUSTAIN-6) · Aroda VR et al. PIONEER 1: Oral Semaglutide Monotherapy vs Placebo in Type 2 Diabetes. Diabetes Care 2019;42:1724–1732. diabetesjournals.org (PIONEER 1) · SUSTAIN programme pooled efficacy (HbA1c/weight ranges). pmc.ncbi.nlm.nih.gov (SUSTAIN pooled) · pmc.ncbi.nlm.nih.gov (SUSTAIN pooled)
5. Sodhi M et al. Risk of Gastrointestinal Adverse Events Associated With GLP-1 Receptor Agonists for Weight Loss. JAMA 2023. doi:10.1001/jama.2023.19574. pmc.ncbi.nlm.nih.gov (Sodhi 2023) · STEP pooled gallbladder analysis. Diabetes Obes Metab. onlinelibrary.wiley.com (STEP gallbladder) · STEP UP thigh muscle / physical-function secondary analysis. Diabetes Obes Metab. onlinelibrary.wiley.com (STEP UP) · Alissou M et al. SEMALEAN. Diabetes Obes Metab 2026;28(1):112–121. doi:10.1111/dom.70141. hmpgloballearningnetwork.com (SEMALEAN)
6. EMA — PRAC concludes NAION is a very rare side effect of semaglutide medicines (6 Jun 2025), via EMA reportage. medicaldialogues.in (EMA PRAC) · WHO — The use of semaglutide medicines and risk of NAION (27 Jun 2025). who.int (NAION alert) · Hathaway et al. 2024 / Cai et al. 2025, JAMA Ophthalmology (NAION), via reviews. sorrell-md.com (NAION review)
7. FDA — Approval history: Drugs.com Ozempic history (approval dates, formulations, MASH/tablet additions). drugs.com (Ozempic history) · FDA compounded-GLP-1 policy / shortage resolution (21 Feb 2025) and salt-form statements, via FDA-policy summaries. weightlossrankings.org (compounded semaglutide) · Oral semaglutide / SNAC mechanism reviews (PIONEER programme). dovepress.com (SNAC review) · Exenatide/exendin-4 vs semaglutide comparative summaries. formblends.com (exendin-4 contrast) · “Ozempic face” dermatology context. henryford.com (Ozempic face)
8. Creator sources, cited as claims graded in the Creator layer, not evidence — Vigorous Steve, “Safest Dose Of GLP (NEXT-LEVEL Appetite Suppression)… Deep-Dive” (verified live). youtube.com (Vigorous Steve) · Transcript. youtubetotranscript.com (transcript) · Big Paul / Anabolic Bodybuilding, podcast feed (verified live). anabolicbodybuilding.podbean.com · Episode listing. listennotes.com (Anabolic Bodybuilding)
9. Creator sources — Dr. Alex Tatem, “Ozempic Doubles Your Risk of Blindness?!” youtube.com (Tatem, NAION short) · Fox News interview. foxnews.com (Tatem interview) · Hunter Williams, “GLP-1s Slow Aging?” (opened/read). huntershealthhacks.beehiiv.com (GLP-1s Slow Aging?) · “Why You're Stuck: GLP-1 Plateaus…” transcript (opened/read). library.hunterwilliamshealth.com (plateaus transcript)
10. Creator sources — TRT Dad Unscripted, Ep. 83 show notes. audible.com.au (TRT Dad Ep. 83) · toppodcast.com (TRT Dad Unscripted) · Dan the Bodybuilder / Steroids Podcast, Ep. 77 (transcript page opened). podscan.fm (Steroids Podcast Ep. 77) · Muscle Gain Formula transcript. podscan.fm (Muscle Gain Formula) · Peptide of the Week, episode/show description. podtail.com (Peptide of the Week)
11. Semaglutide — evidence deep-dive, research report filed 28 Oct 2026. Every figure, attribution and hedge on this page follows that report.