This site already compares peptides with anabolic steroids, and the answer there is blunt: one class was tested properly and works, the other was not tested. SARMs are the comparison people run next, and they produce a different and more useful answer — because in this case somebody did run the trial.
This is research journalism. It reports what the sources state, including the amounts used in trials where those are on the record, each figure carrying its source and date — and it never tells anyone what to take.
What a SARM is meant to be
A selective androgen receptor modulator binds the same receptor testosterone binds, but was designed to act on muscle and bone while doing less in the prostate, the skin and the liver. The whole premise is selectivity: keep the anabolic effect, drop the rest. Enobosarm — developed by GTx as GTx-024, and sold on the grey market as ostarine — is the one that went furthest.
A peptide in the muscle market works nowhere near the androgen receptor. The growth-hormone secretagogues act on the pituitary and raise growth hormone and IGF-1. So this is not a comparison of two doses of the same idea. It is two different mechanisms that were both proposed as ways to add lean mass.
The count, before the result
The registry is the cheapest way to see which class was actually studied. Intervention searches of ClinicalTrials.gov:
| Compound | Class | Registrations | Phase III with a muscle endpoint |
|---|---|---|---|
| Enobosarm (ostarine) | SARM | 17 | 2, completed |
| RAD-140 | SARM | 1 | No |
| Ligandrol (LGD-4033) | SARM | 0 under that name | No |
| Tesamorelin | GHRH analogue | 24 | No |
| Ipamorelin | Ghrelin agonist | 3 | No |
| CJC-1295 | GHRH analogue | 1, terminated | No |
| GHRP-6 | Ghrelin agonist | 0 | No |
| IGF-1 LR3 | IGF analogue | 0 | No |
One molecule on this table has been through two phase III trials with lean body mass written into the protocol as a co-primary endpoint. It is not a peptide.
Tesamorelin's 24 registrations are the peptide side's strongest showing, and none of them is a muscle trial — it is approved for abdominal fat in HIV lipodystrophy, and its lean-mass figure comes from pooled meta-analysis rather than from a trial designed to find it.
What the two phase III trials found
POWER 1 and POWER 2 were identically designed: randomised, double-blind, placebo-controlled, multinational, in patients starting first-line chemotherapy for non-small-cell lung cancer. Each randomised roughly 150 patients to enobosarm 3 mg orally once daily and 150 to placebo, for 147 days, with the co-primary endpoints assessed at day 84 (PMID 27138015).
Crucially, the endpoints were agreed with FDA as a responder analysis, and they were two:
- Lean body mass — the proportion of patients with no loss of lean body mass from baseline
- Physical function — the proportion improving stair climb power by at least 10% from baseline
Both trials posted their results to the registry. Taking them straight from the record:
| Lean body mass responders | Physical function responders | |
|---|---|---|
| POWER 1 (n=321) | enobosarm 41.9% vs placebo 30.4% | enobosarm 29.4% vs placebo 24.2% |
| POWER 2 (n=330) | enobosarm 46.5% vs placebo 37.9% | enobosarm 19.5% vs placebo 24.8% |
Read the two columns side by side. Lean body mass separated from placebo in both trials. Physical function did not — and in POWER 2 the placebo group out-performed the drug.
No SARM has been approved since.
Why this is the most useful trial in either class
This site has been writing the same sentence about the peptide side for a week, from different sources each time. Tesamorelin: +1.42 kg pooled lean mass across two independent 2026 meta-analyses, no strength endpoint met. MK-677: +1.1 kg fat-free mass over twelve months, no significant strength or physical-function gain. Capromorelin: +1.4 kg lean mass, trial terminated early for missing its prespecified effect. A class ceiling near 1.4 kg with nothing to show on function.
The POWER trials are the same result, reached by a completely different mechanism, in a properly powered phase III programme, on endpoints negotiated with the regulator in advance. Tissue mass on a scan moved. What the patient could do did not.
That is the strongest available evidence that the split is not an artefact of small peptide trials or of weak compounds. It survives being tested properly.
Legality: the same shape as the peptide market, with one difference
No SARM is an approved drug in the United States. A Drugs@FDA search returns no matches for enobosarm, ostarine, andarine or ligandrol — the same null this site got for CJC-1295, ipamorelin and IGF-1 LR3.
They are also not lawful dietary ingredients: an article that has been investigated as a drug cannot then be marketed as a supplement, which is why SARMs sold in capsules are being sold outside the framework they claim to be inside. As the legality page sets out for research peptides, the exposure in that structure sits primarily with the seller rather than with possession — the reverse of anabolic steroid law, where the possession itself is the offence. On that axis SARMs sit with the peptides, not with the steroids.
The difference is availability. Enobosarm was manufactured to pharmaceutical standard for a phase III programme, and what is sold under the name ostarine is not that. Identity and purity are a separate question from pharmacology, and nothing on this page speaks to what is in a capsule.
Testing
The published anti-doping literature on SARMs runs to 142 records, which repeats the pattern this site found for ipamorelin and CJC-1295: more published work on catching a compound than on whether it does anything.
The prohibited-list section numbers are not reproduced here. The source could not be opened when this page was written, and a half-remembered section number is not a citation — the same call this site made on the endurance page when the same source failed. What is sourced and published is on the drug-testing page, including detection windows and the screening methods that cover the peptide side.
The honest summary
SARMs have more human muscle data than any peptide in this market, by a wide margin, and that is the fact most comparisons get backwards. The reason to know it is not that it favours SARMs. It is that when the muscle trial the peptide side has never had was finally run — twice, in 651 patients, against placebo, on endpoints a regulator signed off — lean body mass responded and the thing patients could actually do did not.
Anyone choosing between the two classes on the strength of the evidence is choosing between a class that has not been tested and a class that was tested and did not clear the bar.
Limits of this page
Both phase III trials were run in lung cancer patients on chemotherapy, not in healthy trained adults, and a compound can behave differently in the two populations — the reason a null in cachexia is evidence about the endpoint, not proof about a gym. The responder-rate figures are as posted to the registry and are proportions, not mean changes. The registry counts were run on 13 September 2026. No prohibited-list or FDA enforcement claim is made here beyond the Drugs@FDA null, because those sources did not open.
Sources and dates
- ClinicalTrials.gov API v2, intervention searches and posted results for NCT01355484 (POWER 1) and NCT01355497 (POWER 2), retrieved 13 September 2026.
- Crawford J et al. Study design and rationale for the phase 3 clinical development program of enobosarm. Curr Oncol Rep 2016;18(6):37. PMID 27138015
- Srinath R, Dobs A. Enobosarm (GTx-024, S-22): a potential treatment for cachexia. Future Oncol 2014;10(2):187–94. PMID 24490605
- openFDA Drugs@FDA API, generic-name searches for enobosarm, ostarine, andarine and ligandrol, run 13 September 2026 — no matches.
- NCBI PubMed E-utilities, record count for selective androgen receptor modulators in doping control, run 13 September 2026.