There is a specific way a claim goes wrong in the SARMs literature, and it is worth naming before anything else. The claim is about lean body mass. The evidence behind it comes from preclinical and clinical studies pooled in a systematic review that rated its own level of evidence as 4 — on a scale where 5 would be a body of randomised trials with good methodology. And then the same review spends most of its length on something else entirely: fifteen people.
What the review counted
The 2025 systematic review in the American Journal of Sports Medicine searched PubMed, Cochrane and Embase under PRISMA guidelines and found 72 articles from 2003 to 2022 relevant to SARM clinical outcomes, elimination profiles, contamination, safety, prevalence and doping control. It states its own level of evidence as 4.
Then the numbers:
- Prevalence of SARM use among athletes is estimated at 1 % to 3 %.
- SARM preclinical and clinical studies reported significant increases in lean body mass, and side effects including bone remodeling, testosterone suppression, and kidney, liver and prostate enlargement.
- Thirteen case reports described 15 cases. All were men. Median age 32 years, range 19 to 52. Eight of the fifteen were identified as athletes. All took SARMs orally, for a mean course of eight weeks.
Read those four numbers together and the shape of the evidence becomes obvious. The muscle gain is a pooled research finding from preclinical and clinical studies; the harms are a case series of fifteen men. Those are two entirely different kinds of evidence, and the public version of this compound usually presents them in the same font size.
The detail that belongs in every article about this
Five of the fifteen patients explicitly denied illicit drug use. The review draws the obvious conclusion: patients may believe their use to be legal. That is a marketing outcome, not a pharmacological one, and it happened before any trial finished.
It also states where the products came from: athletes most commonly purchased SARMs online, and most of these compounds have been shown to be contaminated with other substances. Whatever a paper says about the compound, it does not automatically describe the bottle.
And there is the amount point, which matters more than any mechanism: athletes reported consuming SARMs at much higher amounts than those clinically studied, which the authors say may increase the risk of the reported side effects — liver injury, impaired insulin sensitivity, cardiovascular events and tendon damage. A toxicity study conducted at one amount does not describe toxicity at another, and the people generating the case reports were not at the studied amount.
The rest of the class, for comparison
The same review does not exist for the androgen class as a whole, but a 2023 narrative review in the Journal of Gerontology covers testosterone, other androgens and nonsteroidal SARMs together, and it is more careful than the marketing on either side of this market. In randomised trials, testosterone treatment increases lean body mass, muscle strength, leg power, aerobic capacity and self-reported mobility. It has not consistently improved walking speed. It increases volumetric and areal bone mineral density and estimated bone strength. It improves sexual desire, erectile function and sexual activity, modestly improves depressive symptoms, and corrects unexplained anaemia in older men with low testosterone.
Two sentences from that review deserve to be read as a pair. It says prior studies have not been of sufficient size or duration to determine testosterone’s cardiovascular and prostate safety. And it says the efficacy of testosterone in reducing physical limitations, fractures, falls, progression to diabetes and correcting late-onset persistent depressive disorder remains to be established. So the strongest-molecule member of the class has a better evidence base than anything in the SARM reviews — and still has endpoints open.
On women, a 2026 review in Life is blunter still. Preclinical data in ovariectomised models indicate SARMs can enhance sexual motivation and improve muscle and bone outcomes, with compound-dependent effects on reproductive tissues. Clinical studies in postmenopausal women show increases in lean body mass with generally limited androgenic effects, but functional benefits are inconsistent and alterations in lipid profiles and liver enzymes have been reported. The conclusion: long-term safety and efficacy remain insufficiently characterized.
Why lean mass is not the claim people are making
The 2023 review ends on something that applies to the whole androgen field: strategies to translate androgen-induced muscle mass and strength gains into functional improvements are needed. Muscle mass is an intermediate. The endpoint a runner, a pensioner or a bedridden patient cares about is stairs, a fall, a fracture, a walk to the shop. A compound that moves lean mass and not function has moved the wrong number.
The systematic review’s own summary sentence puts the risk side in one line: SARM use is associated with increased muscle mass, hepatotoxicity, cardiotoxicity, tendon damage, and androgenic side effects throughout the body — including prostate enlargement and serum testosterone suppression. That sentence is also the reason this class is on a grey-market footing rather than an approved one.
What the evidence does not show
- No efficacy claim for any SARM in a functioning person. The level of evidence is 4 by the review’s own rating, and no trial in that set measured function as the endpoint.
- No safety record. The harm data is fifteen case reports, gathered from a population that mostly bought products online, often contaminated, at amounts far above those studied. Fifteen is not a rate.
- No legibility of the legal question. Five patients denied illicit use, which means the label on the bottle, not the pharmacology, is what they acted on.
- No data in women beyond postmenopausal trials with inconsistent functional benefit and reported lipid and liver enzyme changes.
- No long-term cardiovascular or prostate safety record, in the SARM literature or in the better-studied testosterone literature it is usually compared against.
- Contamination makes every consumer report partly a report on something else. This is the most underweighted caveat in the entire file.
The honest summary is short. Lean mass: probably real, from studies of graded evidence quality, at amounts nobody should be using. Harms: sparse, biased toward detection of the worst cases, and clustered in liver, heart, tendon and testosterone. Function: unmeasured. That combination — an appealing endpoint, an unmeasured one that matters, and a harm file fifteen people long — is what a grey market looks like from inside the literature.
