SARMs explained

There is only one androgen receptor. It is a single gene product, expressed in muscle, bone, prostate, skin, hair follicle, brain and liver, and it is the same protein in all of them. So the obvious question is how any molecule could activate it in one tissue and not another. The answer is the most interesting thing in this category.

When a ligand binds the androgen receptor, the receptor changes shape, moves into the nucleus and recruits a set of coregulator proteins before it can switch genes on. Different ligands produce subtly different receptor conformations, and different tissues express different mixes of those coregulators. A ligand that produces a conformation matching the coregulator profile of muscle, but not the profile of prostate, will act as a strong agonist in one and a weak one in the other. Selectivity is therefore not a property of the drug alone; it is a property of the drug and the cell together ^[1][2].

Two further mechanisms help. Testosterone is converted by 5-alpha reductase into dihydrotestosterone, a considerably more potent androgen, and that enzyme is concentrated in prostate, skin and hair follicle rather than in muscle. So a steroid arrives in prostate tissue amplified. A non-steroidal SARM is not a substrate for that enzyme, so it arrives at the same strength everywhere and the prostate loses its local amplifier. Likewise aromatase cannot convert a non-steroidal molecule into oestradiol, which removes the gynaecomastia and oestrogenic pathway entirely ^[1].

Most SARMs are also partial agonists at the receptor. A partial agonist produces less than the maximal response even at saturating concentration, and in tissues where the natural androgen signal is already high, a partial agonist can even reduce net activity by competing with it. That is a genuine pharmacological basis for a muscle-versus-prostate split, and it is also the reason the ceiling on muscle gain is lower than with a full agonist. The selectivity is real. It is partial, not absolute, and it varies a great deal between members ^[1].

Several SARMs went through real clinical development, which means there is genuine controlled human data here. It is worth knowing what it says, because it is both more solid and more modest than either the marketing or the scepticism suggests.

Ostarine (enobosarm) is the best-studied. In a double-blind, placebo-controlled phase 2 trial in healthy elderly men and postmenopausal women, it produced a dose-dependent increase in total lean body mass of roughly 1.4 kg over 12 weeks against placebo, with an improvement in a stair-climb power test ^[3]. In a phase 2 trial in patients with cancer-related muscle wasting it again increased lean body mass significantly against placebo, though the physical-function endpoint did not separate ^[5].

Then the phase 3 programme. Two trials, POWER 1 and POWER 2, were run in patients with non-small-cell lung cancer, with two co-primary endpoints: lean body mass and physical function measured by stair climb ^[6]. Enobosarm met the lean-mass endpoint and missed the physical-function endpoint. Building tissue was not enough; the drug had to make people able to do something, and it did not demonstrate that. It was never approved.

LGD-4033 (ligandrol) has a phase 1 study in healthy young men. Over three weeks it increased lean body mass dose-dependently, and it was described as well tolerated. It also, in the same three weeks, produced dose-dependent suppression of total testosterone, free testosterone, sex hormone binding globulin and HDL cholesterol, with follicle-stimulating hormone falling as well ^[4]. Three weeks is not a long exposure, and the suppression was already unambiguous.

That is the honest summary of the clinical record: SARMs increase lean mass in controlled trials, by amounts that are real but smaller than a steroid, they suppress the gonadal axis at the same time, and the one programme that tried to demonstrate a functional benefit in a phase 3 setting failed to do so.

The claim SARMs are usually sold on is that they do not suppress natural testosterone. That claim is false, and it was false in the first published human trial ^[4]. The mechanism is unavoidable: the hypothalamus and pituitary detect androgen activity and reduce luteinising hormone and FSH accordingly, and a SARM is an androgen as far as that circuit is concerned. Suppression is dose-dependent and it recovers on stopping in the trial data, but the trials ran for weeks, not the periods these compounds are actually used for.

Lipids move in the wrong direction. HDL cholesterol falls, and it falls quickly and consistently; the same phase 1 study saw it within three weeks ^[4]. This is a shared feature of oral androgens generally and is the main reason cardiovascular risk cannot be dismissed here.

Liver injury is documented, not theoretical. There are published case reports of cholestatic and hepatocellular drug-induced liver injury associated with products containing RAD-140 and LGD-4033, with jaundice and markedly raised bilirubin and liver enzymes lasting weeks to months ^[9][10]. In both of those reports the products were multi-ingredient supplements bought as SARMs, which is itself part of the finding: with an unregulated product it is often impossible to say which molecule caused the injury.

A survey of self-reported SARM users found a pattern consistent with all of the above, including a substantial rate of reported adverse effects and of testicular and libido changes, alongside the reported gains ^[8]. Survey data is weak evidence and should be read as weak evidence; it is included here because it is one of the few sources on what happens outside a trial.

So the accurate framing is not "fewer side effects than steroids" but a different and smaller side-effect profile, over a much shorter human safety record. There is no aromatisation, so no oestrogenic effects. There is no 5-alpha reduction, so the prostate and hair follicle effects are genuinely reduced. There is still androgen-receptor activity everywhere the receptor is expressed, still gonadal suppression, still an adverse lipid shift, and still documented liver injury. And there is no long-term human safety data at all, because the programmes that would have generated it stopped.

The table separates the compounds that are actually SARMs from the ones sold alongside them that are not. That second group causes more confusion than anything else in this corner of the market, and the distinction matters because a PPAR agonist and a growth-hormone secretagogue have nothing to do with the androgen receptor and nothing to do with each other.

Two of the impostors deserve a specific note. Cardarine (GW-501516) is a PPAR-delta agonist developed for dyslipidaemia; development was halted, and it is prohibited in sport as a metabolic modulator rather than as an anabolic agent ^[12]. MK-677 is a ghrelin-receptor agonist that raises growth hormone; see growth hormone and secretagogues. Neither is a SARM by any definition, and calling them one is not a harmless shorthand: it means the safety profile a buyer thinks applies does not.

YK-11 is the odd one out inside the group. It is not non-steroidal at all; it is a steroidal molecule built on a dihydrotestosterone-like core with a distinctive gem-dimethyl ester group, usually described as a myostatin-related agent as well as an androgen-receptor ligand. Its metabolism has been characterised for doping-control purposes, which is essentially the entire published human-relevant literature on it ^[11].

Everything above concerns pure compounds administered in trials. What is sold online is a different object, and this has been measured.

A JAMA analysis bought 44 products marketed as SARMs from the internet and analysed them. Only 52 percent contained a SARM at all. Thirty-nine percent contained a different unapproved drug, most often a growth-hormone secretagogue or a compound like Cardarine. Nine percent contained no active compound whatsoever. Among the products that did contain what they claimed, the amount matched the label in only about 41 percent of cases, and 25 percent contained substances not listed on the label at all ^[7].

That result reframes almost every argument about this category. Debating the tissue selectivity of RAD-140 is beside the point if the bottle in question has a one-in-two chance of containing it, and a one-in-four chance of containing something the buyer does not know about. It also explains the liver-injury case reports: both involved multi-ingredient products, and in that situation the causative agent frequently cannot be identified ^[9][10].

SARMs are also prohibited at all times in tested sport under the anabolic-agent category, and detection methods exist and are sensitive; metabolites of even obscure members have been characterised specifically so that they can be found in urine ^[11]. Contamination of ordinary supplements with SARMs is a documented cause of failed tests, which is a real hazard for anyone who is tested and takes anything at all.

Put next to anabolic steroids, the picture is consistent. SARMs produce smaller lean-mass gains than supraphysiological testosterone, which in a controlled ten-week trial raised fat-free mass by roughly two to four times the amounts reported anywhere in the SARM literature ^[13]. They remove the oestrogenic pathway and substantially reduce the prostate and hair effects, which was the design brief and which the pharmacology genuinely delivers ^[1][14]. They retain the gonadal suppression and the adverse lipid shift, and they add a liver signal that the injectable steroids do not have.

What is not known is the important part and is rarely said out loud. There is no long-term human safety data on any SARM, because no SARM completed a programme that would have produced it. The longest published exposures in healthy volunteers are measured in weeks. Nothing is known about cardiovascular outcomes, about prostate outcomes over years, about fertility after prolonged use, or about what repeated exposure does to the hypothalamic-pituitary-gonadal axis. The absence of that data is not the same as reassurance, and the marketing habit of treating it as reassurance is the single most misleading thing about this category.

It is also worth noting what a positive result would have looked like. Enobosarm had the best shot: a genuine pharmaceutical sponsor, a real unmet need in cancer cachexia, and two phase 3 trials. It grew tissue and did not improve function ^[6]. That is a meaningful scientific finding about the class rather than a regulatory technicality, and it is the reason there is still no approved SARM after roughly a quarter of a century of development.

None of this is medical or performance advice, and this page deliberately contains no protocol.