Cardarine for endurance
Does GW501516 improve endurance performance in humans?
Two corrections before anything else
Cardarine is not a SARM. It is sold as one, discussed as one, and stocked alongside them, but it does not touch the androgen receptor. GW501516 is a PPARδ agonist — a metabolic modulator that acts on an entirely different pathway. Anyone reasoning about it by analogy to ostarine or RAD-140 is reasoning about the wrong drug.
Its development was not merely abandoned. It was terminated over cancer. That is the single most important fact on this page, and by the rules of our own grading scale it does not appear in the marks at the top. We will come back to why.
Where the endurance claim comes from
In 2008, a landmark paper in Cell reported that a PPARβ/δ agonist combined with exercise training "synergistically increase oxidative myofibers and running endurance in adult mice."4 The work was elegant, widely covered, and gave rise to the phrase that still sells this compound: exercise mimetic.
Read the sentence again. Adult mice, on a treadmill.
That is the origin of every endurance claim made for Cardarine, and eighteen years later it remains the strongest evidence for one. The finding has not been reproduced in a human endurance trial, because no human endurance trial has been run.
One detail worth flagging for anyone who has seen "44% increase in endurance" attached to Cardarine: that figure in the paper belongs to AICAR, a different compound tested in the same study, in sedentary mice.4 It is routinely misattributed. We mention it because misattributed numbers are how a mouse result becomes a product claim.
What was actually tested in humans
Human trials of GW501516 do exist — four of them, all conducted during its pharmaceutical development. Every one measured blood lipids. None measured endurance.
| What Cardarine is sold for | What the human trials tested | |
|---|---|---|
| Outcome | endurance, fat loss | HDL, LDL, triglycerides, apolipoproteins3 |
| Population | healthy trainees | patients with low HDL and metabolic syndrome features3 |
| Largest trial | none exists | 268 participants, 12 weeks3 |
| Endurance measured | — | never |
The largest of those trials gave 2.5, 5 or 10 mg daily to 268 patients for twelve weeks. On lipids it worked: HDL cholesterol rose up to 16.9%, triglycerides fell 16.9%, apoB fell 14.9%, free fatty acids fell 19.4%.3 The authors concluded that PPARδ was "a potentially important target for providing cardiovascular protection."
So the compound does something real in people. It changes their lipid panel — a legitimate finding, in patients with a lipid disorder, which is not what anyone is buying it for.
Why the programme stopped
Development did not quietly run out of funding. It was halted after preclinical toxicity work, and the regulatory record on what was found is unusually blunt.
Sport Integrity Australia, the Australian government's anti-doping body, states that GW1516 "is associated with high frequency of carcinogenicity and reproductive toxicity," and that clinical trials "have since been abandoned due to safety concerns."2
Australia's Therapeutic Goods Administration went further than any regulator has for anything else covered on this site. It placed GW1516 in Schedule 10:
Substances of such danger to health as to warrant prohibition of sale, supply and use. TGA Schedule 10, as cited by Sport Integrity Australia2
Schedule 10 is not a restriction on who may prescribe something. It is a category for substances that should not be available at all.
The warning anti-doping authorities almost never give
Anti-doping agencies exist to keep sport clean. They prohibit substances; they do not, as a rule, take responsibility for the wellbeing of people deliberately breaking the rules.
In March 2013, WADA broke that convention for this compound.
"It has come to WADA's attention that the black market substance GW501516 is being sold to and used by some athletes."
"The side effect of this chemical compound is so serious that WADA is taking the rare step of warning 'cheats' to ensure that there is complete awareness of the possible health risks to athletes who succumb to the temptation of using GW501516 for performance enhancement."
"GW501516 was a developmental drug that was withdrawn from research by the pharmaceutical company and terminated when serious toxicities were discovered in pre-clinical studies."
"Clinical approval has not, and will not be given for this substance."1
That last sentence is worth sitting with. Regulators are ordinarily careful never to foreclose a future approval. WADA stated flatly that this one is not coming.
People are still taking it
Thirteen years after that alert, the compound remains in circulation and in use. In 2023, anti-doping testing recorded 54 adverse analytical findings for GW1516 — 17% of every positive test in its category.2
Those are only the people who were tested and caught. It is prohibited at all times as a non-Specified substance, and no therapeutic use exemption would be granted for it under any circumstances.2
Why the marks at the top don't tell you the important part
This breakdown is graded Preliminary · Untested in humans, which is accurate and which badly undersells the situation. It is worth explaining why, because it exposes a real limit of our own scale.
Our rubric grades claims about outcomes in humans. Under version 2.2, mechanism and preclinical findings are reported but never carry marks — a rule we adopted so that a laboratory result could not masquerade as a clinical one.
Here that rule works against the reader. The decisive fact about Cardarine is preclinical: animal carcinogenicity serious enough that a pharmaceutical company terminated a drug it had spent fifteen years developing, and serious enough that a national regulator prohibited its sale outright.
The dots cannot express that, because the dots measure how much we know about what it does in people — and the answer there is genuinely "almost nothing." Both statements are true at once: the human evidence is thin, and the preclinical evidence is alarming.
Read the marks as what they are — a measure of human evidence — and read the safety section as the part that matters here.
Three claims, three answers
The lipid claim reaches Emerging rather than Moderate because all four human trials come from the same pharmaceutical development programme — under our definition that is one research group, not the two independent ones Moderate requires.
Safety and unknowns
-
Adverse events reported
Sport Integrity Australia states that adverse effects have been documented in both clinical trials and case reports, and that GW1516 is associated with a high frequency of carcinogenicity and reproductive toxicity.2 WADA describes the preclinical findings as "serious toxicities."1
-
Longest human exposure studied
Twelve weeks, in the largest lipid trial.3 Cancers in the animal work developed over far longer exposures than any human trial has run — meaning the human studies could not have detected the effect that ended the programme.
-
Known interactions and at-risk groups
Not characterised. Human trial populations were patients with low HDL and metabolic syndrome features, aged well beyond the demographic buying it now.3
-
Regulatory status
Not approved anywhere, and WADA states approval "has not, and will not be given."1 Not on the Australian Register of Therapeutic Goods; placed by the TGA in Schedule 10, substances warranting prohibition of sale, supply and use.2 Prohibited by WADA at all times under S4.4.1 as a non-Specified substance, with no therapeutic use exemption available.2
-
What has not been characterized
Whether the animal carcinogenicity translates to humans at the doses people actually take — nobody will run that study, and the absence of an answer is not reassurance. Endurance effects in humans at any dose. Long-term exposure of any duration. And, as with every research-chemical purchase, what is actually in the vial.
How this was graded
Evidence strength — Preliminary, criteria checked
Trials of a compound's effect on cholesterol do not count toward a claim about running further. This is the same rule that kept thymosin beta-4's eye-drop trials out of TB-500's grade: evidence about a different outcome is evidence about that outcome.
Bottom line
Cardarine is not a SARM, and no human being has ever been tested to see whether it improves endurance. The entire premise rests on a 2008 study of mice on treadmills.
The four human trials that do exist measured cholesterol in patients with metabolic syndrome, and found real effects on their lipid panels. That is the whole of the human record.
What ended the drug was cancer in animals — tumours across multiple organs, serious enough that the developer terminated the programme, serious enough that Australia's regulator placed it in the category reserved for substances too dangerous to sell, and serious enough that WADA took the almost unheard-of step of publicly warning people who were breaking anti-doping rules that they might be harming themselves. In 2023 the tests still caught fifty-four of them.
Sources
- World Anti-Doping Agency. WADA issues alert on GW501516. 21 March 2013. wada-ama.org
- Sport Integrity Australia. GW1516 (GW501516) information. Includes TGA scheduling, WADA classification, and 2023 adverse analytical finding figures. sportintegrity.gov.au
- Olson EJ, Pearce GL, Jones NP, Sprecher DL. Lipid effects of peroxisome proliferator-activated receptor-δ agonist GW501516 in subjects with low high-density lipoprotein cholesterol: characteristics of metabolic syndrome. Arteriosclerosis, Thrombosis, and Vascular Biology. 2012;32(9):2289–2294. NCT00258899, NCT00388180. PMID 22814748
- Narkar VA, Downes M, Yu RT, Embler E, Wang YX, Banayo E, Mihaylova MM, Nelson MC, et al. AMPK and PPARδ agonists are exercise mimetics. Cell. 2008;134(3):405–415. PMID 18674809
Search results for this compound are dominated by vendor pages, several of which reproduce the animal carcinogenicity findings while arguing the doses were too high to matter. Under our sourcing standard none are usable as evidence for any mark, and none were used here.
