Cardarine cancer risk across rodent studies, mouse models and human trials

The sources do not show that cardarine causes cancer in people, but, according to a review, the developer stopped work after rodent studies were linked to tumour development, academic mouse studies reported enhanced tumour formation in some colorectal and pancreatic models while others found no effect, and none of the human trial reports describes a cancer outcome, so the question is unresolved. GW-501516, also sold under the name Cardarine, is a laboratory compound that was developed as a possible treatment for abnormal blood fats and then abandoned. The cardarine cancer question is the one that led the developer to stop work, according to a review (Mackenzie and Lione, 2013), so this guide to GW-501516 research sets out what the published papers actually found in cells, mice and human volunteers, why the developer stopped work after rodent studies, which adverse findings and measured results the literature contains, and how anti-doping bodies treat the compound. It is written for readers who want the evidence rather than the marketing. The descriptions of the 25 PubMed-indexed papers listed under Sources, which include reviews, a product survey, studies of other ligands and four SARM and RAD-140 papers used only in the comparison section, draw on what their abstracts state, together with the PubMed publication-type labels (for example randomised or multicentre); the registry, legislation and UKAD sources are cited separately, and short plain-English glosses and a few labelled editorial notes are added for readability. Listings (laboratory research only, 18+): Cardarine (GW-501516) capsules compared.

Key takeaways

• GW-501516 is a PPAR-delta agonist: a synthetic molecule that switches on a receptor involved in how cells handle fats and sugars. It was developed by GlaxoSmithKline and Ligand as a candidate treatment for dyslipidaemia (Pelton, 2006).
• Four published human trial reports are reviewed here, with treatment periods of 2 to 12 weeks. They measured blood lipids, liver fat, insulin and other laboratory markers, including muscle gene expression; the evidence section sets out what none of the papers reports.
• Development was stopped after rodent studies were linked to widespread tumour development, according to a peer-reviewed review (Mackenzie and Lione, 2013). The original reports are not on PubMed.
• Academic studies reported that GW-501516 enhanced tumour formation in colorectal cancer models in mice and in a pancreatic cancer model. Other studies in cell lines and mice found no such effect. Human relevance is unknown.
• The World Anti-Doping Agency added GW-501516 to its Prohibited List in 2009, and anti-doping laboratories have published urine test methods for it.
• The endurance results come from mice, and none of the human trial reports describes a cancer outcome. Nothing here should be read as advice for any reader.
• Free Muscle's research-use-only Cardarine listings, and its lab testing page, are described under About the Free Muscle Cardarine listings.
• The principal adverse findings in the sources, from the rodent tumour signal to the two multi-substance case reports, are collected in one table under Cardarine side effects reported in the research, and the compound is set against SR9009 and RAD-140 by receptor, evidence base and anti-doping status under Cardarine vs SR9009 vs RAD-140.

What is GW-501516?

GW-501516 is a small synthetic molecule that binds and activates peroxisome proliferator-activated receptor delta, usually written PPAR-delta (or PPAR-beta/delta). PPAR-beta/delta is described as a nuclear receptor (Billin, 2008). An "agonist" is simply a molecule that activates its target receptor.

The compound was developed by GlaxoSmithKline and Ligand as a candidate treatment for dyslipidaemia, meaning an unhealthy balance of blood fats. A 2006 review recorded that phase II clinical trials were under way at that time (Pelton, 2006). ClinicalTrials.gov lists three completed GW501516 studies: NCT00158899, a Phase 2 study in people with low HDL cholesterol; NCT00841217, a Phase 4 study of lipoprotein transport in metabolic syndrome; and NCT00388180, an exploratory body fat and inflammation study. None of the three is registered for athletic performance (ClinicalTrials.gov registry search).

By 2008 one review concluded that it was still not known whether PPAR-beta/delta agonists would be different enough from existing treatments to justify use in dyslipidaemia or type 2 diabetes, and that the path forward was undefined (Billin, 2008). The papers cited here record no regulatory approval. Across these papers, GW-501516 appears in preclinical studies, a small set of lipid trials, anti-doping papers and two case reports. The human data for GW-501516 are therefore limited, and development ended after a reported rodent safety signal.

A note on terminology: GW-501516, GW1516 and Cardarine all refer to the same molecule; one case report writes "GW1516 (cardarine)" (Kintz et al., 2021). It is not a SARM (selective androgen receptor modulator); it acts on a different receptor (PPAR-delta, not the androgen receptor).

How researchers think it works

PPAR-delta is expressed in skeletal muscle and in intestinal and pancreatic tissue in the studies below. Researchers noted that exercise raises its expression in muscle and that it influences how muscle handles fat. The working idea behind GW-501516 research is that activating the receptor shifts cells towards using fatty acids as fuel.

The cell-culture evidence addresses mechanism most directly. In primary cultured human muscle cells, GW-501516 triggered phosphorylation of AMPK (an energy-sensing enzyme) and increased fatty acid transport, fatty acid oxidation and glucose uptake. Fatty acid oxidation depended on both PPAR-delta and AMPK, while the rise in glucose uptake needed AMPK but not PPAR-delta (Krämer et al., 2007). This was cell culture only, so it shows a possible mechanism, not an outcome in a person.

Human trials added indirect support. In one trial, gene expression for fat-handling proteins rose in muscle cells treated in the laboratory, and the exhaled-breath data in another suggested more fat-derived carbon dioxide after treatment (Sprecher et al., 2007; Risérus et al., 2008). In a lipoprotein-kinetics study, treatment changed how lipoprotein particles were cleared and produced (the abstract reports increased hepatic removal of VLDL particles and increased apoA-II production; Ooi et al., 2011). These findings point to two possible mechanisms the authors discuss: muscle fat oxidation and altered lipoprotein handling.

For background only: Narkar et al. (2008) tested pathway-specific drugs on treadmill running in mice. This abstract does not name GW-501516; it refers only to a PPAR-beta/delta agonist, which combined with exercise training synergistically increased oxidative muscle fibres and running endurance, and its headline endurance figure belongs to a different compound, the AMPK agonist AICAR, in sedentary mice. It is therefore not evidence about GW-501516, and mouse data cannot be assumed to transfer to humans.

PPAR-delta is a downstream target of beta-catenin signalling and is upregulated in colorectal cancer (Liu et al., 2019).

These mechanistic findings sit alongside the tumour-promotion findings in mouse models and the rodent signal behind the discontinuation of development, which are set out in the evidence and safety sections below (Wang et al., 2006; Liu et al., 2019; Mackenzie and Lione, 2013).

Key studies at a glance

The evidence, study by study

Four published human trial reports of GW-501516 are reviewed here. Their treatment periods ranged from 2 to 12 weeks in volunteers or patients, and all measured laboratory markers; only Sprecher et al. state in the abstract that the drug was given orally. The Olson paper describes two studies, a main trial and a smaller exploratory one. The papers state the protocol doses, so they are not repeated here. None of the papers cited here reports results from a human trial of body weight, body composition, endurance or athletic performance. The first-in-human and later trials are covered first, followed by preclinical work.

Sprecher et al., 2007: the first report in people

This randomised study gave healthy volunteers placebo (6 people) or one of two treatment levels (9 people each) for 2 weeks while they were hospitalised and sedentary. The authors described it as the first report of a PPAR-delta agonist given to humans. In these hospitalised, sedentary healthy volunteers, HDL cholesterol (often called "good" cholesterol) was higher in both treatment groups than in the placebo group, where it fell by 11.5%. Serum triglycerides trended downward (P=0.08 in the higher treatment group), while triglyceride clearance after fat feeding improved (P=0.02). In cells, the compound raised fatty acid oxidation and doubled expression of the ABCA1 gene (Sprecher et al., 2007).

Limitations: the groups were tiny, the volunteers were healthy and sedentary, and the trial lasted two weeks. The abstract does not describe adverse events.

Risérus et al., 2008: moderately overweight men and a comparator

In a double-blind trial, six moderately overweight people per group received GW-501516, a comparison PPAR-alpha agonist or placebo for 2 weeks. In the six people who received GW-501516 in this double-blind trial, fasting triglycerides fell 30%, apolipoprotein B 26%, LDL cholesterol 23% and insulin 11%, while HDL cholesterol was unchanged. Liver fat fell by 20% and urinary isoprostanes, a marker of oxidative stress, by 30%. Muscle expression of the fat-handling gene CPT1b rose (Risérus et al., 2008).

Limitations: with six people per group, the findings are exploratory, and two weeks says nothing about long-term effects. The abstract reports no adverse-event data. The authors concluded that effects were probably due to increased fat oxidation in muscle, which is an interpretation, not a direct measurement of fat oxidation in muscle.

Ooi et al., 2011: lipoprotein kinetics

This randomised, double-blind crossover trial recruited 13 dyslipidaemic men with central obesity. Each had a 6-week period on placebo and a 6-week period on GW-501516, separated by a 2-week washout. In these 13 men, treatment lowered plasma triglycerides, fatty acids, apoB-100 and apoB-48, and raised HDL cholesterol and apoA-II. Kinetic analysis suggested the liver removed VLDL particles faster and that production of apoA-II rose. Cholesteryl ester transfer protein activity fell (Ooi et al., 2011).

Limitations: only men took part, the sample was small and the study aim was mechanistic. The abstract does not give safety outcomes.

Olson et al., 2012: the largest of the four trials reviewed

This multicentre randomised trial enrolled 268 patients with HDL cholesterol below 1.16 mmol/L and ran for 12 weeks, with placebo and three treatment levels. In the 268 patients, HDL cholesterol rose from baseline by up to 16.9% and apoA-I by up to 6.6%. LDL cholesterol fell 7.3%, triglycerides 16.9%, apoB 14.9% and free fatty acids 19.4%. A second, smaller exploratory study of 37 similar patients reported fewer very-low-density and intermediate-density particles and more medium and large HDL particles (Olson et al., 2012).

Limitations: the abstract reports changes from baseline and does not state whether the figures are placebo-adjusted, and twelve weeks is the longest treatment period among the four trial reports. It reports lipid and metabolic endpoints, not a cancer outcome, and it does not report adverse events. The abstract reports lipid and apolipoprotein measurements, not cardiovascular events.

Krämer et al., 2007: human muscle cells

In primary cultured human muscle cells, GW-501516 increased AMPK phosphorylation, fatty acid transport and oxidation, and glucose uptake. Silencing PPAR-delta with small interfering RNA showed the receptor was needed for fatty acid accumulation and oxidation; the glucose effect required AMPK but not PPAR-delta (Krämer et al., 2007). Limitation: this was cell culture, so it cannot show what happens in a whole person.

Chen et al., 2015: Kunming mice and metabolomics

Chen and colleagues reported that GW-501516 enhanced running endurance and raised the proportion of succinate dehydrogenase-positive muscle fibres in both trained and untrained Kunming mice. It also raised serum unsaturated fatty acids, and the authors suggested the mechanisms behind enhanced running capacity are not identical to those of training (Chen et al., 2015). Limitation: running to exhaustion in mice is not a human performance endpoint, and the abstract gives no safety assessment.

Cancer-related animal and cell studies

Wang and colleagues reported genetic and pharmacological evidence that deleting PPAR-delta reduced intestinal adenoma growth in Apc(Min/+) mice and inhibited the tumour-promoting effects of GW-501516. Activation also increased VEGF in colon carcinoma cells, and the authors flagged concern for patients at high risk of colorectal cancer (Wang et al., 2006).

Liu and colleagues reported in 2019 that GW-501516 enhanced colorectal tumorigenesis in mice carrying an Apc mutation, while the antagonist GSK3787 suppressed it (Liu et al., 2019). A 2022 paper reported that in mutant KRAS(G12D) pancreatic epithelial cells, GW-501516 strongly accelerated progression of precancerous lesions to pancreatic cancer, apparently through an immunosuppressive cell environment; the abstract mentions mouse and human precancerous tissue, but the species used for the GW-501516 experiment is not stated (Liu et al., 2022).

Limitations: these were models that already carried cancer-driving mutations, and the abstracts do not state the route of dosing. They do not tell us what happens in people without such mutations. Separately, a review reports that GSK's own rodent studies, which are not on PubMed, were the stated reason development ended (Mackenzie and Lione, 2013).

Counter-evidence

The literature is not one-sided. Hollingshead and colleagues tested GW0742 and GW-501516 in five human cancer cell lines (HT29, HCT116, LS-174T, HepG2 and HuH7) and in mouse liver, colon and polyps. Neither ligand increased cell growth, Akt phosphorylation, VEGF or COX2, and the authors concluded the ligands did not potentiate tumorigenesis (Hollingshead et al., 2007). A separate mouse study used only GW0742, a different ligand, and found attenuated chemically induced colon carcinogenesis; it also recalled that a PPAR-beta ligand had increased small-intestinal tumours in Apc(min/+) mice (Marin et al., 2006). Because that study did not use GW-501516, it cannot be read as a statement about GW-501516.

Safety signals and what regulators and doping laboratories say

The central safety signal is the rodent one. A peer-reviewed review states that, after clinical trials, GSK announced further research would be discontinued because preclinical rodent studies linked the compound to widespread tumour development (Mackenzie and Lione, 2013). This is a secondary source. The original GSK reports are not on PubMed, so species, doses and tumour sites are not stated here. The academic cancer studies above used models carrying cancer-driving mutations (see limitations above); their relevance to people is unknown.

Two case reports describe people who used products containing the compound. A 43-year-old man presented to an emergency unit with abdominal pain, muscle pain and headache after reporting use of cardarine together with the SARM ostarine. Liver enzymes were markedly raised (ALT up to 922 and AST up to 2558 UI/L), as was creatine phosphokinase (up to 86435 UI/L), a sign of massive muscle breakdown. Cardarine was found in blood at 403 ng/mL and hair testing indicated repeated exposure; he fully recovered by 6 weeks (Kintz et al., 2021). A 40-year-old man developed gynaecomastia and hypogonadotropic hypogonadism after using commercial supplements that contained RAD-140, MK-677 and cardarine, plus undisclosed testosterone, estradiol and growth hormone; symptoms resolved when he stopped (Chong et al., 2024). In both reports several substances were present, so cardarine cannot be proven to be the cause. As editorial context, not a finding of the cited abstracts: case reports are generally regarded as weak evidence, since they can signal a hazard but cannot establish how often a harm occurs.

On product quality, European official medicines control laboratories analysed 324 samples (354 results) from 14 laboratories in 13 countries, most from illegal distribution; cardarine was one of the five most frequently reported of 18 molecules. Of the seized products, 65% were represented as medicines and 24% as dietary supplements. From the limited quantitative data reported, the authors assumed that most samples contained active doses and some were overdosed (Barrios et al., 2025). The figures are not specific to cardarine, and the abstract does not name any seller.

On doping control: Thevis et al. (2010) characterised two major urinary metabolites of GW1516 and developed a urine method to detect GW1516 in doping-control samples, with a detection limit of 0.1 ng/ml. A controlled excretion study of GW1516 and the related compound GW0742 characterised their urinary metabolites for detection purposes; the GW0742 metabolites were detectable for a shorter period than those of GW1516 (Sobolevsky et al., 2012).

What the research does not tell us

Good GW-501516 research needs clear limits, and these are the main ones for the cardarine cancer question and the wider literature.

• No human results on body weight, body composition, endurance or performance in the papers reviewed here. The four human trial reports measured blood lipids and apolipoproteins, liver fat, insulin, glucose, free fatty acids, oxidative-stress markers, lipoprotein kinetics and muscle gene expression; none of the papers reviewed reports body weight, body composition, endurance or performance outcomes (see the evidence section). Mouse endurance results do not establish a human effect.
• No long-term human safety data in the papers reviewed here. The longest treatment period among the four trial reports was 12 weeks, and none describes a cancer outcome.
• Contradictory cancer literature. Tumour promotion appeared in Apc-mutant mouse models and in a mutant-KRAS pancreatic model, while some cell-line and mouse work found no growth effect. Which result applies to people is unknown.
• Missing primary documents. The original GSK rodent carcinogenicity reports were not available, so species, exposure margins and tumour sites are unknown here.
• Route and formulation. Only one of the four trial abstracts (Sprecher et al.) states the route of administration, and the cancer and endurance findings come from mice and cell culture, where the abstracts do not give the route of dosing. Results from these studies cannot be assumed to apply to an oral capsule, and nothing in them establishes that products obtained through illegal distribution match the purity or amount of a clinical-trial product, as the product-quality survey above illustrates.
• Co-ingestion. The case reports involve several compounds at once, so causality for cardarine alone is unproven. The controlled human trials compared the compound with placebo (and, in one trial, a different drug in a separate group) for treatment periods of 12 weeks at most, and none of the abstracts describes a combination arm, so there are no controlled data on combinations or longer exposure.
• Evidence hierarchy (editorial context, not drawn from the cited abstracts). Randomised trials are generally ranked above cohort studies, animal studies, cell culture and case reports. Anecdotal reports, such as forum posts or personal accounts, are not evidence and are not used here.

Cardarine results: what the studies actually measured

This section groups the measured results by outcome rather than by paper, naming the population and design each time, and the adverse findings from the same literature close the section so that neither half is read on its own. The sources contain no human body-composition or performance figures.

Blood lipids in volunteers and patients. The recurring measured result across the four trial reports was a shift in blood lipids. In the randomised two-week study of hospitalised healthy volunteers, with two treatment groups of nine and a placebo group of six, HDL cholesterol rose in both treated groups while the placebo group fell 11.5%, and triglyceride clearance after a fat meal improved (P=0.02) (Sprecher et al., 2007). In the double-blind two-week study of six moderately overweight people per group, fasting triglycerides fell 30%, apolipoprotein B 26%, LDL cholesterol 23% and insulin 11%, with HDL cholesterol unchanged (Risérus et al., 2008). In the crossover study of 13 dyslipidaemic men with central obesity, using six-week periods, triglycerides, fatty acids, apoB-100 and apoB-48 fell, HDL cholesterol and apoA-II rose, and cholesteryl ester transfer protein activity decreased (Ooi et al., 2011). In the 12-week multicentre trial of 268 patients with low HDL cholesterol, HDL cholesterol rose by up to 16.9% and apoA-I by up to 6.6% from baseline, while LDL cholesterol fell 7.3%, triglycerides 16.9%, apoB 14.9% and free fatty acids 19.4%; the 37-patient exploratory study recorded 19% fewer VLDL, 52% fewer IDL and 14% fewer LDL particles and 10% more HDL particles (Olson et al., 2012).

Liver fat and oxidative stress. Only one trial abstract reports these. In the six Risérus participants who received GW-501516, liver fat content fell 20% and urinary isoprostanes 30% (Risérus et al., 2008).

Muscle and cell markers. Human muscle cells treated in the laboratory showed higher fatty acid oxidation, a two-fold rise in ABCA1 and increased CPT1 and CD36 expression (Sprecher et al., 2007), and AMPK phosphorylation with greater glucose uptake (Krämer et al., 2007). In muscle biopsies from the participants of the double-blind, randomised, two-week Risérus study, CPT1b expression rose, and a larger share of their exhaled carbon dioxide came from the fat in a test meal (Risérus et al., 2008).

Endurance, in mice only. In a mouse metabolomics study of trained and untrained Kunming mice, the GW-501516-treated mice, trained and untrained, ran longer before exhaustion and had a higher proportion of SDH-positive fibres, with raised serum galactose, beta-hydroxybutyrate and unsaturated fatty acids (Chen et al., 2015). No human study in the sources measured endurance, strength, body weight or body composition.

Adverse findings and limits from the same literature. None of the four trial abstracts reports adverse events, and 12 weeks is the longest treatment period among the four trial reports. The cardarine cancer evidence sits beside these lipid figures: the developer's rodent studies were linked to widespread tumour development (Mackenzie and Lione, 2013), GW-501516 promoted intestinal and colorectal tumours in Apc-mutant mice (Wang et al., 2006; Liu et al., 2019) and accelerated pancreatic cancer in KRAS-mutant cells (Liu et al., 2022), while five cancer cell lines showed no growth effect (Hollingshead et al., 2007). Two multi-substance case reports describe raised liver enzymes with rhabdomyolysis and a hormonal disturbance (Kintz et al., 2021; Chong et al., 2024). The lipid figures above therefore describe short trials in named populations (13 to 268 participants, treatment periods of 2 to 12 weeks) and nothing else; they cannot be assumed to apply to an oral capsule sold as a research compound, and they are not results anyone should expect.

Cardarine side effects reported in the research

Cardarine side effects cannot be listed the way a licensed medicine's can, because development was discontinued and the sources describe no approved product label with a safety section. The table below collects the principal adverse or safety-relevant findings in the sources, including the counter-evidence, with the population and design beside each row so that the weight of each finding is visible. The cancer findings in the table are all preclinical (rodent studies, mouse models and cell lines), the two case reports describe liver and hormonal findings rather than cancer, and none of the findings comes from a human trial.

What this table can and cannot show. A case report records that a harm occurred in one person after a particular exposure; it cannot establish how often the harm occurs, whether it depends on the amount, or even that the compound caused it, and in both reports here several substances were present, in one case with undisclosed hormones. The rodent tumour signal that ended development is known, in the sources used here, only through a review, and the academic cancer models carried cancer-driving mutations, which is why the counter-evidence row belongs in the same table. The largest gap is that no trial abstract reports adverse events: the human safety record for GW-501516 in the sources is therefore not "nothing was found" but "nothing was reported in the abstract", which is a different statement. None of the abstracts states the dosage form of the products in the case reports, so nothing in the sources links the findings to an oral capsule.

Cardarine vs SR9009 (stenabolic) vs RAD-140: how they differ

The three compounds are often discussed together, which invites the assumption that they are variations on one theme. They are not. They act on three different receptors, carry three very different evidence bases. There is no head-to-head trial of any two of them in the sources used here, so no statement about relative effect is possible and none is made. Comparator statements below come either from PubMed records merged into this article's sources for RAD-140 or from the Free Muscle articles on SR9009 (stenabolic) and RAD-140 (testolone), as indicated in each cell.

Three points stand out. First, the human data differ in kind, not only in quantity: GW-501516 was given to healthy volunteers and lipid patients, RAD-140 to heavily pretreated cancer patients, and SR9009 to no trial population at all, so the three cannot be ranked against each other. Second, the adverse signals differ in kind too: a rodent carcinogenicity signal and tumour promotion in mutant mice for GW-501516, liver enzyme elevation in the one RAD-140 trial in the sources (LoRusso et al., 2022), and case-report liver injury for SR9009. Third, all three are prohibited in sport, and anyone subject to testing should read the current list rather than this summary. Where Free Muscle lists any of them, it is as a research compound for laboratory research use only, to customers aged 18 or over; the different receptor each targets is a reason to read its literature separately, not a basis for treating one as a substitute for another.

Legal and anti-doping status in brief

This is general information, not legal advice. In the UK, the Human Medicines Regulations 2012 provide that a person may not sell or supply, or offer to sell or supply, an unauthorised medicinal product, and possession is prohibited where the person knows or has reasonable cause to believe it is intended for sale or supply (regulation 46). A medicinal product is defined in regulation 2 as a substance presented as treating or preventing disease, or one that may be used with a view to modifying a physiological function through a pharmacological, immunological or metabolic action. Those pages do not name cardarine, so whether the definition applies to it is a legal judgment, not a regulator statement quoted here.

UK Anti-Doping (UKAD) states in its SARMs guidance that sellers often add "for research purposes only" or similar wording to circumvent restrictions, and that products labelled this way are not regulated for safety or composition (UKAD on SARMs). That page concerns SARMs and does not name cardarine, so it is used here only as an analogy for the wording.

In sport, WADA added GW501516 to the Prohibited List in 2009 because of assumed performance-enhancing properties (Pokrywka et al., 2014), and Thevis et al. (2010) record that the list has included GW1516 since January 2009. UKAD explains that the list is updated each year from 1 January, that athletes are strictly liable for anything found in their system, and that a substance missing from a product label is no defence (UKAD Prohibited List guidance). This article does not reproduce the current WADA Prohibited List; check the official WADA list directly for the current position.

Full breakdown with sources: Is cardarine legal in the UK?

About the Free Muscle Cardarine listings

Free Muscle sells research compounds and gym apparel online. Research compounds sold by Free Muscle are supplied for laboratory research use only and are not for human consumption. They are not authorised as medicines by the MHRA, EMA or FDA, are not food supplements, and are sold only to adults aged 18 or over. Nothing on this page is medical or dosing advice.

Two GW-501516 listings are currently on the site: the Cardarine GW-501516 5mg, 60 capsules product and the Cardarine GW-501516 10mg, 60 capsules product. Neither listing publishes any dosing schedule. Both listings can also be reached from the research compounds collection.

• Batch testing: every batch of research compounds is assayed for identity and purity by Janoshik Analytical, an independent laboratory. The Certificate of Analysis is sent on request: email the batch number printed on the label with the product name. See the lab testing page.

A batch assay reports identity and purity for that batch. It says nothing about whether any use of the compound is appropriate or lawful.

Frequently asked questions

What is GW-501516, in one paragraph?

It is a synthetic PPAR-delta agonist, developed by GlaxoSmithKline and Ligand as a possible treatment for dyslipidaemia, as described in a 2006 review (Pelton, 2006). One case report uses the name "GW1516 (cardarine)" for the same compound (Kintz et al., 2021).

Why was GW-501516 discontinued?

According to a peer-reviewed review, GSK discontinued further research because rodent studies linked the drug to widespread tumour development (Mackenzie and Lione, 2013). The original reports are not on PubMed.

Has GW-501516 been tested in humans?

Yes, according to the published GW-501516 research. Four published trial reports cited here describe volunteers or patients with treatment periods of 2 to 12 weeks and measured lipids, liver fat, insulin and other laboratory markers, including gene expression (see the evidence section for what they did not report).

Does GW-501516 cause cancer?

The human trial reports do not describe cancer outcomes, so they cannot answer this. A peer-reviewed review reports that the developer stopped research after rodent studies linked the drug to widespread tumour development (Mackenzie and Lione, 2013). Mouse models with an Apc mutation showed enhanced colorectal tumour development (Liu et al., 2019), and a pancreatic model with mutant KRAS showed accelerated progression of precancerous lesions (Liu et al., 2022), while some cell-line and mouse studies found no such effect (Hollingshead et al., 2007). The question remains open.

Is GW-501516 a SARM?

No. SARMs act on the androgen receptor, while GW-501516 acts on PPAR-delta. The two classes are sometimes found together in the literature: one case report described both in the same patient (Kintz et al., 2021) and a European laboratory survey covered SARMs and metabolic modulators together (Barrios et al., 2025).

Is it prohibited in sport?

WADA added it to the Prohibited List in 2009 (Pokrywka et al., 2014). Check the current official list directly, because UKAD says the list is updated every year from 1 January and that athletes are strictly liable for what is in their system (UKAD).

How do anti-doping laboratories test for it?

Thevis and colleagues characterised two major urinary metabolites of GW1516 and reported a urine screening method with a detection limit of 0.1 ng/ml (Thevis et al., 2010). Sobolevsky and colleagues characterised the urinary metabolites of GW1516 and the related compound GW0742 for detection in human urine (Sobolevsky et al., 2012). In that controlled excretion study a metabolite was reported to remain detectable for weeks after a single exposure, which is a finding about laboratory detection methods and not guidance for any testing situation. None of the abstracts in the sources reports a half-life for GW-501516, so none is given here. UKAD applies strict liability to athletes (UKAD).

Why does "research use only" matter?

It is wording chosen by the seller. Whether a product is a medicinal product is a separate legal question, discussed under Legal and anti-doping status in brief. In its SARMs guidance UKAD notes that sellers often add such wording to circumvent restrictions and that products labelled this way are not regulated for safety or composition (UKAD). That page does not name cardarine, so it is used here as an analogy only. The wording does not make any product suitable for people.

What side effects of cardarine are reported in the research?

In the sources, the adverse findings are a rodent tumour signal reported second-hand (Mackenzie and Lione, 2013), tumour promotion in Apc-mutant and KRAS-mutant models (Wang et al., 2006; Liu et al., 2019; Liu et al., 2022), and two case reports, one of raised liver enzymes with rhabdomyolysis and one of gynaecomastia with hypogonadism, each after exposure to several substances at once (Kintz et al., 2021; Chong et al., 2024). The four human trial abstracts report no adverse-event data. The full list, with the counter-evidence, is in the side-effects table.

What results did the cardarine studies and human trials report?

In cultured human muscle cells it activated AMPK and increased fatty acid transport, fatty acid oxidation and glucose uptake (Krämer et al., 2007). In short randomised trials it changed blood lipids in the volunteers and patients studied (Olson et al., 2012). In trained and untrained Kunming mice it enhanced running endurance (Chen et al., 2015); no study in the sources measured endurance, strength or body composition in a person. The tumour promotion seen in Apc-mutant mice (Liu et al., 2019) belongs to the same answer. In the four human trial reports only laboratory markers were measured, in 13 to 268 participants with treatment periods of 2 to 12 weeks; none reported adverse events in the abstract and none describes a cancer outcome. The figures are grouped by outcome under Cardarine results.

Cardarine vs SR9009 (stenabolic): what is the difference?

Different receptors and different evidence. GW-501516 activates PPAR-delta and has four short human lipid trials plus a rodent cancer signal; SR9009 activates the REV-ERB clock receptors, has no clinical trial and has two liver-injury case reports, as summarised in our SR9009 article. Both are prohibited in sport. There is no head-to-head study of the two; see the comparison section.

Is Cardarine legal in the UK?

Regulation 46 of the Human Medicines Regulations 2012 bars the sale or supply of a medicinal product that has no authorisation, and whether the definition in regulation 2 captures cardarine is a question of legal judgement rather than something a regulator has stated in the sources used here. The legal position, the anti-doping rules and what the research-use-only label does and does not change are examined at length in our companion guide, Is cardarine legal in the UK? This is general information, not legal advice.

Does Free Muscle publish instructions or doses for Cardarine?

No. Free Muscle sells GW-501516 as a research compound for laboratory research use only, to customers aged 18 or over; it is not a medicine or food supplement and is not for human consumption. Neither listing publishes dosing instructions, and this article gives no dose, schedule or protocol; the strengths in the product names describe the capsules, not an amount for anyone to use. The protocols used in the published studies are set out in the papers themselves, which are linked under Sources, and they describe trial and laboratory conditions, not anything intended for a reader.

Related reading

• Short overview: Cardarine (GW-501516): why development stopped and what the data show

Sources

1. Sprecher DL, Massien C, Pearce G, et al. Triglyceride:high-density lipoprotein cholesterol effects in healthy subjects administered a peroxisome proliferator activated receptor delta agonist. Arterioscler Thromb Vasc Biol. 2007;27(2):359-65. doi PMID 17110604
2. Risérus U, Sprecher D, Johnson T, et al. Activation of peroxisome proliferator-activated receptor (PPAR)delta promotes reversal of multiple metabolic abnormalities, reduces oxidative stress, and increases fatty acid oxidation in moderately obese men. Diabetes. 2008;57(2):332-9. doi PMID 18024853
3. Ooi EM, Watts GF, Sprecher DL, Chan DC, Barrett PH. Mechanism of action of a peroxisome proliferator-activated receptor (PPAR)-delta agonist on lipoprotein metabolism in dyslipidemic subjects with central obesity. J Clin Endocrinol Metab. 2011;96(10):E1568-76. doi PMID 21816786
4. Olson EJ, Pearce GL, Jones NP, Sprecher DL. Lipid effects of peroxisome proliferator-activated receptor-delta agonist GW501516 in subjects with low high-density lipoprotein cholesterol: characteristics of metabolic syndrome. Arterioscler Thromb Vasc Biol. 2012;32(9):2289-94. doi PMID 22814748
5. Krämer DK, Al-Khalili L, Guigas B, et al. Role of AMP kinase and PPARdelta in the regulation of lipid and glucose metabolism in human skeletal muscle. J Biol Chem. 2007;282(27):19313-20. doi PMID 17500064
6. Narkar VA, Downes M, Yu RT, et al. AMPK and PPARdelta agonists are exercise mimetics. Cell. 2008;134(3):405-15. doi PMID 18674809
7. Chen W, Gao R, Xie X, et al. A metabolomic study of the PPARdelta agonist GW501516 for enhancing running endurance in Kunming mice. Sci Rep. 2015;5:9884. doi PMID 25943561
8. Pelton P. GW-501516 GlaxoSmithKline/Ligand. Curr Opin Investig Drugs. 2006;7(4):360-70. PMID 16625823
9. Billin AN. PPAR-beta/delta agonists for Type 2 diabetes and dyslipidemia: an adopted orphan still looking for a home. Expert Opin Investig Drugs. 2008;17(10):1465-71. doi PMID 18808307
10. Mackenzie LS, Lione L. Harnessing the benefits of PPARbeta/delta agonists. Life Sci. 2013;93(25-26):963-7. doi PMID 24184294
11. Wang D, Wang H, Guo Y, et al. Crosstalk between peroxisome proliferator-activated receptor delta and VEGF stimulates cancer progression. Proc Natl Acad Sci U S A. 2006;103(50):19069-74. doi PMID 17148604
12. Liu Y, Deguchi Y, Tian R, et al. Pleiotropic effects of PPARD accelerate colorectal tumorigenesis, progression, and invasion. Cancer Res. 2019;79(5):954-969. doi PMID 30679176
13. Liu Y, Deguchi Y, Wei D, et al. Rapid acceleration of KRAS-mutant pancreatic carcinogenesis via remodeling of tumor immune microenvironment by PPARdelta. Nat Commun. 2022;13(1):2665. doi PMID 35562376
14. Marin HE, Peraza MA, Billin AN, et al. Ligand activation of peroxisome proliferator-activated receptor beta inhibits colon carcinogenesis. Cancer Res. 2006;66(8):4394-401. doi PMID 16618765
15. Hollingshead HE, Killins RL, Borland MG, et al. Peroxisome proliferator-activated receptor-beta/delta (PPARbeta/delta) ligands do not potentiate growth of human cancer cell lines. Carcinogenesis. 2007;28(12):2641-9. doi PMID 17693664
16. Kintz P, Gheddar L, Paradis C, et al. Peroxisome proliferator-activated receptor delta agonist (PPAR-delta) and selective androgen receptor modulator (SARM) abuse: clinical, analytical and biological data in a case involving a poisonous combination of GW1516 (cardarine) and MK2866 (ostarine). Toxics. 2021;9(10). doi PMID 34678947
17. Chong S, Woolnough CA, Koyyalamudi SR, Perera NJ. Reversible gynecomastia and hypogonadism due to usage of commercial performance-enhancing supplement use. JCEM Case Rep. 2024;2(8):luae148. doi PMID 39145153
18. Sobolevsky T, Dikunets M, Sukhanova I, Virus E, Rodchenkov G. Detection of PPARdelta agonists GW1516 and GW0742 and their metabolites in human urine. Drug Test Anal. 2012;4(10):754-60. doi PMID 22977012
19. Thevis M, Möller I, Thomas A, et al. Characterization of two major urinary metabolites of the PPARdelta-agonist GW1516 and implementation of the drug in routine doping controls. Anal Bioanal Chem. 2010;396(7):2479-91. doi PMID 19946680
20. Pokrywka A, Cholbinski P, Kaliszewski P, et al. Metabolic modulators of the exercise response: doping control analysis of an agonist of the peroxisome proliferator-activated receptor delta (GW501516) and 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR). J Physiol Pharmacol. 2014;65(4):469-76. PMID 25179079
21. Barrios MM, Deconinck E, Vanhee C, et al. SARMs, metabolic modulators and growth hormone secretagogues in suspected illegal medicines, bought as sport performance enhancers: a retro- and prospective study within the GEON. Drug Test Anal. 2025;17(10):2078-2085. doi PMID 40551438
22. Miller CP, Shomali M, Lyttle CR, et al. Design, synthesis, and preclinical characterization of the selective androgen receptor modulator (SARM) RAD140. ACS Med Chem Lett. 2011;2(2):124-9. doi PMID 24900290 (used in the comparison section only)
23. LoRusso P, Hamilton E, Ma C, et al. A first-in-human phase 1 study of a novel selective androgen receptor modulator (SARM), RAD140, in ER+/HER2- metastatic breast cancer. Clin Breast Cancer. 2022;22(1):67-77. doi PMID 34565686 (used in the comparison section only)
24. Thevis M, Schänzer W. Detection of SARMs in doping control analysis. Mol Cell Endocrinol. 2018;464:34-45. doi PMID 28137616 (used in the comparison section only)
25. Wagener F, Euler L, Görgens C, et al. Human in vivo metabolism and elimination behavior of micro-dosed selective androgen receptor modulator RAD140 for doping control purposes. Metabolites. 2022;12(7). doi PMID 35888790 (used in the comparison section only)
26. Human Medicines Regulations 2012, regulations 46 and 2. legislation.gov.uk.
27. UK Anti-Doping. About SARMs and What's banned in sport: the Prohibited List.