YK-11

Identity, and two structural claims it rules out

Nothing about YK-11's registration is ambiguous. The only thing likely to trip a reader is that the same structure is written out two very different ways. PubChem holds a single record under the name: CID 119058028, CAS registry number 1370003-76-1, molecular formula C25H34O6, molecular weight 430.5, InChIKey KCQHQCDHFVGNMK-PQUNLUOYSA-N. The FDA Global Substance Registration System points at that same CAS number and that same PubChem identifier under UNII Z9748J6B0R, and computes a weight of 430.5348 from the deposited structure. The CAS index name, methyl (17α,20E)-17,20-[(1-methoxyethylidene)bis(oxy)]-3-oxo-19-norpregna-4,20-diene-21-carboxylate, sits in both registries as a deposited synonym. PubChem's own computed IUPAC name renders the identical structure in spiro form, as methyl (2E)-2-[2'-methoxy-2',13-dimethyl-3-oxospiro[dodecahydrocyclopenta[a]phenanthrene-17,5'-1,3-dioxolane]-4'-ylidene]acetate. Read either way, the compound is a four-ring steroid with a ketone at C3 conjugated to a double bond at C4, no methyl at C10, and a spiro orthoester bridging C17 and C20 which carries a methoxy substituent and terminates in a methyl ester.

Two structural claims repeated across secondary pages do not survive those records. YK-11 is almost always introduced as a derivative of dihydrotestosterone. Dihydrotestosterone is an androstane with a fully saturated A-ring and a methyl at C10; the registered YK-11 structure is a 19-norpregnane with a 3-keto A-ring carrying a double bond at C4. The two differ in ring skeleton, in carbon count and in the oxidation state of the A-ring. The second claim is that YK-11 is 17α-methylated, the modification usually invoked to explain hepatotoxicity in orally active androgens. No methyl group sits at C17 in either deposited structure. That position is occupied by an orthoester oxygen. A trade designation for the compound also circulates on vendor product pages; it appears in neither PubChem's twenty-two synonyms for the CID nor the GSRS record, and it is not carried in the identity block here.

Regulatory position is unambiguous. No medicines agency has approved YK-11 for any indication anywhere. The FDA consumer update on selective androgen receptor modulators, content current as of 26 April 2023, states that SARMs are not FDA approved, that although they are often marketed as dietary supplements or sold for research use only they are considered unapproved drugs, and that they cannot be legally marketed in the United States as a dietary supplement or drug. The 2026 WADA Prohibited List names YK-11 under S1.2, Other Anabolic Agents, in a bracketed list of six example SARMs alongside andarine, enobosarm, LGD-4033, RAD140 and S-23. Everything in S1 is prohibited at all times, in and out of competition.

Four cell papers from one laboratory

The compound enters the literature in March 2011, in Biological and Pharmaceutical Bulletin. Kanno and colleagues at Toho University reported that YK11 activated an androgen-response-element luciferase reporter as a partial agonist, accelerated nuclear translocation of the androgen receptor, and did not induce the amino/carboxyl-terminal interaction that full agonists require, while also preventing dihydrotestosterone from inducing it. In MDA-MB 453 breast cancer cells the effect on gene expression relative to a full agonist varied by gene. On that basis the authors proposed the SARM label, which every downstream source has carried since. Concentrations and replicate counts are not in the abstract, and the full text could not be retrieved this session: J-STAGE returned HTTP 500 on repeated requests.

Kanno's 2013 paper is the one the rest of the field rests on. Mouse C2C12 myoblasts were treated with 500 nM YK11, 500 nM dihydrotestosterone or ethanol vehicle in differentiation medium, three replicates per condition. Myosin heavy chain protein rose by day 7 under both compounds. Myf5 and myogenin transcripts on day 4 rose further under YK11 than under dihydrotestosterone, though across a 1 to 500 nM range YK11 needed 100 nM or 500 nM to move them while dihydrotestosterone acted lower. Hydroxyflutamide at 10 µM suppressed the effect. YK11, and not dihydrotestosterone, induced follistatin, and an anti-follistatin antibody reversed the differentiation.

Yatsu and colleagues extended the same in vitro programme in 2018. Treatment with YK11 and treatment with dihydrotestosterone each accelerated proliferation and mineralisation in MC3T3-E1 mouse osteoblasts. Osteoprotegerin and osteocalcin were reported raised in YK11-treated cells against untreated cells; the observations were attenuated by an androgen receptor antagonist, and phosphorylated Akt was increased. Kanno returned to MDA-MB-453 cells in 2022 and found different patterns of receptor recruitment to enhancer regions between the two ligands, with docking suggesting the sterically hindered C17 group displaces helices 11 and 12 and two-hybrid assays suggesting different cofactor recruitment. Four papers, one research group, no animal in any of them.

What has been done in animals, and what has not

Mice inoculated with gram-negative bacteria supply the first appearance in a living organism. Lee and colleagues reported in 2021 that myostatin protein rose in the muscle of septic mice alongside follistatin, NF-κB, myogenin, MyoD, phosphorylated FOXO3a and phosphorylated Smad2, and that YK11 repressed pro-inflammatory cytokines and organ damage markers in blood and in major organs, decreasing mortality. Dose, route and group sizes are absent from the abstract and the full text was not retrievable. The paper's title calls YK11 a myostatin inhibitor. What the experiment measured was cytokines, organ damage markers and survival.

Bone work follows the osteoblast result into an animal. Wang and colleagues reported in 2025 that 0.25 to 4 µM YK11 promoted osteogenic differentiation of bone marrow mesenchymal stem cells, with the effect increasing across that range, and that 2 µM promoted proliferation. The species of those cells is not stated in the abstract, and the publisher full text did not open this session. A YK11-loaded hydrogel at 0.5 and 1 mg/mL promoted repair of cranial defects in rats, which is the part of the study the abstract does place in a species. Inhibiting the androgen receptor suppressed both the differentiation and the repair. The authors implicate BMP2/Smad signalling. Group sizes are not given in the abstract.

Rat neurotoxicology accounts for the remaining in vivo work, and it points the other way. Dahleh and colleagues at the Federal University of Pampa ran a five-week protocol splitting male Wistar rats into control, YK11, swimming exercise and combined groups. The 2023 and 2024 papers both print the dose as 0.35 g/kg, and neither abstract states the route or the group sizes. Hippocampal oxidative stress rose and every mitochondrial function marker was impaired; exercise alone was protective; the combination preserved MnSOD/SOD2 and MTT reduction capacity but reversed nothing else. The 2024 paper adds raised IL-1β and IL-6, lowered IL-10, increased p38 MAPK, an engaged Bax/Bcl-2/cleaved caspase-3 cascade, downregulated BDNF/TrkB/CREB signalling and altered aversive memory consolidation.

Absent from all of it is any measurement of muscle. No published experiment has recorded skeletal muscle mass, lean mass, grip strength, fibre cross-sectional area or body composition in an animal given YK-11. A PubMed search combining YK11 with muscle mass, hypertrophy or body composition returns four records: a supplement-screening method paper, the sepsis study, the osteoblast study and a review. No Hershberger assay, the castrated-rat design that separates anabolic from androgenic activity and is the standard first test for this compound class, has been run on it. A search combining YK11 with prostate, gonadal or testosterone-suppression terms returns nothing at all.

The human record is analytical

ClinicalTrials.gov returns no registered study for YK11 or YK-11, at any phase, under any sponsor. There is no development programme. The human data that exist were generated by anti-doping laboratories, and they exist because the compound turned up in a seized black-market product rather than because anyone set out to study it. Thevis and colleagues characterised its mass spectrometry in 2017, synthesising YK-11 and three stable-isotope-labelled analogues and mapping the dissociation pathways and diagnostic product ions needed to look for it.

Piper and colleagues followed in 2018 with an elimination study using six-fold deuterated YK11. Fourteen deuterated urinary metabolites were detected, unconjugated, glucuronidated and sulfoconjugated. No intact YK11 appeared in any post-administration sample. Unconjugated metabolites disappeared within 24 hours; glucuronidated and sulfated ones remained traceable beyond 48. Two glucuronidated metabolites, 5β-19-nor-pregnane-3α,17β,20-triol and 5β-19-nor-pregnane-3α,17β-diol-20-one, were synthesised in-house, confirmed by NMR and folded into the routine method for anabolic agents. The abstract does not say how many volunteers were dosed, or with how much, and the full text sits behind a paywall that did not open this session. This is the only human administration of YK-11 in the published record.

A 2024 letter records a detection. Sobolevsky and colleagues at the UCLA Olympic Analytical Laboratory published in Drug Testing and Analysis a communication typed by PubMed as a case report and titled as the detection of YK-11 in a doping control sample. No abstract is deposited in PubMed and the publisher returned HTTP 403 to retrieval attempts made for this record, so nothing beyond the title is available here: the sample, the species, the finding and its circumstances are not established by anything this page could read. Set that title against the WADA anti-doping testing figures for 2011 to 2020 as tabulated in Leciejewska's 2024 review: ostarine, ligandrol, RAD-140, andarine, S-23 and S-22 all appear with counts, and YK-11 appears in none of those ten years.

Horses supply what direct measurement exists of the intact molecule in a living body, across two separate oral administration studies run by two laboratories. Harding and colleagues gave two Thoroughbred horses three daily oral doses of 50 mg and analysed urine and plasma. Parent YK-11 and seven phase I metabolites were detected in plasma, largely unconjugated; eleven phase I metabolites appeared in urine, with a mono-O-demethylated breakdown product and di-O-demethylated YK-11 the most abundant. Incubation with equine liver microsomes generated 79 metabolites that barely overlapped the in vivo set. Detection of the parent compound in plasma after oral dosing is the only direct evidence in the published record that intact YK-11 is absorbed in any species.

The second equine study reaches a different matrix. So and colleagues at the Hong Kong Jockey Club Racing Laboratory reported in 2025 that RAD140 and YK-11 were detected in mane hair from horses given the respective drugs orally, YK-11 at 0.2 mg/kg daily for three days in an administration study the same laboratory had run earlier to establish urine and plasma analytes. Segmental analysis indicated that RAD140 was incorporated into hair both internally via the bloodstream and externally through sweat or sebum, whereas YK-11 was incorporated primarily through sweat or sebum. That route makes the hair result a detection finding rather than an absorption measurement. Group sizes are not stated in the abstract.

Adverse-event records, and what they carry

Leciejewska and colleagues searched PubMed between September 2022 and October 2023 and assembled 20 reports of adverse events published since 2020, most of them drug-induced liver injury presenting as cholestatic or hepatocellular injury with jaundice. One tabulated case names YK-11: Lee and colleagues in Military Medicine, a previously healthy active-duty United States Marine with cholestatic liver injury over three months, ALT 148 IU/L, AST 88 IU/L, ALP 151 IU/L and total bilirubin 29.2 mg/dL. The review's extraction table lists the exposure as LGD-4033, RAD-140 and YK-11 together and records that the dose was not specified. The case report abstract itself names no compound. Documented there is an outcome following three compounds, not one.

Spontaneous reports add two more records and no more certainty. The same review queried the FDA's Center for Food Safety and Applied Nutrition Adverse Event Reporting System. The terms YK-11, YK11 and YK 11 returned two reports: a 23-year-old man with hepatic failure logged 16 May 2019, categorised life-threatening with hospitalisation and disability, and a 37-year-old man with cerebrovascular accident and paraesthesia logged 17 December 2020. Neither is a single-product record. The hepatic-failure entry appears identically in the table's LGD-4033 row as well as its YK-11 row; the cerebrovascular entry appears identically under LGD-4033, RAD-140 and YK-11. Each is one report naming more than one product. Spontaneous reports are unverified and carry no denominator.

For the class rather than the compound, the FDA lists increased risk of heart attack or stroke, psychosis and hallucinations, sleep disturbances, sexual dysfunction, liver injury and acute liver failure, infertility, pregnancy miscarriage and testicular shrinkage among the problems associated with SARMs in studies and reports it has received. That is a class statement covering molecules with far larger case literatures than this one. No published case series isolates YK-11 as a single exposure.

The size of the record

Product analysis gives a sense of what circulates under the name. Lee and colleagues at Korea's National Institute of Food and Drug Safety Evaluation validated an LC-MS/MS method for six SARMs and applied it to 60 dietary supplements bought online and directly from international vendors in 2020. SARMs were detected at high concentrations in 20 products advertised as having androgenic properties, andarine at 7.2% in one and GW501516 at 3.49% in another. MK-677 and YK-11 were found in products whose labels did not declare them.

That paper also records a chemical property with practical weight. YK-11 hydrolyses within a few hours, and it did so under the analytical conditions: the protonated ion at m/z 431 was not detected at all, the compound had to be confirmed through fragmentation and a sodiated ion at m/z 453.3, and the acid-hydrolysis product was characterised by NMR. A molecule that decomposes on that timescale complicates any statement about what a preparation contains at the moment of use, and it sits consistently beside Piper's finding that no intact YK11 reached urine.

Twenty PubMed records return for the string YK11. Three of them are not this compound: two are unrelated peptides that happen to share the letters, and one is an unrelated small molecule carrying the same designation. Of the seventeen that are, this page cites sixteen. The one left out is a 2018 review of selective androgen receptor modulator detection in doping control, which surveys the class and reports no primary data on YK-11; it is named here so that the exclusion is recorded rather than silent. Publication types and correction links were checked on every YK-11 record cited: none is typed as a Retracted Publication or an Expression of Concern, and none carries an erratum.

The most recent additions are computational, and there are two rather than one. Dahleh and colleagues docked eight SARMs against the androgen receptor and 5-alpha reductase II in 2025, combining molecular docking, density functional theory and molecular dynamics, and reported that YK11 interacted with both proteins while a different compound in the set returned the lower RMSD values. Mihalev and colleagues docked YK11 and ostarine against five nuclear receptors with AutoDock 4.2 in 2026 and found YK11's predicted binding ordered GR > PR > AR > MR > ER, then concluded that docking energies describe thermodynamic compatibility and should not be read as predictors of tissue selectivity. That caveat is the authors' own. Neither paper measured anything in a cell or an animal.

What is not known

No experiment has measured what this compound does to muscle in a living animal, which is the property it is sold for. There is no pharmacokinetic study in any species: no half-life, no clearance, no volume of distribution, no oral bioavailability figure and no measured plasma concentration in a person. No binding affinity has been published, in the sense of a Ki, an IC50 or an EC50 at the androgen receptor or at any other steroid receptor; the affinity figures in circulation come from docking software, and of the two docking papers on this compound the more recent, from 2026, says explicitly that its energies should not be read as predictors of tissue selectivity. Nothing addresses effects on the hypothalamic-pituitary-gonadal axis, on prostate tissue, on lipids, on haematocrit or on liver enzymes under controlled conditions, in any species. The rodent work that exists concerns the hippocampus and the skull, prints its dose in grams per kilogram, and does not state its route. The equine work establishes metabolite profiles and detection windows, not effects. No dose-response relationship has been established for any endpoint outside cell culture. The compound has never been the subject of a registered clinical trial, is not approved anywhere, and hydrolyses in solution over hours, so material offered under the name has neither an identity standard nor a stability profile behind it.

Questions