SARMs vs. Classic Anabolic Androgenic Steroids: Molecular, Pharmacokinetic and Safety Differences: A Narrative Review

Abstract

Androgens regulate skeletal muscle, bone, erythropoiesis, and male reproductive function via the androgen receptor (AR), a ligand-dependent transcription factor. Pharmacologic modulation of AR has been pursued for clinical and non-medical purposes. Anabolic androgenic steroids (AAS), synthetic testosterone derivatives, act as full AR agonists, broadly activating multiple tissues. While effective in promoting muscle growth and strength, AAS cause well-known adverse effects, including hypothalamic–pituitary–gonadal (HPG) axis suppression, dyslipidemia, hepatotoxicity, cardiovascular disease, tendon injury, and neuropsychiatric disturbances. Selective androgen receptor modulators (SARMs) aim to stimulate AR in muscle and bone while minimizing androgenic effects in prostate and skin. They induce ligand-specific AR conformations, altering coactivator and corepressor recruitment, and avoiding metabolism by 5α-reductase or aromatase. Preclinical studies show favorable anabolic-to-androgenic ratios, but clinical translation is limited. Early human trials report modest lean mass gains, variable functional outcomes, and dose-dependent testosterone suppression. Emerging evidence also suggests cardiotoxicity, tendon injury, and liver toxicity, though long-term effects are unclear. Pharmacokinetically, SARMs have predictable oral absorption and moderate half-lives, enabling once-daily dosing, unlike AAS. This review compares AAS and SARMs in molecular mechanisms, pharmacokinetics, and safety. While SARMs offer partial tissue selectivity and reduced adverse effects, risks remain, and long-term safety is uncertain. Regulatory oversight is limited, and non-medical use is rising. Preclinical and clinical studies are needed to clarify whether SARMs can separate anabolic benefits from androgenic toxicity and inform safe clinical application.

1. Introduction

Androgens are key regulators of multiple physiological systems, including skeletal muscle development, bone integrity, red blood cell production, and male reproductive function. These effects are mediated by the androgen receptor (AR), a ligand-activated nuclear transcription factor expressed widely across tissues [1,2]. Testosterone, along with its more potent derivative dihydrotestosterone (DHT), orchestrates these processes by activating AR-dependent gene transcription, highlighting the receptor’s central role in systemic androgen signaling [3]. Due to these diverse effects, pharmacological modulation of androgen pathways has been explored extensively, both for clinical indications and non-medical enhancement of performance [4].

It is important to distinguish between the androgenic and anabolic effects of AR activation. Androgenic effects involve the development and maintenance of male sexual characteristics, including prostate growth, facial and body hair, and voice deepening. Anabolic effects primarily promote muscle mass increase, bone density, and red blood cell production [5]. This distinction is crucial to understanding the therapeutic potential and adverse effect profiles of anabolic androgenic steroids (AAS) and selective androgen receptor modulators (SARMs).

AAS, which are synthetic derivatives of testosterone, were developed to stimulate muscle growth and strength gains. Although chemical modifications were intended to limit unwanted androgenic effects, AAS typically continue to activate AR in a non-selective manner [6]. Their use has been associated with a broad spectrum of adverse outcomes, including suppression of endogenous gonadotropins, alterations in lipid profiles, liver toxicity, and increased cardiovascular risk [7]. Clinical and epidemiological studies consistently demonstrate both the anabolic effectiveness and long-term health risks of these compounds.

SARMs were introduced as a strategy to overcome the limitations of traditional AAS. By preferentially binding and activating the AR in a tissue-specific manner, SARMs are designed to promote anabolic activity in muscle and bone while reducing androgenic stimulation in organs such as the prostate and skin [8]. Preclinical investigations suggest several SARMs exhibit favorable anabolic-to-androgenic activity ratios, supporting their potential use in conditions such as sarcopenia, cachexia, osteoporosis, and hypogonadism [9].

Despite these promising properties, clinical translation of SARMs has been limited. No SARM has received regulatory approval for long-term therapeutic application, and emerging human data indicate that these compounds can still produce endocrine suppression, metabolic disturbances, or organ-specific toxicity [10]. Moreover, widespread non-medical use has outpaced the accumulation of robust safety data, complicating risk assessment and public health guidance [11]. It remains unclear whether the apparent tissue selectivity observed in humans reflects true receptor-specific activity or is instead due to reduced intrinsic androgenic potency or pharmacokinetic differences [10].

This review aims to provide a critical comparison between SARMs and classic AAS, focusing on molecular mechanisms of AR activation, pharmacokinetic profiles, and safety considerations. By integrating mechanistic, preclinical, and clinical evidence, this manuscript seeks to delineate similarities and differences between these androgenic agents and identify unresolved questions relevant to both clinical development and recreational use.

2. Materials and Methods

This narrative review was conducted to compare the molecular mechanisms, pharmacokinetics, and safety profiles of SARMs and AAS. The objective was to integrate evidence from biological mechanisms, pre-human experimental studies, and clinical research to construct a structured comparative framework.

A structured literature search was performed across PubMed, Google Scholar, and Elsevier’s ScienceDirect databases. The search strategy combined keywords and MeSH terms including “selective androgen receptor modulators,” “SARMs,” “anabolic androgenic steroids,” “AAS,” “androgen receptor mechanism of action,” “tissue selectivity,” “pharmacokinetics,” “side effects,” “endocrine suppression,” “cardiovascular risk,” and “hepatotoxicity.” No restrictions were applied regarding publication date to ensure inclusion of both foundational mechanistic studies and recent translational research.

Articles were considered eligible if they were published in English and addressed molecular, pharmacological, translational, or clinical aspects of SARMs or AAS. Both preclinical and human studies were included, encompassing randomized controlled trials, observational studies, mechanistic experiments, toxicology investigations, and relevant review articles. Preclinical research was primarily incorporated to clarify mechanistic pathways or to provide context in areas where human data remain limited.

Titles and abstracts were initially screened for relevance, followed by full-text review when necessary. Study selection and data extraction were performed by the author. Priority was given to peer-reviewed studies with clearly described methodology, appropriate controls, and clinically relevant endpoints. Greater interpretive weight was placed on controlled human trials and well-characterized epidemiological evidence compared with single case reports or uncontrolled experiments.

Findings were synthesized descriptively and organized thematically, focusing on androgen receptor activation mechanisms, pharmacokinetic characteristics, and safety outcomes. In line with the narrative review design, no formal risk-of-bias assessment or quantitative meta-analysis was performed. Instead, this approach sought to integrate mechanistic, preclinical, and clinical data to identify areas of agreement, divergence, and remaining uncertainty.

3. Molecular Mechanisms of Androgen Receptor Activation: SARMs Versus Classic Anabolic Androgenic Steroids

3.1. Androgen Receptor Structure and Signaling

The AR is a ligand-activated transcription factor in the nuclear receptor superfamily, comprising an N-terminal transactivation domain, a DNA-binding domain, a hinge region, and a C-terminal ligand-binding domain [12]. Ligand binding triggers conformational changes that lead to dissociation from heat shock proteins, receptor dimerization, nuclear translocation, and binding to androgen response elements in target genes [13]. AR signaling primarily occurs via genomic transcriptional regulation, though non-genomic pathways via membrane-associated interactions exist [14]. Transcriptional outcomes depend on ligand structure, receptor conformation, and tissue-specific coactivator/corepressor balance.

3.2. Molecular Action of Classic AAS

Classic AAS are full AR agonists that strongly recruit coactivators across androgen-responsive tissues, producing broad, non-selective activation in skeletal muscle, prostate, skin, liver, and CNS. Representative compounds such as testosterone enanthate, nandrolone, stanozolol and oxandrolone exhibit these properties, with varying degrees of anabolic potency and androgenic activity depending on structural modification. Many AAS undergo metabolic conversion: 5α-reduction to DHT amplifies androgenic effects in the prostate and skin, while aromatization to estradiol benefits bone but may cause gynecomastia or fluid retention [6]. It should be noted that aromatization of androgens produces several other estrogens like estrone and to a lesser extent, estriol, which can also contribute to both therapeutic and adverse effects. Structural modifications reduce but do not eliminate androgenic adverse effects, so anabolic and androgenic actions remain mechanistically linked (

Table 1. Molecular mechanisms of androgen receptor activation (AAS vs. SARMs).

Feature	Classic AAS	SARMs
Chemical structure	Steroidal, testosterone-derived	Non-steroidal
AR agonism	Full agonists	Partial or tissue-biased agonists
AR conformational effect	Strong, canonical AR activation	Ligand-specific, altered AR conformations
Coregulator recruitment	Broad coactivator recruitment across tissues	Differential coactivator/corepressor recruitment
Tissue selectivity	Largely non-selective	Relative, functional selectivity (↑ activity in muscle/bone, ↓ in the prostate)
5α-reductase metabolism	Yes, into DHT	No
Aromatization	Yes, into estrogens	No
Anabolic vs. androgenic separation	Mechanistically inseparable	Partially dissociable, context-dependent

Abbreviations: AAS—Anabolic androgenic steroids; SARMs—Selective androgen receptor modulators; DHT—Dihydrotestosterone; AR—Androgen receptor. Symbols: ↑—Increased activity or effect; ↓—Decreased activity or effect.

).

3.3. Molecular Action of SARMs and Mechanistic Basis of Tissue Selectivity

SARMs are primarily non-steroidal AR ligands that exploit ligand-dependent receptor signaling differences. Prototypical SARMs such as ostarine (enobosarm, MK-2866), ligandrol (LGD-4033), and testolone (RAD-140) have been extensively characterized in preclinical and early clinical studies. They bind the AR ligand-binding domain with high affinity but induce conformations distinct from testosterone or DHT [15]. Preclinical data suggest SARMs act as partial or tissue-selective agonists, promoting transcription in skeletal muscle and bone while eliciting weaker activation in prostate and other androgen-sensitive tissues [16] (Table 1).

Tissue selectivity arises from receptor conformations favoring recruitment of anabolic coactivators, limited corepressor recruitment, local coregulator expression, and pharmacokinetic factors such as oral bioavailability and tissue penetration [15]. Unlike AAS, SARMs are not substrates for 5α-reductase or aromatase, avoiding metabolic amplification of androgenic effects (Table 1) [17].

4. Translational Evidence and Limitations of Tissue Selectivity

4.1. Clinical Evidence for AAS

Consistent with their full agonist activity, classic AAS produce widespread effects across multiple tissues, which underlies both their anabolic efficacy and side effects. They have been studied in clinical contexts, particularly in chronic disease and wasting conditions. Clinical reviews and trials have evaluated agents such as testosterone, nandrolone, and oxandrolone for their anabolic effects in patients with muscle catabolism associated with HIV infection, chronic obstructive pulmonary disease, severe burns, renal or hepatic failure, and other chronic illnesses [4]. In these settings, AAS administration has been associated with increases in lean body mass and improvements in nitrogen balance compared with baseline. However, their use is accompanied by well-recognized androgenic and metabolic adverse effects, along with elevated cardiovascular risk, which limit broader therapeutic application [18]. Because anabolic efficacy depends on systemic AR activation, dissociation of anabolic benefits from androgenic risk is not achievable with conventional steroidal ligands [19].

4.2. Clinical Evidence for SARMs

Early phase human trials of SARMs such as ostarine and ligandrol generally demonstrate modest increases in lean body mass. However, the magnitude and clinical relevance of these effects remain under active evaluation. Increases in lean mass observed in several studies do not consistently translate into proportional improvements in strength, physical performance, or long-term musculoskeletal outcomes, highlighting the distinction between surrogate and functional endpoints [20,21]. For example, placebo-controlled Phase II studies of ostarine in cancer-associated muscle wasting reported dose-dependent increases in lean mass, although improvements in functional endpoints such as stair-climb power were inconsistent [22]. Phase I clinical trial of ligandrol in healthy young men and older adults showed dose-dependent gains in lean body mass, while functional outcomes were variable [23]. Recently (in 2025), a clinical trial (NCT 06282458) evaluating the effect of ostarine on preventing muscle mass loss in patients receiving semaglutide was completed, although the results have not yet been reported [24]. In another recent clinical trial (NCT05573126), researchers are evaluating the safety and efficacy of the SARM RAD-140 (testolone, vosilasarm, EP0062) in advanced or metastatic AR+/ER+/HER2− breast cancer, and the study is still recruiting participants [25].

Moreover, clinical responses appear context-dependent. Age, baseline hormonal milieu, degree of muscle wasting, and other individual factors influence responsiveness to AR modulation, suggesting that tissue selectivity arises from an interaction between ligand, receptor, and host physiology rather than an intrinsic property of the drug alone [26].

Finally, most available trials rely on short-term surrogate endpoints rather than durable clinical outcomes such as fracture prevention, disability reduction, or disease-specific morbidity. Consequently, while mechanistic models predict preferential anabolic signaling, the extent to which this translates into meaningful long-term therapeutic benefit in diverse populations remains uncertain [16]. Addressing these gaps will require integrated trials combining pharmacodynamic biomarkers, mechanistic assays, and meaningful clinical endpoints to determine whether a true separation of anabolic benefit from androgenic risk is achievable [10].

5. Pharmacokinetic Differences

5.1. Pharmacokinetics of Classic AAS

The pharmacokinetic profile of classic AAS varies with molecular structure and administration route. Oral 17α-alkylated AAS (e.g., Methandrostenolone, Oxandrolone, Stanozolol) avoid extensive first-pass hepatic metabolism, allowing systemic bioavailability but producing relatively short elimination half-lives (several hours) and pronounced peak–trough fluctuations [5]. Transdermal testosterone therapies, such as patches and gels, deliver the hormone directly through the skin into the systemic circulation, bypassing hepatic first-pass metabolism (e.g., AndroGel, Androderm). This route provides relatively steady daily plasma concentrations and reduces peak–trough variability compared with many oral formulations [27] (

Table 2. Pharmacokinetic characteristics of classic AAS and SARMs.

Parameter	Classic AAS	SARMs
Route	Oral (17α-alkylated), intramuscular (esterified), transdermal (gel or patch)	Oral (main); injectable in preclinical studies
Oral bioavailability	Oral requires 17α-alkylation; injectables bypass first pass	Designed for reliable oral absorption
Half-life	Short (hours) oral; prolonged, ester-dependent (days–weeks) injectable	Moderate (~12–36 h), supporting once-daily dosing
Hepatic exposure	High for oral; extensive metabolism	Moderate; compound-dependent
Active metabolites	Yes (e.g., DHT, estradiol)	Mainly parent compound; limited active metabolites
Exposure control	Injectable relatively stable; oral shows peak–trough fluctuations; supraphysiologic doses increase variability	Predictable daily exposure; accumulation depends on elimination
Protein binding	High, SHBG/albumin; may saturate at high doses	Moderate; less SHBG-dependent; intercompound variability
Long-term data	Extensive	Limited, especially chronic exposure
Endocrine suppression	Correlates with cumulative systemic exposure	Dose-dependent; long-term effects unclear

Abbreviations: AAS—Anabolic androgenic steroids; SARMs—Selective androgen receptor modulators; DHT—Dihydrotestosterone; SHBG—Sex hormone–binding globulin.

).

Injectable AAS are usually administered as intramuscular, esterified depot formulations in oil, thereby avoiding first-pass hepatic metabolism (e.g., Nandrolone decanoate, Testosterone enanthate, Testosterone cypionate, Testosterone propionate). Ester hydrolysis governs slow systemic release, with longer ester chains prolonging elimination and sustaining plasma concentrations [28]. While injectable preparations provide more stable exposure than oral agents, supraphysiologic dosing particularly outside clinical contexts can still result in variable circulating androgen levels (Table 2).

AAS are highly protein-bound, mainly to sex hormone–binding globulin (SHBG) and albumin [29]. At supraphysiologic concentrations, partial SHBG saturation may disproportionately increase the free biologically active fraction [30]. Hepatic metabolism involves reduction, oxidation, and conjugation prior to renal excretion of metabolites. Variability in systemic exposure reflects differences in hepatic enzyme activity, SHBG concentrations, body composition, and formulation (Table 2) [31].

Endocrine suppression correlates more closely with cumulative systemic exposure reflecting integrated androgen receptor activation than with peak concentration alone, highlighting the importance of total exposure for hypothalamic–pituitary–gonadal axis effects (Table 2) [5].

5.2. Pharmacokinetics of SARMs

Favorable pharmacokinetic characteristics of SARMs particularly predictable oral absorption and elimination profiles, have been reported in early-phase human studies. Clinical studies of ligandrol and related agents, peak plasma concentrations are typically reached within hours of oral dosing [23]. Elimination half-lives generally range from 12 to 36 h, supporting once-daily administration in phase I/II evaluations. For example, ligandrol exhibits a half-life of ~24–36 h in healthy adults [23,32], while ostarine has a similar profile (~24 h) [33]. Under controlled dosing, this allows relatively stable plasma concentrations over 24 h (Table 2).

Most SARMs (e.g., RAD-140) undergo hepatic metabolism, often via cytochrome P450 pathways, though differences exist between compounds [34,35]. Protein binding is moderate and appears less dependent on SHBG than steroidal androgens, with intercompound variability likely [36]. Available clinical data suggest pharmacologic activity is largely mediated by the parent compound. However, detailed metabolite characterization and long-term pharmacokinetic data, especially under chronic or supratherapeutic exposure remain limited (Table 2).

6. Safety Profiles and Adverse Effects

6.1. Well-Characterized Risks of AAS

The safety profile of classic AAS is well documented through decades of clinical observation, epidemiological studies, and investigational research. Many adverse effects arise from sustained, non-selective AR activation combined with supraphysiologic systemic androgen exposure and compound-specific metabolic issues [37].

Endocrine suppression is a central and predictable consequence of AAS use. Exogenous AR agonism suppresses gonadotropin-releasing hormone secretion via negative feedback at the hypothalamic–pituitary level, leading to reduced luteinizing hormone and follicle-stimulating hormone release [38]. This effect is well described with commonly used agents such as testosterone esters (e.g., testosterone enanthate) and nandrolone decanoate. The resulting suppression of endogenous testosterone production is frequently associated with testicular atrophy, impaired spermatogenesis, and infertility. Recovery of the hypothalamic–pituitary–gonadal (HPG) axis following cessation may be prolonged or incomplete, particularly after long-term or high-dose exposure (

Table 3. Comparative safety profiles of classic AAS and SARMs.

System	Classic AAS	SARMs
Endocrine (HPG axis)	Marked LH/FSH suppression; infertility; recovery often prolonged	Dose-dependent suppression; short-term reversible; long-term unknown
Cardiovascular	↓ HDL, ↑ LDL; HTN; thrombosis; cardiomyopathy; ↑ SCD risk	↓ HDL; possible ↑ LDL; long-term CV risk unknown
Hepatic	Oral: cholestasis, tumors. Injectables: lower risk	↑ Transaminases; DILI reported
Musculoskeletal	Tendon rupture; premature epiphyseal closure	Tendon rupture cases; long-term effects unknown
Dermatologic	Acne, alopecia, striae	Limited data
Neuropsychiatric	Mood changes; aggression; depression; dependence	Limited data; possible CNS effects
Prostate	Hypertrophy; ↑ PSA; possible long-term risk	Small PSA changes; long-term unknown
Thyroid	↓ TBG; ↓ total T4/T3;	No consistent significant effect
Glucose metabolism	Insulin resistance; adverse metabolic profile	Limited data; long-term risk unclear
Reversibility	Often prolonged/incomplete	Short-term reversible; long-term unknown
Evidence base	Extensive clinical/epidemiologic data	Short-term trials; emerging reports

Abbreviations: AAS—Anabolic-androgenic steroids; SARMs—Selective androgen receptor modulators; HPG axis—Hypothalamic–pituitary–gonadal axis; LH—Luteinizing hormone; FSH—Follicle-stimulating hormone; HDL/LDL—High/Low-density lipoprotein cholesterol; HTN—High blood pressure (hypertension); SCD—Sudden cardiac death; DILI—Drug-induced liver injury; TBG—Thyroxine-binding globulin; T4/T3—Thyroid hormones thyroxine/triiodothyronine; PSA—Prostate-specific antigen. Symbols: ↑—Increase/elevated; ↓—Decrease/reduced.

) [38].

Cardiometabolic risk represents another major safety concern. AAS consistently induce adverse alterations in lipid profiles, characterized by reductions in high-density lipoprotein cholesterol and increases in low-density lipoprotein cholesterol [39]. These effects are especially pronounced with 17α-alkylated oral AAS such as stanozolol and methandienone (Dianabol). The abovementioned changes are compounded by endothelial dysfunction, increased arterial stiffness, prothrombotic effects, and direct myocardial remodeling [40]. Agents such as trenbolone and high-dose testosterone have been implicated in adverse cardiac remodeling. Together, these mechanisms contribute to an elevated risk of hypertension, atherosclerotic disease, cardiomyopathy, and sudden cardiac death, particularly with chronic use (Table 3) [41].

Hepatic toxicity is most prominently associated with orally administered, 17α-alkylated AAS, including oxymetholone, stanozolol, and methandienone. First-pass hepatic exposure leads to a spectrum of adverse effects ranging from transient transaminase elevations to cholestasis, peliosis hepatis, hepatic adenomas, and, rarely, hepatocellular carcinoma [42,43]. Injectable AAS largely avoid first-pass metabolism but are not entirely devoid of hepatic risk, particularly with prolonged exposure.

AAS misuse leads to musculoskeletal complications such as tendon rupture, altered muscle-tendon stiffness, and premature epiphyseal closure in adolescents, increasing the risk of long-term joint and growth issues. These effects have been reported in association with potent anabolic agents such as trenbolone and nandrolone. Chronic AAS abuse may also disrupt normal bone remodeling, potentially reducing bone density and increasing fracture risk over time (Table 3) [44,45].

Psychiatric and behavioral effects are also well recognized. AAS use has been linked to mood disturbances, irritability, aggression, anxiety, and depressive symptoms, as well as dependence and withdrawal syndromes [46]. Such effects are frequently reported with high-dose testosterone, trenbolone, and fluoxymesterone, and likely reflect both direct AR-mediated actions in the central nervous system and indirect psychosocial factors related to use patterns and body image concerns (Table 3) [46].

AAS use can alter thyroid function and glucose metabolism. Androgens such as testosterone and nandrolone reduce hepatic thyroid-binding globulin (TBG), lowering total T4 and T3 levels [47]. Effects on glucose homeostasis depend on dose and duration. Although increased lean mass may improve glucose uptake, supraphysiologic androgen exposure, particularly with agents including oxymetholone and high-dose testosterone, is associated with reduced insulin sensitivity, especially with dyslipidemia and visceral adiposity. Impaired insulin signaling and increased hepatic glucose output may promote insulin resistance and elevate cardiometabolic risk with chronic AAS use (Table 3) [18].

AAS misuse is linked to a range of cutaneous manifestations, including acne vulgaris and seborrhea due to sebaceous gland stimulation, as well as acne fulminans in severe cases [48]. These effects are commonly associated with androgenic compounds such as testosterone, trenbolone, and stanozolol. Users also show a higher incidence of acne and bacterial skin infections compared with controls [49]. Histopathologic studies have demonstrated enlarged sebaceous units, oil-rich skin, striae distensae, hirsutism, and androgenetic alopecia, especially with dihydrotestosterone-derived AAS, including stanozolol and drostanolone, in steroid users (Table 3) [50].

6.2. Emerging Safety Data on SARMs

In contrast to AAS, the safety profile of SARMs remains incompletely characterized, with most available data derived from short-duration, early-phase clinical trials and post-marketing reports [11]. While SARMs were designed to improve the anabolic-to-androgenic ratio, accumulating evidence indicates that many adverse effects observed with AAS are attenuated rather than eliminated. Cases involving mainly the compounds ostarine and ligandrol have been reported in both controlled clinical trials and real-world post-marketing use, including case reports and adverse event reports.

Testosterone suppression has emerged as a consistent finding across multiple short-term human and animal studies. Despite lacking conversion to dihydrotestosterone or estradiol, SARMs suppress gonadotropin secretion in a dose-dependent manner, confirming that central AR activation alone is sufficient to inhibit the HPG axis [23]. This effect has been demonstrated in clinical studies with ligandrol [23]. Although suppression is often reversible after discontinuation, recovery kinetics following prolonged or repeated exposure are poorly defined (Table 3).

Alterations in lipid profiles have also been reported. Several SARMs reduce high-density lipoprotein cholesterol and, in some cases, increase low-density lipoprotein cholesterol, raising concerns regarding long-term cardiometabolic risk. Such changes have been described with compounds including ligandrol and GSK-2881078 [51,52]. While the magnitude of these changes is generally smaller than that observed with potent AAS, their clinical relevance over extended exposure remains uncertain (Table 3).

Beyond metabolic alterations, emerging preclinical data suggest potential tissue-specific toxicity. Rodent studies have reported cardiotoxic effects following ostarine exposure, including evidence of myocardial remodeling and increased fibrotic markers, with possible sex-specific differences [53]. Additional experimental work has demonstrated increased myocardial collagen deposition and cardiac hypertrophy after several weeks of administration of ostarine [54]. While the translational relevance of these findings remains uncertain, they raise concerns regarding potential long-term cardiac remodeling with chronic SARM exposure (Table 3).

Musculoskeletal complications have also been described. A case report described asynchronous bilateral Achilles tendon ruptures in a competitive powerlifter using the SARMs [55]. Although AAS are well known to impair tendon integrity, this report suggests that SARMs may similarly predispose users to tendon injury, possibly through disproportionate muscle strength gains relative to tendon adaptation or direct effects on connective tissue structure (Table 3).

Hepatic safety has become an area of increasing concern. Although SARMs are not 17α-alkylated steroids, reports of drug-induced liver injury (DILI) have been documented in both clinical trials and real-world use. Presentations range from asymptomatic transaminase elevations to cholestatic or mixed-pattern liver injury. Cases have been reported in association with agents such as ostarine and RAD-140 [56,57,58]. The mechanisms underlying these effects are unclear and may involve idiosyncratic toxicity, dose-related exposure, or product contamination in non-pharmaceutical preparations (Table 3) [56,57,58].

A critical limitation in interpreting SARM safety is the lack of long-term outcome data. Most trials have durations measured in weeks to months and exclude populations at higher baseline risk. Consequently, the cumulative effects of chronic exposure on cardiovascular, hepatic, endocrine, musculoskeletal, and neuropsychiatric systems remain largely unknown.

6.3. Regulatory and Public Health Considerations

From a regulatory perspective, SARMs remain investigational compounds without approval for clinical use in most jurisdictions. None have received authorization for non-research indications, and their distribution outside clinical trials occurs largely through unregulated channels. Despite this status, SARMs are widely marketed via the internet and often misrepresented as dietary supplements or “research chemicals” [59]. Furthermore, the U.S. Food and Drug Administration has explicitly warned against the use of SARMs in body-building products, highlighting that they are not approved for human consumption and may cause serious health risks, including liver toxicity and cardiovascular injury [60].

Mislabeling and contamination present substantial public health risks. Analytical studies of commercially available SARM products frequently demonstrate inaccurate dosing, substitution with other pharmacologically active agents, or contamination with classic AAS [61]. These practices undermine any presumed safety advantage and complicate attribution of adverse effects.

Risk perception surrounding SARMs is often discordant with available evidence. Marketing narratives commonly emphasize “selectivity” and “non-steroidal” structure as indicators of safety, leading to underestimation of endocrine, hepatic, and cardiometabolic risks. The FDA warning reinforces that claims of safety or “selectivity” are misleading, and that real-world adverse effects have been reported despite such marketing narratives. Current evidence suggests that SARMs retain many class-related toxicities associated with AR activation, albeit often at reduced magnitude and with substantial uncertainty regarding long-term outcomes [56].

7. Conclusions and Future Directions

SARMs were designed to improve the anabolic-to-androgenic ratio of androgenic therapies by inducing ligand-specific AR conformations, differential coregulator recruitment, and avoiding 5α-reduction and aromatization. Mechanistically, this provides a rationale for relative tissue selectivity in muscle and bone compared with classic AAS, which act as full, non-selective AR agonists across multiple tissues. Translational and clinical data, however, indicate that this selectivity is partial rather than absolute. While SARMs exhibit more predictable oral pharmacokinetics and generally attenuated short-term androgenic effects, they still suppress gonadotropins, alter lipid profiles, and have emerging signals of hepatic and cardiac toxicity. In contrast, the safety profile of AAS is well characterized and includes sustained HPG axis suppression, cardiometabolic risk, hepatotoxicity, musculoskeletal injury, dermatologic effects, and neuropsychiatric consequences.

The current evidence base for SARMs is limited by short-duration trials, small sample sizes, and reliance on surrogate endpoints. To acquire missing knowledge on their long-term efficacy and safety, a coordinated approach is required, including adequately powered, longer-duration randomized controlled trials with clinically meaningful outcomes such as strength, fracture prevention, cardiovascular events, and reproductive function. Mechanistic studies integrating pharmacodynamic biomarkers, omics profiling, and direct comparison with physiologic testosterone therapy can clarify whether observed tissue selectivity is intrinsic or context dependent. In addition, longitudinal observational cohorts, rigorous post-marketing pharmacovigilance, and independent verification of product identity and purity are essential to capture rare or delayed adverse events and reduce confounding from contaminated or mislabeled products.

In summary, SARMs represent a refinement rather than a departure from classical androgen biology, offering potential anabolic benefits with partially attenuated androgenic risks. However, their long-term risk–benefit profile remains incompletely defined, and true tissue selectivity in humans is not yet fully validated. Closing these knowledge gaps through integrated clinical, mechanistic, and real-world studies is essential to determine whether SARMs can safely achieve meaningful anabolic outcomes, inform therapeutic development, and guide responsible regulatory and public health strategies.