S23 Benefits: What Studies Report

What this page does and does not say

S23 is an investigational non-steroidal compound described in the fitness and research-chemical world as a selective androgen receptor modulator (SARM). It is not an approved medicine, and materials carrying the label are handled under research-use-only terms rather than as pharmaceuticals. People encounter long lists of S23 "benefits" in forums and product copy. This page does not repeat those lists as facts. Instead, it sets out what a benefit claim has to look like before it can be attributed to published research, and it states plainly where this page's verified citation set contains nothing on S23.

That last point matters more than any individual claim: the verified reference set available to this page includes work such as a preclinical study of Ling-Gui-Zhu-Gan decoction and hepatic steatosis through ALKBH1-mediated DNA N(6)-methyladenosine modification (PMID 41237871) and a network meta-analysis comparing different doses of oral and ocular propranolol for retinopathy of prematurity (PMID 39112827), but it contains no trial, cohort, animal experiment or cell study of S23. Nothing on this page should be read as a finding about S23, because no such finding is cited here. For background on the compound itself, see the S23 overview.

This page is for educational purposes only and is not medical advice; consult a licensed physician before making any health decision.

How benefit claims are graded

Readers comparing claims benefit from knowing which study type produced a number. The tiers below are illustrated with real published examples from unrelated fields, purely to show what each design can and cannot establish.

In-vitro and cell-model work

Cell-based experiments identify mechanisms, not patient outcomes. Researchers characterised helicid as a novel anti-Staphylococcus aureus adjuvant in laboratory work published in Biochimie (PMID 39681184), and a separate group reported that co-targeting MRPS7-23 synergistically enhanced cisplatin efficacy against nasopharyngeal carcinoma growth and metastasis (PMID 41522354). Both are mechanistic contributions; neither establishes a clinical benefit in people, and the same limitation would apply to any mechanistic claim made for a SARM.

Animal and preclinical models

Animal studies add whole-organism physiology but still leave species differences unresolved. The study of Ling-Gui-Zhu-Gan decoction reported improvement in hepatic steatosis via ALKBH1-mediated DNA N(6)-methyladenosine modification in a preclinical model (PMID 41237871). A reader encountering an S23 claim sourced from rodent work should expect the same caveat: an effect observed in animals is a hypothesis about humans, not a demonstration.

Observational clinical data

Observational studies describe what happens in patient populations without assigning treatment. Researchers examining steatosis in cirrhosis reported it as a prognostic marker for liver-related outcomes in metabolic-dysfunction associated steatotic liver disease (PMID 42533609). Prognostic association is not the same as a treatment effect, a distinction that is frequently lost when benefit claims circulate online.

Controlled trials, syntheses and guidelines

Comparative dosing questions are answered by trials and by syntheses of trials. The network meta-analysis of oral and ocular propranolol doses for retinopathy of prematurity pooled trial data specifically to rank dose options (PMID 39112827), and a systematic review produced management guidelines for scrotal inguinal hernias (PMID 38312421). No equivalent trial, synthesis or guideline for S23 appears in this page's verified citation set, which is why no dose, duration or effect size for S23 is stated anywhere on this page.

Outcome domains commonly attached to S23

The domains below are the ones most often named in circulating claims. For each, the honest statement is the same: no study in this page's verified citation set measured it for S23.

Lean mass and body composition

Claims of rapid lean-mass accrual are the most common. Supporting such a claim would require a controlled human trial with a validated body-composition measure and a comparator arm — the kind of pooled, comparator-based design used in the propranolol dose network meta-analysis (PMID 39112827). No such trial of S23 is cited here, and anecdotal reports are not a substitute for one.

Strength and performance

Strength claims face the additional problem of training confounding: untrained participants improve on almost any programme. Separating a compound effect from a training effect requires randomisation and blinding, which is precisely why syntheses such as the scrotal inguinal hernia systematic review and guideline grade underlying study quality before issuing recommendations (PMID 38312421). For S23, this page cites no strength data of any tier.

Fat loss and metabolic endpoints

Metabolic endpoints are measurable but easy to overstate. Published metabolic work in this reference set concerns other interventions entirely — hepatic steatosis improvement in a preclinical model (PMID 41237871) and steatosis as a prognostic marker in cirrhosis (PMID 42533609). Neither involved S23, and neither can be transferred to it.

Bone density

Bone endpoints need imaging over months to years. No bone-density study of S23 is cited on this page, and nothing in the verified reference set addresses skeletal outcomes for any androgen-receptor-targeting compound.

Fertility and hormonal effects

Claims in this area are frequently asserted in both directions. Hormone-related research in the verified set used laboratory models rather than clinical endpoints: researchers applied a proteomic analysis in neuronal models to examine hormonal contraceptives and depression (PMID 40931400). That is a mechanistic study in a different domain; it says nothing about S23, and no fertility or hormonal outcome for S23 is cited here.

Adverse Events: What Studies Report

No adverse-event dataset for S23 appears in this page's verified citation set, so no frequency, severity or reversibility figure for the compound is stated here. What the cited literature does illustrate is how adverse and prognostic outcomes are captured when they are studied properly. Researchers reported steatosis in cirrhosis as a prognostic marker for liver-related outcomes in metabolic-dysfunction associated steatotic liver disease (PMID 42533609), and a preclinical study reported improvement in hepatic steatosis through ALKBH1-mediated DNA N(6)-methyladenosine modification (PMID 41237871). Together these show that liver endpoints require defined populations, defined measurement and follow-up — the elements absent from anecdotal safety reports about SARM-labelled products.

Comparative harm and dose-related tolerability are likewise established through pooled trial evidence, as in the network meta-analysis of different oral and ocular propranolol doses for retinopathy of prematurity (PMID 39112827), and through formal review, as in the systematic review and guidelines for management of scrotal inguinal hernias (PMID 38312421). Nothing comparable exists in this citation set for S23.

Why absence of evidence is reported as absence

A page that lists hypothetical mechanisms in benefit language creates the impression of support that does not exist. Two habits reduce that risk. First, every effect statement should name the study design, population and direction of effect in the same sentence as its citation — as done above for the preclinical steatosis work (PMID 41237871) and the in-vitro adjuvant characterisation (PMID 39681184). Second, guidance is provisional and can be formally withdrawn, as recorded in a published guideline retirement notice (PMID 36241341). Claims about an uninvestigated compound deserve at least the same humility that organisations apply to their own published guidance.

What would change this page

1. Publication of controlled human trials in S23 reporting pre-specified outcomes with comparator arms, in the manner of pooled dose comparisons such as the propranolol network meta-analysis (PMID 39112827).
2. Peer-reviewed preclinical work reporting direction of effect and dose-response in a named species, comparable in reporting standard to the hepatic steatosis model study (PMID 41237871).
3. Systematic review or guideline development that grades the available evidence rather than summarising it, as in the scrotal inguinal hernia guideline process (PMID 38312421).
4. Prospective safety follow-up with defined clinical endpoints, of the kind used in prognostic liver outcome research (PMID 42533609).

Until such work exists and is indexed, the accurate description of S23 benefits is that they are claimed rather than reported. Readers with questions about androgen-receptor-targeting compounds, hormonal health or laboratory monitoring should direct them to a licensed clinician.

References

• Ling-Gui-Zhu-Gan decoction improves hepatic steatosis through ALKBH1-mediated DNA N(6)-methyladenosine modification (Journal of Ethnopharmacology, 2026)
• Helicid: A novel Anti-Staphylococcus aureus adjuvant (Biochimie, 2025)
• Steatosis in Cirrhosis: A Prognostic Marker for Liver-Related Outcomes in Metabolic-Dysfunction Associated Steatotic Liver Disease (Liver International, 2026)
• Systematic Review and Guidelines for Management of Scrotal Inguinal Hernias (Journal of Abdominal Wall Surgery, 2023)
• Co-targeting MRPS7-23 synergistically enhances cisplatin efficacy to suppress nasopharyngeal carcinoma growth and metastasis (International Journal of Biological Sciences, 2026)
• Hormonal Contraceptives and Depression: A Proteomic Analysis Using Neuronal Models (Proteomics Clinical Applications, 2026)
• Comparison of Different Doses of Oral and Ocular Propranolol for Retinopathy of Prematurity: A Network Meta-Analysis (Paediatric Drugs, 2024)
• SOGC Guideline Retirement Notice No. 2 (Journal of Obstetrics and Gynaecology Canada, 2022)