NAC (N-acetylcysteine)
A genuine, life-saving antidote in its own indication — carrying the strongest evidence of anything in the supplement column. Also: an evidence-free habit in the exact role it is most used for here, protecting a steroid-stressed liver.
The short version
- As an acetaminophen antidote NAC is standard of care — the accepted treatment for that overdose and the reference DILI antidote in 2026 reviews. DATA
- It measurably raises glutathione: systematic review of 9 studies — blood and CSF glutathione up, consistently. DATA
- Zero evidence for liver protection alongside anabolic steroids — no trial, no cohort, nothing. DATA (an absence, verified by search)
- Where it was properly tested beyond its antidote role, results are sobering: 2400 mg/day for 24 weeks did nothing for cognition in a randomized trial. DATA
What it actually is
A glutathione prodrug with an antidote pedigree DATA
Cysteine donor → deacetylated in the gut and liver → rate-limiting substrate for glutathione synthesis. That mechanism made it the acetaminophen-overdose antidote (replenishing the GSH that detoxifies NAPQI) decades ago, and 2026 reviews still name it the most accepted treatment in that overdose and in drug-induced liver injury support generally.
The glutathione effect is measured, in humans, in multiple compartments DATA
Systematic review, 9 studies, n=196: NAC supplementation raises reduced and oxidized glutathione in blood and cerebrospinal fluid. This is the one supplement claim in this whole category where the surrogate endpoint has actually been shown to move in humans.
Where the evidence runs out
No AAS-hepatoprotection study exists DATA
Systematic negative: nothing — trial, cohort, or case series — tests NAC as protection against anabolic-steroid liver injury. The practice is a mechanism borrowed from a different disease (acetaminophen) applied to an injury type (cholestatic 17aa) where glutathione replenishment has no demonstrated role.
Properly tested elsewhere, big promises shrink DATA
Randomized, placebo-controlled, n=59, 2400 mg/day for 24 weeks: no cognitive benefit versus placebo (p=0.8). Exercise data: acute-dosing tolerance effects only, with prolonged-use benefits uncertain (Mănescu 2026). The pattern is consistent — real pharmacology, real surrogate effects, and nulls whenever a hard endpoint gets measured blind.
Context worth knowing
Even the antidote role has competition DATA
A 2026 RCT of fomepizole added to NAC after acetaminophen overdose cut the NAPQI-derived adduct burden a further 60–70% — evidence the field is still optimizing NAC's own flagship indication, not a reason to doubt it.
Why people still take it on-cycle — and what that is worth MODEL
Premise (data): NAC is cheap, measurably raises glutathione, and is safe at studied doses. Premise: 17aa liver injury is primarily cholestatic, a phenotype NAC has never been shown to prevent. Logic: as cheap insurance against unmeasured oxidative components it is defensible; as a reason to feel safe about oral steroids it is borrowing certainty that does not exist. The honest sentence is “harmless and unproven here,” not “liver protection.”
What does not exist
No AAS trial · no liver-endpoint RCT in healthy adults · no dose-response for the surrogate DATA
Verified by search (2026-08-31): no NAC study with any anabolic steroid involved, no randomized trial with a liver-injury primary endpoint in healthy adults, and no established dose-response curve for glutathione elevation across supplement-range doses. The 600–2400 mg range in circulation traces to the antidote protocol and the null cognitive trial, not to an optimization study.