A cluster analysis of 466 patients demonstrates that glutathione supplementation could preferentially benefit advanced unstable cirrhosis phenotype rather than stable cirrhosis.

Background: Chronic liver disease is characterized by progressive hepatic glutathione depletion and oxidative stress, yet antioxidant therapies have historically shown disappointing clinical efficacy in unselected populations. This therapeutic paradox may reflect inadequate patient stratification and failure to distinguish between acute reversible oxidative injury and chronic persistent inflammation. We hypothesized that glutathione supplementation may preferentially benefit distinct cirrhosis phenotypes with active systemic inflammation. Methods: We conducted a retrospective cohort study of 466 patients with chronic liver disease who received oral glutathione supplementation (500 mg daily, median 30 days) at a tertiary care center. Primary endpoints were all-cause mortality and change in Model for End-Stage Liver Disease 3.0 (MELD-3) score. We employed conventional statistical methods, survival analysis, multivariable regression, machine learning feature importance, and unsupervised K-means clustering to identify distinct patient phenotypes. Results: Overall mortality was 10.9% over median 417-day follow-up. At the population level, MELD-3 worsened significantly (mean ΔMELD-3 +2.69, p < 0.001), though 33.5% of patients achieved MELD-3 improvement with low mortality (1.9%). Treatment duration showed no dose-response relationship. Counter-intuitively, patients with higher baseline MELD, bilirubin, INR, and C-reactive protein demonstrated better treatment response. K-means clustering identified four phenotypes, with the "Potential High-Risk Responder" cluster (10.9% of cohort, median MELD-3 27.9) showing highest improvement rate (52.9%) and the only mean MELD-3 improvement (ΔMELD-3-0.85), despite highest mortality (23.5%). An "ideal responder profile" comprising younger patients with alcohol-associated liver disease, elevated baseline MELD, preserved albumin, and elevated inflammatory markers achieved 70.5% improvement rates. Conclusion: These findings demonstrate that glutathione supplementation may preferentially benefits patients with advanced, acutely unstable cirrhosis characterized by active systemic inflammation rather than stable compensated disease, challenging indiscriminate antioxidant use and supporting a precision-medicine approach targeting inflammation-rich, treatment-responsive phenotypes in chronic liver disease.