A serum protein–to–bilirubin ratio predicts immune-mediated hepatotoxicity in lung cancer patients treated with immune-checkpoint inhibitors

Abstract

Immune checkpoint inhibitors (ICIs) have transformed outcomes for patients with advanced non-small cell lung cancer (NSCLC) but are associated with immune-related adverse events, including hepatotoxicity. Immune-mediated liver injury can necessitate treatment interruption and immunosuppression, potentially compromising oncological benefit. Validated pre-treatment biomarkers to predict hepatic toxicity are currently lacking. We conducted a retrospective cohort study of patients with advanced NSCLC treated with ICIs across multiple tertiary centres in New South Wales, Australia. Patients with normal baseline liver function and no liver metastases were included. Longitudinal liver biochemistry was analysed to characterise the incidence, timing, and severity of hepatotoxicity (elevation of liver transaminases ALT and AST above the normal limit). Pre-treatment clinical and laboratory parameters, including body mass index, neutrophil-to-lymphocyte ratio, and non-invasive fibrosis indices (APRI and FIB-4), were assessed. Exploratory analyses of baseline liver-related laboratory ratios were performed to identify predictive markers. Findings were validated in an independent NSCLC cohort. Among 100 eligible patients, 20% developed elevated liver transaminases following ICI initiation, with 80% occurring within the first three months. Immune-mediated hepatitis (grade ≥ 2) occurred in 6% of patients. Conventional predictors, including BMI, neutrophil-to-lymphocyte ratio, APRI, and FIB-4, were not associated with hepatotoxicity. In contrast, a low baseline serum protein-to-total bilirubin (SP/TB) ratio was significantly associated with subsequent transaminase elevation (AUC 0.81 95% CI: 0.71 to 0.92, p = 0.0001; negative predictive value 98.5%). This association was confirmed in a second cohort (AUC 0.72 95% CI: 0.66 to 0.83, p = 0.0017). Immune-mediated liver injury is a common early toxicity in NSCLC patients receiving ICIs. The baseline SP/TB ratio is a novel, inexpensive biomarker that may enable pre-treatment risk stratification and personalised monitoring strategies. Clinical trial registration The study was registered at clinicaltrials.gov on the 07/Jan/2020 with the ID number NCT04595734.