Affiliation:
1. Department of Clinical Pharmacology Zhongshan Hospital, Fudan University Shanghai China
2. Department of Biostatistics Zhongshan Hospital, Fudan University Shanghai China
3. Department of Gastroenterology Zhongshan Hospital, Fudan University Shanghai China
Abstract
AimsProgrammed cell death receptor (ligand)‐1 inhibitors (PD‐(L)1), as the preferred immunotherapy, have been widely used in the Chinese mainland and drug‐induced liver injury (DILI) has been reported. The study aimed to investigate the clinical features or risk factors for immunotherapy‐related DILI.MethodsPatients who received PD‐(L)1 inhibitors from January 2020 to July 2021 were retrospectively reviewed. The likelihood of DILI was adjudicated by the Roussel‐Uclaf causality assessment.ResultsA total of 1175 patients were included in the study and 89 patients (7.6%) developed DILI, of which 12 (13.5%) progressed to acute liver failure (ALF) and three (3.4%) died. Among the DILI population, 56 (62.9%) had a cholestatic pattern and exhibited a prolonged treatment course and duration for resolution compared to the hepatocellular and mixed patterns. Hepatocellular carcinoma (HCC), hepatitis B virus (HBV) and abnormal baseline of alkaline phosphatase (ALP) had increased risks of DILI by 2.1‐fold (95% confidence interval [CI], 1.231–3.621), 1.9‐fold [95% CI, 1.123–3.325] and 2.1‐fold [95% CI, 1.317–3.508], respectively. The model for end‐stage liver disease (MELD) score had a c‐statistic of 0.894 (95% CI, 0.778–1.000) with a sensitivity of 67% and a specificity of 95% for poor outcomes. COX analysis showed that the MELD ≥ 18 was predictive of immunotherapy‐related ALF or death.ConclusionsPD‐(L)1 inhibitor‐related liver injury manifests primarily as a cholestatic pattern, on which corticosteroid treatment has minimal effect compared to hepatocellular and mixed patterns. MELD score ≥ 18 at the time of liver injury performed best in the prediction of ALF or death in immune checkpoint inhibitor (ICI)‐related DILI.
Funder
National Natural Science Foundation of China