Risk and impacts of drug-induced liver injury (DILI) due to immune-checkpoint inhibitors (ICI) on overall survival in patients with newly diagnosed cancer: A retrospective cohort study.

Abstract

12049 Background: The risk and impact of drug-induced liver injury (DILI) due to ICI treatment on newly diagnosed cancer patients have not been fully investigated. This study aims to assess the incidence and impact of ICI related DILI on overall survival. Methods: We included patients with newly diagnosed cancer treated with ICIs between 01/2012 and 12/2020 at the Ohio State University Comprehensive Cancer Center. DILI was defined using ICD-9 and ICD-10 codes for abnormal results of liver function, acute hepatitis, drug-induced liver injury, or toxic liver disease. History of Liver Disease (HLD) before ICI infusion was defined using ICD-9 and ICD-10 codes for any liver disease, including primary and secondary liver malignant neoplasms. To examine the causal effects of ICI-related DILI, patients who developed hepatocellular carcinoma or metastatic liver disease were excluded. Kaplan Meier method was used to estimate DILI risk post-ICI infusion in patients with and without HLD. Crude and Adjusted Cox proportional models were used to examine the association between DILI, modeled as time-dependent, and overall survival. Results: A total of 2,816 patients were included in the study, with a median age of 62 years, 58% male, and 90% white non-Hispanics. An estimated 11% of patients developed DILI during follow-up. The median (interquartile range) follow-up time from ICI infusion to DILI was 8.7 (3.0–20.1) months. The overall risk of DILI was 25.3%, which was significantly higher in patients with HLD than in those without HLD (17.8% vs. 30.8%, Log-Rank P <0.001). The median (interquartile range) survival time was 10.4 (3.7–22.7) months. During follow-up, a total of 1390 (49.4%) patients died. DILI was significantly associated with a higher risk of all-cause mortality in all regression models. The crude risk of all-cause mortality was 40% higher in patients with DILI than in those without DILI [HR: 1.40; 95% CI: 1.15–1.70]. In the fully adjusted models, DILI was associated with a 63% higher risk of all-cause mortality [adjusted HR: 1.63; 95% CI: 1.34–1.99]. DILI's effects on the overall survival risk remained significant when we further adjusted for HLD status [HR: 1.57; 95% CI: 1.29–1.93]. Conclusions: The findings from this retrospective cohort study highlight the significant risk and detrimental impact of DILI on overall survival in patients treated with ICIs. Notably, the incidence of DILI was significantly higher among patients with HLD, underscoring the elevated vulnerability of this subgroup. Furthermore, DILI was a significant predictor of increased all-cause mortality regardless of HLD status. These results underscore the need for vigilant monitoring and management of liver function in patients receiving ICIs, particularly in those with pre-existing liver conditions, to mitigate the risk of DILI and improve survival outcomes.