The risk factors of immune-mediated liver injury caused by immune checkpoint inhibitors.

Abstract

e14587 Background: Despite the fact that the majority of immune-mediated liver injuries caused by immune checkpoint inhibitors (ILICI) clinically present as mild-to-moderate elevations in transaminase levels, severe symptoms of ILICI such as fulminant hepatitis and acute liver failure have been reported. Consequently, the effective identification of risk factors for the occurrence of ILICI will help screen individuals at high risk of ILICI before treatment, thereby assisting clinicians in evaluating the benefit/risk ratio of immunotherapy for cancer patients effectively. This study was to construct a clinical prediction model for ILICI. Methods: A retrospective analysis was performed on the clinical data of 1432 pan-cancers Chinese patients treated with immune checkpoint inhibitors (ICIs) from January 2019 to December 2021. Univariate and multivariate logistic regression analyses were used to identify the independent risk factors for the occurrence of ILICI, and a nomogram model for predicting the risk of ILICI was constructed subsequently. The discrimination, calibration, and clinical utility of the nomogram model were comprehensively evaluated by the receiver operating characteristic curve (ROC), calibration curve, and decision curve analysis (DCA). Results: Multivariate logistic regression analysis showed that the use of imported PD-1 inhibitors (OR, 6.45, 95% CI 3.60-11.57; p < 0.001), baseline aspartate aminotransferase (AST) level ≥ 26.04 U/L (OR, 2.27; 95% CI 1.33-3.87; p = 0.003), and baseline prognostic nutritional index (PNI) < 38.75 (OR, 2.21; 95% CI 1.22-4.02; p = 0.009) were independent risk factors for the development of ILICI. Based on these identified variables, a nomogram model was developed and underwent validation through diverse analytical methods. The results demonstrated that this nomogram model could serve as a reliable and valid predictive tool for ILICI. Conclusions: In this study, we found that the use of imported PD-1 inhibitor, high baseline AST level (AST ≥ 26.04U/L) and low baseline PNI level (PNI < 38.75) were independent risk factors for the occurrence of ILICI. By integrating these three identified factors, an effective and convenient nomogram model was developed for accurate ILICI risk prediction, which is expected to enable early identification of individuals with increased risk of ILICI and early intervention, thereby potentially improving the prognosis of patients receiving ICIs therapy.