(W-049) Exposure-Response Analysis for Hepatotoxicity of Rivoceranib Monotherapy and in Combination with Camrelizumab in Patients with Various Cancers, Including Advanced Hepatocellular Carcinoma
Objectives: Rivoceranib (rivo) is a vascular endothelial growth factor receptor-2 tyrosine kinase inhibitor. Camrelizumab (cam) is a humanized monoclonal antibody that specifically binds to programmed cell death receptor 1 (PD-1). The combination (combo) of rivo 250 mg once daily (QD) and cam 200 mg once every 2 weeks (Q2W) has shown favorable clinical efficacy for the treatment of patients (pts) with advanced hepatocellular carcinoma (HCC) in a pivotal Phase 3 study (Study CARES-310). Hepatotoxicity was one of the adverse events for rivo+cam combo therapy. The purpose of this study was to assess exposure-response (E-R) relationship for hepatotoxicity and identify significant covariates.
Methods: The E-R relationship of hepatotoxicity was developed using the data pooled from various dose regimens of rivo monotherapy, cam monotherapy, and rivo+cam combo therapy across 11 studies in 1208 pts with various cancer types, including HCC from CARES-310. Exposure metrics were generated from 2 independently developed population PK models for rivo+cam. After an exploratory Kaplan-Meier plot by quartiles of rivo or cam exposures, parametric Time-to-Event analysis and logistic regression methodology were used to characterize the E-R relationship between rivoand/or cam exposure and the incidence of hepatotoxicity. Significant covariates were identified via developing a full model with all covariates followed by a backward elimination using a significance level of p < 0.05 as the inclusion criteria. The final logistic regression model was used to predict the impact of drug exposures and the significant covariates on the probability of hepatotoxicity.
Results: Log-linear values of rivo Cmax prior to the event or end of treatment had the most correlation with the probability of hepatotoxicity. None of cam exposure metrics had statistically significant relationship with hepatotoxicity. The predicted incidence of any grade hepatotoxicity within the first 2 months after start of treatment was 46.2% in reference subjects who were non-HCC Asian pts with normal hepatic function and median rivo Cmax. In contrast, the incidence of any grade hepatotoxicity was predicted to be 80.2% for pts with mild hepatic impairment, 67.5% for pts with HCC, 20.6% for non-Asian pts, 53.3% for pts with >100 ng/mL than median rivo Cmax. In the population of pts who received the combo therapy of rivo 250 mg QD and cam 200 mg Q2W, both rivo and cam exposures had no statistically significant effect on the incidence of hepatoxicity.
Conclusions: Time-to-Event and logistic regression models well characterized the impact of rivo and/or cam exposure and other covariates on the incidence of hepatotoxicity. Baseline hepatic function (mild impairment vs normal function) was the most significant covariate to affect the incidence of hepatotoxicity. There appeared to be no clinically meaningful association between the incidence of hepatotoxicity and drug exposures at the therapeutical dose regimens.
