219O Safety and preliminary efficacy of nivolumab in patients with advanced hepatocellular carcinoma: interim analysis of the phase 1/2 CheckMate-040 study

Abstract

Background: Median overall survival (OS) for first‐line treatment of advanced hepatocellular carcinoma (HCC) with sorafenib is up to 11 mo and is 7‐8 mo with best supportive care after sorafenib failure. Nivolumab, a fully human IgG4 monoclonal antibody inhibitor of the programmed death‐1 (PD‐1) receptor, was evaluated in a phase 1/2 study in patients with advanced HCC. After multiple ascending doses were evaluated in the dose‐escalation phase, a dose‐expansion phase followed. Interim results are presented. Methods: CheckMate 040 (NCT01658878) enrolled patients with histologically confirmed advanced HCC and Child‐Pugh scores <7 (dose escalation) or <6 (dose expansion). The dose‐escalation phase included patients who progressed on sorafenib or who refused, or were intolerant of sorafenib; patients received nivolumab 0.1‐10 mg/kg Q2 W across three cohorts: uninfected, HCV‐infected, and HBV‐infected. In the dose‐expansion phase, patients received nivolumab 3 mg/kg Q2 W across four cohorts: uninfected sorafenib naive/intolerant, uninfected sorafenib progressors, HCV‐infected, and HBV‐infected. Primary endpoints were safety and tolerability (dose escalation) and objective response rate (ORR) by RECIST v1.1 (dose expansion). Other endpoints included OS, duration of response, and assessment of programmed death‐ligand 1 (PD‐L1) expression. Result: Between November 26, 2012, and March 15, 2016, 48 patients in the dose‐escalation phase and 214 patients in the doseexpansion phase were treated with nivolumab. At baseline, 85 and 70% had Child‐Pugh scores of 5, 77 and 75% had extrahepatic metastases, and 73 and 66% had prior sorafenib treatment in the dose‐ escalation and dose‐expansion phases, respectively. The safety profile in the dose‐expansion phase was similar to that of the dose‐escalation phase. In the dose‐expansion phase, treatment‐related adverse events (TRAEs) occurred in 65% of patients; 18% had grade 3/4 TRAEs. The most common TRAEs were fatigue (21%), pruritus (15%), rash (12%), and diarrhea (9%); the most common grade 3/4 TRAEs were increases in aspartate aminotransferase (4%), alanine aminotransferase (3%), lipase (3%), and amylase (2%). Efficacy data are presented in the Table. Responses occurred regardless of underlying HCC etiology or PD‐L1 expression on tumor cells. Conclusion: Nivolumab was well tolerated in patients with advanced HCC. The ORR and OS rate for the dose‐escalation phase are favorable to historic best supportive care data. Tolerability and efficacy profiles are consistent between the dose‐escalation and doseexpansion phases of this ongoing study. (Table Presented).