Cabozantinib (C) versus placebo (P) in patients (pts) with advanced hepatocellular carcinoma (HCC) who have received prior sorafenib: Results from the randomized phase III CELESTIAL trial.

207 Background: C, an inhibitor of MET, VEGFR, and AXL, has previously shown clinical activity in pts with advanced HCC. This phase 3 trial (NCT01908426) evaluated C vs P in previously treated pts with advanced HCC. Methods: In this double-blind, global, phase 3 trial, pts were randomized 2:1 to receive C (60 mg qd) or matched P stratified by disease etiology (HBV, HCV, other), geographic region (Asia, other), and presence of extrahepatic spread and/or macrovascular invasion (EHS/MVI). Eligible pts had pathologic diagnosis of HCC, Child-Pugh score A, ECOG PS ≤1, and must have received prior sorafenib. Pts received up to two lines of prior systemic therapy for HCC and must have progressed following at least one. The primary endpoint was overall survival (OS). Secondary endpoints were investigator-assessed progression-free survival (PFS) and objective response rate (ORR) per RECIST 1.1. The study was designed to detect a hazard ratio (HR) for OS of 0.76 (90% power, 2-sided α = 0.05) at the final analysis with two prespecified interim analyses at 50% and 75% of the planned 621 events. Results: As of 1 Jun 2017, 707 pts were randomized, and 484 deaths had occurred (317 out of 470 for C; 167 out of 237 for P). Baseline characteristics were balanced between the two arms: median age was 64 years, 82% were male, 38% had HBV, 24% had HCV, 25% enrolled in Asia, 78% had EHS, 30% had MVI, 85% had EHS/MVI, and 27% had received two prior systemic therapy regimens for advanced HCC. The study met the primary endpoint at the second planned interim analysis with median OS 10.2 mo for C vs 8.0 mo for P (HR 0.76, 95% CI 0.63-0.92; p = 0.0049). Median PFS was 5.2 mo for C vs 1.9 mo for P (HR 0.44, 95% CI 0.36-0.52; p < 0.001), and ORR was 4% vs 0.4% (p = 0.0086). The most common grade 3/4 adverse events (predominantly grade 3) with higher incidence in the C vs P arm included hand-foot skin reaction (17% vs 0%), hypertension (16% vs 2%), increased aspartate aminotransferase (12% vs 7%), fatigue (10% vs 4%), and diarrhea (10% vs 2%). Conclusion: C significantly improved OS and PFS vs P in previously treated pts with advanced HCC. Adverse events were consistent with the known safety profile of C. Clinical trial information: NCT01908426.