Phase 3 (COSMIC-312) study of cabozantinib in combination with atezolizumab vs sorafenib in patients with advanced hepatocellular carcinoma (aHCC) who have not received previous systemic anticancer therapy

Introduction: Cabozantinib inhibits tyrosine kinases involved in tumor growth, angiogenesis, and immune regulation, including MET, VEGFR, and TAM kinases (Tyro3, AXL, MER). Cabozantinib is approved for the treatment of aHCC after prior sorafenib based on improved overall survival (OS) vs placebo in the phase 3 CELESTIAL trial (Abou‐Alfa NEJM 2018). Standard of care for first‐line treatment of aHCC is tyrosine kinase inhibition with sorafenib or lenvatinib, and phase 3 trials of immune checkpoint inhibitors (ICIs) in first‐ and second‐ line aHCC are ongoing. Cabozantinib may promote an immune‐permissive tumor environment, which could enhance response to ICIs. Cabozantinib is being evaluated in combination with the anti‐PD‐L1 antibody atezolizumab in multiple tumor types including HCC in a phase 1 study; and dose, preliminary clinical activity, and safety have been established in aRCC (Agarwal Ann Oncol 2018). Atezolizumab in combination with bevacizumab, an anti‐VEGF antibody, has shown preliminary clinical activity in first‐line aHCC (Pishvaian Ann Oncol 2018). Here, we present the study design of a phase 3 trial of cabozantinib + atezolizumab vs sorafenib in patients with aHCC who have not received prior systemic therapy. Methods: This international, randomized, open‐label phase 3 trial (NCT03755791) is evaluating the efficacy and safety of cabozantinib + atezolizumab vs sorafenib as firstline treatment for aHCC. Eligibility criteria include age ≥18 years, BCLC stage B or C, Child‐Pugh A, ECOG PS 0 or 1, and measurable disease per RECIST 1.1. Patients are randomized 6:3:1 to an experimental arm of cabozantinib (40 mg qd) + atezolizumab (1200 mg infusion q3w), a control arm of sorafenib (400 mg bid), and an exploratory arm of cabozantinib monotherapy (60 mg qd). 640 patients are planned at∼200 sites globally. Randomization is stratified by disease etiology (HBV [with or without HCV], HCV [without HBV], or other), region (Asia, other), and the presence of extrahepatic disease and/or macrovascular invasion (yes, no). OS and progression‐free survival are co‐primary endpoints and objective response rate is a secondary endpoint. Additional endpoints include safety, pharmacokinetics, and correlation of biomarker analyses with clinical outcomes. Enrollment in COSMIC‐312 is ongoing.