Hepatocellular carcinoma (HCC) is a complex disease that is often diagnosed at advanced unresectable stages where, if the liver function allows, systemic therapy is the only option to improve survival. The multikinase inhibitors, sorafenib and lenvatinib, are current standard of care first-line therapy for patients with unresectable HCC, based on modest survival benefits seen with sorafenib versus placebo in phase III trials (SHARP and Asia-Pacific HCC) and the non-inferiority of lenvatinib compared to sorafenib (REFLECT). However, these agents are associated with considerable toxicity that may have an adverse impact on quality of life. Immunotherapy is also being evaluated as first-line therapy based on promising activity and a favorable toxicity profile in the second-line setting. Although recent data from the phase III CheckMate 459 trial showed a trend towards improved overall survival (OS) with the PD-1 inhibitor, nivolumab, compared with sorafenib, this was not statistically significant.

As many HCCs are hypervascular and overexpress vascular endothelial growth factor (VEGF) and immune checkpoints such as PD-L1, there is a strong rationale for combining angiogenesis targeting agents and immune checkpoint inhibitors. Indeed, a phase Ib study of a combination of atezolizumab plus bevacizumab in patients with untreated, unresectable HCC showed an acceptable safety profile and promising efficacy with objective response rate (ORR) of 36% and a median progression free survival (PFS) of 7 months. Thus, this combination was further evaluated in a global, phase III IMbrave150 trial that randomized 501 patients with previously untreated, unresectable HCC in a 2:1 ratio to atezolizumab (1,200 mg intravenously every 3 weeks) plus bevacizumab (15 mg/kg intravenously every 3 weeks) or sorafenib (400 mg twice daily). Results from this study were recently published in the New England Journal of Medicine and clearly indicate the superiority of the atezolizumab plus bevacizumab over sorafenib. At the time of the primary analysis, median OS had not yet been reached for patients treated with atezolizumab plus bevacizumab, and was 13 months for those who received sorafenib (hazard ratio [HR] 0.58; P < 0.001). At 6 months, OS was significantly longer in the combination arm (84.8%) versus sorafenib (72.2%), and at 12 months, OS was 67.2% versus 54.6%. Median PFS was 6.8 months with atezolizumab plus bevacizumab and 4.3 months with sorafenib (HR 0.59; P < 0.001). In general, the OS and PFS benefits were consistent across all patient subgroups. In additon, according to independent assessment using RECIST 1.1 criteria, ORR was significantly better with the combination versus sorafenib (27.3% vs 11.9%; P < 0.001). Furthermore, atezolizumab plus bevacizumab substantially delayed deterioration in quality of life versus sorafenib. The occurrence of grade 3/4 adverse events (AEs) were similar in both arms (56.5% vs 55.1%). Hypertension was the most common AE in the combination arm (15.2%), which is consistent with bevacizumab’s known safety profile. Serious AEs were noted in 38% of the patients who received the combination therapy, although no new or unexpected AEs were observed. Approximately 15% of patients discontinued treatment in the combination arm, mainly due to gastrointestinal events, versus 10% in the sorafenib arm.

In an accompanying editorial, Robin K. Kelley, MD (University of California, San Francisco) acknowledged that ‘’the combination of atezolizumab plus bevacizumab has become the new benchmark for first-line therapy in advanced HCC’’. However, he highlighted that since the trial only enrolled patients with preserved liver function who have decreased risk of variceal bleeding, there is a need to “investigate the extent of bleeding risks and implications of this regimen for a broader population”. He also emphasized the unmet need to identify biomarkers of response to immune checkpoint inhibitors in HCC, to determine which patients would benefit most from immunotherapy as monotherapy or as a combination regimen. 

Reference

Finn RS, et al. N Engl J Med. 2020; 382:1894-905.

Kelley RK. NEJM N Engl J Med. 2020; 382: 1953-1955.