Although early-stage hepatocellular carcinoma (HCC) can be treated with curable intent, the majority of patients present with unresectable disease and subsequently receive a poor prognosis. For these patients, the multikinase inhibitors sorafenib and lenvatinib are licensed in the 1st-line setting, although overall survival (OS) benefits are modest at best, and come with considerable toxicity. The overexpression of several intrinsic immune-evasion pathways in HCC, including vascular endothelial growth factor (VEGF), contribute to immunosuppression and tumour survival. Suppression of VEGF overexpression may potentiate the efficacy of anti-PD-1/L1 immunotherapy.
In this light, the phase III IMbrave 150 trial randomised (2:1) 501 treatment-naïve HCC patients with unresectable locally advanced or metastatic disease to receive atezolizumab plus bevacizumab (ATEZO: 1200mg every 3 weeks, BEVA: 15 mg/kg intravenously every 3 weeks), or sorafenib monotherapy (400 mg twice daily, orally) (SORA) until unacceptable toxicity or loss of clinical benefit. The study had co-primary endpoints of OS and progression-free survival (PFS), with key secondary endpoints of toxicity.
Baseline characteristics were well balanced between the two study arms. The median age of participants was 64 and 66 years of age in the ATEZO plus BEVA arm and sorafenib arms, respectively. Approximately 82% in both arms were male. At a median-follow up of 8.6 months, 28.6% of ATEZO-BEVA patients had died, compared to 39.4% in the SORA arm. OS was significantly improved with ATEZO-BEVA, compared to SORA, with a median OS that was not reached vs. 13.2 months with SORA (HR[95%CI]: 0.58[0.42-0.79], P< 0.001). 6- and 12-month OS rates were 84.8% and 67.2% with ATEZO-BEVA and 72.2% and 54.6% with SORA, respectively. Median PFS was also extended with ATEZO-BEVA, compared to SORA (6.8 months vs. 4.3 months. HR[95%CI]: 0.59[0.47-0.76], P< 0.001), with 6-month PFS rates of 54.5% and 37.2%, respectively. Confirmed objective response rates (ORR) followed a similar, positive trend, with ATEZO-BEVA resulting in an ORR of 27.3%, including 5.5% of these patients achieving a complete response (CR), compared to an ORR of 11.9% with SORA, of which no patients achieved a CR. Furthermore the OS- and PFS-benefit observed with the ATEZO-BEVA combination was generally consistent across clinically relevant subgroups.
These clinically relevant benefits were further supported by encouraging patient-reported quality of life (QoL) outcomes, with the ATEZO-BEVA combination conveying a median time to deterioration of 11.2 months, compared to 3.6 months with sorafenib (HR[95%CI]: 0.63[0.46-0.85]). Median time to the deterioration of physical functioning (13.1 months vs. 4.9 months. HR[95%CI]: 0.53[0.39-0.73]) and role functioning (9.1 months vs. 3.6 months. HR[95%CI]: 0.62[0.46-0.84]) were also improved.
However, a higher proportion of patients receiving ATEZO-BEVA discontinued either treatment due to adverse events (AEs), compared to those who received sorafenib monotherapy (15.5% vs. 10.3%). The most common grade ≥ 3 AEs were hypertension (15.2% vs. 12.2%), fatigue (2.4% vs. 3.2%), proteinuria (3.0% vs. 0.6%), aspartate aminotransferase increase (7.0% vs. 5.1%), diarrhoea (1.8% vs. 5.1%) and decreased appetite (1.2% vs. 3.8%).
In conclusion, the phase III IMbrave 150 trial successfully demonstrated the superiority of the anti-VEGF-PD-1 combination of atezolizumab and bevacizumab for treatment-naïve unresectable advanced HCC. Although some grade ≥ 3 AEs were higher, quality of life actually improved with combination in comparison to sorafenib monotherapy. These results support the 1st line use of this combination, as a result.
Finn R et al., Atezolizumab plus Bevacizumab in Unresectable Hepatocellular Carcinoma. New Eng J Med. 2020; 382: 1894-1905.