Atezolizumab plus Bevacizumab: A Novel Breakthrough in Hepatocellular Carcinoma

In this issue of Clinical Cancer Research, Casak and colleagues summarize the data that led to the FDA's approval of atezolizumab (a PD-L1 inhibitor) and bevacizumab (a VEGF-A inhibitor) for the treatment of advanced hepatocellular carcinoma (HCC) in the front-line setting (1). The approval is based on the recently reported IMbrave150 study (2), an international, multicentre, phase III trial that randomized 501 patients with advanced HCC in a 2:1 ratio to atezolizumab and bevacizumab (336 patients) or sorafenib (165 patients). This combination was the first to show an improvement in overall survival (OS) when compared with sorafenib, with a median OS of 19.2 months in the experimental arm versus 13.4 months in the sorafenib arm [HR 0.66; 95% confidence interval (CI), 0.52–0.85; P = 0.0009] (3). Moreover, the combination improved progression-free survival (PFS; median 6.8 vs. 4.3 months, respectively, P < 0.001). Treatment with atezolizumab and bevacizumab is also the first regimen to improve patient-reported outcomes. Specifically, it significantly delayed median times to deterioration in quality of life (11.2 months for the combination regime vs. 3.6 months for sorafenib) as assessed by the EORTC QLQ-30 questionnaire (2). Regarding safety, while atezolizumab and bevacizumab showed a manageable toxicity profile, grade 3 or 4 treatment-related adverse events occurred in 36% compared with 46% with sorafenib. Adverse events leading to discontinuation of treatment occurred in 15.5% of patients (vs. 10.3% with sorafenib). Because of the increased risk of upper gastrointestinal bleeding linked to bevacizumab administration in several previous trials, assessment for esophageal varices by esophagogastroduodenoscopy was mandatory in the 6 months prior to treatment initiation. Therefore, this constitutes a practice changing requirement that will have to be implemented before treatment of patients with advanced HCC. In fact, presence of any contraindication for the combination regime (i.e., autoimmune disease) or untreated varices will render sorafenib or lenvatinib still as first-line treatment options. At this point, it is unknown what would be the percentage of patients suitable for the combination regime versus tyrosine kinase inhibitor (TKI) in front line. All in all, the results were robust and support the use of atezolizumab and bevacizumab as the new standard, first-line therapy in patients with advanced HCC with preserved liver function (Fig. 1A; ref. 1).

The IMbrave150 trial has not only changed the treatment landscape of advanced HCC management but also paved the way for the new era of combination therapies in HCC. Indeed, despite the encouraging preliminary efficacy results of checkpoint inhibitor monotherapies in the treatment of advanced HCC, phase III trials evaluating the PD-1 inhibitors nivolumab and pembrolizumab failed to reach their primary endpoint, in the first- and second-line settings, respectively (4). The prevailing rationale for these results is that the proportion of responding patients is too small (15%–20%) to elicit an advantage for the entire population, particularly when compared with an active treatment improving survival such as sorafenib. Unfortunately, so far efforts to identify biomarkers predicting treatment response are not robust enough to be translated into clinical practice. The IMbrave150 trial, on the other hand, demonstrated that a response rate of approximately 30% along with a disease control rate of approximately 75% led to a significant benefit among all treated patients. A recent phase III, open-label, ORIENT-32 trial in China also showed similar results combining sintilimab (PD-1 inhibitor) and bevacizumab.

Combining an inhibitor of the PD-1/PD-L1 axis with other drugs that may expand the immune-sensitive HCC population, and thus increase the proportion of patients with objective response is a common theme in ongoing trials. The most promising synergistic partners currently being explored in HCC are VEGF inhibitors, TKIs, and other checkpoint inhibitors (i.e., anti-CTLA4). The rationale for these strategies is sound. The VEGF pathway promotes local tumor suppression through inhibition of antigen-presenting cells and activation of immunosuppressive cells such as myeloid-derived suppressor cells (MDSC) and regulatory T cells (Treg; Fig. 1B). Therefore, an inhibitor of the PD-1/PD-L1 axis combined with a VEGF pathway inhibitor present complementary and synergistic mechanisms of action (5). Indeed, atezolizumab increases the recruitment and activation of effector CD8 T cells, dendritic cells, and natural killer cells, and promotes an M1 antitumoral macrophage phenotype. Simultaneously, bevacizumab reduces the infiltration and activity of MDSCs and Tregs, and additionally reduces M2 macrophage polarization. Therefore, while checkpoint inhibitor monotherapies may be active only in inflamed tumors, combining a VEGF pathway inhibitor could modulate a suppressive microenvironment to increase immune infiltration and response to checkpoint inhibitors. In this line, a phase III trial testing the combination of camrelizumab (PD-1 inhibitor) and apatinib (VEGFR-2 inhibitor) is ongoing (NCT03764293).

In addition to anti-VEGF antibodies, mechanistic insights suggest that combining checkpoint inhibitors with TKI will further enhance the immune response (5). TKIs approved for the treatment of advanced HCC (sorafenib, lenvatinib, regorafenib, and cabozantinib) are targeting multiple kinases (Fig. 1A). For instance, lenvatinib is an inhibitor of VEGFR 1, 2, and 3 but also targets FGFR 1–4, RET, and platelet-derived growth factor receptor. Therefore, these agents can potentially decrease the infiltration of MDSCs and Tregs in the tumor microenvironment and increase the infiltration of CD8 T cells and dendritic cells, thereby turning a suppressive microenvironment into an inflamed, antitumorigenic one. This approach was recently explored in a phase Ib, open-label, multicentre study that treated 100 patients with advanced HCC with lenvatinib and pembrolizumab in the first-line setting (7). This combination rendered a response rate of 46% with a disease control rate of 88% according to modified RECIST criteria (mRECIST), numbers unprecedented in HCC. As a result, these combinations are being explored in phase III randomized controlled trials, such as LEAP-002 (NCT03713593, testing lenvatinib and pembrolizumab) or COSMIC-312 (NCT03755791, testing the TKI cabozantinib and atezolizumab; Fig. 1B).

Finally, combining checkpoint inhibitors with complementary mechanisms of action may further enhance the antitumor immune response and offer durable clinical benefits. CTLA-4 is an essential molecule involved in regulating the priming phase of the immune response. Moreover, Tregs constitutively express CTLA-4, which competitively inhibits the binding of CD80/86 to the costimulatory receptor CD28 in the tumor microenvironment, therefore promoting an immunosuppressive microenvironment. The results of the phase I/II, multicentre, open-label CheckMate-040 trial (NCT01658878) assessing the combination of nivolumab and ipilimumab (a CTLA-4 inhibitor) showed a response rate of 34% according to mRECIST, with a median duration of response of 17.5 months. Further combinations of checkpoint inhibitors are also being explored in phase III randomized controlled trials, such as the CheckMate 9DW (NCT04039607, assessing nivolumab and ipilimumab) and the HIMALAYA trial [NCT03298451, assessing the combination of durvalumab (PD-L1 inhibitor) and tremelimumab (CTLA-4 inhibitor)].

These combination strategies are reshaping the treatment landscape for advanced HCC, but are also being moved to the intermediate and (neo)adjuvant settings (5). In the former, locoregional therapies release antigens and proinflammatory cytokines that can enhance the antitumor immune response and boost the effect of immunotherapy-based combinations. Such strategies are being explored by combining TACE with durvalumab and bevacizumab (EMERALD-1, NCT03778957), lenvatinib and pembrolizumab (LEAP-012, NCT04246177), or nivolumab and ipilimumab (CheckMate 74W, NCT04340193). In the adjuvant setting, the immune context of HCC is known to influence prognosis and risk of recurrence. Indeed, an inflamed microenvironment with abundant immune effector cells is associated with a better prognosis while a high infiltration of Tregs or MDSCs determines a higher risk of recurrence. This suggests that the immune context plays a fundamental role in early stages of HCC, and enhancing these responses at earlier stages constitutes an attractive therapeutic strategy to reduce the high recurrence rates observed after curative treatment. Aside from adjuvant trials assessing nivolumab or pembrolizumab as standalone agents, more recent phase III investigations are exploring the combination of atezolizumab-bevacizumab or the combination of durvalumab and bevacizumab.

Therefore, FDA approval of atezolizumab-bevacizumab has led to a change in the treatment paradigm of HCC (1). This feature has been already captured by a consensus document on trial design (4), which establishes atezolizumab-bevacizumab as the new standard-of-care comparator arm in clinical trials in the first-line setting (4). Moreover, the therapeutic armamentarium in subsequent lines of treatment has increased, with the incorporation of sorafenib and lenvatinib in the second-line setting and regorafenib, ramucirumab, and cabozantinib thereafter. Therefore, very potent drugs will be required to convey meaningful differences in OS in the first-line setting, not only due to the more potent comparator arm but also due to the fact that subsequent lines of treatment may mitigate any differences observed in OS. In this line, PFS with substantial magnitude of benefits is now proposed as a surrogate for overall survival and a primary endpoint for future clinical trials (4).

Overall, atezolizumab-bevacizumab combination therapy points to the likely end of exploring single agents for first-line treatment for advanced HCC, which started in 2007 with sorafenib. More importantly, it marks the dawn of a new and exciting era in HCC management.

J. Llovet reports grants and personal fees from Bayer Pharmaceuticals, Eisai, Bristol Myers Squibb, Boehringer-Ingelheim, and Ipsen; and personal fees from Celsion Corporation, Exelixis, Eli Lilly, Roche, Genentech, Glycotest, Nucleix, AstraZeneca, Merck, Sirtez, and Mina Alpha Ltd outside the submitted work. No disclosures were reported by the other authors.

F. Castet is supported by a Clínico Junior grant from Fundación Científica AECC. C.E. Willoughby is supported by a Sara Borrell Fellowship (CD19/00109) from the Instituto de Salud Carlos III (ISCIII) and the European Social Fund. P.K. Haber is supported by a grant from the German Research Foundation (DFG, HA 8754/1-1). J.M. Llovet is supported by grants from the European Commission (EC) Horizon 2020 Program (HEPCAR, proposal number 667273-2), the U.S. Department of Defense (CA150272P3), the NCI (P30 CA196521), the Samuel Waxman Cancer Research Foundation, the Spanish National Health Institute (MICINN, PID2019-105378RB-I00), through a partnership between Cancer Research UK, Fondazione AIRC and Fundación Científica de la Asociacion Española Contra el Cáncer (HUNTER, Ref. C9380/A26813), and by the Generalitat de Catalunya (AGAUR, SGR-1358).