Dissecting the Tumor Microenvironment to Predict Immunotherapy Response in Hepatocellular Cancer

Zhu AX, Abbas AR, de Galarreta MR, et al. Molecular correlates of clinical response and resistance to atezolizumab in combination with bevacizumab in advanced hepatocellular carcinoma. Nat Med 2022;28:1599–1611. Immunotherapy development for hepatocellular carcinoma (HCC) faces unique challenges owing to the peculiarities of this cancer type, which arises on a chronically active yet dysfunctional immune background. Biomarkers with an established predictive role in other inflammation-mediated tumors, such as programmed-death ligand 1 expression or tumor mutational burden have failed to identify responders to immune checkpoint inhibitors in HCC, leaving the provision of immunotherapy in routine practice and the development of novel regimens completely unsupported by reproducible biologic correlates. Recent studies demonstrate that combination immuno- and anti-angiogenesis therapy (atezolizumab plus bevacizumab) achieve an unprecedented survival advantage in patients with unresectable HCC, lending a median overall survival of 19.2 months in patients whose prognosis was historically limited to <6 months. However, measurable responses were achieved in a minority of patients and treatment-related toxicity can be life threatening. Therefore, the study from Zhu et al provides a significant step forward in addressing the acute unmet need for predictive biomarkers of response to immunotherapy in HCC. By presenting a comprehensive series of integrated analyses on patient samples collected within the phase Ib GO30140 trial and the phase III IMbrave150 trial, the authors provide practice-informing data demonstrating that the existence of preexisting T-cell immunity is a key phenotypic characteristic that correlates with response to atezolizumab plus bevacizumab. By integrating the results of whole genome RNA sequencing with multiplex immunohistochemistry of pre-treatment tissue samples, the authors demonstrate that an imbalance of the tumor microenvironment (TME) favoring the enrichment of the effector T-cell response over immunosuppressive regulatory T cells identifies patients who achieve significantly improved survival from the combination. To complement translational data, the authors provide further mechanistic insight into the synergistic role of anti-vascular endothelial growth factor and programmed death-1 pathway blockade in murine models, documenting how the addition of bevacizumab can enhance atezolizumab-mediated remodeling of immune-suppressive signals within the TME. Taken together, these findings ascribe HCC as an oncological diagnosis in which the presence of spontaneous immunogenicity is key to immunotherapy efficacy. Although histopathologic assessment confirms itself as an irreplaceable source of predictive information, TME immunogenicity is a function of density, functional polarization, and distribution of the infiltrate. The heterogeneity of the HCC TME and need for more readily applicable biomarkers than complex high-throughput transcriptomics remains a challenge for immediate clinical translation. Despite these limitations, the study by Zhu et al establishes a rationale to prospectively assess the TME as source of clinical stratification in trial participants. The interest of this approach extends beyond prediction of response to existing therapies. Identifying nonimmunogenic HCC, for instance, can lead to targeted testing of novel combinations modulating prognostically adverse traits such as regulatory T-cell infiltration, therefore broadening the reach of effective cancer immunotherapy to a wider proportion of patients.