Hepatocellular carcinoma (HCC) is a global health problem, with chronic hepatitis B and C viral infection accounting for well over half the cases worldwide [1]. NASH is emerging as the most rapidly rising cause for developing HCC in the West. The vast majority of patients with NASH-related HCC present at an advanced stage, when curative options, such as ablation, resection, or liver transplantation, are unavailable. Therefore, optimizing the strategic use of loco-regional and systemic therapy serve a vital role in improving long-term outcomes for patients with HCC. Although recent data on the use of combinations of checkpoint inhibitors and anti-angiogenic antibodies are encouraging, there remains concern that not all tumors are equally responsive to checkpoint inhibition.
1IntroductionThe prognosis of patients with early-stage HCC has improved over the past two decades following advances in diagnosis and treatment, including curative options including surgical resection, ablation and liver transplantation and the adoption of multidisciplinary management strategies by tumor boards [2,3]. Despite these advances, the vast majority of patients with HCC are unfortunately diagnosed with the advanced-stage disease when curative options are not feasible. Those patients for whom cure is not possible are often managed with palliative non-surgical liver-directed therapies and/or systemic therapies. It is estimated that over half of all patients with HCC will receive systemic therapy at some point during the course of their illness [4]. It is, therefore, imperative to determine the most effective systemic treatment strategies early in the disease course to improve clinical outcomes for patients with HCC.
2Systemic TherapyHistorically, HCC has been a relatively chemotherapy-refractory tumor [5]. Until 2008, no systemic treatment options were found to provide a survival benefit in advanced disease or following liver-directed therapy as investigated in randomized controlled trials. The SHARP trial, a phase III randomized controlled trial, compared sorafenib, an oral multi-kinase inhibitor, to a placebo in patients with HCC and well-compensated cirrhosis but who were unsuitable for surgical resection [6]. This trial demonstrated the superiority of sorafenib over placebo, with an improvement in median overall survival of 10.7 months in the treatment arm, up from 7.9 months in the placebo group. This represented a significant improvement leading to FDA approval as a first-line agent for the treatment of advanced-stage HCC. Following the approval of sorafenib for unresectable HCC, other multi-kinase inhibitors have been approved. Unfortunately, the overall survival benefit of sorafenib and the other multi-kinase inhibitors when introduced into clinical practices has been less satisfactory, with significant rates of adverse effects being common and necessitating withdrawal of therapy in approximately 15% of patients [7–12].
3Immune Checkpoint InhibitorsThe microenvironment of many cancers has repressed T-cell responses. Immune checkpoint inhibitors (ICIs) target specific inhibitory sites within the inflammatory cascade, including representative examples of the programmed death-1 (PD-1), programmed death-ligand 1 (PD-L1), and cytotoxic T-lymphocyte antigen 4 (CTLA-4). The inhibition of these targets using monoclonal antibody therapy has been shown across cancer types to restore anti-tumor T-cell response, including melanoma, lung, renal, and bladder cancer [13]. The efficacy of ICI in combination with anti-angiogenic agent for advanced HCC not amenable to potentially curative regimens was tested in phase III IMbrave150 trial. The trial was designed as an open-label study, randomized patients to sorafenib ('the standard of care') or a combination of a ICI (atezolizumab) and an anti-angiogenic antibody directed at the vascular endothelial growth factor (bevacizumab) as first-line therapy for advanced HCC [14]. Updated survival analysis revealed an improvement in median patient survival from 13.4 months in those receiving sorafenib to 19.2 months in the patients in the combination arm [15]. Combination therapy also prolonged progression-free survival (PFS) (6.9 months versus 4.3 months for the sorafenib arm). IMbrave150 was a landmark study that established atezolizumab/bevacizumab as the new standard-of-care first-line therapy for patients with advanced HCC not amenable to curative therapy without prior exposure to systemic therapy.
4Limitations of Immune Checkpoint InhibitorsThere is mounting clinical evidence, however, that ICIs may be less effective in patients with non-virally-associated HCC. In the IMbrave150 trial, nearly 70% of the study population had viral HCC as an international trial design. Subgroup analysis of this trial showed a limited benefit of a combination atezolizumab/bevacizumab in overall survival in patients with non-viral HCC compared to sorafenib with median overall survival of 17 months vs. 18.1 months, within the limitations of this particular subgroup with 30 events [16]. A recent meta-analysis of three phases III trials studying the efficacy of ICIs in HCC, including IMbrave150, KEYNOTE-240 and CheckMate-459, totaling 1656 patients, showed that ICIs improved overall survival in patients with viral HCC but not those with non-viral HCC [17]. Another large meta-analysis of eight randomized controlled trials with a total of 3739 patients showed that ICIs are less beneficial in patients with non-viral HCC, whereas the efficacy of tyrosine-kinase inhibitors or anti-VEGF-agents was unaffected by the underlying driver of HCC [18]. It is important to note that patients with non-viral HCC in these meta-analyses were not differentiated further and included patients with HCC related to NASH, alcoholic liver disease, and other non-viral etiologies.
These differences in the efficacy of ICIs based on the etiological driver of HCC may at least partly be explained by inherent alterations in the tumor micro-environment, as suggested by in vivo studies. Recently, Pfister et al. reported that despite increased CD8+ PD-1+ cells in the NASH-related HCC mice model, anti-PD1/PD-L1 immunotherapy did not reduce tumor incidence, size, or number but rather progressed liver fibrosis in these livers [17]. In contrast, tumors in the non-NASH HCC models regressed using the same treatment [17]. Further, treatment with anti-CD8 antibody and anti-PD-1 monoclonal antibody appeared to ameliorate the incidence of HCC development as well as liver fibrosis. To assess if these results can be extrapolated to human NASH, they found similar profiles of CD8+ PD-1+ T cells in the liver biopsies of patients with NAFLD and NASH. Separate in vitro studies imply that ICIs impair HCC development in a non-NASH mouse model [19] but not intrahepatic melanoma metastases in the NASH model [20]. Taken together, clinical and pre-clinical data would appear to suggest that ICIs may not have the same success in patients with NASH-related HCC as observed with virally mediated disease and that these findings may be unique in the context of HCC.
Despite this evidence, it is premature to discount immunotherapy's role in managing patients with NASH-related or non-viral HCC. The heterogeneous nature of HCC does not afford itself to readily actionable targets outside of a clinical trial. It is important to acknowledge the heterogeneity inherent in 'non-viral HCC' subgroups included in the meta-analyses were not further stratified by the cause of liver disease. The subgroup analysis reported in IMbrave150 was not designed to address the clinical benefit of this question. There remain many unanswered questions about the influence of the immune microenvironment on the development of HCC. The immune microenvironment may be further influenced by the gut microbiome, which is thought to play a key role in the pathogenesis of NAFLD and NASH [21,22]. Clinical data with nivolumab, an anti-PD1 antibody, as well as pembrolizumab, antibody directed against PD-1, have shown both agents have similar efficacy in inducing durable HCC response in a subset of patients with viral-HCC and non-viral HCC [17]. Further, in the IMbrave150 trial, progression-free survival in the combination of atezolizumab/bevacizumab arm in non-viral HCC patients was comparable to that in patients with viral HCC (6.3 months in non-viral HCC, 4.5 months in HBV-HCC, and 6.6 months in HCV-HCC) [21]. The PFS in patients with non-viral HCC treated with combination atezolizumab/bevacizumab was significantly longer than in patients treated with sorafenib, suggesting that the addition of VEGF inhibitor may ameliorate reduced efficacy of ICIs in patients with non-viral HCC [23].
5ConclusionsHCC has been defined as a chemotherapy-resistant tumor requiring the use of multi-kinase inhibition. However, immunotherapy has emerged as a mainstay of systemic therapy for patients with advanced disease. There is mounting evidence that cirrhosis's underlying etiology may influence immune-based therapies' efficacy. These differences are most likely accounted for by variations in the tumor micro-environment conferring differences in the host immune response and local contributions from local niches, including the native liver and associated microbiome. Given the molecular heterogeneity of HCC among differing disease etiologies, future clinical trials would benefit from patient stratification based on disease etiology and/or the use of biomarker-based treatment algorithms to optimize and personalize therapeutic strategies for patients with advanced-stage HCC.
