Hepatocellular carcinoma is one of the most common causes of cancer-related deaths worldwide. In most patients, hepatocellular carcinoma arises in conjunction with liver cirrhosis and is attributed to several risk factors, including infection (usually with hepatitis B or C viruses), excessive alcohol consumption, and non-alcoholic fatty liver disease.1, 2 Surgical resection, transplantation, and ablation are potentially curative treatment options for patients with early-stage disease; however, most patients present with advanced, unresectable disease.1 These patients are typically treated worldwide with locoregional and systemic therapies.
Sorafenib, a multi-kinase inhibitor, remains the standard first-line systemic treatment for advanced hepatocellular carcinoma, and lenvatinib has been shown to be non-inferior to sorafenib in overall survival3, 4, 5 and is approved for treatment in Japan. All other treatments that have been assessed as first-line therapy, and several other drugs that have been evaluated as second-line therapies have not shown improved survival over sorafenib or placebo.6 Regorafenib, another anti-angiogenic multi-kinase inhibitor, is the only second-line therapy that has been globally approved for use after treatment with sorafenib in patients with hepatocellular carcinoma.7 In 2018, cabozantinib was shown to improve overall survival compared with placebo in patients with advanced hepatocellular carcinoma who were intolerant to or had progressive disease after sorafenib treatment.8 A phase 3 study9 (REACH-2) that evaluated ramucirumab as a second-line therapy in patients with hepatocellular carcinoma who had blood concentrations of alpha-fetoprotein of at least 400 ng/mL also showed improved overall survival compared with placebo. Although these inhibitors have led to improved survival, the benefits remain modest and many of these drugs have prohibitive side-effects. Novel treatment strategies for these patients are desperately needed.
Research in context
Evidence before this study
Few treatment options are available for patients with advanced hepatocellular carcinoma. First-line sorafenib and second-line regorafenib therapies have been approved; however, these treatments have modest efficacy and associated toxicity. Thus, there is an unmet need to develop additional therapeutic strategies and identify biomarkers that are predictive of therapeutic response in these patients. Immunotherapies, including checkpoint blockade therapies, have shown promising preliminary results in patients with hepatocellular carcinoma, and the programmed cell death protein-1 (PD-1) inhibitor nivolumab has received accelerated approval for second-line treatment in the USA. We searched PubMed from Sept 1, 2017, to Feb 22, 2018, with the search terms “advanced hepatocellular carcinoma AND treatment”, “immunotherapy AND hepatocellular carcinoma”, “anti-PD-1 OR pembrolizumab OR MK-3475 OR nivolumab AND hepatocellular carcinoma”, and “biomarkers AND immunotherapy AND hepatocellular carcinoma”. This search was limited to articles and abstracts in English. We found few studies that evaluated immunotherapy in hepatocellular carcinoma; in one study, the cytotoxic T lymphocyte antigen inhibitor, tremelimumab, showed antitumour activity and manageable toxicity in patients with advanced hepatocellular carcinoma. Combination therapy with the programmed death-ligand 1 (PD-L1) inhibitor durvalumab and tremelimumab also showed encouraging preliminary results and no unexpected safety signals in patients with hepatocellular carcinoma. The only study that has described treatment of patients with hepatocellular carcinoma with an anti-PD-1 inhibitor was that of nivolumab, which showed promising outcomes and a safety profile generally consistent with that reported for nivolumab in other tumour types. Apart from the nivolumab study, which suggested that PD-L1 expression in tumour cells was not significantly associated with a clinical response, no relevant studies have evaluated biomarkers of response to immunotherapy in hepatocellular carcinoma, as was done in our study.
Added value of this study
To our knowledge, the KEYNOTE-224 study is the first to show the clinical efficacy and safety of pembrolizumab in patients with advanced hepatocellular carcinoma who were previously treated with sorafenib. A substantial proportion of patients had objective responses and both median progression-free survival and overall survival were promising after treatment with pembrolizumab. These results are also consistent with findings for nivolumab in these patients, indicating the importance of anti-PD-1 therapy as a potential treatment option for these patients. The study also showed that PD-L1 expression, as shown by combined positive score (a measure of PD-L1-positive immune and tumour cell number), was associated with response to pembrolizumab; however, further study is needed to determine the clinical usefulness of PD-L1 in predicting response to anti-PD-1 therapy in patients with advanced hepatocellular carcinoma.
Implications of all the available evidence
Overall, this study indicated that pembrolizumab could be an additional treatment option for patients with advanced hepatocellular carcinoma whose disease progressed on or were intolerant to previous sorafenib therapy; further assessment is ongoing in the phase 3 KEYNOTE-240 and KEYNOTE-394 trials in patients with second-line advanced hepatocellular carcinoma. PD-L1 expression, as assessed by the combined positive score, appeared to be associated with response to pembrolizumab in these patients; however, additional studies are needed to better understand the associations between biomarkers and response to anti-PD-1 therapy in hepatocellular carcinoma.
The liver maintains a balance between activation of and tolerance by immune cells in response to antigenic hyperstimulation, and dysregulation of this tightly controlled immunological network leads to chronic liver disease and hepatocellular carcinoma.2 Hepatocellular carcinoma has been shown to be associated with inflammation and a suppressed immune environment.2, 10 An inflammatory gene-expression signature was shown to be predictive of lower overall survival in liver tissue adjacent to tumours in patients with hepatocellular carcinoma,11 and high expression of programmed death-ligand 1 (PD-L1) in tumours correlates with a poorer prognosis than lower expression of PD-L1 in patients with resected hepatocellular carcinoma.12 Upregulation of programmed cell death protein-1 (PD-1) and PD-L1 expression on T cells is also associated with a more advanced disease stage and higher recurrence rates in patients with hepatocellular carcinoma.13 Signature genes in subsets of infiltrating regulatory T cells and exhausted CD8+ cells of patients with hepatocellular carcinoma have been identified, including layillin (LAYN), which might be linked to immune suppression by these cells.14 These immunological findings suggest that immunotherapy approaches could benefit these patients.
Immunotherapy with checkpoint blockade inhibitors has shown promising preliminary results in patients with advanced hepatocellular carcinoma, either as monotherapy or combination therapy.2 The anti-PD-1 inhibitor nivolumab has shown encouraging outcomes and a safety profile that is generally consistent with that reported in other tumour types and, as such, has received accelerated approval in the USA for the treatment of hepatocellular carcinoma that has previously been treated with sorafenib.15, 16 Pembrolizumab, an anti-PD-1 monoclonal antibody, has shown antitumour activity and a manageable safety profile in several cancers, including melanoma, non-small cell lung cancer, head and neck cancer, squamous cell carcinoma, gastric and urothelial cancers, and classical Hodgkin's lymphoma.17, 18
We present the results of KEYNOTE-224, an open-label, phase 2 trial, in which we aim to evaluate the efficacy and safety of pembrolizumab in patients with advanced hepatocellular carcinoma who have been previously treated with sorafenib, and to assess the association between PD-L1 expression and clinical outcomes.