The immune system performs a complex role in the development and growth of hepatocellular carcinoma (HCC), according to evidence from preclinical and clinical studies that are beginning to unravel the cellular components that link cytokines and inflammation with hepatocarcinogenesis.
Several types of immune cells are present in the liver, including macrophages, natural killer (NK) cells, natural killer T cells (NKT cells), and CD4+ and CD8+ T cells. These immune cells not only regulate liver homeostasis and help to fight microorganisms but also maintain liver regeneration through the growth of hepatic progenitor cells and hepatocytes.1
As with many other tumors, inflammation is also a key contributor to carcinogenesis for those with HCC. Continuous inflammation and the corresponding proliferative stimuli not only promote liver fibrosis and cirrhosis, but also the transformation of hepatocytes to a malignant phenotype.2
Lymphocytes Play Complex Role
Studies conducted on in vivo models have found that lymphocytes are critical regulators of hepatocyte damage, liver fibrosis, and tumor development, according to findings reported by Jessica Endig, PhD, with the Hannover Medical School in Hannover, Germany, and colleagues in Cancer Cell.3
For this study, the investigators utilized a fumarylacetoacetate hydrolase (Fah)-deficient murine model, which reliably mimics the inflammatory environment of chronic liver disease. Liver injury was induced with fumarylacetoacetate (FAA), a toxic metabolite that accumulates with Fah deficiency. This model has been used extensively to delineate hepatocellular carcinogenesis and regeneration.
Repeated flares of liver injury with FAA resulted in strong increases in CD3+ CD4+ and CD3+ CD8+ T cells. In contrast, treatment with the drug NTBC (2-[2-nitro-4-(trifluoromethyl)benzoyl] cyclohexane-1,3-dione, or Nitisinone), which blocks FAA formation and prevents liver injury in Fah-deficient mice, abrogated this effect. Chronic, long-term FAA-induced liver injury in Fah-/- mice resulted in progressive biliary fibrosis.
Generation of mice lacking mature B and T lymphocytes or CD8+ T and NKT cells resulted in significantly decreased median overall survival (OS; 18.5 days and 20 days vs 31 days for Fah-/- mice), indicating that CD8+ T cells protect Fah-/- mice from acute-on-chronic liver failure.
Importantly, blockage of lymphotoxin-beta (LTβ), a member of the tumor necrosis family, using a lymphocyte-beta receptor (LTβR) immunoglobulin fusion protein resulted in significantly suppressed, FAA-induced tumor formation compared with controls (50% vs 8.3%; P ≤.05) and reduced tumor count (2.0 vs 1.0) in Fah-/- mice.
Alternating liver injury flares with recovery and regeneration, a process that mimics human liver diseases, produced multifocal HCCs in Fah-/- mice within 4 months. Array comparative genomic hybridization (aCGH) analysis of tumors showed elevated chromosomal aberrations that reflected increased genomic instability, and were consistent with chromosomal gains and losses in human HCCs.
To identify important gene expression changes associated with the immune response, microarray analysis was performed. In comparison with controls, the most significant expression changes in Fah-/- mice were associated with the cell cycle.
“Our results reveal that lymphocytes significantly contribute to liver damage and hepatocarcinogenesis, but also protect mice from acute-on-chronic liver failure and support liver progenitor cell proliferation during chronic liver injury,” the authors concluded. “These findings emphasize the fundamental requirement that the immune system needs to be tightly regulated in a context-specific fashion to balance immune surveillance and cancer risk. Moreover, inhibiting LTβR signaling might be an interesting approach to prevent tumor development in patients with chronic liver diseases accompanied by high levels of LTβ.”
A Novel Immunomodulatory Effect of Sorafenib
One hallmark of the tumorigenic phenotype is the expression of immune checkpoints, which typically stop an immune response after antigen stimulation. The immune checkpoint receptor programmed death-1 (PD-1) is highly expressed on T cells with chronic stimulation, and functions to downregulate the antitumor T-cell response when stimulated by its ligand PD-L1. This mechanism is beginning to be explored more thoroughly in HCC, using direct inhibitors and available therapies.
In a recent report, Suresh Kalathil, PhD, and coauthors with the Roswell Park Cancer Institute, Buffalo, NY, showed for the first time that sorafenib (Nexavar) has an immunomodulatory effect that reduces the number of PD-1+ T cells and regulatory T cells (Tregs), resulting in a survival benefit. The results were published in JCI Insight.
“From the literature, we knew that both immune cells and tumor cells expressed the VEGF receptor, which is a target of sorafenib,” Yasmin Thanavala, PhD, the senior author and professor of Oncology, described. “We then rationalized there is the possibility that sorafenib could also modulate immune cells via the same VEGF receptor. That is the question that we addressed.”
Re-activating the immune response has been of much interest in a number of tumor types, and novel antibody therapies have targeted the immune checkpoint pathways that inhibit T cells.4,5 Although the impact of PD-1 inhibition in HCC was not clearly understood, studies have indicated that the antitumor T-cell response is significantly impaired in advanced HCC.
In the JCI Insight analysis, blood samples from 19 patients with HCC were obtained before treatment initiation and after 4 weeks of sorafenib. The frequency and phenotype of PD-1+ T cells, Tregs, and myeloid-derived suppressor cells (MDSCs) were analyzed by flow cytometry.
Following sorafenib, a reduction in the absolute number of CD4+PD-1+ T cells and CD8+PD-1+ T cells was observed. Reductions in these cell types after therapy correlated with better OS compared with a lesser decline. Similarly, patients with greater numbers of CD4+PD-1+ T cells and CD8+PD-1+ T cells prior to sorafenib treatment also showed greater improvements in OS compared with patients who had lower numbers of these phenotypes, suggesting that patients with higher levels of PD-1+ T cells were more responsive to therapy.
In addition to a decrease in the number of T cells, the frequency and absolute number of Foxp3+ Tregs was significantly decreased, while the ratio of CD4+CD127+PD-1 T effector cells to CD4+Foxp3+PD-1+ Tregs was significantly increased after sorafenib therapy. The decrease in Foxp3+ Tregs also correlated with a significant improvement in OS.
In vitro analysis demonstrated a reduction in frequency of CD4+CD127+PD-1+ T cells, with a concomitant increase in the frequency of CD4+CD127+PD-1 T cells with sorafenib, indicating that the sorafenib benefit was due to specific downregulation of PD-1 expression on CD4+ T cells rather than nonspecific cytotoxicity.
Ultimately, the results indicate that PD-1+ T cells and Tregs may act as predictive biomarkers, and could be used to identify patients who may benefit the most from sorafenib.
“Checkpoint inhibitors bring down the level of immunosuppression and allow the patient’s own immune system to either target the tumor or make a more favorable environment for other kinds of immunotherapy, such as vaccination,” said Thanavala. “If you have a therapy that reduces immunosuppression, such as sorafenib, and bring in another therapy that activates the patient’s immune system, then patients may experience a greater therapeutic benefit.”
Direct PD-1 Targeting in HCC
Ongoing clinical trials are evaluating the antiPD-1 antibody nivolumab (Opdivo) in patients with HCC. Although approved by the FDA for a number of tumor types, including unresectable or metastatic melanoma, metastatic non–small cell lung cancer, and advanced renal cell carcinoma, nivolumab has not been extensively studied in HCC. Nivolumab blocks the interaction between PD-1 and its ligand PD-L1, enhancing the antitumor immune response.
Early clinical trials of nivolumab in HCC have been promising. In an interim analysis of the phase I/II, dose-expansion CheckMate-040 study, 206 patients with Child-Pugh class A HCC were evaluated. Bruno Sangro, MD, PhD, from the Clinica Universitaria de Navarra in Spain and colleagues presented findings from the dose escalation cohorts at the 2016 ASCO Annual Meeting.6
Patient cohorts consisted of: (1) those who were uninfected sorafenib-naïve/intolerant; (2) those who were uninfected sorafenib progressors; (3) those with hepatitis C virus (HCV); and (4) those with hepatitis B virus (HBV), with dose expansion of nivolumab at 3 mg/kg. Over half of patients (64%) had prior sorafenib treatment, 75% had extrahepatic metastasis, and 7% had vascular invasion. The primary endpoint was confirmed overall response rate (ORR) by RECIST v1.1 criteria, and secondary endpoints were OS, progression-free survival (PFS), time to progression, and biomarker assessments.
Of the 174 evaluable patients, 39% had a reduction in tumor burden. In a preliminary analysis, 55% had at least 18 weeks of follow-up and/or progressive disease. Among those patients, the ORR was 9% across etiologies and the 6-month OS rate was 69%. Nivolumab response was seen in patients with and without quantifiable PD-L1 expression as measured by immunohistochemistry.
Patients received a median of 5 to 6 doses of nivolumab (range, 1-19). Treatment-related adverse events (trAEs) occurred in half of patients, with the most frequent being fatigue (17%) and pruritus (12%). Grade 3/4 trAEs were observed in 14% of patients, and increased alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels were the most common (3% each).
A further analysis of patients from the CheckMate-040 study with 3 years of follow-up was also presented at the 2016 ASCO Annual Meeting by Anthony El-Khoueiry, MD, phase I program director at the University of Southern California Norris Comprehensive Cancer Center, and colleagues.7
Eligible patients had a Child-Pugh Score ≤7 and had previously failed or were intolerant of sorafenib. Nivolumab dosing ranged from 0.1 to 10 mg/kg for up to 2 years. The primary endpoint was safety, and secondary endpoints were antitumor activity and duration of response. Among the 51 patients who were enrolled and treated with nivolumab, 73% had prior sorafenib therapy, 76% had extrahepatic metastasis, and 12% had vascular invasion. Most discontinuations (35/41) were due to progressive disease.
Nivolumab treatment exhibited durable responses and disease stabilization across all dosages and cohorts. The median duration of response was 23.7 months, and median OS was 15.1 months. The OS rates at 6, 9, 12, and 18 months were 67%, 67%, 59%, and 48%, respectively. The complete response (CR) and partial response (PR) rates were 6% and 8%, respectively. Half of patients achieved stable disease.
Nivolumab exhibited a manageable safety profile. Treatment-related AEs occurred in 77% of patients, most commonly rash and AST increase (20% each). Grade 3/4 trAEs occurred in 20% of patients, most commonly increases in AST (10%), lipase (6%), or ALT (6%). The maximum tolerated dose was not reached.
“Hepatocellular carcinoma is an aggressive and fatal cancer, comprising 90% of all liver cancer in adults worldwide with limited therapeutic options for patients with advanced stage disease; no treatment advances have been made for patients who fail to respond or progress on the current standard of care,” said El-Khoueiry in a press release.8“These preliminary data are encouraging and support the ongoing evaluation of nivolumab in this patient population, as they show promising preliminary survival data, and durable partial or complete response in one out of five nivolumab-treated patients, with many others experiencing stable disease.”
A randomized phase III study, CheckMate-459, is currently recruiting participants and will evaluate nivolumab as a first-line therapy in patients with advanced HCC (NCT02576509). In the open-label study, the efficacy of nivolumab will be compared with sorafenib. Eligible patients have 1 or more lesions, Child-Pugh class A liver function, ECOG status of 0 or 1, and no prior systemic therapy.
Over 700 patients will be randomized in a 1:1 ratio to receive nivolumab or sorafenib until disease progression or unacceptable toxicity. The primary objectives are time to progression and OS. The secondary endpoints are ORR, PFS, and PD-L1 expression. Patient data will be stratified by vascular invasion and/or extrahepatic spread, etiology, and geography. The estimated primary completion date is July 2017.
Vaccine Therapy Under Exploration for HCC
Other novel approaches are also being explored that target immunoregulatory cells, with one such trial exploring a vaccine therapy in combination with sorafenib. In a case study reported by Yongxiang Yi et al in Molecular Clinical Oncology,9a patient with metastatic HCC achieved a CR after treatment with sorafenib, focused ultrasound knife therapy, and dendritic cell (DC) DRibbles vaccine.
DRibbles are autophagosomes containing a variety of antigens from antigen donor cells. They stimulate antigen-specific CD8 T cells ex vivo and induce a T-cell response against HCC in vivo. The case report described a 42-year-old male with HBV-associated HCC. The patient underwent curative resection, but in less than 2 months after surgery, multiple intrahepatic and extrahepatic lesions were detected, and the patient was treated with sorafenib. Because alpha-fetoprotein (AFP) levels continued to rise, focused ultrasound knife therapy was initiated, as was DC-DRibbles therapy.
Eight months after combination therapy, AFP levels returned to normal and remission of the liver and lung metastases was observed. Two years after the combination treatment, metastatic lesions were undetectable by computed tomography (CT) scan. No serious AEs were reported.
“Our results demonstrated that antigen-specific T-cell response aimed to treat recurrent HCC may be induced through stimulation by the DC-DRibbles vaccine,” the authors concluded. The case supports a combination strategy for patients with recurrent HCC after resection.
An ongoing, open-label, phase II trial will investigate the safety and efficacy of the individualized anticancer vaccine CRCL-AlloVax in approximately 30 patients with advanced HCC after prior sorafenib therapy (NCT02409524).
The novel AlloVax therapy is a vaccine consisting of chaperone-rich cell lysate (CRCL) antigens, derived from a patient’s own tumor cells, in combination with AlloStim cells, which are activated allogenic Th1 memory T cells that contain cytolytic granules.10The primary outcome measure is survival compared with historical controls. The estimated primary completion date is July 2017.
In conclusion, recent studies support the vital importance of immunosuppression in HCC and key molecular components that contribute to hepatocarcinogenesis. While novel immunomodulatory therapies are on the horizon, sorafenib may reduce the immunosuppressive network in HCC. Further work is needed to fully capitalize on the immune system as an anticancer therapy for HCC.