Immunotherapy for Hepatobiliary Malignancies

Immunotherapy for Hepatobiliary Malignancies


Dr. Thomas Karasic

Dr. Arturo Loaiza-Bonilla

By Thomas Karasic, MD, and Arturo Loaiza-Bonilla, MD, MSEd

Article Highlights

  • Hepatobiliary cancers, predominantly hepatocellular carcinoma (HCC), are the second leading cause of cancer deaths worldwide. Most cases are diagnosed at an advanced stage, when therapeutic options are limited.
  • The ever-expanding success of checkpoint inhibitors has renewed interest in immunotherapy in HCC, and a promising phase I/II study of nivolumab in advanced HCC was presented during the 2015 ASCO Annual Meeting.
  • Other novel approaches include modified oncolytic viruses, TGF-b inhibition, and combination therapies with checkpoint bloImmunotherapy stands poised to become a potent weapon in the limited arsenal of HCC therapy if the early success of nivolumab is confirmed.

Hepatobiliary cancers, predominantly hepatocellular carcinoma (HCC), are the second leading cause of cancer deaths worldwide—accounting for approximately 750,000 deaths across the world in 2012.1 In the United States in 2016, an estimated 39,000 patients will be diagnosed with hepatobiliary cancers, and 27,000 patients will die of the disease.2 The death rate from hepatobiliary tumors in the United States has risen steadily since 1980, even as the death rates of more common cancers have declined.

Despite universal screening guidelines for patients with cirrhosis and for high-risk patients with hepatitis B, most HCC cases are diagnosed at an advanced stage, where curative therapy with resection or liver transplant is not feasible.2 For these patients with advanced HCC, systemic therapy options are limited; sorafenib is the only U.S. Food and Drug Administration (FDA)–approved therapy. Sorafenib conveys only a modest survival benefit, with an improvement in median overall survival (OS) from 7.9 months to 10.7 months.3 Despite intense interest, multiple pivotal phase III clinical trials have failed to demonstrate either improvement over sorafenib in the first-line setting or benefit after sorafenib in the second-line setting.4 The recent success of immunotherapeutic approaches in an ever-growing list of cancers—including melanoma, renal cell carcinoma, and non–small cell lung cancer—has opened new avenues for the treatment of this difficult disease.

HCC: An Immunogenic Tumor in an Immunotolerant Environment

The liver’s normal immune environment is skewed toward immune tolerance in order to accommodate the repeated exposure to intestinal pathogens and exogenous antigens from the digestive system.5 Many of the antigen-presenting cells of the liver, such as liver sinusoidal endothelial cells and Kupffer cells, express high levels of inhibitory immune signaling molecules such as PD-L1, preferentially activate inhibitory regulatory T cells (Tregs), and produce inhibitory cytokines such as interleukin-10.6 The presence of chronic inflammation from viral hepatitis or fatty liver disease further diminishes immune reactivity through induction of T-cell exhaustion,7 a state in which CD8-positive T cells can recognize foreign antigens and cancer neoantigens but are unable to effect a cytotoxic response.8 These exhausted T cells overexpress the inhibitory proteins PD-1 and CTLA-4.9,10

HCC arises in this permissive immune environment despite frequent recognition by CD8-positive T cells of cancer-specific antigens such as AFP,11 GP3,11 and hTERT.12 Researchers have been trying different approaches to augment this latent immune response in HCC for years, with mechanistic evidence of immune activation but limited clinical success. Interferon-based approaches, which have held a long-standing role in viral hepatitis therapy, demonstrate occasional marked and durable responses both as monotherapy and in combination with cytotoxic chemotherapy.13-19 However, they benefit only a small number of patients and ultimately proved too toxic for widespread use. Likewise, vaccine therapies directed against tumor lysates or tumor-specific antigens, such as GP3 or AFP, have demonstrated proof of principle, with target-specific immune activation and AFP responses in a subset of patients, but they have not demonstrated clear clinical benefit.20-26

Checkpoint Inhibition in HCC

The remarkable and ever-expanding success of checkpoint inhibitors has renewed interest in immunotherapy in HCC. These approaches use antibodies to target inhibitory signaling proteins such as CTLA-4 (e.g., ipilimumab and tremelimumab), PD-1 (e.g., pembrolizumab, nivolumab, and durvalumab), or PD-L1 (e.g., atezolizumab), resulting in activation of the adaptive immune response—especially cytotoxic CD8-positive T cells. To date, the only study published on checkpoint inhibition in HCC is a phase II trial using the CTLA-4 inhibitory antibody tremelimumab. The trial enrolled 21 patients with hepatitis C–related cirrhosis. Patients with Child-Pugh A (57%) and Child-Pugh B (43%) cirrhosis were enrolled, and prior systemic therapy, including sorafenib, was permitted. Among 17 evaluable patients, three (18.0%) had a partial response, and an additional 10 (58.8%) had stable disease, with a median time to progression of 6.5 months. Nearly half (45%) of all patients experienced grade 3 or higher toxicity, primarily transaminase elevations. Typical immune-related toxicities such as rash (65%) and diarrhea (30%) were common, but no patients required corticosteroid treatment for toxicity.27

The recent success of immunotherapeutic approaches in an ever-growing list of cancers has opened new avenues for the treatment of hepatobiliary malignancies.

A phase I/II study of nivolumab in advanced HCC, presented during the 2015 ASCO Annual Meeting, is more promising (NCT01658878).28 In this trial, 51 patients with Child-Pugh A cirrhosis received nivolumab at doses from 0.1 to 10.0 mg/kg for up to 2 years. Most patients had extrahepatic metastases (76%) and prior sorafenib treatment (73%). Those patients with active viral hepatitis were excluded, but patients with chronic hepatitis B and C were allowed. Of 48 evaluable patients, seven (15%) experienced a response based on RECIST criteria (three [6%] complete responses and four [8%] partial responses), and an additional 24 patients (50%) had stable disease. Most notably, median OS was 15.1 months, which compares favorably with the median survival of 10.7 months of patients receiving sorafenib in the SHARP trial.3 Toxicity was mild, consisting primarily of transaminase elevations.

Based on these promising results, an FDA registration phase III trial comparing nivolumab with sorafenib in the frontline setting for advanced HCC is currently enrolling patients (NCT02576509). Additional phase II trials of the PD-1 inhibitor pembrolizumab in the second-line setting are also ongoing (NCT02658019, NCT02702414, and NCT02702401), as is a combination trial with tremelimumab and durvalumab (NCT02519348).

Additional Immunotherapy Approaches in HCC

TGF-b regulates the tumor immune microenvironment through induction of Tregs, fibroblasts, and inhibitory inflammatory signals.29 Galunisertib, a kinase inhibitor of TGF-b 1 receptor, has demonstrated promise in a subset of patients with sorafenib-refractory disease, with 24% experiencing an AFP response and a median OS of 96 weeks for responders, compared with 30 weeks for nonresponders.30 As of yet, no clear biomarker has been found to determine which patients will respond. Additional studies are ongoing in a frontline trial of a randomized combination of galunisertib with sorafenib (NCT01246986) and galunisertib with nivolumab in the second-line setting (NCT02423343).

Another novel approach has been the use of modified oncolytic viruses. This strategy has been successful in melanoma, as evidenced by the recent approval of intratumoral injections of talimogene laherparepvec (T-VEC)—a modified herpes simplex virus that produces GM-CSF. In HCC, JX-594—a modified poxvirus that also produces GM-CSF and selectively reproduces in tumor cells as a result of thymidine kinase deficiency—demonstrated activity via both intratumoral injection31 and intravenous infusion.32 In a phase II dose-finding trial in patients who were heavily pretreated, median survival was 14.1 months in the high-dose arm.33 Despite this promise, the TRAVERSE trial, a randomized phase IIb trial of JX-594 in the second-line setting, failed to demonstrate a survival benefit and has yet to be published (NCT01387555). A phase III randomized open-label trial of sorafenib with JX-594 compared with placebo is ongoing (NCT02562755), as is a trial of T-VEC via intratumoral injection (NCT02509507).

Combination Approaches With Standard Therapies

The early success of checkpoint inhibition has sparked interest in combination approaches with local therapies, such as transarterial chemoembolization (TACE), radiofrequency ablation (RFA), and transarterial radioembolization (TARE). These strategies attempt to harness the antigen release from tumor cell death from the local therapy to prime the immune system.34 Active trials using this approach are combining tremelimumab with TACE or RFA (NCT01853618), nivolumab with TARE (NCT02837029), and tremelimumab and durvalumab with TACE or RFA (NCT02821754).

Preclinical data on the immunogenic effects of sorafenib are more mixed, with evidence of suppression of effector T cells,35 natural killer cells,36 and dendritic cells,37 but also reduction in tumor-infiltrating Tregs.38 The combination of sorafenib and a PD-L1 antibody showed activity in mouse xenografts,39 and combination trials with PD-1 inhibitors and antiangiogenic agents such as ramucirumab (NCT02572687) and bevacizumab (NCT02715531) are ongoing.

Immunotherapy in Cholangiocarcinoma

In contrast to HCC, immunotherapy approaches in cholangiocarcinoma have been limited and largely unsuccessful to date, consisting primarily of vaccine-based therapies.40,41 However, a high frequency of tumor-infiltrating lymphocytes and PD-L1 expression42 suggest that checkpoint inhibition may prove effective. Indeed, the activity of immune therapy in cholangiocarcinoma has been demonstrated by marked and durable responses using a personalized peptide vaccine43 and stimulation of tumor-infiltrating lymphocytes.44 We are unaware of any trials evaluating checkpoint blockade specifically in cholangiocarcinoma, but biliary tumors are being included in basket immunotherapy trials targeting a variety of rare tumors with either PD-1 blockade (KEYNOTE-158; NCT02628067) or combination PD-1 and CTLA-4 blockade (NCT02834013).

There is a potential benefit from assessing the DNA mismatch repair (MMR) system as a potential predictive biomarker of response to checkpoint inhibitors. MMR system abnormalities (i.e., MMR-deficient tumors) are known to drive genomic microsatellite instability that may produce neoantigens that can be recognized and targeted by T cells in several noncolorectal cancer models, including cholangiocarcinoma.45 One potential hurdle for many patients is the high prevalence of primary sclerosing cholangitis, which is associated with 30% of cholangiocarcinomas46 and may be a contraindication to checkpoint blockade.

Conclusions and Future Directions

Immunotherapy stands poised to become a potent weapon in the limited arsenal of HCC therapy if the early success of nivolumab is confirmed in the ongoing pivotal trial.47 Checkpoint blockade may also breathe new life into thus far unsuccessful approaches such as cytotoxic therapies, antiangiogenic agents, vaccines, oncolytic viruses, or cellular therapies. And these combinations could improve the first-ever reported, yet modest, 15% response rate of nivolumab. Success in advanced disease should also prompt additional investigation into the adjuvant setting, which has no proven therapies.

Most important, though, is the potential for immunotherapy to combine with locoregional therapies in early-stage disease to allow for additional avenues toward curative therapy with resection or liver transplant. After nearly a decade of stalled progress, HCC appears primed for a big leap into the modern era of immuno-oncology.

About the Authors: Dr. Karasic is a fellow in hematology and oncology at the University of Pennsylvania. Dr. Loaiza-Bonilla is an assistant professor of clinical medicine at the University of Pennsylvania Perelman School of Medicine and co-leader of the hepatobiliary malignancies program at the Abramson Cancer Center of the University of Pennsylvania.