IMMUNOTHERAPY FOR EBV+ GASTRIC CANCER

Gastric adenocarcinoma is the second leading cause of cancer death worldwide, following lung cancer. The five-year survival rate of advanced metastatic disease is low. Combination chemotherapy results in improvement in overall survival in comparison with single-agent treatment or supportive care.^[1,2] Limited progress has been made by adding targeted therapy to the treatment of gastric cancer. The only two targeted agents approved by the US Food and Drug Administration (FDA) are ramucirumab,^[3,4] a VEGF antibody, and trastuzumab,^[5] which is indicated only in tumors with amplified human epidermal growth factor receptor 2 (Her-2). Both drugs have a modest impact on overall survival.

For the last 10 years, immunotherapy has made a clinically significant impact on the treatment of several different types of cancer. Several classes of drug have been approved for the treatment of melanoma, renal cell carcinoma, non–small cell lung cancer, bladder cancer, and head and neck squamous cell carcinoma, among others. Antibodies blocking the immune checkpoint cytotoxic T lymphocyte antigen–4 (CTLA-4), programmed death–1 (PD-1), and programmed death–1–ligand 1 (PD-L1) antigens have been tested in several gastrointestinal malignancies. Although promising, no immunotherapy drugs have been approved by the FDA to treat gastrointestinal malignancies outside of the recent approval of pembrolizumab for the treatment of patients with colon cancer with microsatellite instability (MSI).^[6]

Anti-PD-1 or PD-L1 antibodies alone or in combination with CTLA-4 were shown to induce objective responses in approximately 10% to 20% of patients with gastric and gastroesophageal adenocarcinoma in small nonrandomized trials.^[7,8] Many of the responses were durable.

In the recently presented study, KEYNOTE-059,^[9] 259 patients with advanced refractory gastric cancer received pembrolizumab 200 mg every three weeks until progression. The overall response rate was 11.2% (95% confidence interval [CI] = 7.6% to 15.7%) in the entire cohort and 15.5% (95% CI = 10.1 to 22.4) in PD-L1-positive patients vs 5.5% (95% CI = 2.0% to 11.6%) in PD-L1-negative patients.

The first randomized phase III trial to investigate a checkpoint inhibitor in gastric cancer was the ONO-4358 study.^[9] In this placebo controlled, double-blind trial, nivolumab reduced the risk of death by 37% (hazard ratio [HR] = 0.63, 95% CI = 0.50 to 0.78, P < .0001) in patients with previously treated advanced gastric cancer compared with placebo; however, median progression-free survival (PFS) was 1.61 months with nivolumab vs 1.45 months with placebo (HR = 0.60, 95% CI = 0.49 to 0.75, P < .0001).

Identifying potential biomarkers of response to immunotherapy in different malignancies including gastric cancer has been a daunting task. Expression of PD-L1 in tumors is a predictive biomarker that enriches for clinical responsiveness; however, it is far from ideal because PD-L1-negative tumors may respond to immunotherapy and positive tumors may not.

It has been also hypothesized that mismatch repair–deficient/microsatellite-unstable tumors are more responsive to PD-L1 blockade. In a phase II study of pembrolizumab in patients with microsatellite-high (MSI-H) solid tumors, the objective response rate was 39.6% (95% CI = 31.7% to 47.9%), including 7.4% complete responses (CRs) and 32.2% partial responses (PRs).^[10] Based partly on this study, pembrolizumab was recently approved by the FDA for treatment of advanced MSI-H tumors. Other factors have been suggested to correlate with response to immunotherapy based on clinical and preclinical studies, including elevated type 1 helper T cell (Th1) tumoral gene signature,^[11] and IFN-γ expression, possibly attributable to its ties to other proteins and cytokines within the tumor microenvironment.^[12]

Viral infections have been implicated in the pathogenesis of several malignancies including head and neck malignancies, among others. Epstein-Barr virus (EBV)–positive gastric carcinoma represents one of the four subtypes of gastric carcinoma recently defined by The Cancer Genome Atlas. EBV-positive tumors have a higher frequency of Crohn's disease–like lymphoid reaction that correlates with an improved prognosis in patients with gastric cancer.^[13]

In this issue of the Journal, Panda et al. report a case of a patient with metastatic gastric cancer with a meaningful response to the anti–programmed death–ligand 1 (PD-L1) antibody avelumab.^[14] Interestingly, the tumor treated by the authors showed low mutation burden and had proficient mismatch repair phenotype, but was positive for EBV. PD-L1 expression was high.

The authors go on to suggest, from their single patient report, as well as review of The Cancer Genome Atlas, that microsatellite-stable (MSS) tumors that are EBV-positive are a subset of patients who might benefit from checkpoint inhibitors. These findings are yet to be confirmed in prospective clinical trials.

The authors confirm that despite the success of checkpoint inhibitors in treating several cancer types, identifying a population subset that is most likely to benefit from immunotherapy is challenging. The complex nature of the interaction between cancer and the immune system makes it difficult to identify a single predictive biomarker.

EBV infection has been implicated in the pathogenesis of a subset of gastric cancer; EBV-associated tumors represent a distinct biological and clinical entity. The benefits of immunotherapy in this subset of patients are still not clear. In the head and neck squamous cell carcinoma population, responses in patients treated with checkpoint inhibitors were observed in both human papillomavirus–positive (HPV+) and HPV– patients, but overall survival was better in HPV+ patients.^[15] Data on other virus-associated tumors such as anal cancer and hepatocellular carcinoma have not been conclusive.

Other factors including validation and standardization of any biomarker, tumor heterogeneity, and technical challenges associated with more advanced signature like (IFN-γ) related mRNA-based signatures and T cell receptor sequencing should be taken into account as well. Well-designed personalized and randomized trials are still needed.