EBVaGC is one of the four major genomic gastric cancer types defined by The Cancer Genome Atlas (TCGA) in 2014, associated with 8.7% of gastric carcinoma worldwide (8). Compared with EBV-non associated GC (EBVnGC), EBVaGC is characterized by recurrent PIK3CA mutations, extreme DNA hypermethylation, and amplification of JAK2, CD274 (also known as PD-L1) and PDCD1LG2 (also known as PD-L2) according to TCGA subtype. Usually, patients with EBVaGC present a best prognosis for both recurrence-free survival and overall survival, and they seemed to be chemotherapy-resistant and the ORR for first-line chemotherapy were 29.0%, nearly half of that of EBV-negative GC(15). Extensive lymphocyte infiltration and constitutive expression of immunosuppressive molecules PD-L1/PD-L2 in EBVaGC highlight the potential of EBV status as a novel biomarker for immunotherapy efficacy in gastric cancer. In 2018, Kim et al conducted a single-center, phase 2 trial in which 61 unselected patients with metastatic GC received pembrolizumab monotherapy and performed integrated molecular profiling(10). Dramatic responses of EBVaGC to pembrolizumab were observed with an ORR of 100%, but the study only included six patients. In contrast, another four studies showed the ORR was only 16.7%-28.6% for EBVaGC patients received anti-PD-1 antibody alone after first-line therapy (11–14). In this article, we reported the efficacy of ICIs in a case series of metastatic EBVaGC patients in our institute and then performed a pooled analysis. The ORR was 60% for the five cases in our hospital, while 48.6% for all patients in the pooled analysis. Moreover, based on the results of pooled analysis, the ORR of ICIs monotherapy on EBVaGC was better than that of ICIs monotherapy on unselected GC in ATTRACTION-2 trial (46.7% vs 11.9%) 3. These results demonstrated that an EBV positive status was a reliable biomarker for immunotherapy in metastatic gastric cancer. ICIs combined with chemotherapy seemed to have a better response than immunotherapy alone with an ORR of 60% and 46.7% respectively, indicating immunotherapy combined with chemotherapy maybe a better strategy than immunotherapy alone for the first line treatment of metastatic EBVaGC.
Besides EBV status, other potential biomarkers of ICIs used in clinical practice include MSI, TMB, PD-L1(3, 12, 16–19).Microsatellite instability (MSI) status and tumor mutation burden (TMB) have been widely accepted as potential biomarkers for response to anti-PD-1/PD-L1 antibody in many types of cancer. PD-L1 expression was also usually associated with good response rate for immunotherapy in gastric cancer. However, these biomarkers could not completely cover EBV positive gastric cancer, which also benefits from immunotherapy. First, MSI-H is mutually exclusive with EBV positivity (10). Second, majority of EBVaGC are TMB-L (20), such as the patient No.1 in our case series. Third, Although EBVaGC is characterized by marked high expression of PD-L1(12), a substantial proportion of it does not express high levels of PD-L1(14). Moreover, the relationship between PD-L1 expression and efficacy of ICIs was inconsistent across different trials (21), and even PD-L1 positive patients were possible to resistant to immunotherapy. In short, TMB-H, MSI-H, PD-L1 or EBV positive patients would be more likely to benefit from immunotherapy. With the current understanding of cancer immunology, a single parameter may not be sufficient to predict immunotherapy efficacy. Therefore, the future research of biomarker should address the integration of multi-omics biomarkers with an interdisciplinary approach.
It is accepted that ICIs require an endogenous adaptive immune response to be effective and EBVaGC is considered beneficial owing to presumably activated T cells in response to viral antigens (22). However, large heterogeneity exists in the immune contexture of GC even in its subtype; favorable factors and downsides were found in types traditionally insensitive and sensitive to ICIs (23). This kind of heterogeneity partly explains why patients displayed diverse response to ICIs in our cases and it is reasonable to relate efficacy of immunotherapy to tumor immune environment (TME) (24). Researchers proposed a way of categorizing patients under four TME groups with potential implications for mechanism and therapy according to PD-L1 expression and the presence of TILs in tumor biopsies (25), (26). About 65% of EBVaGC cases fell into type I (TILs+ PD-L1+) or type IV (TILs+ PD-L1-) microenvironments (27). In our cases, genetic testing of patient No.2 and No.4 were consistent with this finding since they were type IV and I respectively. Overregulation of activated TILs due to the effect of the B7-H1/PD-1 pathway and potentially other T cell regulatory pathways, dysfunctional TILs in the TME due to suppression by molecular pathways may explain the antitumor immune defect in EBVaGC patients (28). More basic and clinical researches are in need of exploration.
Ongoing clinical trials about ICIs efficacy in GC including EBV positive subtypes are all enumerated in the Table 3. They are phase II or I/II which objective population are not limited to advanced patients. PD-1/PD-L1 antibody plus chemotherapy is the main condition instead of monotherapy; some drugs are designed in combination with CTLA-4 inhibitors or antiangiogenic therapy. We are convinced that with more prospective large-scale clinical tests, a better standard treatment for EBVaGC would be finally found.
Table 3
Ongoing Trials Evaluating Immune Checkpoint Inhibitors efficacy related to EBVaGC
Study identifier | Status | Stage | Drug treatment | Outcome Measures | Phases | Study Designs |
NCT04675866 | Not yet recruiting | Advanced | Camrelizumab+S-1+Albumin-bound paclitaxel | ORR/DCR/OS/PFS | II | Single Group Assignment |
NCT04202601 | Recruiting | Advanced | Sintilimab+IBI310(CTLA-4 antibody) | pCR/ORR | I/II | Single Group Assignment |
NCT04152889 | Recruiting | III | Camrelizumab+S-1+Docetaxel | DFS | II | Single Group Assignment |
NCT03755440 | Enrolling by invitation | Metastatic | SHR-1210(PD-1 antibody) | RR/PFS/OS/DCR | II | Single Group Assignment |
NCT03257163 | Recruiting | IB-IIIC | Capecitabine, Pembrolizumab | RFS rate | II | Single Group Assignment |
NCT04250948 | Recruiting | Locally Advanced | JS001(PD-L1 antibody) +Oxaliplatin+S1+Capecitabine | TRG0/1/pCR/R0 resection rate /RFS/ORR/DCR/OS | II | Parallel Assignment |
NCT04632459 | Not yet recruiting | Metastatic | Pembrolizumab +ramucirumab | ORR | II | Single Group Assignment |
NCT03751761 | Recruiting | Metastatic | durvalumab/tremelimumab (CTLA-4 antibody) +paclitaxel | RR | I/II | Single Group Assignment |
KCT0004186 | Recruiting | Advanced | paclitaxel+nivolumab | MTD; RP2D; PFS | Ib/II | Single Group, Blinding |
RFS, recurrence-free survival; MTD, maximal tolerated dose; RP2D, recommended phase 2 dose; GEJ, gastroesophageal junction; pCR, pathologic complete response |