1 Introduction

Gastric cancer (GC) is the fifth most common cancer; approximately 1,090,000 GC cases are newly diagnosed globally [1]. GC remains in the third palce in number of new cases and mortality, wheras in China most patients are being diagnosed with a locally advanced stage [2]. Combining surgery, chemotherapy, and radiotherapy has been the primary treatment approach for these patients with advanced gastric cancer (AGC). At the same time, the prognosis remains poor, with a median survival of around 12–15 months [3]. Therefore, there is a need for the development and introduction of newer treatment options for the management of this condition. Over the years, treatment strategies for gastric cancer have developed and they have helped to improve the overall prognosis of gastric cancer patients. Immune checkpoint inhibitors (ICIs) have been revealed as a new standard treatment in some malignancies or metastatic tumors with great clinical benefit in several populations [4]. The investigation of combination immunotherapy has emerged as a research focus, particularly in the context of with or without combination of chemotherapy for the treatment of human epidermal growth factor receptor 2 (Her2) positive and Her2 negative advanced gastric cancer (AGC) [5]. This research area is of great interest given the potential benefits that combination therapies may offer, including improved treatment outcomes and reduced toxicity. Overall, the exploration of combination immunotherapy represents a promising area of research for the management of AGC.

2 Curative/perioperative therapeutic model

The Cancer Genome Atlas (TCGA) study identified four molecular subtypes of gastric cancer based on the analysis of somatic copy numbers, whole-exome sequencing (WES), DNA methylation profiling, and microRNA sequencing [6, 7]. These subtypes are microsatellite instability (MSI), Epstein-Barr virus (EBV), genomically stable (GS), and chromosomal instability (CIN) [4]. The TCGA study found that MSI-H tumors have high infiltration of CD8 + T cells, which is attributed to the recognition of a large number of neoantigens and the corresponding expression of immune checkpoints in the tumor microenvironment [8]. The poorly differentiated adenocarcinoma has been defined as EBV-positive tumors [9]. GS tumors typically display diffuse histology and harbor mutations in CDH1 and RHOA or a CLDN18-ARHGAP fusion [10]. CIN tumors are frequently found at the gastroesophageal junction/cardia and are characterized by recurrent TP53 mutations and relatively high amplifications of genes encoding receptor tyrosine kinases (RTKs) [6]. This study suggests that there may be a significant association between shorter progression-free survival (PFS) on first-line cytotoxic chemotherapy in AGC patients with MSI-H/MMR-D. However, it also indicates that subsequent anti-PD-1 therapy may lead to durable responses in these patients. Further research is needed to confirm these findings and explore potential therapeutic strategies for AGC patients with MSI-H/MMR-D [11]. The findings of this study are interesting as they suggest that Asian gastric cancer samples may exhibit higher expression of T cell markers (CD3, CD45R0, and CD8) and lower expression of the immunosuppressive T regulatory cell marker FOXP3 compared to non-Asian tumors. However, further research is needed to confirm these results and investigate potential underlying mechanisms that may contribute to these differences in immune marker expression between Asian and non-Asian gastric cancer samples [12].

The publication of the results of the MAGIC and FNCLCC/FFCD9703 studies established the treatment strategy for advanced gastric cancer in European countries [13]. However, clinical studies on perioperative chemotherapy for gastric cancer were repeatedly frustrated in the Asian population for a long time. Until 2012, the RESOLVE study, initiated by researchers at Peking University Cancer Hospital, a multi-center, three-arm, randomized study, showed that the perioperative SOX regimen significantly improved R0 resection rate, D2 lymph node dissection, and 3-year DFS compared with the postoperative adjuvant chemotherapy group, reducing the risk of tumor recurrence by 23% among the 1094 gastric cancer patients enrolled and followed up. For the Chinese gastric cancer patient population, the SOX regimen is non-inferior to the CAPOX regimen as postoperative adjuvant chemotherapy [14]. The results have established a milestone in China's new neoadjuvant therapy for advanced gastric cancer. They have been included in the CSCO guidelines as level I evidence for recommendation.

The role of perioperative treatment in advanced gastric cancer has already been recognized, while the mode of treatment and drug selection for different populations remain research hotspots. With the refinement of molecular subtypes of gastric cancer, clinical studies on targeted therapy and immunotherapy have gradually been carried out [15]. However, it is worth exploring how to define the benefiting population in the real world accurately. Multi-index and combined predictive models may achieve better predictive performance. Several multi-center, randomized phase II/III clinical trials explore the value and advantages of perioperative immunotherapy combined with chemotherapy, for instance, HLX10-006-GCneo, Keynote585, SHR-1701-III-308, and NCT04354662 [16].

3 Palliative therapy

With the significant breakthrough brought by immunotherapy in gastric cancer treatment, the population of advanced gastric cancer patients has also received unprecedented benefits. Compared to monotherapy, combining anti-PD-1 drugs with chemotherapy is more effective for treating advanced gastric or gastroesophageal junction adenocarcinoma [17]. The CheckMate-649 study showed that the combination of nivolumab and chemotherapy significantly prolonged OS and PFS compared to chemotherapy alone in patients with advanced gastric or gastroesophageal junction adenocarcinoma (CPS score ≥ 5), while no survival benefit from immunotherapy was observed in patients with CPS score ≥ 1 [18]. Based on the above results, nivolumab was approved as the first PD-1 inhibitor for first-line treatment of advanced or metastatic gastric cancer, gastroesophageal junction adenocarcinoma in China in 2021.The latest results of the clinical trial for first-line treatment of advanced gastric or gastroesophageal junction adenocarcinoma with the AK104 combination chemotherapy were presented at ASCO GI 2022 which enrolled a total of 98 gastric cancer patients as of December 2021. The ORR reached 68.1%, while the DCR was 92.3%. The median PFS of the overall population was 7.2 months, and the median OS was 17.08 months, showing survival benefits over chemotherapy and PD-1 monotherapy. Meanwhile, the incidence of grade 3 or higher drug-related adverse reactions was 62.5%, and severe adverse reactions were 41.7%.No significant difference in adverse events was observed across the varying doses of 4 mg/kg, 6 mg/kg, 10 mg/kg, and 15 mg/kg, and no additional safety concerns were identified. Based on the excellent Safety and efficacy data in stage Ib/II, combining anti-PD-1 and anti-CTLA-4 may provide better treatment options for advanced gastric cancer patients [19]. The KEYNOTE-590 clinical trial, reported at ESMO 2020, showed that in 749 patients with locally advanced or metastasized squamous cell carcinoma of the esophagus (PEC, n = 73%) and adenocarcinoma of the gastroesophageal junction (n = 25%, Siewert type 1), patients receiving pembrolizumab plus chemotherapy (cisplatin, 5-FU) had a superior overall survival (OS) compared to patients receiving chemotherapy alone, regardless of CPS score and tumor histology. The combination therapy showed a median OS of 12.4 vs. 9.8 months (HR 0.73 (95% CI 0.62–0.86), p < 0.0002) and median PFS of 6.3 vs. 5.9 months (HR 0.65 (95% CI 0.55–0.76)) in all patients. Squamous cell and adenocarcinoma patients with CPS ≥ 10 especially benefited from the combination therapy (PEC: median OS 13.9 vs. 8.8 months, HR 0.57 (95% CI 0.43–0.75); adenocarcinoma: median OS 12.1 vs. 10.7 months, HR 0.83 (95% CI 0.52–1.34)). The objective response rate (ORR) was 45% in the combination of immune and chemotherapy (95% CI, 40–40) versus 29% (95% CI, 25–34) in the chemotherapy alone [20].

In summary, in European patients with CPS ≥ 5, the immune checkpoint inhibitor nivolumab combined with chemotherapy (platin/fluoropyrimidine) has been established as an approved feasible immune therapy for first-line therapy adenocarcinoma of the stomach and gastroesophageal junction. In contrast, in combination with chemotherapy, pembrolizumab is approved for EGC patients with CPS ≥ 10. In the USA, the FDA has approved the combination of pembrolizumab, trastuzumab, and chemotherapy in a first-line setting for patients with Her2-positive EGC based on an interim analysis of the KEYNOTE-811 trial.

Currently, drug therapy for gastric cancer mainly includes chemotherapy, molecular targeted therapy, and immune checkpoint inhibitors in China. Chemotherapy has strong evidence-based medicine support and rich clinical experience [21]. There are numerous studies on gastric cancer targeted therapy, and currently approved indications in China are limited to anti-HER2 drugs trastuzumab and pertuzumab, as well as some anti-angiogenic pathway drugs. Immune therapy has made groundbreaking progress in treating advanced gastric cancer, with the PD-1 monoclonal antibody approved for third-line treatment. However, monotherapy with immune therapy has suboptimal efficacy, and PD-1 monoclonal antibody combined with chemotherapy has become the new first-line standard for treating advanced metastatic gastric cancer. Gastric cancer exhibits strong spatiotemporal heterogeneity and a complex tumor microenvironment. There are differences between the gastric cancer populations of Estern countries and Wester countries regarding epidemiological characteristics, clinical and pathological features, biological behavior, treatment modes, and drug selection. Therefore, more large-scale multi-center clinical research data is required to provide high-level evidence-based medicine.

4 Molecular biomarkers in advanced gastric cancer

According to the NCCN Clinical Practice Guidelines in Gastric Cancer, HER2 overexpression status has been established as a predictive biomarker for treating advanced gastric cancer. Further investigation of other biomarkers is needed to identify molecular characteristics of patient subgroups and to find individualized anti-tumor treatment beyond conventional chemotherapy.

4.1 PD-L1 CPS

PD-L1 CPS includes the total number of positively stained tumor cells and mononuclear immune cells (including lymphocytes, macrophages, and dendritic cells, with simultaneous consideration of membranous and cytoplasmic staining) divided by the total population of viable tumor cells [22]. In gastric cancer, a PD-L1 CPS score of ≥ 1 is generally considered to define PD-L1-positive tumors. The optimal PD-L1 CPS score for predicting the efficacy of immunotherapy is still under investigation in clinical trials, with cut-off values ranging from CPS ≥ 1/CPS ≥ 5 to CPS ≥ 10. Pembrolizumab trials have demonstrated a significant benefit for patients with a CPS score of ≥ 10. The current understanding is that the benefit level and prolongation of overall survival with immunotherapy depend on the amount of PD-L1 CPS positivity. Specifically, the higher the CPS score, the greater the benefit in terms of overall survival. This has been confirmed in the data from the CHECKMATE-649 trial (higher benefit in overall survival with CPS ≥ 5 compared to CPS ≥ 1) and the KEYNOTE-062/061 trials (higher benefit in overall survival with CPS ≥ 10 compared to CPS ≥ 1). A recent analysis of selected clinical trials (KEYNOTE-059, KEYNOTE-061, KEYNOTE-062) administering pembrolizumab in patients with CPS ≥ 10 further supports these findings [23].

4.2 MSI

As mentioned above, the microsatellite instability (MSI) subgroup of gastric cancer patients resembles one of the four molecular subgroups proposed by the TCGA, along with chromosomally unstable tumors, Epstein-Barr virus-infected tumors, and genomically stable tumors. A meta-analysis was conducted to compare randomized trials investigating the efficacy of treatment with or without a PD-1 inhibitor for advanced gastric cancer, and reported outcomes based on MSI status (including KEYNOTE-062, KEYNOTE-061, CHECKMATE-649, JAVELIN Gastric 100). Data were selectively analyzed based on the role of MSI in response to immunotherapy. A total of 2545 patients with evaluable MSI status were included, of whom 123 were MSI-high GC patients (4.8%). In MSI-high tumors, the hazard ratio (HR) for OS benefit with anti-PD-1 therapy was 0.34 (95% CI 0.21–0.54), compared to 0.85 (95% CI 0.71–1.00) for MSS (microsatellite stable) tumors. The HR for PFS in MSI-high tumors was 0.57 (95% CI 0.33–0.97, p = 0.04), and the odds ratio for response was 1.76 (95% CI 1.10–2.83, p = 0.02) [24]. The results of this meta-analysis provide further support for the hypothesis that MSI-high gastric cancer patients constitute a highly immune-sensitive population that responds particularly well to immunotherapy. The following clinical question that needs to be addressed is the further evaluation of the therapeutic efficacy of single or combination immunotherapy in patients with high levels of microsatellite instability or deficient mismatch repair (dMMR).

The GERCOR NEONIPIGA phase II trial investigated the efficacy of neoadjuvant immune checkpoint inhibition with nivolumab plus ipilimumab and adjuvant nivolumab in localized MSI-high/dMMR EGC patients, with the primary endpoint being pathological complete response rate (pCRR). The first results, presented at ASCO GI 2022, showed that neoadjuvant immune checkpoint inhibition is a feasible therapeutic option, with a pCRR of 59% (17/29 patients) and 94% of patients being free of gastric cancer related events after 12 months of follow-up [25]. Considering the research data mentioned above, we strongly recommend the routine testing of MSI status, including PD-L1 CPS, in gastric cancer patients before treatment. Moreover, we need more high-quality clinical studies to support the incorporation of immunotherapy into perioperative treatment strategies for gastric cancer, particularly for patients with MSI-high/dMMR tumors.

4.3 EBV

Based on available data, it has been observed that latent Epstein-Barr virus (EBV) infection is found in approximately 9% of patients diagnosed with gastric adenocarcinoma [26]. This finding is important as it provides insights into the potential underlying causes of this type of cancer and may have implications for the development of targeted therapies. The expression of immune checkpoint genes, such as PD-1 or CTLA-4, and the level of lymphocytic infiltration are higher in GC patients with EBV-positive tumors than in those with MSS (microsatellite stable) tumors [22]. A clinical study conducted by the Asian Cancer Research Group (ACRG) involving 300 gastric cancer patients revealed that a PD-L1 CPS score of ≥ 1 (59.3%, 178 patients) was significantly associated with MSI-high tumors and positive EBV status. These results suggest that gastric cancer patients with a positive EBV status are likely to benefit considerably from immunotherapy [27].

5 Conclusions

Recent pivotal clinical trials have shown that combining anti-PD-1 antibodies and conventional agents demonstrates clinical activity for HER2-negative and HER2-positive advanced gastric cancer (AGC). This finding is significant as it offers a potential treatment option for patients with advanced gastric cancer who have previously had limited treatment choices. This has led to a change in the standard of care for first-line treatment of AGC. However, the efficacy of nivolumab plus chemotherapy in HER2-negative AGC with higher PD-L1 expression (CPS ≥ 5) remains unclear. It is also unknown whether this combination can be adopted irrespective of PD-L1 CPS or only for the CPS ≥ 5 population. While the FDA and regulatory agencies in Asian countries have approved the combination of nivolumab and chemotherapy irrespective of PD-L1 CPS, the EMA and NCCN guidelines have approved or recommended it for AGC with CPS ≥ 5. Investigation of PD-L1 CPS and MSI status is necessary, as these biomarkers are associated with clinical outcomes. Chemotherapy alone should be considered an option for patients with CPS < 5, especially those with autoimmune disease, poor general condition, and lack of family support, due to increased treatment-related adverse events (TRAEs) in anti-PD-1 antibodies plus chemotherapy compared with conventional agents. Given the limited number of patients who have benefited from immune checkpoint inhibitors (ICIs), developing new immunotherapy approaches is urgently needed. Early clinical trials have shown promise in combining ICIs with other targeted agents, including multikinase inhibitors and CLDN18.2-specific CAR-T cell therapies, for the treatment of certain cancers. These findings suggest that combination therapies may hold potential for improving patient outcomes and provide a basis for further evaluation in subsequent studies. However, further research is needed to fully evaluate the safety and efficacy of these treatment approaches. If future studies confirm these early results, this could represent a significant step forward in the development of more effective and personalized cancer treatments.