Safety and clinical efficacy of immune checkpoint inhibitors in advanced gastric cancer in the real world

According to the data related to 2020, gastric cancer is the fifth most common cancer worldwide and the fourth most common cause of cancer deaths (Sung et al. 2021). China has one of the highest incidence rates of gastric cancer (Smyth et al. 2020). Palliative chemotherapy is the corner-stone of treatment for patients with advanced gastric cancer who have a poor prognosis with a median overall survival of 10 to 12 months (Digklia and Wagner 2016). However, the success of immune checkpoint therapies targeting cytotoxic T-lymphoid-associated protein-4 (CTLA-4), programmed death receptor-1 (PD-1), and programmed death-ligand 1 (PD-L1) has led to a breakthrough in tumor immunotherapy (Li et al. 2019), numerous clinical trials have also successfully revealed that PD-1/PD-L1 immune checkpoint inhibitors exhibit favorable therapeutic effects (Wang et al. 2024).

Relevant trials on immunotherapy have been conducted in the clinic, confirming that immunotherapy for gastric cancer has excellent therapeutic prospects. One is based on the ATTRACTION-2 trial, in which nivolumab has been approved in Japan for the third-line treatment of recurrent or metastatic adenocarcinoma of the stomach or gastroesophageal–geal junction (Allemani et al. 2018). Another anti-PD-1 monoclonal antibody, pembrolizumab, was previously approved by the US Food and Drug Administration (FDA) for the treatment of gastric cancer (PD-L1-positive) in the third or backline, based on the results of a significant phase II trial (Kawazoe et al. 2021). In addition, based on the results of the CheckMate 649, the first-line trial showed that nivolumab, in combination with the chemotherapy group, also better prolonged OS and PFS in patients with a CPS of 5 (Moehler et al. 2020). All these indicate that this immunotherapy is promising and will bring new hope to patients with advanced gastric cancer.

Rashes, colitis, hepatitis, hypo-or hyperthyroidism, and pneumonia are common irAEs (Friedman et al. 2016). The prevalence of irAEs worldwide may differ due to various immunotherapy studies, and there needs to be more epidemiologic data on irAEs and other influencing factors in China. The study examined the clinical features, incidence, treatment approaches, and the effectiveness and Prognosis of ICIs in actual adverse events (irAEs) in the Chinese population.

Immunotherapy can produce unique inflammatory toxicity while activating the immune system, which is called immune-related adverse effects (Yu et al. 2021). The incidence of irAEs has been reported in the literature to be 55%-60% (Wang et al. 2018), and the incidence of irAEs of different types and degrees occurred in 191 (47.51%) patients in the present study, which is not significantly different from that reported in the literature. In the CheckMate 649 study, the nivolumab combination chemotherapy group for advanced gastric cancer/gastroesophageal junction cancer/esophageal adenocarcinoma showed the most common adverse reactions were nausea (39%), diarrhea (28%), peripheral neuropathy (24%), and vomiting (23%) (Janjigian et al. 2021) In the phase I study, NCT02937116 trial, sintilimab in combination with CapeOx for the treatment of adenocarcinoma of the gastric/gastroesophageal junction produced the main adverse reactions associated with rash (25%) and hypothyroidism (30%) (Jiang et al. 2020), and in the present study were immune-related skin toxicity (23.9%), immune-related endocrine toxicity (23.4%), and immune-related gastrointestinal toxicity (19.9%) was higher in this study, and the main manifestation of immune-related skin toxicity was rash, immune-related endocrine toxicity was hypothyroidism, and immune-related gastrointestinal toxicity was diarrhea, which was not much different from that reported in literature. In clinical work, we will focus on common adverse reactions such as diarrhea, vomiting, and other gastrointestinal reactions, while rare adverse reactions such as neurological toxicity and cardiotoxicity are mainly judged based on the laboratory tests and examination indicators. Some of the safety differences may be related to various factors such as the quality of retrospective data, sample size, physical condition of patients, dosage of medication, and combination therapy programs. There was relevant literature about ICIs-related irAEs reported a median onset time of 42 days (Chen et al. 2021), and the median onset time of toxicity of irAEs in the present study was 49 days, with which the results of the present study are generally consistent. The immunotoxicity in this study was mainly concentrated in grades 1–2, similar to that reported in the literature (Pauken et al. 2019; Postow et al. 2018). In the real world, irAEs are largely relieved with relevant treatment and symptoms. The occurrence of different PD-1 irAEs varies, and a larger sample is needed for analysis.

Studies have demonstrated that the sequence of targeted therapy versus immunotherapy is associated with the incidence of irAEs: one study included 126 NSCLC patients with EGFR mutations who were treated with ICI and EGFR TKIs. Treatment with ICI followed by EGFR TKIs was associated with the emergence of irAEs, but treatment with EGFR TKIs before ICI was not related to the emergence of irAEs (Schoenfeld et al. 2019). While in this paper, prior targeted therapy (OR: 4.360, 95% CI 1.692–11.238) was a risk factor influencing the occurrence of irAEs, the incidence of irAEs was higher in patients with advanced gastric cancer who had received prior targeted therapy before receiving immunotherapy. We already know that targeted therapy sequenced with immunotherapy not only has an impact on efficacy (one study showed that the addition of immunotherapy to targeted therapy improved the prognostic outcome of patients with HER2-positive gastric cancer (Killock 2022). It may also have an impact on the emergence of irAEs. This result is associated with the study with less literature and also needs to do prospective experiments to support it.

In this study, we analyzed the association between the occurrence of irAEs and hypertension. Subgroup analyses showed that the presence or absence of hypertension in oncology patients had a significant between-group difference in the occurrence of irAEs (P = 0.008), and multifactorial logistic regression analyses further revealed a higher risk for the occurrence of irAEs in patients suffering from hypertension (OR: 2.759, 95% CI 1.206-6.310), indicating a positive correlation between hypertension and the occurrence of irAEs. In addition, univariate COX regression analysis in hypertension and patients’ prognosis pointed out that patients suffering from hypertension had a lower risk of death as compared to controls (HR: 0.594, 95% CI 0.418–0.844). And multifactorial COX regression analysis showed that hypertension was an independent risk factor for the prognosis of patients with gastric cancer, in which patients with hypertension had a lower risk of death than the control group (HR: 0.680, 95% CI 0.474–0.973), and these differences were statistically significant (P-values were all less than 0.05). In another metabolic disease, diabetes, a significant between-group difference in the occurrence of irAEs was similarly found. Although there is no direct evidence in the literature we reviewed that elucidates the relationship between hypertension, diabetes, and immune efficacy, there are studies that point to the fact that hypertension and diabetes do have an impact on the immune system. For example, certain studies have revealed interactions between hypertension, cytokines, and T cells (Guzik et al. 2007; Singh et al. 2014). In addition, some studies have shown that the combination of antihypertensive drugs and immune checkpoint blockers can enhance tumor response and prolong survival in mouse models of breast cancer. This is because cancer cells, in order to survive and proliferate in the body, construct a microenvironment conducive to their growth by compressing blood vessels, altering the body’s pH, and creating a hypoxic environment, which in turn suppresses the function of immune cells. Antihypertensive drugs have the potential to destroy this high-pressure environment created by cancer cells (Chauhan et al. 2019). Also, it has been demonstrated that oncology patients with hypertension are more prone to cardiac irAEs (Chennamadhavuni et al. 2022), while hyperglycemic states similarly affect immune cell function (Daryabor et al. 2020). A study in patients with type 2 diabetes (as compared to nondiabetic patients) also found that PD-1 expression was reduced on peripheral T cells in patients with type 2 diabetes, which may affect the effectiveness of immune checkpoint inhibitors (Sun et al. 2019). In addition, drugs used to treat metabolic diseases, such as metformin, have been shown to have anti-inflammatory effects (Cameron et al. 2016). These findings suggest that hypertension and diabetes may affect the expression of irAEs and the efficacy of immunotherapy. Although these data provide some theoretical support, more prospective studies are needed to further validate these preliminary observations in order to determine the exact relationship between these chronic diseases and immunological efficacy.

Several studies have shown that immunotherapy has become a new option for the treatment of advanced gastric cancer. A CheckMate 649-based study evaluated the efficacy of nivolumab in combination with chemotherapy versus chemotherapy alone in gastric and esophageal adenocarcinomas and showed a 44% reduction in the risk of death in the nabumab-combination-chemotherapy group as compared to the chemotherapy group in patients with a PD-L1 CPS ≥ 5. In all randomized patients, the risk of death was reduced by 38% in the nivolumab combination chemotherapy group (mOS: 14.3 months vs.10.3 months; HR: 0.620) (Janjigian et al. 2018). An ORIENT-16-based assessment of the efficacy of the sintilimab combination chemotherapy group versus chemotherapy alone in patients with locally advanced or metastatic gastro-gastric/gastro-esophageal junction adenocarcinoma demonstrated a significant reduction in the risk of death by 38% in the sintilimab combination chemotherapy group (mOS) for all patients (15.2 months vs. 12.3 months; HR: 0.766); and a reduction in the risk of mortality by 38% in the sintilimab combination chemotherapy group (mOS: 14.3 months vs. 10.3 months; HR: 0.620) in the sintilimab combination chemotherapy group (mPFS: 7.1 months vs. 5.7 months; HR: 0.636) (Xu et al. 2021). In addition, there is the ATTRACTION 4 trial, an Asian population study in patients with HER2-negative, advanced, or recurrent gastric/esophagogastric union cancer, which was conducted in Japan, South Korea, and Taiwan, China. The trial evaluated the efficacy and safety of nivolumab in combination with chemotherapy (SOX or XELOX) versus chemotherapy alone as a first-line treatment. The results showed that first-line treatment with nivolumab in combination with chemotherapy had a mPFS of 10.5 months and a mOS of 17.5 months (Kang et al. 2022). One study summarized the survival time of pembrolizumab monotherapy in Japanese patients, in which, in KEYNOTE-062, the mOS was 20 months in the pembrolizumab group and 18 months in the chemotherapy group (HR: 0.760, 95% CI 0.43–1.33). mPFS (pembrolizumab group vs. chemotherapy group) was 6 vs 7 months (HR: 1.030, 95% CI 0.61–1.74); ORR was 29% vs 34%, respectively (Muro et al. 2023). In our study, the mPFS in the immunotherapy first-line irAEs group was 188 days (95% CI 136.193–239.807), and the mOS was 605 days (95% CI 516.492–693.510). The mPFS in the immunotherapy first-line non-iraes group was 152 days (95% CI 111.733–192.227), and the mOS was 416 days (95% CI 357.866–474.136). The overall immunotherapy efficacy of mOS was 572 days (95% CI 474.71–579.290) and mPFS was 159 days (131.711–186.289). In our study, OS was within the range of prognostic outcomes of immunotherapy. The study design was based on the patient population that met the standard entry criteria to ensure the reliability and comparability of the results. However, there may be some variability in OS due to the practicalities of individualized combination therapy (e.g., in the real world, immunotherapy is often used in conjunction with chemotherapy, targeted therapy, and radiation), as well as differences in clinical patient status and subgroup analysis of clinical patients.

Clinical data shows that immunologic efficacy is reduced in patients with liver metastases (Lee et al. 2022), and Tumeh et al. (2015) first observed a weaker response to PD-1 blockade in patients with liver metastases in 2015. In our clinical work, we have observed that ICIs are less effective in patients with liver metastases. In the present study, we found that patients with liver metastases exhibited a trend toward progression with an elevated risk of death (OS: HR: 6.110, 95% CI 3.354–11.131) (PFS: HR: 3.031, 95% CI 1.755–5.236), while there was little correlation with other metastatic pathways situations.

It has been demonstrated that the higher the ECOG score, the worse the physical status of the patients and the worse the efficacy of immunotherapy (Dall'Olio et al. 2020); our study found that ECOG score ≥ 2 (HR: 1.885, 95% CI 1.061–3.347) The higher risk of death when compared with the control group was still negatively correlated with the prolongation of OS also confirmed this view. Another relevant experimental study demonstrated that VEGFR-2 antibody treatment led to intratumoral immunomodulation and enhanced the antitumor efficacy of PD-L1 blockers in a homozygous mouse tumor model (Li et al. 2022). Our study found that VEGFR2-positive (HR: 0.537, 95% CI 0.313–0.921) had a lower risk of progression as compared to controls, which was still positively associated with prolonged PFS, possibly due to patients receiving VEGFR-2 antibody therapy.

A positive correlation between irAEs and the development of ORR, PFS, and OS in patients treated with ICIs has been reported in the literature (Hussaini et al. 2021; Zhao et al. 2022). The results of this paper showed that the occurrence of irAEs was associated with patients’ OS. The occurrence of irAEs could prolong patients’ OS. Since most of the real-world patients with advanced gastric cancer were treated with first-line therapy, patients’ overall immunologic efficacy was poor, and the sample size of patients treated with backline therapy was small, we did not observe that the occurrence of irAEs was associated with patients’ PFS. Exploring the correlation between irAEs and disease prognosis may be because one of the main reasons for the occurrence of irAEs in ICIs-treated patients is the similarity between antigens presented by tumor cells and normal cells (Hasan Ali et al. 2016), and antigen sharing or cross-reactivity may lead to T cell-mediated responses not only to tumor cells but also to normal cells, and the activated immune system may also target these nontumor sites (Hasan Ali et al. 2016). However, the occurrence of fewer irAEs may lack clinical efficacy, and the occurrence of more irAEs may predict clinical efficacy, which may also create new dilemmas for clinical treatment, and the monitoring of irAEs, as well as their timely treatment, becomes more critical.

The mechanism of the occurrence of irAEs needs to be further investigated by researchers, and whether the occurrence of irAEs can be used as an alternative marker for ICIs needs to be verified in prospective experiments. irAEs treatment is still relatively unitary, and although the safety of irAEs is controllable, serious irAEs can be life-threatening, so the clinical practice needs to manage and control irAEs promptly. Targeting TGF-β and PD-L1 with a bispecific antibody for synergistic cancer immunotherapy. These modified antibodies have shown strong antitumor effectiveness in preclinical and clinical trials, outperforming anti-PD-1/PD-L1 monotherapies (Niu et al. 2023). Recently, Merck announced the discovery of M7824, a bifunctional antibody that suppresses TGF-β and PD-L1 at the same time (Lan et al. 2018). M7824 combines the PD-L1 antibody with a trap structure targeting TGF-β, acting as a neutralizing receptor for TGF-β. Phase I clinical studies show that M7824 therapy has tolerable side effects and has shown therapeutic efficacy in a variety of cancer types (Strauss et al. 2018). Later, other bispecific antibodies (BsAbs) are created; they include YM101 and BiTP, which show strong antitumor effects in preclinical and clinical research (Yi et al. 2022, 2021). Further exploratory analysis can also be conducted on this in the future.

In summary, the incidence of irAEs in the real world is still relatively common, but the safety can be controlled with treatment. By retrospectively analyzing the data of patients with advanced gastric cancer treated with ICIs previously admitted to our medical center, we found some clinical characteristic factors associated with the occurrence of irAEs as well as the efficacy and prognosis: the presence or absence of hypertension, whether or not to receive targeted therapies can predict the occurrence of irAEs, and the more the presence of irAEs, the better the prognosis. These can help clinicians in clinical drug selection. To obtain more accurate data, it is necessary to expand the sample size further or conduct prospective experiments.