目的·探索应用程序性死亡蛋白-1（programmed death-1，PD-1）抑制剂行免疫治疗的进展期胃癌患者的免疫相关不良反应（immune-related adverse events，irAEs）的特征及预测因素，并分析irAEs与患者预后的相关性。方法·选择2018年6月—2021年10月于上海交通大学医学院附属仁济医院应用PD-1抑制剂治疗的进展期胃癌患者140例。根据患者有无irAEs发生，将其分为irAEs组和非irAEs组。收集并分析2组患者的临床特征、irAEs的表现及预后情况。采用多因素Logistic回归模型分析影响irAEs发生的相关因素，并建立irAEs的预测模型。运用受试者操作特征曲线（receiver operating characteristic curve，ROC curve，ROC曲线）对不同指标预测irAEs发生的能力进行评估。采用Kaplan-Meier生存曲线分析irAEs与预后的相关性。运用Cox比例风险模型分析影响患者预后的相关因素。结果·共计132例患者完成随访，其中有63例（47.7%）患者发生irAEs。比较2组患者的临床特征的结果显示，年龄≥65岁、Ki-67指数、白细胞计数、中性粒细胞计数、调节性T细胞（regulatory T cell，Treg）计数的组间差异具有统计学意义（均P<0.05）。多因素Logistic回归分析显示，Treg计数为影响irAEs发生的保护因素（P=0.030）。ROC曲线提示，Treg+Ki-67+年龄（≥65岁）联合指标可较好地预测irAEs的发生（AUC=0.753，95% CI 0.623~0.848，P=0.000）。Kaplan-Meier生存曲线的结果显示，irAEs组患者的无进展生存期（progression-free survival，PFS）较非irAEs组有所延长（P=0.001）。Cox比例风险回归分析提示，irAEs是患者PFS的独立影响因素（P=0.006）。结论·Treg计数是行PD-1抑制剂免疫治疗的进展期胃癌患者发生irAEs的独立影响因素，且irAEs的发生可延长患者的PFS；Treg+Ki-67+年龄（≥65岁）联合指标可对该不良反应的发生进行较好的预测。

Objective ·To explore the characteristics and predictive factors of immune-related adverse events (irAEs) in advanced gastric cancer patients treated with programmed death-1 (PD-1) inhibitors, and analyze the correlation between irAEs and prognosis. Methods ·A total of 140 patients with advanced gastric cancer treated with PD-1 inhibitors in Renji Hospital, Shanghai Jiao Tong University School of Medicine from June 2018 to October 2021 were selected. They were divided into irAEs group and non-irAEs group according to whether they had irAEs or not. The clinical characteristics, irAEs occurrence and prognosis of the patients in the two groups were collected and analyzed. Multivariate Logistic regression model was used to analyze the related factors affecting the occurrence of irAEs, and the prediction model of irAEs was established. The receiver operating characteristic curve (ROC curve) was used to evaluate the ability of different indicators to predict the occurrence of irAEs. Kaplan-Meier survival curve was used to analyze the correlation between irAEs and prognosis. Cox proportional hazard model was used to analyze the related factors affecting the prognosis of patients. Results ·A total of 132 patients completed the follow-up, of which 63 patients (47.7%) developed irAEs. In the comparison of clinical characteristics between the groups, it was found that there were statistically significant differences in age≥65 years, Ki-67 index, leukocyte count, neutrophil count and regulatory T cell (Treg) count (all P<0.05). Multivariate Logistic regression analysis showed that Treg count was the protective factor for affecting the occurrence of irAEs (P=0.030). ROC curve showed that the combined index of Treg+Ki-67+Age (≥65 years) could better predict the occurrence of irAEs (AUC=0.753, 95% CI 0.623?0.848, P=0.000). Kaplan-Meier survival curve showed that the progression-free survival (PFS) of patients in the irAEs group was longer than that in the non-irAEs group (P=0.001). Cox proportional hazard regression analysis showed that irAEs was an independent influencing factor of PFS (P=0.006). Conclusion ·Treg count is an independent influencing factor for the occurrence of irAEs in patients with advanced gastric cancer undergoing PD-1 inhibitor immunotherapy, and its occurrence can prolong the PFS of patients. The combined index of Treg+Ki-67+Age (≥65 years) can better predict the occurrence of irAEs.