【摘要】 目的 探讨中性粒细胞-淋巴细胞比值(NLR)与血小板-淋巴细胞比值(PLR)对晚期胃癌患者一线免疫检查点抑制剂(ICIs)临床疗效的预测价值。方法 回顾性收集2017-01-01-2024-06-30在中国医科大学附属第一医院接受一线ICIs联合或不联合化疗的111例晚期胃癌患者的临床资料。主要研究终点为无进展生存期(PFS)和总生存期(OS),次要研究终点包括疾病控制率(DCR)和安全性。采用受试者工作特征曲线(ROC)确定NLR和PLR预测疗效的最佳截断值，Kaplan-Meier法进行生存分析，单因素及多因素Cox回归模型分析预后因素，并构建联合生物标志物模型(高NLR+高PLR)评估其预测效能。结果 全组患者中位PFS为5.0个月(95%CI:3.71～6.29),中位OS为16.0个月(95%CI:12.37～19.60),DCR为89.19%(99/111)。ROC曲线分析确定NLR和PLR预测12个月PFS的最佳截断值分别为2.02和189.78。多因素Cox回归分析显示，高NLR(HR=1.25,95%CI:1.03～1.45,P=0.045)和高PLR(HR=1.01,95%CI:1.01～1.01,P<0.001)是PFS的独立危险因素；高NLR(HR=1.24,95%CI:1.04～1.53,P=0.045)、高PLR(HR=1.01,95%CI:1.01～1.01,P=0.002)及高LDH(HR=1.01,95%CI:1.01～1.01,P=0.034)是OS的独立危险因素。在疗效预测方面，低NLR组(≤1.97)的DCR为96.49%(55/57),高于高NLR组(>1.97)的81.48%(44/54),P=0.011;低PLR组(≤135.32)的DCR为98.04%(50/51),高于高PLR组(>135.32)的81.67%(49/60),P=0.006。安全性方面，33.33%(37/111)的患者出现免疫相关不良事件，其中Ⅰ～Ⅱ级占25.23%(28/111),Ⅲ～Ⅳ级占8.11%(9/111),发生免疫相关不良事件的患者PFS优于未发生者，P=0.006。高危组(高NLR+高PLR)在PFS和OS上均差于低危组，均P<0.05。该联合模型预测12个月PFS的曲线下面积为0.851(95%CI:0.782～0.920),灵敏度为85.0%,特异度为76.3%。结论 在真实世界中，ICIs一线治疗晚期胃癌具有良好的疗效和可控的安全性。治疗前基线高NLR和高PLR是PFS与OS的独立危险因素，且与近期疗效下降显著相关，二者联合构建的模型能有效识别预后不良的高危患者，预测效能良好，是潜在的疗效预测生物标志物。更多还原

【Abstract】 Objective To investigate the predictive value of the neutrophil-to-lymphocyte ratio(NLR) and platelet-to-lymphocyte ratio(PLR) on the clinical efficacy of first-line immune checkpoint inhibitors(ICIs) in patients with advanced gastric cancer.Methods Clinical data of 111 patients with advanced gastric cancer who received first-line ICIs with or without chemotherapy at the First Affiliated Hospital of China Medical University between January 1,2017,and June 30,2024,were retrospectively collected.The primary endpoints were progression-free survival(PFS) and overall survival(OS).The secondary endpoints included disease control rate(DCR) and safety.Receiver operating characteristic(ROC)curve analysis was used to determine the optimal cutoff values of NLR and PLR for predicting efficacy.Survival analysis was performed using the Kaplan-Meier method.Univariate and multivariate Cox regression models were used to analyze prognostic factors.A combined biomarker model(high NLR+high PLR) was constructed accordingly to assess its predictive performance.Results Among the 111 patients with advanced gastric cancer receiving first-line ICIs,the median PFS was 5.0 months(95 % CI:3.71-6.29),the median OS was 16.0 months(95 % CI:12.37-19.60),and the DCR was 89.19%(99/111).ROC curve analysis determined the optimal cutoff values for NLR and PLR to be 2.02 and189.78,respectively.Multivariate Cox regression analysis showed that high baseline NLR(HR=1.25,95%CI:1.03-1.45,P=0.045) and high PLR(HR=1.01,95 % CI:1.01-1.01,P<0.001) were independent risk factors affecting PFS and OS;high NLR(HR=1.24,95%CI:1.04-1.53,P=0.045),high PLR(HR=1.01, 95%CI:1.01-1.01,P=0.002),and high lactate dehydrogenase(HR=1.01,95%CI:1.01-1.01,P=0.034) were independent risk factors for OS.In terms of efficacy prediction,the DCR in the low NLR group(≤1.97) was 96.49%(55/57),which was higher than that in the high NLR group(>1.97) at 81.48%(44/54),P=0.011.The DCR in the low PLR group(≤135.32)was 98.04%(50/51),which was higher than that in the high PLR group(>135.32) at 81.67%(49/60).Regarding safety,33.33%(37/111) of patients experienced immune-related adverse events(irAEs),of which 25.23%(28/111)were grades Ⅰ-Ⅱ and 8.11 %(9/111) were grades Ⅲ-Ⅳ.Patients who experienced irAEs had better PFS than those who did not,P=0.006.The high-risk group(high NLR+high PLR) exhibited significantly worse PFS and OS compared to the low-risk group(both P<0.05).The area under the curve for predicting 12-month PFS by the combined model was 0.851(95%CI:0.782-0.920),with a sensitivity of 85.0% and a specificity of 76.3%.Conclusions In a real-world setting,first-line ICIs demonstrated favorable efficacy and manageable safety in patients with advanced gastric cancer.High baseline NLR and PLR are independent risk factors for PFS and OS and are significantly associated with poorer short-term treatment response.The combined model constructed from these two markers effectively identifies high-risk patients with poor prognosis and shows good predictive performance,representing a potential biomarker for predicting immunotherapy efficacy.更多还原