炎症细胞和临床特征对免疫治疗非小细胞肺癌预后的预测价值研究

Predictive value of inflammatory cells and clinical features in prognosis for non-small cell lung cancer immunotherapy

目的分析炎症细胞和临床特征对免疫检查点抑制剂(ICIs)治疗非小细胞肺癌(NSCLC)预后的预测价值。方法收集2017年1月1日至2022年12月31日在该院行ICIs治疗的163例Ⅲ期和Ⅳ期NSCLC患者资料。治疗6~8周进行CT检查,根据预后情况将患者分为客观缓解组[完全缓解(CR)+部分缓解(PR)患者]和非客观缓解组[疾病稳定(SD)+疾病进展(PD)患者]、疾病控制组[CR+PR+SD患者]和非疾病控制组[PD患者]、持久临床获益(DCB)组和非DCB组。比较不同组间的临床特征和炎症细胞指标的差异。采用受试者工作特征(ROC)曲线评价炎症细胞指标对DCB的预测效能。无进展生存(PFS)时间、总生存(OS)时间的影响因素分析采用Cox回归分析。结果疾病控制组淋巴细胞计数(ALC)高于非疾病控制组,中性-淋巴细胞比率(NLR)、血小板-淋巴细胞比率(PLR)、单核-淋巴细胞比率(MLR)低于非疾病控制组(P<0.05)。DCB组鳞癌、TNM分期Ⅲ期、东部肿瘤协作组(ECOG)评分0~1、有不良反应的患者比例高于非DCB组(P<0.05),PLT、NLR、PLR、MLR低于非DCB组(P<0.05)。ROC曲线分析结果显示PLT、NLR、PLR、MLR可作为预测DCB的指标,ROC曲线下面积(AUC)分别为0.633、0.602、0.635、0.604,最佳截断(cut off)值分别为187×109/L(P=0.004)、5.0(P=0.026)、235(P=0.003)、0.35(P=0.024)。多因素Cox回归分析显示,非鳞癌[包括腺癌(HR=1.565,95%CI:1.057~2.316)和其他(HR=2.285,95%CI:1.326~3.936)、ECOG评分2~3分(HR=2.375,95%CI:1.652~3.415)、AMC≥0.65×109/L(HR=1.847,95%CI:1.160~2.938)、PLR≥235(HR=1.557,95%CI:1.016~2.386)是PFS时间较短的独立危险因素(P<0.05);ECOG评分2~3分(HR=4.615,95%CI:2.882~7.391)、AMC≥0.65×10^(9)/L(HR=5.161,95%CI:2.984~8.925)、PLR≥235(HR=1.732,95%CI:1.059~2.833)是OS时间较短的独立危险因素(P<0.05),有不良反应(HR=0.472,95%CI:0.294~0.757)是OS时间较短的独立保护因素(P<0.05)。结论PLT、AMC、NLR、MLR、PLR较低,ALC较高,鳞癌,TNM为Ⅲ期,ECOG评分0~1分,有免疫治疗相关不良反应可提示ICIs治疗晚期NSCLC预后较好。PLT、NLR、PLR、MLR可作为预测DCB的指标。

Objective To investigate the predictive value of inflammatory cells and clinical features in the prognosis of immune checkpoint inhibitors(ICIs)treating non-small cell lung cancer(NSCLC).Methods The data of 163 cases of stageⅢandⅣNSCLC patients treated with the ICIs in this hospital from January 1,2017 to December 31,2022 were collected.The CT examination was conducted after 6-8 weeks treatment.The patients were divided into the objective remission group[complete remission(CR)+partial remission(PR)]and non-objective remission group[stable disease(SD)+progressed disease(PD)],disease control group(CR+PR+SD)and non-disease control group(PD),persistent clinical benefit group(DCB)and non-DCB group.The differences in clinical features and inflammatory cells indicators were compared among the different groups.The receiver operating characteristic(ROC)curve was adopted to evaluate the predictive efficiency of the inflammatory cells indicators for DCB.The influencing factors analysis of progression free survival(PFS)time and overall survival(OS)time adopted the Cox regression analysis.Results The lymphocyte count(ALC)in the disease control group was higher than that in the non-disease control group.The neutrophil to lymphocyte ratio(NLR),platelet-lymphocyte ratio(PLR)and mononuclear lymphocyte ratio(MLR)were lower than those in the non-disease control group.The proportions of squamous cell carcinoma,stageⅢ,ECOG score 0-1 point,adverse reactions in the DCB group were higher than those in the non-DCB group(P<0.05),the PLT count,NLR,PLR and MLR were lower than those in the non-DCB group(P<0.05).The ROC curve analysis results showed that PLT,NLR,PLR and MLR could serve as the indicators for predicting DCB,the area under of ROC curve(AUC)was 0.633,0.602,0.635 and 0.604 respectively,the optimal cut off values were 187×10^(9)/L(P=0.004),5.0(P=0.026),235(P=0.003)and 0.35(P=0.024)respectively.The multivariate Cox regression analysis showed that non-squamous carcinoma including adenocarcinoma(HR=1.565,95%CI:1.057-2.316)and other pathologic types(HR=2.285,95%CI:1.326-3.936),ECOG score 2-3 points(HR=2.375,95%CI:1.652-3.415),AMC≥0.65×109/L(HR=1.847,95%CI:1.160-2.938)and PLR≥235(HR=1.557,95%CI:1.016-2.386)were the in dependent risk factors for short PFS.The ECOG score 2-3 points(HR=4.615,95%CI:2.882-7.391),AMC≥0.65×109/L(HR=5.161,95%CI:2.984-8.925)and PLR≥235(HR=1.732,95%CI:1.059-2.833)were the independent risk factors for short OS(P<0.05),and having adverse reactions(HR=0.472,95%CI:0.294-0.757)was the independent protective factor for short OS(P<0.05).Conclusion Lower PLT,AMC,NLR,MLR and PLR,higher ALC,squamous cell carcinoma,TNM stageⅢ,ECOG score 0-1 point and immunotherapy related adverse reactions could prompt that the prognosis is good in ICIs treating advanced NSCLC.PLT,NLR,PLR and MLR could serve as the indicators for predicting DCB.