错配修复正常型晚期结直肠癌免疫联合靶向治疗患者预后因素分析及列线图模型的构建

Analysis of prognostic factors for patients with mismatch repair⁃proficient advanced colorectal cancer treated by immunotherapy combined with targeted therapy and its prediction nomogram construction

目的探讨影响错配修复正常(mismatch repair‐proficient,pMMR)型晚期结直肠癌(colorectal cancer,CRC)免疫联合靶向治疗的预后因素并建立列线图预测模型。方法回顾性分析2020年1月至2023年1月在广西医科大学附属肿瘤医院末线接受免疫联合靶向治疗的pMMR型晚期CRC患者的临床资料。采用多因素Cox回归分析影响预后的因素。构建列线图模型预测无进展生存期(progress‐free survival,PFS)、总生存期(overall survival,OS),采用受试者工作特征(receiver operating characteristic,ROC)、一致性指数(C‐index)和校准曲线评估模型效能。计算患者列线图预测得分,采用X‐tile软件将患者分为高风险组和低风险组,绘制Kaplan‐Meier曲线并用Log‐rank检验比较两组患者的生存差异。结果本研究共纳入116例患者。Cox回归分析显示,体质量指数(body mass index,BMI)≥25 kg/m2是OS的独立保护因素(HR=0.431,95%CI:0.187~0.990),而癌症‐炎症预后指数(cancer‐inflammation prognostic index,CIPI)≥828.8则是OS的独立危险因素(HR=1.820,95%CI:1.012~3.271);癌胚抗原(carcinoembryonic antigen,CEA)>5 ng/mL(HR=2.448,95%CI:1.121~5.345)和系统免疫炎症指数(systematic immune‐inflammation index,SII)≥663.9(HR=2.730,95%CI:1.205~6.183)是PFS的独立危险因素。在PFS列线图模型中,训练集、验证集的C‐index分别为0.625、0.601,训练集3、6、12个月的AUC分别为0.666、0.639、0.811,验证集分别为0.696、0.647、0.588。在OS列线图模型中,训练集、验证集的C‐index分别为0.663、0.662,训练集6、12个月的AUC分别为0.689、0.761,验证集分别为0.726、0.691。校准图中OS率曲线与45°对角线贴合较高,均体现该列线图模型具有较好的区分度。Log‐rank检验显示,低风险组的PFS率和OS率均高于高风险组(均P<0.001)。结论基于BMI、CIPI、SII、CEA等预后影响因素建立的列线图模型可以对pMMR型晚期CRC免疫联合靶向治疗患者的预后进行有效预测,BMI、CIPI、SII、CEA可能是pMMR型晚期CRC免疫联合靶向治疗预后评估的潜在标志物。

Objective To investigate the prognostic factors affecting immunotherapy combined with targeted therapy for mismatch repair‐proficient(pMMR)advanced colorectal cancer(CRC)and to establish a nomogram prediction model.Methods The clinical data of patients with pMMR advanced CRC who received immunotherapy combined with targeted therapy at the end line of the Guangxi Medical University Cancer Hospital from January 2020 to January 2023 were retrospectively analyzed.Multivariable Cox regression analysis was performed to identify the factors affecting prognosis.The nomogram models were constructed to predict progress‐free survival(PFS)and overall survival(OS).Receiver operating characteristic(ROC),consistency index(C‐index)and calibration curve were used to evaluate the performance of the model.The scores of nomogram prediction were calculated,and the patients were divided into the high‐risk group and the low‐risk group by X‐tile software.Kaplan‐Meier curve was drawn and Log‐rank test was conducted to compare the survival differences between the two groups.Results A total of 116 patients were included in this study.Cox regression analysis showed that body mass index(BMI)≥25 kg/m2 was an independent protective factor for OS(HR=0.431,95%CI:0.187-0.990),while the cancer‐inflammation prognostic index(CIPI)≥828.8 was an independent risk factor(HR=1.820,95%CI:1.012-3.271).The carcinoembryonic antigen(CEA)>5 ng/mL(HR=2.448,95%CI:1.121-5.345)and systematic immune‐inflammation index(SII)≥663.9(HR=2.730,95%CI:1.205-6.183)were independent risk factors for PFS.In the PFS nomogram model,the C‐index of the training set and the validation set were 0.625 and 0.601,respectively.The AUCs of the training set at 3,6,and 12 months were 0.666,0.639,and 0.811,respectively,and those of the validation set were 0.696,0.647,and 0.588,respectively.In the OS nomogram model,the C‐index of the training set and the validation set were 0.663 and 0.662,respectively.The AUCs of the training set at 6 and 12 months were 0.689 and 0.761,respectively,and those of the validation set were 0.726 and 0.691,respectively.The OS rate curve in the calibration plot was fitted well with the 45°diagonal,indicating that the nomogram model had good discrimination.The Log‐rank test showed that the PFS rate and OS rate of the low‐risk group were higher than those of the high‐risk group(all P<0.001).Conclusions The nomogram model based on prognostic factors such as BMI,CIPI,SII and CEA can effectively predict the prognosis of patients with CRC immunotherapy.BMI,CIPI,SII and CEA may be potential markers for evaluating the prognosis of CRC immunotherapy.