基于炎症反应评分系统的宫颈癌患者预后不良风险列线图预测模型构建及验证  

Construction and validation of a nomogram prediction model for the risk of adverse prognosis in cervical cancer patients based on the inflammatory response scoring system

作　　者：李峥 李宁[2] 单梦杰 马梅 王立群 LI Zheng;LI Ning;SHAN Mengjie;MA Mei;WANG Liqun(Bengbu Medical University,Bengbu 233000,China;Department of Gynecology,Suzhou Hospital of Anhui Medical University,Suzhou 234000,China;Department of Gynecology,the First Affiliated Hospital of Bengbu Medical University,Bengbu 233004,China)

机构地区：[1]蚌埠医科大学,安徽蚌埠233000 [2]安徽医科大学附属宿州医院妇科,安徽宿州234000 [3]蚌埠医科大学第一附属医院妇科,安徽蚌埠233004

出　　处：《陕西医学杂志》2025年第5期618-621,626,共5页Shaanxi Medical Journal

基　　金：国家自然科学基金资助项目(81973658)。

摘　　要：目的:探讨基于炎症反应评分系统构建及验证宫颈癌患者预后不良风险列线图预测模型。方法:选取175例宫颈癌患者,并分为建模组(123例)和验证组(52例)。术后随访3年,根据有无复发、远处转移、死亡将建模组患者分为预后不良组(37例)和预后良好组(86例)。比较不同预后患者全身炎症评分(SIS)、中性粒细胞与淋巴细胞比值(NLR)、血小板与淋巴细胞比值(PLR)、纤维蛋白原(FIB)以及C-反应蛋白(CRP)。宫颈癌患者预后不良的影响因素采用Logistic回归分析;R软件构建列线图风险预测模型;列线图模型的预测效能通过绘制受试者工作特征(ROC)曲线、校准曲线、决策曲线进行分析。结果:不同预后患者基线资料比较差异无统计学意义(均P>0.05)。预后不良组SIS评分、NLR、PLR、CRP高于预后良好组,FIB低于预后良好组(均P<0.05)。SIS评分、NLR、PLR、FIB、CRP是宫颈癌患者预后不良的独立影响因素(均P<0.05)。构建列线图风险预测模型,该模型预测验证组患者预后不良的曲线下面积(AUC)、敏感度及特异度分别为0.899、94.23%、70.16%。校正曲线与理想曲线(对角线)拟合效果理想,斜率为0.978,行Hosmer-Lemeshow拟合优度检验无统计学差异(P>0.05)。决策曲线显示该模型在0.12~0.98阈值范围内有良好的临床有效性。结论:SIS评分、NLR、PLR、FIB、CRP是宫颈癌患者预后不良的影响因素,基于这些指标构建列线图风险预测模型对患者预后具有较好的预测效能。Objective:To construct and validate a nomogram prediction model based on inflammatory response scoring system for assessing the risk of poor prognosis in cervical cancer patients.Methods:A total of 175 cervical cancer patients were selected and divided into a modeling group(123 cases)and a validation group(52 cases).Patients were followed up for 3 years after surgery and categorized into poor-prognosis(37 cases)and good-prognosis groups(86 cases)based on recurrence,distant metastasis,or death.Comparisons were made between patients with different prognoses regarding the SIS,NLR,PLR,FIB and CRP.Logistic regression analysis was used to identify risk factors for poor prognosis in cervical cancer patients.The nomogram risk prediction model was constructed using R software,and its predictive performance was evaluated using receiver operating characteristic(ROC)curves,calibration curves and decision curve analysis.Results:There were no significant differences in baseline characteristics between patients with different prognoses(all P>0.05).The poor-prognosis group had higher SIS scores,NLR,PLR and CRP levels and lower FIB levels compared to the good-prognosis group(all P<0.05).SIS score,NLR,PLR,FIB and CRP were identified as independent risk factors for poor prognosis(all P<0.05).The nomogram model demonstrated an AUC of 0.899,sensitivity of 94.23%,and specificity of 70.16%in predicting poor prognosis in the validation group.The calibration curve showed good agreement with the ideal curve(slope=0.978),and the Hosmer-Lemeshow goodness-of-fit test revealed no significant differences(P>0.05).Decision curve analysis indicated that the model had good clinical validity within the threshold range of 0.12 to 0.98.Conclusion:SIS score,NLR,PLR,FIB and CRP are risk factors for poor prognosis in cervical cancer patients.The nomogram prediction model based on these indicators provides effective prognostic prediction for patients.

关 键 词：宫颈癌 全身炎症评分 中性粒细胞与淋巴细胞比值 血小板与淋巴细胞比值 纤维蛋白原 C-反应蛋白 预测模型 

分 类 号：R737.33[医药卫生—肿瘤]