食管鳞癌同步放化疗患者血小板严重减少的风险模型构建及应用价值  

Construction and application value of a risk model for severe thrombocytopenia in patients with esophageal squamous cell carcinoma undergoing synchronous radiotherapy and chemotherapy

作　　者：杨英花 代丽 叶骉飞 杨植 党升强 郭愿愿 Yang Yinghua;Dai Li;Ye Biaofei;Yang Zhi;Dang Shengqiang;Guo Yuanyuan(Department of Oncology,Chang'an Hospital,Xi'an 710016,China;Department of Oncology Radiation,Chang'an Hospital,Xi'an 710016,China;Department of Oncology,Xi'an Third Hospital,Affiliated Hospital of Northwestern University,Xi'an 710021,China)

机构地区：[1]长安医院肿瘤科,西安710016 [2]长安医院肿瘤放疗科,西安710016 [3]西北大学附属西安市第三医院肿瘤内科,710021

出　　处：《中华消化病与影像杂志(电子版)》2025年第2期125-129,共5页Chinese Journal of Digestion and Medical Imageology(Electronic Edition)

摘　　要：目的构建食管鳞癌同步放化疗患者血小板严重减少的风险模型,并探索其应用价值。方法选取长安医院2016年1月至2024年1月收治的150例食管鳞癌患者,按照7∶3比例分为训练集(n=105)、验证集(n=45)。根据训练集患者同步放化疗期间血小板严重减少发生情况,将其分别纳入发生组、未发生组,对比2组患者临床资料,使用Logistic多因素回归分析归纳风险因素,并使用Nomogram建立预测模型,绘制受试者工作特征曲线(ROC)评价模型效能。使用验证集行模型验证。结果训练集105例患者中,24例发生血小板严重减少,发生率为22.86%。Logistic多因素回归分析示,年龄≥65岁、合并糖尿病、诱导化疗周期3周期是影响食管鳞癌患者同步放化疗期间血小板严重减少的独立危险因素(P<0.05)。以总分≥83分为截断值,Nomogram预测模型预测训练集患者发生血小板严重减少的AUC为0.886,其灵敏度、特异性分别为91.67%、87.65%,阳性预测值、阴性预测值分别为68.75%、97.26%。训练集45例患者中,10例发生血小板严重减少,发生率为22.22%。模型预测验证集患者发生血小板严重减少的AUC为0.822,其灵敏度、特异性分别为86.36%、78.26%,阳性预测值、阴性预测值分别为79.17%、85.17%。结论食管鳞癌同步放化疗患者血小板严重减少发生风险较高,且与患者年龄、合并症、诱导化疗周期有关,基于上述风险因素建立的预测模型能够为患者血小板严重减少风险评估提供可靠参考。Objective To establish a risk model of severe thrombocytopenia in patients with esophageal squamous cell carcinoma after concurrent radiotherapy and chemotherapy,and to explore its application value.Methods A total of 150 patients with esophageal squamous cell carcinoma admitted to Chang’an Hospital from January 2016 to January 2024 were selected,and divided into training set(n=105)and verification set(n=45)according to the ratio of 7∶3.According to the occurrence of severe thrombocytopenia during postoperative concurrent radiotherapy and chemotherapy,the patients in the training group were divided into the occurrence group and the non-occurrence group,and the clinical data of the two groups were compared.Logistic multivariate regression analysis was used to summarize the risk factors,and Nomogram was used to establish a prediction model,and receiver operating characteristic(ROC)curve was drawn to evaluate the effectiveness of the model.Validation set was used for model validation.Results Among the 105 patients in the training set,24 patients suffered from severe thrombocytopenia,with an incidence rate of 22.86%.Logistic regression analysis showed that age≥65 years,diabetes mellitus and three cycles of induction chemotherapy were independent risk factors for severe thrombocytopenia in patients with esophageal squamous cell carcinoma during concurrent radiotherapy and chemotherapy(P<0.05).Taking the total score≥83 as the cut-off value,the AUC predicted by Nomogram prediction model for patients with severe thrombocytopenia in the training set was 0.886,with sensitivity and specificity of 91.67%and 87.65%,positive predictive value and negative predictive value of 68.75%and 97.26%respectively.Among the 45 patients in the training set,10 patients suffered from severe thrombocytopenia,with an incidence rate of 22.22%.The AUC of patients with severe thrombocytopenia in the model prediction validation set was 0.822,and its sensitivity and specificity were 86.36%and 78.26%,respectively.The positive predictive valu

关 键 词：食管鳞癌 同步放化疗 血小板严重减少 风险模型 

分 类 号：R735.1[医药卫生—肿瘤]