中晚期宫颈癌同步放化疗致重度骨髓抑制的危险因素分析及预测模型的构建  

Risk factors and predictive model for severe myelosuppression due to concurrent chemoradiotherapy in advanced cervical cancer

作　　者：叶海主 王洁 彭小保 孔萍 YE Haizhu;WANG Jie;PENG Xiaobao;KONG Ping(Department of Oncology,Nanjing Gaochun People's Hospital,Nanjing 211300,China)

机构地区：[1]南京市高淳人民医院肿瘤科,南京211300 [2]南京市高淳人民医院药学部,211300

出　　处：《临床肿瘤学杂志》2025年第1期49-54,共6页Chinese Clinical Oncology

摘　　要：目的探讨中晚期宫颈癌患者行同步放化疗致严重骨髓抑制的危险因素,同时构建预测模型并进行验证。方法回顾性分析2016年7月至2024年3月行同步放化疗的112例中晚期宫颈癌患者,分为重度骨髓抑制组(n=43)和非重度骨髓抑制组(n=69)。采用Logistics回归分析筛选重度骨髓抑制的危险因素并构建列线图预测模型,并验证模型的区分度、校准度及临床实用性。结果112例患者中,重度骨髓抑制的发生率为38.4%(43/112)。年龄、BMI、化疗方案、放疗范围、治疗前血红蛋白水平及治疗前白细胞计数与发生重度骨髓抑制有关(P<0.05);Logistics多因素分析显示,年龄、BMI、化疗方案、治疗前血红蛋白水平和治疗前白细胞计数为影响重度骨髓抑制发生的独立因素(P<0.05)。采用R软件构建列线图,受试者工作特征(ROC)曲线显示,模型的曲线下面积为0.907,明显优于单一指标,预测准确度高。校准曲线与理想曲线相近,表明模型具有较好的校准能力。决策曲线(DCA)提示模型净获益较好,具有临床实用性。结论成功构建列线图预测模型,该模型具有良好的预测效能,对预测行同步放化疗的中晚期宫颈癌患者是否出现重度骨髓抑制有一定的参考价值。Objective To investigate the risk factors for severe myelosuppression in patients with advanced cervical cancer undergoing concurrent chemoradiotherapy and further construct and validate a predictive model.Methods The clinical data of 112 patients with advanced cervical cancer who underwent concurrent chemoradiotherapy from July 2016 to March 2024 were retrospectively reviewed.Paitents were diveided into severe myelosuppression group(n=43)and non severe myelosuppression group(n=69).Logistic regression analysis was used to screen risk factors of severe myelosuppression.These factors were used to construct a nomogram prediction model and validate the discrimination,calibration and clinical utility of the model.Results The incidence of severe myelosuppression was 38.4%(43/112)among 112 cervical cancer patients.Age,BMI,chemotherapy regimen,radiation field,pre-treatment hemoglobin count and pre-treatment white blood cell count were related to severe myelosuppression(P<0.05).Multivariate logistic regression showed that age,BMI,chemotherapy regimen,pre-treatment hemoglobin count and pre-treatment white blood cell count were predictive factors for severe myelosuppression(P<0.05).R software was used to construct a nomogram model,and ROC curve showed that the AUC of the model was 0.907,significantly better than that of a single model,which indicated high prediction accuracy of the model.The calibration curve was close to the ideal curve,indicating that the model had good calibration ability.The decision curve(DCA)indicated that the model had good clinical net benefits and clinical practicality.Conclusion A nomogram model is successfully constructed,which has good predictive performance.It has important reference value for clinical physicians to predict whether severe myelosuppression will occur in patients with advanced cervical cancer undergoing concurrent chemoradiotherapy.

关 键 词：宫颈癌 同步放化疗 骨髓抑制 列线图 

分 类 号：R737.33[医药卫生—肿瘤]