摘要
目的构建可个体化预测胸部肿瘤患者发生放射性肺损伤(radiationinduced lung injury,RILI)的风险nomgram模型,并对模型的预测效能进行验证。方法选取2018年9月-2020年11月间于我院行放射治疗的胸部肿瘤155例患者作为研究对象,采用整群分组法将患者分为训练集(n=128)和验证集(n=27),分析训练集患者的参数资料,分别使用单因素和Logistic回归多因素分析发生≥2级RILI的独立危险因素,并建立相关nomogram预测模型。结果年龄≥70岁、同期化疗、MLD≥17Gy、V _(10)≥33%、V 20≥29%是老年胸部肿瘤患者发生≥2级RILI的独立危险因素(P<0.05)。基于以上5项独立危险因素建立预测老年胸部肿瘤患者发生≥2级RILI的nomogram模型,并对该模型进行验证。结果显示:训练集和验证集的C-index分别为0.882(95%CI:0.845~0.917)和0.811(95%CI:0.784~0.845),两集的校正曲线均与理想曲线拟合反映较好,两集ROC曲线的AUC分别为0.934和0.852,表明该nomogram模型具有良好的预测能力。结论导致RILI的危险因素较多,本研究基于危险因素建立的nomogram模型具有良好的预测能力,可为临床筛查高风险患者和进一步改进治疗计划提供参考依据。
Objective To construct a Nomgram model that can predict the risk of radiation-induced lung injury(RILI)in patients with thoracic cancer,and to verify the effectiveness of the model.Methods From September 2018 to November 2020,155 patients with thoracic tumor who received radiotherapy in our hospital were selected as the research objects.The patients were divided into the training group(n=128)and the validation group(n=27).The parameters data of the patients in the training group were analyzed.The independent risk factors of grade 2 RILI were analyzed by single factor and logistic regression,and the related Nomogram prediction model was established.Results Age≥70 years old,concurrent chemotherapy,MLD≥17Gy,V _(10)≥33%,and V 20≥29%were independent risk factors for grade 2 RILI in elderly patients with thoracic tumor(P<0.05).Based on the above five independent risk factors,a Nomogram model was established to predict the occurrence of RILI≥2 in elderly patients with thoracic tumor.The results showed that the C-index of the training group and the validation group were 0.882(95%CI:0.845~0.917)and 0.811(95%CI:0.784~0.845),respectively.The calibration curves of the two groups were better than the ideal curve fitting response.The AUC of the two ROC curves are 0.934 and 0.852,indicating that the Nomogram model had good prediction ability.Conclusion There are many risk factors leading to RILI.The Nomogram model based on risk factors has good predictive ability,which can provide reference for clinical screening of high-risk patients and further improvement of treatment plan.
作者
洪文翠
毕金玲
梁伟
HONG Wen-cui;BI Jin-ling;LIANG Wei(Fuyang Hospital Affiliated to Anhui Medical University,Fuyang,Anhui 236000,China;the First Affiliated Hospital of Anhui Medical University,Hefei,Anhui 230022,China)
出处
《临床肺科杂志》
2021年第9期1384-1390,共7页
Journal of Clinical Pulmonary Medicine
基金
安徽省重点研究与开发计划项目(No.201904a07020024)
安徽医科大学校科研基金项目(No.2019xkj088)。
