难治性小细胞肺癌的预测模型

发布时间：2018-08-08 20:56

【摘要】：目的:通过分析临床因素与难治性SCLC的关系,建立难治性SCLC的预测模型。方法:1.研究因素:回顾性分析2008年4月-2014年9月年期间,就诊于吉林大学第一医院肿瘤中心的329例小细胞肺癌患者。根据一线化疗结束后到疾病进展的时间将患者分为敏感性SCLC组和难治性SCLC组,纳入的329例患者,敏感性SCLC组有239例,难治性SCLC组有90。研究因素包括年龄、性别、吸烟指数、临床分期、上腔静脉综合征、化疗方案、化疗疗周期及2疗程化疗疗效等。2.建立预测模型:通过应用logistic回归进行单因素及多因素回归分析确定影响难治性SCLC的独立危险因素,对独立危险因素进行logistic多因素分析时产生一个新的联合预测变量PRE-1并建立回归方程,用受试者工作特性曲线(ROC)对新预测变量的预测能力进行分析,获得曲线下面积,根据约登指数(敏感性+特异性-1)找出最佳诊断界点。以P0.05为差异有统计学意义。结果:1.单因素logistic回归分析主要是确定各项因素对难治性SCLC的相关性。有统计学意义的因素主要有性别、吸烟指数、上腔静脉综合征、临床分期、化疗方案和胸部放疗。进一步多因素logistic回归分析显示,吸烟指数400支*年、有上腔静脉综合征、广泛期、与EP方案相比EC方案和无胸部放疗是难治性SCLC的独立危险因素。2.预测模型ROC曲线下面积(AUC)为0.721,95%可信区间为0.657-0.784,取约登指数最大值所对应的分界点为诊断点,即PRE_1为0.306,其灵敏度为61.1%,特异度为79.9%。结论:1.吸烟指数400、有上腔静脉综合征、广泛期、与EP方案相比EC方案和无胸部放疗是难治性SCLC的独立危险因素。2.根据独立危险因素建立的难治性SCLC的预测模型,具有较高的灵敏度、特异度,可辅助临床医生对一线治疗后的SCLC进行预测难治性SCLC。
[Abstract]:Objective: to establish a predictive model of refractory SCLC by analyzing the relationship between clinical factors and refractory SCLC. Method 1: 1. Study factors: a retrospective analysis of 329 patients with small cell lung cancer (SCLC) from April 2008 to September 2014 was conducted in the Cancer Center of the first Hospital of Jilin University. The patients were divided into sensitive SCLC group and refractory SCLC group according to the time from the end of first-line chemotherapy to the progression of the disease. There were 239 cases in the sensitive SCLC group and 90 cases in the refractory SCLC group. The factors included age, sex, smoking index, clinical stage, superior vena cava syndrome, chemotherapy regimen, chemotherapy cycle and 2 courses of chemotherapy. The prediction model was established: the independent risk factors affecting refractory SCLC were determined by single factor and multivariate regression analysis with logistic regression. A new joint predictive variable (PRE-1) was generated by logistic multivariate analysis of independent risk factors and a regression equation was established. The predictive ability of the new predictive variable was analyzed by using the operating characteristic curve (ROC) of the subjects, and the area under the curve was obtained. According to the Yorden index (sensitivity specificity-1) to find out the best diagnostic threshold point. P0.05 as the difference was statistically significant. The result is 1: 1. Single factor logistic regression analysis was mainly to determine the correlation between the factors and refractory SCLC. Statistically significant factors were gender, smoking index, superior vena cava syndrome, clinical staging, chemotherapy and chest radiotherapy. Further multivariate logistic regression analysis showed that smoking index was 400 years, had superior vena cava syndrome, extensive stage, EC regimen and non-chest radiotherapy were independent risk factors of refractory SCLC compared with EP regimen. The area (AUC) under the ROC curve of the prediction model was 0.721% 95% confidence interval 0.657-0.784. The dividing point corresponding to the maximum value of the Jorden index was taken as the diagnostic point, that is, the PRE_1 was 0.306, the sensitivity was 61.1% and the specificity was 79.9%. Conclusion 1. Smoking index was 400, had superior vena cava syndrome, extensive stage, EC regimen and non-chest radiotherapy were independent risk factors of refractory SCLC compared with EP regimen. The predictive model of refractory SCLC based on independent risk factors has high sensitivity and specificity and can assist clinicians in predicting refractory SCLC after first-line treatment.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：R734.2

【参考文献】