非小细胞肺癌放化疗后复发转移的预测模型

摘要

Objective To explore the factors influencing recurrence and metastasis of non-small cell lung cancer (NSCLC) after radiotherapy and chemotherapy,and establish predictive model of recurrence and metastasis. Methods Clinical data of 187 pa-tients with NSCLC treated with radiotherapy and chemotherapy from January 2013 to January 2017 were collected. All patients were di-vided into non-recurrent group and relapsed group,as well as non-metastasis group and metastasis group. The relationship of clinico-pathological features (gender,age,hemoglobin,platelet,CEA,clinical stage,T stage,N stage,vascular invasion,pathological type, differentiation,AFP,chemotherapy cycles,treatment patterns)with recurrence and metastasis were analyzed. Univariate and multiva-riate Logistic regression analysis were performed to determine the influencing factors for recurrence and metastasis of NSCLC after radio-therapy and chemotherapy. The receiver operating characteristic (ROC)curve was generated for multivariate Logistic regression analy-sis. Results Among 187 patients,there were 146 relapses (78.1%)and 105 metastases (56.1%). Hemoglobin,platelet,CEA, clinical stage,T stage,N stage,vascular invasion,pathological differentiation degree,chemotherapy cycles and treatment patterns were related to recurrence and metastasis. Univariate Logistic regression analysis showed that hemoglobin,CEA,clinical staging,T stage,N stage,vascular invasion,differentiation,chemotherapy cycles and treatment methods were factors influencing the recurrence of NSCLC;platelets,CEA,clinical stage,T stage,N stage,vascular invasion,differentiation,chemotherapy cycles and treatment patterns influenced the metastasis of NSCLC. Multivariate Logistic regression analysis showed that CEA,clinical stage,differentiation, chemotherapy cycles and treatment patterns were independent factors influencing recurrence;CEA,clinical stage,differentiation and chemotherapy cycles were independent factors influencing metastasis of NSCLC. The AUC of NSCLC recurrence and metastasis was 0.867 (0.738-0.913)and 0.813 (0.722-0.895). Conclusion CEA,clinical stage,differentiation,chemotherapy cycles and treat-ment patterns are closely related to the recurrence and metastasis of NSCLC. Preventive measures should be taken to reduce the occur-rence and metastasis previously.%目的 探讨影响非小细胞肺癌(NSCLC)放化疗后复发和转移的因素,并建立预测模型.方法 收集2013年1月至2017年1月187例NSCLC放化疗患者的临床资料,按照是否复发和是否转移分为未复发组和复发组,以及未转移组和转移组.分析NSCLC临床病理特征(性别、年龄、血红蛋白、血小板、CEA、临床分期、T分期、N分期、血管侵犯、病理类型、分化程度、AFP、化疗周期数、治疗方式)与复发和转移的关系.单因素和多因素Logistic回归分析影响NSCLC复发和转移的因素,并根据多因素分析结果绘制受试者工作特征(ROC)曲线.结果 随访至2018年3月,146例复发(78.1%),105例转移(56.1%).血红蛋白、血小板、CEA、临床分期、T分期、N分期、血管侵犯、分化程度、化疗周期数、治疗方式与复发和转移有关(P＜0.05).单因素Logistic回归分析显示,血红蛋白、CEA、临床分期、T分期、N分期、血管侵犯、分化程度、化疗周期数、治疗方式是影响NSCLC复发的因素;血小板、CEA、临床分期、T分期、N分期、血管侵犯、分化程度、化疗周期数、治疗方式为影响NSCLC转移的因素.多因素Logistic回归分析显示,CEA、临床分期、分化程度、化疗周期数、治疗方式是影响NSCLC复发的独立因素;CEA、临床分期、分化程度、化疗周期数是影响NSCLC转移的独立因素.影响NSCLC复发和转移的ROC曲线下面积(AUC)分别为0.867(0.738～0.913)和0.813(0.722～0.895).结论 CEA、临床分期、分化程度、化疗周期数、治疗方式与NSCLC放化疗后复发与转移有关,采取针对性的预防措施,从而降低复发和转移的发生.