非小细胞肺癌术后早期复发风险列线图模型的建立与验证

Construction and validation of a Nomogram model for postoperative early recurrence in patients with non-small cell lung cancer

目的探讨非小细胞肺癌(NSCLC)患者术后早期复发的风险因素,并构建一种新的列线图模型。方法回顾性分析2021年1—8月在安徽医科大学附属阜阳医院及安徽医科大学附属阜阳市人民医院接受手术治疗的236例NSCLC患者的临床病理资料,将所有患者按7∶3的比例随机分为建模组(n=165)与验证组(n=71)。采用单因素、多因素Cox回归分析筛选NSCLC术后早期复发的独立风险因素,并构建列线图模型。通过一致性指数(C-index)、校准曲线和受试者工作特征(ROC)曲线评价模型的预测能力。结果236例NSCLC患者术后早期复发率为17.4%(41/236)。单因素、多因素Cox分析显示,淋巴结转移(HR=2.342,95%CI:1.214~4.517,P=0.011)、胸膜侵犯(HR=2.738,95%CI:1.443~5.196,P=0.002)、脉管侵犯(HR=3.526,95%CI:1.802~6.899,P<0.001)及血浆D-二聚体(HR=3.656,95%CI:1.265~10.561,P=0.017)是NSCLC患者术后早期复发转移的独立预测因子。基于上述4个变量构建列线图模型,结果显示该模型在建模组与验证组中的C-index分别为0.769(95%CI:0.661~0.879)和0.790(95%CI:0.682~0.897);该模型预测建模组患者1、2年无复发生存(RFS)的曲线下面积(AUC)为0.817和0.792,预测验证组患者1、2年RFS的AUC为0.782和0.771。校准曲线表明模型在2组中的预测概率与实际复发风险具有良好的一致性。结论该列线图模型对NSCLC术后早期复发具有良好的预测价值,对协助临床医生准确识别高危复发人群具有重要意义。

Objective To explore the risk factors of postoperative early recurrence in patients with non-small cell lung cancer(NSCLC)and establish a new Nomogram model.Methods The clinicopathological materials of 236 NSCLC patients with surgical resection in Fuyang Hospital Affiliated to Anhui Medical University and Fuyang City People′s Hospital Affiliated to Anhui Medical University from January to August 2021 were retrospectively analyzed,and all the patients were randomly divided into a modeling group(n=165)and a validation group(n=71)with a ratio of 7 to 3.The independent risk factors of postoperative recurrence for NSCLC patients were determined by the univariate and multivariate Cox regression analyses,and a Nomogram model was constructed.The consistency index(C-index),calibration curve and receiver operating characteristics(ROC)curve were used to evaluate the predictive ability of the Nomogram model.Results The early recurrence rate of 236 NSCLC patients after surgery was 17.4%(41/236).Univariate and multivariate Cox analyses indicated that lymph node metastasis(HR=2.342,95%CI,1.214 to 4.517,P=0.011),pleural invasion(HR=2.738,95%CI,1.443 to 5.196,P=0.002),vascular invasion(HR=3.526,95%CI,1.802 to 6.899,P<0.001)and serum D-dimer level(HR=3.656,95%CI,1.265 to 10.561,P=0.017)were the independent predictors of early recurrence and metastasis for NSCLC patients.Based on the above four variables,a Nomogram model was constructed,and the result showed that the C-index of this model in the modeling group and validation group were 0.769(95%CI,0.661 to 0.879)and 0.790(95%CI,0.682 to 0.897)respectively;the area under the curve(AUC)of this model in predicting recurrence free survival(RFS)for patients in the modeling group at 1 year and 2 years was 0.817 and 0.792 respectively,while the AUC for patients in the validation group at 1 year and 2 years was 0.782 and 0.771 respectively.The calibration curve indicated that the predicted probability of this model was consistent with the actual recurrence risk in both groups.Conclusion This Nomogram model has good predictive value for early postoperative recurrence of NSCLC,and is of great significance for assisting clinical doctors in accurately identifying high-risk recurrence populations.