Construction of Predictive Nomogram Model of NSCLC Based on Serological Indicators and Clinical Data

Ge Hui

doi:10.3969/j.issn.1004-5775.2025.14.014

Abstract：Objective:To construct a nomogram model for postoperative prognosis prediction of non-small cell lung cancer(NSCLC)based on serological indexes CA125,CEA,NSE,CYFRA21-1,CA15-3 and clinical data.Methods:126 patients with NSCLC who underwent surgical treatment in the hospital from June 2019 to March 2022 were retrospectively collected.The serum tumor markers CA125,CEA,NSE,CYFRA21-1 and CA15-3 were determined by ELISA kit,and the general data of patients were collected.They were followed up for 24 months and divided into survival group and death group according to whether they survived or not.Logisitic regression methods were used to analyze the influencing factors of patients'prognosis,nomogram model was constructed according to the independent risk factors of prognosis,and the model was verified by Bootstrap self-sampling method.Results:After the follow-up,among 126 NSCLC patients,28 died and 98 survived.Compared with the patients in the survival group,the age,operation time,intraoperative blood loss,proportion of hyperlipidemia,proportion of smoking history,and proportion of lobectomy in the death group were all higher,and the differences were statistically significant(t=2.562,4.622,4.112;P＜0.05;χ2=9.404,12.204,5.878;P＜0.05).The indicators of serum CA125,CEA,NSE,CYFRA21-1 and CA15-3 in NSCLC patients of the death group were all higher than those in the survival group,and the differences were statistically significant(t=6.479,7.392,4.022,9.945,11.066;P＜0.05).The results of logistic multivariate regression analysis showed that age＞60 years old,operation time＞2.5 hours,CA125,CEA,NSE,CYFRA21-1,CA15-3 are independent risk factors for poor prognosis in NSCLC patients(OR=1.118,1.108,1.037,1.749,1.212,1.311,1.108;P＜0.05).The risk value for predicting poor prognosis of NSCLC by the nomogram model is 0.902.The AUC of the model was 0.861(95%CI:0.782～0.922,P＜0.001),the consistency index was 0.873,and the discrimination was good.The calibration curve showed that the predicted probability fit well with the reference probability.Conclusion:Old age,long-term operation,high levels of serum index are independent influencing factors of postoperative poor prognosis of NSCLC patients.The nomogram model based on this has high clinical value and can provide theoretical support for the prevention and treatment of postoperative poor prognosis of NSCLC patients.