A nomogram to predict outcomes of lung cancer patients after pneumonectomy based on 47 indicators

Abstract

AIMS: We aimed to establish a nomogram for lung cancer using patients' characteristics and potential hematological biomarkers. METHODS: Principle component analysis was used to reduce the dimensions of the data, and each component was transformed into categorical variables based on cutoff values obtained using the X-tile software. Multivariate analysis was used to determine potential prognostic biomarkers. Five components were used in the predictive nomogram. Internal validation of the model was performed by bootstrapping of samples, while external validation was performed on a separate cohort from Shandong Cancer Hospital. The predictive accuracy of the model was measured by concordance index and risk group stratification. Decision curve analysis was performed to evaluate the net benefit of the models. RESULTS: One hundred patients in the Discovery group and 111 patients in the Validation group were retrospectively analyzed in this study. Forty-seven indexes were sorted into eight subgroups. Five components based on cox regression analysis were enrolled into the predictive nomogram. The nomogram prediction of the probability of 3- and 5-year overall survival was in great concordance with the actual observations. Of interest, the nomogram allowed better risk stratification of patients and better accuracy in predicting patients' survival compared with pathological tumor-node-metastasis staging system. CONCLUSION: A nomogram was established for prognosis of lung cancer, which can be used for treatment selection and clinical care management.