Aim: Prognosis of patients with metastatic non-small-cell lung cancer differ widely. Methods: All patients were randomly divided into training or validation cohort. Cox-regression analyses were conducted to select independent predictors. We built a nomogram by R code and evaluated the accuracy and the reliability of the model using C-index, calibration curves and decision curve analyses. We made a risk classification system based on the nomogram. Results: In the validation cohort, C-index was 0.729 and 0.738 for 1- and 2-year overall survival. Calibration plots and decision curve analyses presented great prognostic accuracy and clinical applicability. Its prognostic accuracy preceded the American Joint Committee on Cancer staging with evaluated integrated discrimination improvement. Conclusion: The model can be a practical tool in treatment decision and individual counseling.
Keywords: SEER database; metastatic NSCLC; nomogram; overall survival; prognosis.