Construction of a nomogram to predict the prognosis of non-small-cell lung cancer with brain metastases

Patients with non-small-cell lung cancer (NSCLC) often have a poor prognosis when brain metastases (BM) occur. This study aimed to evaluate the prognostic factors of BM in newly diagnosed NSCLC patients and construct a nomogram to predict the overall survival (OS). We included NSCLC patients with BM newly diagnosed from 2010 to 2015 in Surveillance, Epidemiology, and End Results database. The independent prognostic factors for NSCLC with BM were determined by Cox proportional hazards regression analysis. We then constructed and validated a nomogram to predict the OS of NSCLC with BM. We finally included 4129 NSCLC patients with BM for analysis. Age, race, sex, liver metastasis, primary site, histologic type, grade, bone metastasis, T stage, N stage, surgery, chemotherapy, and lung metastasis were identified as the prognostic factors for NSCLC with BM and integrated to establish the nomogram. The calibration, receiver operating characteristic curve, and decision curve analyses also showed that the clinical prediction model performed satisfactorily in predicting prognosis. A clinical prediction model was constructed and validated to predict individual OS for NSCLC with BM. The establishment of this clinical prediction model has great significance for clinicians and individuals.