Purpose: Identi cation of a truly high-risk group of brain metastases (BM) in patients with non-small cell lung cancer (NSCLC) could lead to early interventions and probably better prognosis. The objective of the study was to identify this group by generating a multivariable model with recognized and accessible risk factors.Methods: A retrospective cohort from patients seen at a single center during 2010-2020, was divided into a training (TD) and validation (VD) datasets, associations with BM were measured in the TD with logit, variables signi cantly associated were used to generate a multivariate model. Model´s performance was measured by 10-Fold cross validation of the TD and in the VD with the AUC/C-statistic, Akaike information index, and Press´s Q precision test.Results: From 570 patients with NSCLC with complete records a TD and VD were formed, no signi cant differences were found amid both datasets. Variables associated with BM in the multivariate logit analyses were age [P 0.021, OR 0.97 (95%CI 0.94-0.99)]; male gender [P 0.010, OR 2.44 (95%CI 1.24-4.82)]; mutational status [P 0.03, OR 2.1 (95%CI 1.07-4.12); and carcinoembryonic antigen levels [P 0.002, OR 1.002 (95%CI 1.001 -1.003). BM were diagnosed in 23% of the whole cohort. Strati cation into low, medium, or high-risk groups after simpli cation of the model, displayed a frequency of BM in the VD of 8%, 16% and 40% respectively (P 0.027). Conclusion:A multivariate model comprising age, gender, CEA, and mutation status allowed the identi cation of a truly high-risk group of BM in NSCLC patients.