Introduction: Management of brain metastases (BMs) depends on important prognostic factors such as age, performance status (PS), primary cancer and the status of extracranial metastases. At this moment, the size of brain metastasis (BM) is used to decide the therapeutic approach but not as a prognostic factor. Methods: A retrospective and single center study, of patients with primary head and neck cancer, melanoma or sarcoma with diagnosis of BMs from 2006 to 2015 were analysed. Patients were selected by number of BMs (≥ 4) and/or by size of the largest BM (≥ 3 cm). The aim was to evaluate the association between the size of BM and overall survival (OS), in months, and the second outcome was to evaluate the association between whole brain radiotherapy (WBRT) and OS, in months. We compared the association by baselines covariates using log-rank test and Cox proportional hazards regression. Results: 66 patients were collected (median age 63 years old) with primary head and neck cancer, melanoma or sarcoma. In univariate analysis, there was no difference in OS between gender, number of BMs and primary cancer. However, there was a statistical difference in OS when the diameter of the largest metastasis is ≥ 3 cm (difference in median OS=1.3 months, p<0.05). There was also an increase in OS among patients treated with WBRT (difference in median OS=2.5 months, p<0.05). In a multivariate analysis, there was no difference in association between the size of BM and OS (Hazard Ratio (HR)=1.6, 95% CI: 0.82-3.2), p=0.16). Conclusion: This retrospective study shows no association between BMs’ size and OS in patients with primary head and neck cancer, melanoma or sarcoma.