With the progressive aging of the global population, age-related diseases are expected to become increasingly common, such as dementias, which are characterized by the acquired loss of cognitive abilities with significant functional impairment. Thus, there is a clear need for studies on new biomarkers, specifically tailored to the context of the Global South, in order to contribute to the diagnosis, treatment, and prognosis of dementias. In this context, the present work investigated a brain age prediction model as a potential biomarker for neurodegeneration, evaluating its performance in a Brazilian clinical dementia sample. The final prediction model, composed of fifteen predictor variables, showed performance consistent with current literature. Additionally, through the brain age gap (BAG) -calculated as the difference between predicted age and chronological age - there was significant discrimination of the Alzheimer s disease group relative to the Cognitively Normal group, as well as of the Cognitively Normal group relative to the non-Alzheimer dementia group. Non-Alzheimer dementias were also significantly differentiated from the Mild Cognitive Impairment group. Socioeconomic variables such as education and socioeconomic status were not significant in explaining the brain age gap. Overall, these findings indicate the effectiveness of this brain age prediction model in distinguishing between dementia and non-dementia groups. The need to integrate this paradigm with other clinical and research biomarkers, as well as to further explore socio-environmental factors, is highlighted.