BSBM-21 MACHINE LEARNING IDENTIFIES EXPERIMENTAL BRAIN METASTASIS SUBTYPES BASED ON THEIR INFLUENCE ON NEURAL CIRCUITS

A high percentage of patients with brain metastases frequently develop neurocognitive symptoms, however understanding how brain metastasis co-opt the function of neuronal circuits beyond a mass effect remains unknown. We report a comprehensive multidimensional modelling of brain functional analysis in the context of brain metastasis. By testing different pre-clinical models of brain metastasis from various primary sources and oncogenic profiles we dissociated the heterogeneous impact on brain function that we detected from the homogeneous inter-model tumor size or glial response. In contrast we report a potential underlying molecular program responsible for impairing neuronal crosstalk in a model-specific manner. Additionally, measurement of various brain activity readouts matched with machine learning strategies confirmed model-specific alterations that could help to predict the presence and subtype of metastasis. We envision that our findings not only increase our knowledge on the molecular basis of neurocognitive impairment associated with brain metastases but they are also the first step towards new therapeutic strategies to prevent or stop the decline in quality of life associated with these symptoms. In addition, our computational findings exploiting electrophysiological profiles suggest the possibility to exploit them as novel biomarkers.