Spinal Muscular Atrophy (SMA) is caused by the homozygous loss of the Survival Motor Neuron (SMN) 1 gene, which leads to degeneration of spinal alpha-motor neurons and muscle atrophy. Three treatments have been approved for SMA but the available data show interpatient variability in therapy response and, to date, individual factors such as age or SMN2 copies,cannot fully explain this variance. The aim of this project is: * collect clinical data and patient-reported outcome measures (PROM) from patients treated with nusinersen, risdiplam, onasemnogene abeparvovec, * identify novel biomarkers and RNA molecular signature profiling, * develop a predictive algorithm using artificial intelligence (AI) methodologies based on machine learning (ML), able to integrate clinical outcomes, patients' characteristics, and specific biomarkers. This effort will help to better stratify the SMA patients and to predict their therapeutic outcome, thus to address patients towards personalized therapies.