Analisi proteomica in pazienti con atrofia muscolare spinale trattati con Nusinersen.

Spinal muscular atrophy (SMA) is an autosomal recessive neurodegenerative disease caused by mutations in the SMN1 gene and characterized by marked clinical heterogeneity. The availability of new therapies and the advent of newborn screening programs have made the identification of reliable biomarkers essential for improving phenotypic stratification and monitoring therapeutic response. The aim of this study was to identify prognostic protein biomarkers in the cerebrospinal fluid (CSF) of SMA patients at baseline (T0) and to characterize treatment-induced changes in the proteomic profile following ten months of nusinersen therapy (T302), with particular attention to the biological processes and molecular pathways involved, including neuroinflammatory processes. A total of 61 SMA patients (19 SMA type 1, 19 SMA type 2, and 23 SMA type 3) treated with nusinersen were enrolled from four Italian tertiary referral centers specialized in neuromuscular disorders. Proteomic analysis was performed using an untargeted approach based on liquid chromatography–mass spectrometry (LC-MS). Data analysis was conducted using the Random Forest (RF) machine learning algorithm and biological pathway enrichment analysis. Proteomic analysis identified 1625 proteins at T0 and 1399 at T302. Baseline proteomic profiles revealed significant differences among SMA subtypes, identifying nine key proteins capable of distinguishing the most severe form from the milder phenotypes. Analysis of proteomic profile changes identified a total of 147 differentially expressed proteins after nusinersen treatment in SMA1, 135 in SMA2, and 289 in SMA3. Nusinersen treatment induced significant modifications in the CSF proteome, with a shared response characterized by the regulation of pathways associated with axonogenesis, as well as severity-specific modulations, including alterations in pathways involved in energy metabolism and oxidative stress response in SMA1, and in the coagulation and complement cascades in SMA2 and SMA3. This study demonstrates that the CSF proteomic profile represents a potential source of prognostic and therapeutic response biomarkers in SMA. The integration of untargeted proteomics with machine learning techniques enabled the identification of protein differences in naïve patients and the characterization of biological changes induced by nusinersen therapy, providing new evidence for the involvement of neuroinflammatory processes in treatment response.