Longitudinal cohort study of muscle function and metabolic biomarkers in disease-modifying treatment of spinal muscular atrophy

Disease-modifying treatments (DMT) for spinal muscular atrophy (SMA) developed in the past decade have improved the prognosis of patients with this severe condition. However, treatment effects vary, highlighting the need for sensitive predictive biomarkers to identify individuals more likely to respond to DMT. A previous study detected changes in mitochondrial oxidative capacity in SMA arm muscles, using 31-phosphorus magnetic resonance spectroscopy (³¹P MRS). Here, we investigated whether ³¹P MRS can distinguish responders from non-responders to DMT. In this longitudinal observational cohort study, patients with SMA were assessed at baseline, two months, and ten months after initiation of DMT treatment. We used a 7 Tesla MR scanner equipped with a custom-built ergometer that allowed for individual mechanical loading and metabolic analysis of the upper arm muscles. Maximal voluntary contraction (MVC) force and motor functional scores were measured before each MR examination. Responders to DMT were defined as those who gained ≥20% in MVC after ten months of treatment. Seven non-ambulatory and five ambulatory patients with SMA starting DMT were included. Four patients fulfilled the MVC response criterion. While significant correlations were observed between MVC force and <i>in vivo</i> metabolic readouts of muscle phenotype, the latter could not longitudinally discriminate responders from non-responders to DMT. Our MR platform for single-arm muscle exercise provides a non-invasive tool for monitoring treatment response in both ambulatory and non-ambulatory patients. Further studies involving larger cohorts will be necessary to establish the diagnostic value of ³¹P MRS in evaluating DMT efficacy in SMA and other neuromuscular diseases.