L'ipermetabolismo del muscolo scheletrico predice la sopravvivenza in pazienti con Sclerosi Laterale Amiotrofica: un approccio computazionale alle immagini FDG PET/TC.

Abstract

Caricato come file

Abstract Purpose Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease leading to neuromuscular palsy and death. We propose a computational approach to [18F]-fluorodeoxyglucose (FDG) PET/CT images to analyze the structure and metabolic pattern of skeletal muscle in ALS and its relationship with disease aggressiveness. Materials and Methods A computational 3D method was used to extract whole psoas muscle’s volumes and average attenuation coefficient (AAC) from CT images obtained by FDG PET/CT performed in 62 ALS patients and healthy controls. Psoas average standardized uptake value (normalized on liver, N-SUV) and its distribution heterogeneity (defined as N-SUV variation coefficient, VC-SUV) were also extracted. Spinal cord and brain motor cortex FDG uptake were also estimated. Results As previously described, FDG uptake was significantly higher in spinal cord and lower in the brain motor cortex, in ALS compared to controls. While psoas AAC was similar in patients and controls, in ALS a significant reduction in psoas volume (3.6±1.02 vs 4.12±1.33 mL/Kg; p<0.01) and increase in psoas N-SUV (0.45±0.19 vs 0.29±0.09; p<0.001) were observed. Higher heterogeneity of psoas FDG uptake was also documented in ALS (VC-SUV 8±4%, vs 5±2%, respectively, p<0.001) and significantly predicted overall survival at Kaplan- Meyer analysis. VC-SUV prognostic power was confirmed by univariate analysis, while multivariate Cox regression model identified the spinal cord metabolic activation as the only independent prognostic biomarker. Conclusion: The present data suggest the existence of a common mechanism contributing to disease progression through the metabolic impairment of both second motor neuron and its effector.