Neurofilament light chain and profilin-1 dynamics in 30 spinal muscular atrophy type 3 patients treated with nusinersen.

ABSTRACT

BACKGROUND AND PURPOSE: The aim was to investigate whether neurofilament light chain (NfL) and profilin-1 (PFN-1) might qualify as surrogate disease and treatment-response biomarkers by correlating their concentrations dynamic with clinical status in a cohort of 30 adult spinal muscular atrophy type 3 patients during nusinersen therapy up to 34 months. METHODS: Neurofilament light chain was measured in cerebrospinal fluid at each drug administration with a commercial enzyme-linked immunosorbent assay (ELISA); PFN-1 concentrations were tested in serum sampled at the same time points with commercial ELISA assays. Functional motor scores were evaluated at baseline, at the end of the loading phase and at each maintenance dose and correlated to biomarker levels. The concurrent effect of age and clinical phenotype was studied. RESULTS: Neurofilament light chain levels were included in the reference ranges at baseline; a significant increase was measured during loading phase until 1 month. PFN-1 was higher at baseline than in controls and then decreased during therapy until reaching control levels. Age had an effect on NfL but not on PFN-1. NfL was partially correlated to functional scores at baseline and at last time point, whilst no correlation was found for PFN-1. CONCLUSION: Cerebrospinal fluid NfL levels did not qualify as an optimal surrogate treatment biomarker in adult spinal muscular atrophy patients with a long disease duration, whilst PFN-1 might to a greater extent represent lower motor neuron pathological processes. The observed biomarker level variation during the first 2 months of nusinersen treatment might suggest a limited effect on axonal remodeling or rearrangement.