Evaluation of plasma neurofilament light chain levels in patients with spinal muscular atrophy treated with onasemnogene abeparvovec: association with motor outcomes
Abstract
Background: Spinal muscular atrophy (SMA) is an autosomal recessive disease of the lower motor neurons characterized by progressive hypotonia and weakness. Onasemnogene abeparvovec is a gene replacement therapy that restores SMN protein levels in motor neurons. Neurofilament light chain (NfL) is a biomarker of neuroaxonal injury measurable both in plasma and cerebrospinal fluid, and may serve as a biomarker of disease activity. Objective: We evaluated the association between post-treatment plasma NfL levels and motor outcomes in SMA patients receiving onasemnogene abeparvovec. Methods: We retrospectively analyzed 21 patients with genetically confirmed SMA who received onasemnogene abeparvovec and had post-treatment plasma NfL levels recorded. NfL concentrations were measured using a chemiluminescence assay. Motor outcomes were assessed by comparing milestones achieved before therapy and at the most recent follow-up. Covariates included prior disease-modifying treatment (nusinersen or risdiplam), age at infusion, SMN2 copy number, and time from therapy to NfL measurement. Results: Twenty one patients were eligible for analysis (11 male; median age at infusion, 17 months; range, 3–45 months). Thirteen patients had SMA type 1, seven had type 2, and one was presymptomatic. Eight patients were treated with nusinersen prior to gene therapy, twelve with risdiplam and one did not receive any prior treatment. Post-treatment plasma NfL values ranged from 7,5 to 195,6 pg/mL. The reference ranges, between 10th percentile and 90th percentile, by age are: Age 0-1 (5,6-16,9 pg/ml) Age 1-2 (5,1-14,7 pg/ml) Age 2-4 (4,7-12,7 pg/ml) Age 4-6 (3,9-9,6 pg/ml). Three patients classified as SMA 1 showed no motor milestone gains; their NfL levels were 43,37 pg/ml, 142.6 pg/mL and 195,6 pg/mL. Using Spearman’s rank correlation test to examine the relationship between NfL levels and motor milestone gain, we found rho value of -0,462 (p=0.035), indicating that persistently elevated plasma NfL after gene therapy was associated with poorer motor outcomes. The identified cutoff was 55.55 pg/mL, yielding an AUC of 0.821, indicating good discriminative capacity. At this threshold, sensitivity was 75.0% and specificity was 85.7%. Conclusion: In this cohort of 21 SMA patients treated with gene therapy, higher persistent post‑treatment plasma NfL levels were associated with poorer motor outcomes (Spearman rho = −0.462, p = 0.035), with three non‑responders showing markedly elevated NfL (43.4–195.6 pg/mL). ROC analysis identified a cutoff of 55.55 pg/mL (AUC 0.821), with 75% sensitivity and 85.7% specificity, suggesting good discriminative value of post‑treatment NfL for predicting motor milestone gain. These findings support plasma NfL as a promising prognostic biomarker after gene therapy in SMA, though prospective validation in larger cohorts is warranted before routine clinical implementation.