Serum neurofilament light-chain levels and long-term treatment outcomes in relapsing-remitting multiple sclerosis patients: A post hoc analysis of the randomized CombiRx trial
Gary Cutter, Richard A. Rudick, Carl de Moor, Carol M. Singh, Elizabeth Fisher, Thijs Koster, Fred Lublin, Jerry S. Wolinsky, Henry F. McFarland, Steven Jacobson, Maria L. Naylor
Abstract
Background: CombiRx was a randomized, double-blind, placebo-controlled phase 3 trial in treatment-naive relapsing-remitting multiple sclerosis (RRMS) patients randomized to intramuscular interferon beta-1a (IM IFN beta-1a), glatiramer acetate (GA), or both therapies. Objective: This analysis investigated changes in serum neurofilament light-chain (sNfL) levels in response to treatment and assessed baseline sNfL as a predictor of relapse. Methods: RRMS patients treated with IM IFN beta-1a 30 µg weekly + placebo (n = 159), GA 20 mg/mL daily + placebo (n = 172), or IM IFN beta-1a + GA (n = 344) were included. A linear mixed model compared sNfL values over time. Cox regression models analyzed baseline sNfL and gadolinium-enhancing (Gd+) lesions as predictors of relapse. Results: In all treatment arms, the proportion of patients with sNfL ≥16 pg/mL decreased significantly from baseline to 6 months and was maintained at 36 months. A significantly higher percentage of patients with both baseline sNfL ≥16 pg/mL and ≥1 Gd+ lesion experienced relapses within 90 days compared to patients with sNfL <16 pg/mL and/or no Gd+ lesions. Conclusion: sNfL levels were reduced within 6 months and remained low at 36 months. Results suggest that the combination of lesion activity and sNfL was a stronger predictor of relapse than either factor alone.