Longitudinal short-term assessments of single- and multi-protein disease activity biomarkers to determine reliability and the minimum detectable change in people with clinically stable RRMS

Abstract

Objectives:

This study aims to evaluate the ability of a multi-protein biomarker score to characterize clinical stability in relapsing-remitting multiple sclerosis (RRMS) compared to that of single-protein markers.

Methods:

Serum was collected from 62 out of 86 MOVING MS participants over 12 months (N_samples = 186). Multiple Sclerosis Disease Activity (DA) scores were calculated from 18 proteins, sex, and age. In a clinically stable subcohort, the DA score reliability was compared to single biomarkers, and the minimum detectable change at 95% confidence (MDC95) was estimated. An independent cohort (N = 128) was used to assess whether change in the DA score > MDC95 was associated with odds of gadolinium-enhancing (Gd+) lesions.

Results:

In 34 clinically stable MOVING MS participants, the DA score demonstrated directionally stronger test–retest reliability (r_XX′ = 0.92, 95% CI = [0.79, 0.97]) than NfL (r_XX′ = 0.75, 95% CI = [0.47, 0.91]) and GFAP (r_XX′ = 0.71, 95% CI = [0.46, 0.89]). The DA score MDC95 was 1.24 units (95% CI = [0.74, 1.73]). In the independent cohort, DA score increases that were greater/less than the MDC95 were significantly associated with increased/decreased odds of Gd+ lesions across the DA score range.

Conclusions:

The DA score directionally outperformed NfL and GFAP in test–retest reliability. The MDC95 of 1.24 units provides a threshold for interpreting DA score changes, supporting its use for monitoring disease activity in RRMS.