NfL correlated with disease activity and was shown to be a treatment response marker mainly in active SPMS (aSPMS), whereas GFAP was responsive to treatment effects in both aSPMS and non-active SPMS (naSPMS) patients. Therefore, the combined information of both markers may enable to refine the identification of patients at risk for progression.

In this post-hoc analysis of the EXPAND study (core+extension), we quantified baseline pNfL and pGFAP levels using Single Molecule Array technology. Time-to-3-month confirmed disability worsening (3mCDW), time-to-Expanded Disability Status Scale (EDSS)7.0, and time-to-1-point-sustained EDSS worsening by baseline pNfL and pGFAP categories (pNfL: low [<30pg/mL] versus high [≥30pg/mL]; pGFAP: low [<130pg/mL] versus high [≥130pg/mL) were assessed using a Cox-regression model adjusted for age, gender, disease duration, treatment, relapses in the 24months before study initiation, and baseline EDSS. Analyses were conducted in patients with aSPMS versus naSPMS.

Samples from 1369/1651 patients were analyzed. In naSPMS patients (n=704), the high-high signature showed the higher hazard ratios (HRs) versus the low-low signature across all disability outcomes: time-to-EDSS 7.0 (2.65, p=0.0014), time-to-1-point-sustained EDSS worsening (1.57, p=0.0176), and time-to-3mCDW (1.45, p=0.0151). Higher HRs were also found with the high-low signature (range: 2.09-1.17 across all three outcomes; p<0.05, except for time-to-EDSS 7.0). In aSPMS patients (n=665), the higher risk of disability worsening was associated with the high-low signature for time-to-3mCDW (HR, 1.65; p<0.05) and time-to-1-point-sustained EDSS worsening (HR, 1.66; p<0.05), and the high-high signature for time-to-EDSS 7.0 (HR, 1.38; p>0.05).