sNfL a biomarker of future disease activity: APLIOS


Periodic assessment of serum neurofilament-light chain (sNfL) may be useful to identify subclinical disease activity, according to an analysis from the APLIOS study (Bar-Or et al. Neurol Ther 2023;12:303-317).

APLIOS was an open-label phase II trial of ofatumumab in 284 patients with relapsing MS. The study collected 14 blood samples over the 12-week observation period (Bar-Or et al. Mult Scler 2022;28:910-924). MRIs were obtained every four weeks. Frequent testing enabled the group to evaluate the temporal relationship between sNfL and disease activity.

The median sNfL level at baseline was 9.1 pg/mL. Comparisons were made between groups above vs. below the median. Patients in the above-median group were older (38.6 vs. 36.1 years), with a lower BMI (24.58 vs. 26.42 kg/m2), a longer disease duration (10.19 vs. 8.40 years), and a higher number of Gd+ T1 lesions at baseline (2.6 vs. 0.5). A high sNfL level at baseline was prognostic of relapses and/or Gd+ lesions (hazard ratio 2.81 vs. low sNfL).

The risk of disease activity was similar in the subgroup with high sNfL and no Gd+ lesions at baseline (HR 2.48). In this group, sNfL levels increased in the month before a new Gd+ lesion, reached a maximum a few weeks later then declined thereafter. Thus, sNfL changes appear to have a longer duration than Gd+ lesions, enabling clinicians to detect new inflammatory activity that would have been otherwise missed without frequent MRIs.

Disease activity was also analysed according to temporal changes in NfL. The proportion of patients who relapsed was higher in those with sNfL consistently above the median (15%) compared to those with levels below the median (2.6%) or who crossed the median (2.7%). The proportion that developed Gd+ lesions also differed among the groups (64.7% vs. 16.1% vs. 31.1% at week 4).

Similarly, the rate of no evidence of disease activity (NEDA) was lower in the group with high sNfL levels (21.7%) compared to those with low sNfL (65.4%) or those who crossed the median (50.0%).

High sNfL and number of Gd+ lesions at baseline were similarly prognostic of disease activity even during treatment with a high-efficacy therapy. Where NfL levels appeared especially useful were in predicting relapses/MRI activity in patients without baseline Gd+ lesions. The authors further noted that NfL remained predictive of disease activity during therapy, suggesting a role for serial NfL assessments throughout the treatment course. Further work is needed to determine an optimal cut-off value for identifying patients with high baseline sNfL.

Recommend to a Colleague

Related Posts

Go back to home page