Clinical application of age-derived cut-offs for plasma neurofilament light chain in multiple sclerosis

Clinical use of neurofilament light chain (NfL) requires cut-off values that reflect disease status independently of confounding factors, such as age, hemodilution, and cardiovascular risk factors. We compared the performance of different previously suggested cut-offs in separating MS cases and controls, and in identifying different MS clinical features, across age groups.

In this cross-sectional study, we included people with MS (n = 312) and age-, sex-, and eGFR-matched controls (n = 236). For MS cases, we collected descriptor of disease progression (relapsing or progressive), EDSS, and evidence of disease activity in the previous year (including relapses, active MRI, and EDSS progression) and disease duration. Plasma NfL (pNfL) was evaluated using Lumipulse™ fully automated chemiluminescent enzyme immunoassay. We then classified both MS cases and controls based on pNfL suggested by Simrèn et al. (specific for different age ranges), Vermunt et al. (age-derived percentiles), and Benkert et al. (age- and BMI-derived percentiles).

In individuals aged 18-50 years, the three suggested pNfL cut-offs provided high specificity (> 85%) in discriminating MS cases and controls (AUC = 0.73; 95%CI = 0.67, 0.78; p = 0.028). In the MS population, the three suggested pNfL cut-offs provided high sensitivity (> 75%) in discriminating relapsing and progressive cases (AUC = 0.70; 95%CI = 0.63, 0.77; p = 0.034), patients with EDSS ≥ 4.0 and EDSS < 4.0 (AUC = 0.69; 95%CI = 0.63, 0.76; p = 0.032), and patients with EDSS ≥ 6.0 and EDSS < 6.0 (AUC = 0.70; 95%CI = 0.62, 0.78; p = 0.040). The three suggested pNfL cut-offs provided lower accuracy in age groups older than 50 years.

Previously validated cut-offs provided similar sensitivity and specificity in separating MS cases and controls and in identifying MS clinical features across different age groups, with the best performance before 50 years.