Impact of MRI Activity and Medical Comorbidities on Serum Neurofilament Light Chain in Multiple Sclerosis
Hira Chouhdry1, Katherine Schleiss2, Matthew Baker3, Pearse Morris4, W. Tobin5, Eoin Flanagan6, Orhun Kantarci5, Mark Keegan5, M. Paz Soldan5, Sean Pittock6, Samantha Banks5
1Mayo Clinic Department of Neurology, 2Mayo Clinic Alix School of Medicine, 3Mayo Clinic Department of Neurologic Surgery, 4Mayo Clinic Department of Radiology, 5Mayo Clinic Department of Neurology, Center for MS and Autoimmune Neurology, 6Mayo Clinic Department of Neurology, Center for MS and Autoimmune Neurology, Division of Laboratory Medicine and Pathology
Objective:
Investigate comorbidities that may influence differential changes in sNfL among a clinical cohort of pwMS.
Background:
Serum neurofilament light chain (sNfL) is a marker of axonal injury, yet some people with multiple sclerosis (pwMS) with enhancing MRI lesions show no sNfL elevation. Comorbidities including renal disease and diabetes mellitus may elevate sNfL, whereas obesity may decrease it, potentially confounding interpretation.
Design/Methods:
Retrospective study of Mayo Clinic pwMS ≥20 years who underwent clinical sNfL testing using Simoa assay. Those with elevated sNfL for age, based on population reference data, were age- and sex-matched 1:1 to pwMS with normal sNfL. Obesity was defined as body mass index (BMI) ≥30 and renal disease as CKD stage 1 or worse. Continuous and binary variables were compared using Wilcoxon signed-rank and McNemar’s tests, respectively.
Results:
169 pwMS with elevated sNfL compared to 169 age- and sex-matched pwMS with normal sNfL (median age at time of sNfL draw 39 years, 65% female). Median sNfL values were 23.1 pg/mL (elevated group) and 6.6 pg/mL (normal group). Patients with elevated sNfL were more likely to have enhancing lesions on brain MRI (67/132[51%] vs. 23/132[17]%, p<0.0001) or spinal cord MRI (22/74[30%] vs. 9/74[12%], p=0.01) within 3 months of sNfL draw. Positive predictive value of elevated sNfL for any enhancing lesion was 57% (sensitivity 72%), while negative predictive value of normal sNfL without enhancing lesion was 78% (specificity 64%). No differences were found in BMI (median elevated sNfL 27.7 vs. 29.0, p=0.22), obesity (65/167[39%] vs. 72/167[44%], p=0.43), renal disease (12/151[8%] vs. 16/151[11%], p=0.5), or diabetes (8/49[16%] vs. 6/49[12%], p=0.75).
Conclusions:
In this age- and sex-matched clinical cohort, elevated sNfL was correlated with but only moderately predictive of active lesions on MRI. Comorbidities known to alter sNfL did not account for observed differences in NfL elevation.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.