LOS ANGELES – Serum neurofilament light chain shows promise as a biomarker for disease severity and response to treatment with autologous bone marrow transplant in patients with multiple sclerosis, according to findings from an analysis of paired samples.
Serum neurofilament light chain (Nf-L) levels were found to be significantly elevated in both the serum and cerebrospinal fluid (CSF) of 23 patients with aggressive multiple sclerosis (MS), compared with samples from 5 controls with noninflammatory neurologic conditions such as chronic fatigue syndrome or migraine. The levels in the patients with MS were significantly reduced following autologous bone marrow transplant (BMT),, reported in a poster at the annual meeting of the American Academy of Neurology.
Nf-L has been shown to be a biomarker for neuronal damage and to have potential for denoting disease severity and treatment response in MS; for this analysis, samples from patients were collected at baseline and annually for 3 years after transplant. Samples from controls were obtained for comparison.
“Our objective, really, was [to determine if we could] detect a treatment response in neurofilament light chain, and secondly, did the degree of neurofilament light chain being high at baseline predict anything about subsequent disease outcome?” Dr. Thebault, a third-year resident at the University of Ottawa, reported in the poster.
Indeed, a treatment response was detected; baseline levels were high, and levels in both serum and CSF were down significantly (P = .05) at both 1 and 3 years following bone marrow transplant, and they stayed down. “In fact, they were so low, they were not significantly different from [the levels] in noninflammatory controls,” he said, noting that serum and CSF NfL levels were highly correlated (r = .833; P less than .001).
Study subjects were patients with aggressive MS, characterized by early disease onset, rapid progression, frequent relapses, and poor treatment responses. Such patients who have received autologous BMT represent an interesting group for examining treatment responses, he said.
At baseline, these patients presumably have a high burden of tissue injury, which would be representative of high levels of Nf-L; good, prospective data show these patients have no evidence of ongoing inflammatory disease following an intensive regimen that includes chemoablation and reinfusion of autologous stem cells, which is representative of a significant reduction of neurofilament levels, he explained.
Importantly, serum and CSF levels were correlated in this study, he said, noting that most prior work has involved CSF, but “the real clinical utility in neurofilament light chain is probably in the serum, because patients don’t like having lumbar punctures too often.”
With respect to the second question pertaining to disease outcomes, the study did show that patients with baseline Nf-L levels above the median for the group had worse T1 lesion volumes at baseline and after transplant (P = .0021 at 36 months), and also had worsening brain atrophy even after transplant (P = .0389 at 60 months).
The curves appeared to be diverging, suggesting that, in the absence of inflammatory disease activity, patients with high Nf-L levels at baseline continue to have ongoing atrophy at a differential rate, compared with those with low baseline levels, Dr. Thebault said.
Other findings included better Expanded Disability Status Scale outcomes after transplant in patients with lower baseline Nf-L, and a trend toward worsening outcomes among those with higher baseline Nf-L levels, although the study may have been underpowered for this. Additionally, lower baseline Nf-L predicted significantly lower T1 and T2 lesion volume, number of gadolinium-enhancing lesions, and a reduction in brain atrophy, whereas higher baseline Nf-L predicted the opposite, he noted, adding that N-acetylaspartate-to-creatinine ratios also tended to vary inversely with Nf-L levels.
The most important findings of the study are “the sustained reduction in Nf-L levels after BMT, and that higher baseline levels predicted worse MRI outcomes,” Dr. Thebault noted. “Together these data add to a growing body of evidence that suggests that serum Nf-L has a role in monitoring treatment responses and even a predictive value in determining disease severity.”
This study was supported by Quanterix, which provided Nf-L assay kits. Dr. Thebault reported having no disclosures.
SOURCE: Thebault S et al. Neurology. 2018 Apr;90(15 Suppl.):