a new study has shown.
The study found that current serum NfL levels predicted relapses, disability worsening, and MRI activity in the following year independent of standard metrics for treatment monitoring, such as relapse rate, disability worsening, and MRI findings. The biomarker also detected subclinical disease activity in patients with no evidence of disease activity (NEDA3), as measured by absence of previous relapses, worsening score on the Expanded Disability Status Scale (EDSS), or brain lesion formation on MRI.
“Our data in this well-characterized large real-world cohort supports the value of serum NfL levels for treatment monitoring in MS clinical practice,” lead author Özgür Yaldizli, MD, concluded.
Dr. Yaldizli, who is a consultant neurologist at University Hospital Basel (Switzerland), presented the findings at the Joint European Committee for Treatment and Research in Multiple Sclerosis–Americas Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS-ACTRIMS) 2020, this year known as MSVirtual2020.
“This is the first study to compare NfL simultaneously with other markers of disease progression, such as MRI lesions and relapse rate in treated patients. We show that NfL gives a unique signal that is not captured by other markers,” Dr. Yaldizli said.
“This is likely the largest study of NfL in MS to date, with more than 7,000 samples from well-characterized MS patients followed longitudinally for more than 5 years of sampling and including high quality data on MRI and clinical examinations. It is the first time all these factors have been combined so that we can see how NfL compares with other markers of disease progression in predicting clinical events and monitoring treatment efficacy,” said senior author Jens Kuhle, MD, PhD, also from University Hospital Basel.
Large normative database for reference
The researchers also reported a large normative database of NfL values with data from more than 8,000 healthy controls. “This is the largest normative database to date, that gives us reliable reference values for NfL across a range of ages and comorbidities,” Dr. Kuhle noted.
Dr. Jens Kuhle
In his presentation, Dr. Yaldizli explained that NfL is a neuronal cytoskeletal protein released into the cerebrospinal fluid and blood following neuroaxonal injury. Although numerous studies have shown that serum NfL is associated with clinical and MRI disease activity and treatment response, it is not clear whether serum NfL under established disease-modifying therapy (DMT) can identify patients with suboptimal treatment response, compared with standard clinical and MRI activity measures.
This study addressed that question in the large real-world Swiss MS cohort.
The study involved 1,366 patients (88.8% with relapsing remitting MS [RRMS], 5.4% with secondary progressive MS, and 5.8% with primary progressive MS) receiving DMT for at least 3 months from seven MS centers. The median disease duration was 7.2 years. Serum NfL was measured every 6 or 12 months with NF-Light assay on the latest-generation HDX platform (blinded for clinical and MRI data). The median follow-up was 4.9 years. There was an average of five samples per patient, with a total of 7462 samples.
Results showed that NfL levels were higher in older patients (14.5% per 10 years), those with secondary progressive MS (12.4% vs. RRMS), those with primary progressive MS (14.4% vs. RRMS), and in those who had a relapse in the last 4 months (53.4%).
NfL levels were 13.4% lower in patients receiving oral DMT (vs. untreated patients) and 17.7% in patients receiving monoclonal antibodies (vs. untreated patients).
In the large cohort of healthy controls, NfL levels also increased with age, but levels in patients with MS were higher than in controls across the whole age spectrum.
To obtain a measure of deviation from normal, the authors converted NfL levels to z score, which express how much (in terms of number of standard deviations) a measurement differs from mean values found in healthy controls of the same age. Effects were more pronounced with use of z score derived from the normative database than with use of absolute NfL levels even after adjustment for age.
In the univariate analysis, serum NfL z score predicted relapse or EDSS worsening in the following year: The higher the z score, the higher the risk for relapse or EDSS worsening. Patients with an NfL z score greater than 1 had a 41% higher risk for relapse or EDSS worsening in the following year, compared with those whose z score was less than 1 (odds ratio, 1.41).
Patients with an NfL z score exceeding 1.5 had an 80% higher risk for relapse or EDSS worsening in the following year than did those whose score was below 1.5 (OR, 1.8).
Patients with an NfL z score greater than 2 had a 2.3 times higher risk for relapse or EDSS worsening in the following year versus those with a score below 2. (P < 0.001 for all comparisons.)