Highlights from

AAN 2019

American Academy of Neurology annual meeting

Philadelphia, USA 4-10 May 2019

Clinical relevance of neurofilament light chain levels

Serum neurofilament light chain (NfL) reflects neuroaxonal damage. It holds promise as a quick and easy biomarker of MS disease course and treatment response.

Data from the ozanimod phase 3 studies suggested that serum NfL levels may be a marker of acute disease in relapsing MS, correlate with the number of Gd+ and T2 lesions, and potentially predict future relapse [1]. As was shown by studies presented at the AAN 2019 meeting, it is also predictive of ongoing disease activity in radiologically isolated syndrome and of cognitive impairment in secondary progressive MS (SPMS):

  • In 62 patients with radiologically isolated syndrome from 4 French centres, elevated serum and CSF NfL, IgG oligoclonal bands (OCB), and 2005 MRI criteria were found to be predictive of evidence of disease activity (HR 1.87, P=0.047; HR 2.19, P=0.019; and HR 2.21; P=0.025; respectively) [2]. However, only CSF NfL levels predicted clinical conversion during follow-up (log rank, P=0.033). The authors concluded that if these outcomes can be replicated in larger datasets, these biomarkers may inform on treatment decisions in this population.

  • In an analysis of 1,397 SPMS patients, high serum NfL levels at baseline were associated with an elevated risk for prolonged cognitive worsening speed [3]. This was the case regardless of the presence of Gd+ T1 lesions, but the risk of worsening was more pronounced with Gd+ T1 lesions present at baseline. The cohort was randomised to siponimod or placebo in the phase 3 EXPAND trial. Patients with high baseline NfL levels had a 41.4% higher risk of reaching time to 6-month confirmed worsening on SDMT (by 4 points from baseline, 6mCWSDMT) compared to patients with low NfL levels (P=0.0103). In patients without Gd+ T1 lesions at baseline (n=1,060) this percentage was 34.2% (P=0.061). Patients with Gd+ T1 lesions and high baseline NfL at baseline (n=296) had 135.2% higher risk of 6mCWSDMT (P=0.0289).

  • Integration of serum NfL into clinical practice requires a standardised validated assay and establishing clinically meaningful cut-offs. Using serial samples from >1,400 patients in four phase 3 trials (ADVANCE, CHAMPS, MSCRG, and SENTINEL), feasibility of establishing relevant cut-offs for disease severity stratification and treatment monitoring in RRMS patients was demonstrated [4]. In the analysed trials, serum NfL was measured with a SIMOA NF-light® Advantage Kit (Quanterix) or laboratory method, with the following results:

  • baseline serum NfL levels were associated with number of enhancing lesions and accumulation of new T2 lesions (P<0.0001);

  • serum NfL >16 pg/mL indicated high probability of disease activity over the following year (positive predictive value 91%);

  • combining three serum NfL measurements improved positive predictive value from 91% to 94%;

  • serum NfL>16 pg/mL was associated with long-term with worse outcomes;

  • serum NfL levels were lowered by disease-modifying treatments; natalizumab reduced serum NfL to <16 pg/mL in 96% of patients.

  1. Harris S, et al. AAN 2019, S56.006.
  2. Thouvenot E, et al. AAN 2019, S37.003.
  3. Kuhle J, et al. AAN 2019, S12.009.
  4. Calabresi PA, et al. AAN 2019, S26.001.

The content and interpretation of these conference highlights are the views and comments of the speakers/authors.