pNF-H Has Potential As Biomarker of Spinal Muscular Atrophy Disease Activity

Basil Darras, MD, Associate Neurologist-in-Chief; Chief, Division of Clinical Neurology; Director, Neuromuscular Center and Spinal Muscular Atrophy Program; Professor of Neurology, Harvard Medical School

Darras Basil, MD

An observed and significant decline in plasma phosphorylated neurofilament heavy chain (pNF-H) and the following stability as a result of treatment with nusinersen (Spinraza, Biogen) in infant patients with spinal muscular atrophy (SMA) has suggested that pNF-H may have a role as a biomarker of disease activity and treatment response in the genetic neuromuscular disease.¹

Ultimately, treatment with nusinersen was associated with a rapid and solid decline in pNF‐H levels, down 71.9% at 2 months and down 90.1% at 10 months compared to sham-treated infants who experienced declines of 16.2% and 60.3% at 2 months and 10 months, respectively. Levels were measured in 117 patients with SMA from the ENDEAR (NCT02193074) study, with mean baseline SMA levels being 15,400 pg/mL (range, 2390 to 50,100), in comparison to 167.0 pg/mL (range, 7.46 to 7,030) in children without SMA.

“Disease biomarkers can serve many purposes; they can inform prognosis, signal disease progression, predict response to therapy, and/or mark a key process of disease that is influenced by therapy,” the investigators wrote. “Baseline correlations between pNF‐H levels and clinical characteristics indicative of disease severity in the ENDEAR population suggest that elevated levels of pNF‐H may have prognostic value because they correlate with disease course.”

Led by Basil Darras, MD, Associate Neurologist-in-Chief; Chief, Division of Clinical Neurology; Director, Neuromuscular Center and Spinal Muscular Atrophy Program; Professor of Neurology, Harvard Medical School, the study found that ultimately, the median baseline levels of pNF-H were approximately 10&#8208;fold higher than that of age&#8208;matched infants without SMA (P <.0001) and about 90&#8208;fold higher than in children without SMA (P <.0001).

“Together these data suggest plasma pNF&#8208;H is a promising marker of disease activity/treatment response in infants with SMA,” Darras and colleagues detailed. They did acknowledge, though, that there is still a presently unknowable time-based association between the onset of pNF-H’s release into the interstitial fluid and the irreversible development in disease-related neurodegeneration. Until the cascade of events which either precede, overlap, or trail the potential biomarker’s release are understood, this will remain unknown, they noted.

“Nonetheless, the early, marked elevation of pNF&#8208;H suggests that levels may also be elevated in pre-symptomatic children, such as those identified during newborn screening, indicating an underlying degenerative process and justifying treatment initiation,” they wrote. “In those identified during newborn screening who have ≥ 4 SMN2 copies, monitoring pNF&#8208;H levels may facilitate decisions on the appropriate time to initiate therapy to achieve the optimal outcome.”

All told, the average level of pNF-H was 536.6 pg/mL (standard error [SE], ±231.04), with the lowest levels observed in the youngest—aged <1 year&mdash;patients (was 1510 pg/mL; range, 579 to 7030; n = 6), while those who were older than 1 year but younger than 18 years had mean levels of with 124.5 pg/mL (range, below the limit of quantification [BLQ] to 395; n = 28) in those aged 1 to 18 years (P = .0002).

Notably, despite the use of various methods and assays in measurement of these levels, the pNF-H values were comparable with previously published results, Darras and colleagues wrote.

At the end of the 302-day treatment period, the nusinersen-treated patients had lower absolute levels of pNF-H (1465.5 pg/mL; range 41 to 5180; n = 30) than the sham group (5664.7 pg/mL; range, 1100 to 10,600; n = 15). Although, they were elevated compared to age-matched infants without SMA (2516.3 pg/mL; range, 579 to 7030; n = 6). At Day 302, 13% participants (n = 2) in the sham control group and 73% of patients (n = 22) in the nusinersen&#8208;treated group had pNF&#8208;H levels <1940 pg/mL, which 75% of those without SMA younger than 2 years (n = 9) were below.

REFERENCE

1. Darras BT, Crawford TO, Finkel RS, et al. Neurofilament as a potential biomarker for spinal muscular atrophy. Ann Clin Transl Neur. Published online April 17, 2019. Accessed April 24, 2019. doi: 10.1002/acn3.779.