Neurofilament light chain as a biomarker for monitoring response to change in treatment in hereditary ATTR amyloidosis.

Abstract

Hereditary ATTR (ATTRv) amyloidosis is a fatal autosomal dominant disorder in which variants in the transthyretin (TTR) gene cause systemic deposition of amyloid fibrils. ATTRv amyloidosis is characterised by progressive lengthdependent sensorimotor neuropathy, autonomic neuropathy, and non-neuropathic manifestations [1]. Several effective pharmacological therapies are available, including TTR tetramer stabilisers (diflunisal and tafamidis) and nucleic acid-based medications such as RNA interference therapeutics (patisiran and vutrisiran) and anti-sense oligonucleotide (inotersen) [1,2]. However, evaluation and monitoring of disease progression and treatment response are challenging in clinical practice, because existing assessment methods are not sensitive enough to detect changes in disease activity. Neurofilament light chain (NfL) is an integral component of the neurons, and has been described as a biomarker of neuroaxonal injury across several central and peripheral nervous system diseases. Additionally, recent analyses have identified NfL as a potential biomarker of neuronal injury in ATTRv amyloidosis [3,4]. In this analysis, we evaluated the value of NfL in monitoring treatment response in patients with ATTRv amyloidosis with polyneuropathy whose treatment was switched from tafamidis to patisiran. Eleven patients (10 male) with ATTRv amyloidosis with polyneuropathy who switched treatment from tafamidis (20mg oral once daily) to patisiran (0.3mg/kg IV once in every three weeks) were included. The reason for switch was worsening of polyneuropathy. Serum NfL levels of patients were measured at baseline (while receiving tafamidis, just before switch), one and two years after switch to patisiran. In eight of the eleven patients, NfL levels were available at two years following switch. Patients were also assessed for neurologic impairment with neuropathy impairment score (NIS) at baseline and one and/or two years following switch. Since the patients were assessed during routine visits, NfL and NIS measurements at 1-year and 2-year post-switch occurred within 3months of those time points, except one patient whose 2-year measurements were delayed by four months. NfL levels were measured using the QuanterixR Simoa platform. The Wilcoxon signed-rank test was used to compare NfL levels and NIS values before, and one and two years after switch to patisiran. p< 0.05 was considered statistically significant. This study was approved by the Ethical Committee of Shinshu University School of Medicine (No. 4852) and written informed consent was obtained from patients. Mean (±SD) age at disease onset and study inclusion were 39.5 ± 12.6 and 46.9 ± 14.0 years, respectively. TTR variants were V30M (p.V50M) (n1⁄4 8), A36P (p.A56P), S50A (p.S70A), and I107V (p.I127V), (n1⁄4 1 each). NfL and NIS measurements at baseline were available for all 11 patients. Serum NfL levels were available for 11/11 and 8/11 patients at one and two years following switch, respectively. NIS measurements were conducted in 9/11 and 8/11 patients at one and two years after switch, respectively. Serum NfL levels significantly decreased at one and two years following switch from tafamidis to patisiran (Figure 1). The mean (±SD) NfL level at baseline, before switch, was 106.4 (±50.7) pg/mL and it decreased to 72.6 (±47.3) pg/mL one year after switch (p1⁄4 0.001). In patients whose NfL levels were available at two years post-switch, mean (±SD) value decreased from 92.8 (±47.0) pg/mL before switch to 55.9 (±36.7) pg/mL following two years of patisiran treatment (p1⁄4 0.008). Additionally, NfL levels increased while on tafamidis in more than half of seven patients with NfL measurements available from before switch (Figure 1). No significant change was observed in NIS values at one and two years following switch. The mean (±SD) NIS values were 40.4 (±43.7) before switch and 46.3 (±49.6) at one year after switch to patisiran treatment (p1⁄4 0.180). In the patients with two-year assessment, the values before and