Neurofilament light chain – Can it be measured in urine?

Abstract

Objective

This exploratory study investigates if neurofilament light chain (NfL) is excreted in the urine and whether this depends on plasma NfL (pNfL) levels and kidney function in terms of eGFR and urine albumin-creatinine ratio (uACR).

Methods

Using a computer algorithm, we identified excess urine and plasma from routine testing of uACR and eGFR in patients 45–50 years old. Up to 17 paired urine-plasma samples in each of six categories of kidney function defined by uACR and eGFR were analysed for NfL, and the urinary NfL-creatinine ratio (uNCR) was calculated to correct for urine dilution.

Results

In the 35 subjects with normal eGFR (>90 ml/min/1.73 m²) and varying degrees of albuminuria, uNfL was only above the lower limit of quantification in one subject with microalbuminuria (uACR 30–300 mg/g), and in none of the subjects with uACR < 30 mg/g. In the 47 subjects with impaired eGFR (15–60 ml/min/1.73 m²), the percentage of subjects with detectable uNfL and the average level of uNCR increased with increasing albuminuria. However, multiple regression analysis revealed that uNCR only significantly correlated with pNfL, not eGFR and uACR. pNfL correlated inversely with eGFR, but not uACR.

Conclusions

Our results show that the urinary NfL excretion in subjects with normal kidney function and normal pNfL levels is very low. Thus, the increased pNfL often observed in patients with low eGFR seems not to be explained by impaired urinary NfL excretion, and urine is generally not a suitable matrix for NfL measurements.