We developed and validated a Simoa NfL assay using the capture monoclonal antibody (mAB) 47:3 and the biotinylated detector mAB 2:1 from UmanDiagnostics,31 transferred onto the Simoa platform. mAB 47:3 was buffer exchanged and diluted to 0.3mg/ml. Paramagnetic beads (4 × 106; Quanterix Corporation, Lexington, MA) were buffer exchanged and activated using 0.5mg/ml 1‐ethyl‐3‐(3‐dimethylaminopropyl) carbodiimide (Quanterix), followed by a 30‐minute incubation at room temperature (RT; HulaMixer; Thermo Fisher Scientific, Waltham, MA). During a 2‐hour incubation at RT (HulaMixer) the diluted capture mAB was conjugated with the washed and activated beads. Subsequently, the beads were washed and blocked. After 3 washes, the conjugated beads were suspended and stored at 4°C. Biotinylated mAB 2:1 was obtained from UmanDiagnostics and stored at 4°C pending analysis.
The assay was run on a Simoa HD‐1 instrument (Quanterix) using a 2‐step Assay Neat 2.0 protocol; 100μl of calibrator/sample (diluent: Tris‐buffered saline [TBS], 0.1% Tween 20, 1% milk powder, 400μg/ml Heteroblock [Omega Biologicals, Bozeman, MT]), 25μl conjugated beads (diluent: TBS, 0.1% Tween 20, 1% milk powder, 300μg/ml Heteroblock), and 20μl of mAB 2:1 (0.1μg/ml; diluent: TBS, 0.1% Tween 20, 1% milk powder, 300μg/ml Heteroblock) were incubated for 47 cadences (1 cadence = 45 seconds). After washing, 100μl of streptavidin‐conjugated β‐galactosidase (150pM; Quanterix) was added, followed by a 7‐cadence incubation and a wash. Prior to reading, 25μl Resorufin β‐D‐galactopyranoside (Quanterix) was added. Calibrators (neat) and samples (serum: 1:4 dilution; CSF: 1:10 dilution) were measured in duplicates. Bovine lyophilized NfL was obtained from UmanDiagnostics. Calibrators ranged from 0 to 2,000pg/ml for serum and from 0 to 10,000pg/ml for CSF measurements. Batch prepared calibrators were stored at −80°C.
Intra‐ and interassay variability of the assay was evaluated with 3 native serum and 3 native CSF samples in 22 and 12 consecutive runs on independent days, respectively. For serum, the mean coefficients of variation (CVs) of duplicate determinations for concentration were 5.6% (13.3pg/ml, sample 1), 6.9% (22.5pg/ml, sample 2), and 5.3% (236.5pg/ml, sample 3). In CSF, the mean intra‐assay CVs were 2.5% (572.6pg/ml, sample 1), 0.7% (1,601.8pg/ml, sample 2), and 3.8% (6,110.2pg/ml, sample 3). Interassay CVs for serum were 11.3% (sample 1), 9.3% (sample 2), and 6.4% (sample 3). In CSF, interassay CVs were 10.1% (sample 1), 6.2% (sample 2), and 15.5% (sample 3). We used the concentration of the lowest calibrator fulfilling acceptance criteria (accuracy = 80–120%, CV of duplicate determination ≤ 20%) as an estimate of the analytical sensitivity.32 The analytical sensitivity was 0.32pg/ml. All samples produced signals above the analytical sensitivity of the assay. Few samples with intra‐assay CVs > 20% were repeat measured. Recovery rates ([concentration of spiked sample − concentration of native sample]/spiked concentration × 100) were tested in 4 serum and 4 CSF samples from HC spiked with 5, 50, and 200pg/ml and 500 and 2,000pg/ml of NfL, respectively. The mean recovery after spiking was 107% for serum and 121% for CSF. Parallelism and linearity of the assay for serum and CSF were confirmed by serial dilution experiments.32
Disanto G., Barro C., Benkert P., Naegelin Y., Schädelin S., Giardiello A., Zecca C., Blennow K., Zetterberg H., Leppert D., Kappos L., Gobbi C., Kuhle J., Kuhle J., Lorscheider J., Yaldizli Ö., Derfuss T., Kappos L., Disanto G., Zecca C., Gobbi C., Benkert P., Achtnichts L., Nedeltchev K., Kamm C.P., Salmen A., Chan A., Lalive P.H., Pot C., Schluep M., Granziera C., Du Pasquier R., Müller S, & Vehoff J. (2017). Serum Neurofilament light: A biomarker of neuronal damage in multiple sclerosis. Annals of Neurology, 81(6), 857-870.