Analytical validation of an ultrasensitive multiplexed immunoassay for blood-based biomarkers GFAP, NF-L and tau.

BACKGROUND

Blood based biomarkers are increasingly used for non-invasive diagnostics, prognostics, stratification, and therapeutic monitoring, however clinical validation of new biomarkers is often limited by inconsistent assay methods. Assay developers can enable more reliable clinical studies by conducting analytical evaluation to characterize assay performance. In the context of brain injury and neurodegeneration, GFAP, NF-L and tau have emerged as potentially useful biomarkers for stratifying injury severity. Here we demonstrate analytical validation of the MSD® S-PLEX Human Neurology Panel 1, an ultrasensitive multiplex immunoassay for GFAP, NF-L and total tau.

METHODS

Analytical validation testing consisted of multi-site, multi-lot precision, accuracy, stability, dynamic range, and interference. All testing was conducted with plasma, serum, and diluent-based control samples. The precision study was conducted at three sites: Meso Scale Diagnostics, LLC. (Rockville, MD), Johns Hopkins University (Baltimore, MD) and All Children's Hospital (Tampa, FL). Three kit lots were tested at all three sites for a fully crossed study design. All runs contained an 8-point duplicate calibrator curve, duplicate sample measurements, and control samples. Data were analyzed at each site and then transmitted to MSD for combined analysis.

RESULTS

There were no significant differences between reagent-lot measurements. Intra-plate %CV averaged less than 10%. Between-site comparison of GFAP, NF-L and tau measurements revealed a small but significant difference in measurements across study sites. Accuracy reported that native samples recover within ± 20% of expected concentration through 20-fold dilution, and spiked samples recover within ± 20% through 256-fold dilution. No substances in a panel of potential interferents affected quantitation by more than ± 20% except for erythropoietin. Kits are shelf stable through 24 months, and GFAP, NF-L and tau in samples can withstand at least five freeze/thaw cycles, storage for 24 hours at 4°C, and storage for four hours at ambient temperature.

CONCLUSIONS

The MSD S-PLEX® Human Neurology Panel 1 has been analytically validated for research use biomarker studies measuring GFAP, NF-L and tau. Characterization of sources of variability have identified minor to moderate inter-site variability, but little to no inter-lot variability. Further work is needed to identify an international reference standard and to evaluate use in specific research contexts.