Table II.
Method Development Aspects and Improvements to the Assay
| Parameter | Per INNOTEST Aβ1-42 kit protocol | Modified Aβ42 method | Advantage of modification |
|---|---|---|---|
| Calibration range | 2,000–125 pg/mL | 1,000–40 pg/mL | Better fit to reference range |
| Number of calibrators | 6 | 10 | Improved definition of 4-PL regression |
| Non-linear regression method | Sigmoidal curve fit recommended | 4-PL with 1/Y weighting | Increased accuracy at low analyte concentration |
| Quantitative range | Undefined | 375–4,500 pg/mL | Provides a defined range with which to plan clinical sample analysis |
| MRD | 1 (analyze sample neat) | 6 | Increases accuracy of quantitation |
| Sensitivity | LOD is ±50 pg/mL | LLOQ = 62.5 pg/mL (after MRD, 375 pg/mL) | Established sensitivity defines assay utility in particular clinical populations |
| VS/QCs | None | 5 VS levels/3 QC levels (composed of AD donor CSF) | Provides ability to monitor assay performance and establish validation parameters |
| Sample controls | None | 2 individual AD donor CSF samples | Ensures the robustness of the assay with unadulterated CSF |
| Sample pre-treatment | None | Spike sample with Tween-20 to 0.2 % final | Mitigates potential adsorption of analyte to sample collection tube |
| Assay acceptance criteria | None | 4–6–20 acceptance criterion on QCs | Establishes mechanism for the unbiased rejection of poorly performing plates |