On behalf of the team at Foundation for the National Institutes of Health (FNIH) we inform you that a recent review of the protocol and internal standards used in Dr. Bateman’s lab discovered that the concentration of Aβ42 and Aβ40 provided to FNIH are around 3-fold high.
The stock concentration of internal standard containing synthetic abeta isoforms, universally labeled with a stable isotope of nitrogen (15N) to quantify Aβ38, Aβ40 and Aβ42 in CSF protocol is: 10 ng of Aβ 40, 1 ng of Aβ42 and 1.5 ng of Aβ38 per sample respectively. To quantify Aβ in plasma the dilution is needed to reflect levels of endogenous proteins in human plasma. 1:20 dilution of internal standards was used for initial 2mL plasma protocol. To calculate concentration, we are utilizing Tableau workbook that applies following formula: ((endogenous peptide signal / internal standard signal) * amount of internal standard) / sample volume. Amount of internal standard implemented as parameter that can be set/selected by user depending on actual protocol. With method development progression, we have moved to half mL plasma protocol switching the dilution of our internal standard to 1:60. We had ran a couple of most recent studies with that protocol. We had mistakenly omitted correcting respective parameters as it does not affect the ratio of Aβ42:40 and our standard data release format does not include reporting individual concentrations for isoforms. Therefore, the individual Aβ concentration we provided before were 3 fold higher.
We have updated the automatic algorithm with 1:60 dilution in the lab and provided the corrected Aβ42 and Aβ40 concentration (variable names: AB42, AB40, standardized_plasmaAB42, standardized_plasmaAB40).
Note that only the Aβ42 and Aβ40 concentration were corrected, there is no change in the Aβ42/Aβ40 ratio or other variables.
The updated dataset is named FNIH Biomarkers Consortium Plasma Abeta Project: Wash U[ADNI1,GO,2], version 2021-11-05, and can be downloaded from the Biospecimen Results section of the ADNI Archive.
