AbstractBackgroundWe have completed the analytical validation of an antibody‐free, HPLC‐MS/MS method (ABtest‐MS), to quantitate specifically Aβ1‐40 and Aβ1‐42 in plasma with high precision (inter‐assay %CV 6.2‐7.0% for Aβ1‐40 and 6.1%‐11.0% for Aβ1‐42) and accuracy (%RE ‐0.6 to 0.3% for Aβ1‐40 and ‐1.5 to 1.3% for Aβ1‐42).Previous studies have shown that the Aβ42/Aβ40 ratio in plasma predicts β‐amyloid status reliably, suggesting that the assessment of plasma Aβ levels could be a useful screening tool for Alzheimer’s Disease (AD) clinical trials, reducing the need for invasive or complex procedures.In this work we explore the ability of Aβ42/Aβ40 plasma ratio to predict the Aβ‐PET status at baseline visit in the ABvac40 (an active vaccine against Aβ40) phase II clinical trial.MethodAβ40 and Aβ42 plasma levels from 73 amnestic‐MCI (a‐MCI) participants, who had previously given their consent, were quantitated with ABtest‐MS. Levels of p‐tau181 were assessed with Simoa® p‐tau181 V2 Advantage Kit (Quanterix). The ability of plasma Aβ42/Aβ40 ratio or p‐tau181 to detect Aβ‐PET positivity was assessed as the area under the ROC curve (AUC).ResultAβ42/Aβ40 plasma ratio values were significantly lower in the Aβ‐PET positive than in the Aβ‐PET negative group (p<0.001). The opposite trend was observed for p‐tau181 plasma levels, which were significantly higher in the Aβ‐PET positive than in the Aβ‐PET negative group (p<0.001). The Aβ42/Aβ40 plasma ratio discriminated Aβ‐PET positivity with an AUC value of 0.87, performing identically to p‐tau181 plasma levels.A negligible increment in the AUC was observed after adjusting for ApoE genotype. However, inclusion of both biomarkers in the predictive model increased AUC value up to 0.94, outperforming each biomarker alone.ConclusionAmong the a‐MCI participants in the ABvac40 phase II clinical trial who had previously consent, the Aβ42/Aβ40 plasma ratio discriminates Aβ‐PET positivity with a high AUC, performing as well as p‐tau181. These results demonstrate that the Aβ42/Aβ40 ratio in plasma could be used to predict cerebral Aβ burden.