Abstract Body

To enable secondary prevention trials in asymptomatic, amyloid positive elderly individuals (preclinical AD), we have established the Japanese trial ready cohort for preclinical and prodromal AD (J-TRC). To clarify the utility of combination of plasma amyloid-β42 (Aβ42)-related biomarkers and phosphorylated tau at threonine 217 (p-tau217) to predict Aβ-PET positivity, 474 non-demented individuals (CDR 0: 331, CDR 0.5: 143) from the J-TRC cohort were analyzed. Participants underwent cognitive assessments, APOE genotyping, plasma Aβ and p-tau217 measurement and Aβ-PET imaging by florbetapir or flutemetamol. Participants were divided into CDR 0 or CDR 0.5 groups and PET images were assessed by visual examination. Plasma Aβ biomarkers were measured using immunoprecipitation-MALDI TOF mass spectrometry (Shimadzu), and an immunoassay for p-tau217 on the Meso Scale Discovery platform (Lilly). Discriminative accuracy measured by AUC with ROC analysis, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) of each plasma biomarker (p-tau217/Aβ42, p-tau 217 with Aβ42/40) and its combinatory modelling for the Aβ-PET positivity were examined. 81 subjects in J-TRC were positive for Aβ-PET. AUCs of variables for Aβ42 and p-tau217 individually showed optimal accuracy, which were improved by combination and modelling with multiple variables. The best AUC was obtained by p-tau217/Aβ42 with APOE, age, sex for the total participants (AUC: 0.936, sensitivity: 0.8271, specificity: 0.9211, PPV: 0.6836, NPV: 0.9627), p-tau217 with the Aβ42/40 and APOE, age sex for the CDR 0 group (0.948, 0.9523, 0.8131, 0.4255, 0.9915, respectively), and p-tau217/Aβ42 with APOE, age, sex for the CDR 0.5 group (0.955, 0.9230, 0.9134, 0.800, 0.9694, do.). In conclusion, combinations of plasma Aβ42-related biomarkers and p-tau217 are potent in the prediction of Aβ-PET positivity and may be instrumental in the prescreening for clinical trials of DMTs for preclinical and prodromal AD.