Effect of temporal meta‐ROI and Braak1&2 ROI on the dissociation between plasma pTau181 and PET statuses.

Background: Tau is usually assessed through cerebrospinal fluid or Positron Emission Tomography (PET). However, those methods are expensive and not readily available. Research is now focusing on cost‐effective blood‐based biomarkers to assess Alzheimer's disease (AD) pathophysiologies. A novel immunoassay for plasma pTau181 has been created, and even though there are strong associations between plasma pTau181 and PET, no study has yet investigated which cortical area better reflects plasma tau. Method: We assessed 269 individuals from the TRIAD cohort (164 CU, 60 MCI and 45 AD) that underwent plasma pTau181 assessment, [18F]MK6240 tau‐PET scan, [18F]AZD4694 amyloid‐PET scan to assess their amyloid‐b (Ab) status, an MRI and a neuropsychological evaluation. We conducted statistical analyses using R, first to calculate the threshold values for positivity in both tau assessing methods and to make comparisons between the groups. SUVR values for [18F]MK6240 were either based on Braak1 SUVR or temporal meta‐ROI SUVR. Result: Using Receiver Operating Characteristic (ROC) curves, we established that the cut‐off for positivity for plasma pTau181 was 11.1060 pg/mL. When using Braak1 mask, the cut‐off for positivity was 1.040 SUVR (B1+/‐), while with temporal meta‐ROI it was 1.1827 (TMR+/‐). In both cases, we observed a majority of concordant individuals. However, we obtained 87 discordant individuals in the plasma/TMR, mostly plasma+/TMR‐, and 78 in the plasma/B1 analyses, mostly plasma‐/B1+ (Figure 1). Concomitant status of plasma/TMR was usually accompanied by a similar Ab status. However, in the Braak1 analyses, Ab status seemed to closely follow the PET status (Figure 2). Conclusion: The cut‐off values for positivity using Braak1 mask is significantly smaller than when using temporal meta‐ROI. Using the temporal meta‐ROI mask, we had more discrepant individuals that were plasma+/TMR‐, with plasma‐/TMR+ individuals close to the threshold values. It was accurate at predicting negative biomarker statuses in CU participants, and positive in AD participants. Conversely, we had more plasma‐/B1+ individuals when using Braak1 mask, with values more spread around the plot. Due to low threshold values, portion of CU and MCI individuals were categorized as positive to both biomarkers. Finally, B1 status seemed to follow closely the Ab levels. Plasma pTau181 thus seems to reflect better temporal meta‐ROI tau. [ABSTRACT FROM AUTHOR]