In‐vivo plasma biomarker concentrations correlate with postmortem AD‐neuropathology in the oldest old Authors: Linda MC Lorenz, Susan K Rohde, Maruelle C Luimes, Patricia Fierro‐Hernández, Annemieke JM Rozemuller, Netherlands Brain Bank, Jeroen JM Hoozemans, Charlotte Teunissen, Henne Holstege Published: 2025 ABSTRACT AbstractBackgroundThere has been ongoing uncertainty to what extent plasma biomarkers are representative of neuropathological markers of AD pathology in the oldest old. A complicating factor is that with increasing age, Aβ‐ and Tau pathologies accumulate in the brains of cognitively healthy individuals to highly variable levels (Zhang & Ganz et al., 2022). The unique availability of both antemortem plasma biomarker concentrations and postmortem neuropathology in self‐reported cognitively healthy oldest old, offers a chance to explore whether levels of plasma biomarkers can serve as in‐vivo proxy of AD neuropathology.MethodsWe evaluated postmortem neuropathology of 48 centenarian brains, with a median MMSE score of 26 (range:16–30) at blood draw. We quantified plasma biomarkers Aβ40, Aβ42, Aβ42/40, pTau181, NfL and GFAP on the Simoa platform. The median interval between blood‐sample collection and brain donation was 1.3 (range:0–5.1) years. In addition to conventional spatio‐temporal staging systems such as Thal‐Aβ phase (median 2, range:0‐5) and Braak‐NFT stage (median 3, range:2‐5), we also quantified pathological load by measuring the percentage area of immunopositivity for total Aβ (range:0–9.0%), Aβ40 (range:0–3.3%), Aβ42 (range:0–11.5%), pTau‐217 (range:0–1.8%) and AT8 (pTau‐202/205; range:0–49.2%) in the cortex of the temporal pole. Partial Spearman correlations were conducted adjusted for age, sex, and time interval.ResultsPlasma pTau‐181 correlated with Aβ pathology loads, namely Aβ40 (r=0.48, p<0.001), Aβ42 (r=0.41, p=0.005) and total Aβ (r=0.48, p<0.001), as well as with spatio‐temporal Thal‐Aβ phase (r=0.45, p=0.003). Plasma pTau‐181 correlated with quantitative pTau pathology loads, specifically pTau‐217 (r=0.48, p=0.001) and AT8 (r=0.41, p=0.005), but not with spatio‐temporal Braak‐NFT stage (r=0.17, p=0.290). Furthermore, both plasma NfL and plasma GFAP also correlated with pTau‐217 load in the brain (r=0.30, p=0.0496; r=0.30, p=0.044, respectively). Plasma Aβ40, Aβ42, Aβ42/40 did not correlate with any neuropathology measure.ConclusionsResults reveals plasma pTau‐181, but not plasma Aβ, as a reliable indicator for both Aβ and pTau pathology loads in the temporal pole of the oldest old. The 100‐plus study cohort provides a promising opportunity to study the correlation between different pTau epitopes and the presence of pTau levels, as well as the presence of Tau pathology, including NFTs, neuropil threads, pre‐tangles and ARTAG.