Combining p‐tau217 with Other Blood Biomarkers to Enhance Prediction of Cognitive Decline: A Large Memory Clinic Cohort Study

TL;DR

This study shows that combining the blood biomarker p-tau217 with other biomarkers improves the prediction of cognitive decline in Alzheimer's disease.

Contribution

The study demonstrates the synergistic effect of combining p-tau217 with other blood biomarkers for better clinical prediction of cognitive decline.

Findings

High p-tau217 levels are strongly associated with faster cognitive decline.

Combining p-tau217 with GFAP or NfL provides the greatest predictive power for cognitive decline.

Elevated NfL is most predictive in individuals with low p-tau217 levels.

Abstract

Plasma p‐tau217 has emerged as one of the most promising biomarkers for Alzheimer's disease (AD). However, it remains largely unexplored whether integrating p‐tau217 with other emerging AD blood biomarkers (BBMs) will enhance its clinical performance. This research aims to address this question by analyzing a large memory clinic cohort with over three decades of longitudinal follow‐up.

This study utilized participants enrolled at the University of Pittsburgh Alzheimer's Disease Research Center who underwent baseline blood collection and longitudinal Clinical Dementia Rating‐Sum of Boxes (CDR‐SB) based cognitive functional assessments over thirty years. A sub‐cohort with 11C‐Pittsburgh Compound‐B (PiB) amyloid PET was used to determine the cut‐off for p‐tau217 (0.5471 pg/ml), according to the Youden index. Plasma levels of p‐tau217, p‐tau181, brained‐derived tau (BD‐tau), GFAP, and NfL were measured using the SIMOA assay. Statistical analysis was conducted using Cox Proportional Hazards Regression and the Kaplan‐Meier method, with events defined as an increase in CDR global score during follow‐up.

We included 4382 participants (57.0% female; 85.8% self‐identified non‐Hispanic White), aged 71.9 ± 9.8 years, with 2127 being non‐demented (CDR≤0.5) at baseline. High p‐tau217 levels were associated with a significantly faster cognitive decline, with a hazard ratio (HR) of 3.16 (CI: 2.82 – 3.53) and a median survival time of 4.0 years compared to 10.0 years for those with low p‐tau217. Elevated GFAP was associated with the greatest hazards when combined with high p‐tau217, HR 1.61 (1.41 – 1.84; p < 0.0001), followed by 1.46 (1.28 – 1.66; p < 0.0001) for NfL, 1.31 (1.14 – 1.50; p =  0.0002) for p‐tau181, and 1.29 (1.13 – 1.48; p =  0.0002) by BD‐tau. Among individuals with low p‐tau217 levels, elevated NfL was associated with the greatest hazards, HR 3.53 (2.64 – 4.72; p < 0.0001), followed by 1.77 (1.32 – 2.39; p < 0.0001) for GFAP and 1.33 (1.01 – 1.77; p =  0.029) for p‐tau181. Elevated BD‐tau in individuals with low p‐tau217 was not associated with greater hazards.

Our study underscores the synergetic joint effect between p‐tau217 and other BBMs, and that integrating p‐tau217 with these BBMs can enhance tailored clinical management strategies of AD.