An examination of baseline plasma Aβ42/40 and intra‐individual cognitive variability (IICV) associations with longitudinal cognitive change in a Black Cohort: Data from the African Americans Fighting Alzheimer's in Midlife (AA‐FAIM) study. (20th December 2022)

• Record Type: Journal Article
• Authors: Gleason, Carey E.; Koscik, Rebecca Langhough; Zuelsdorff, Megan; Norton, Derek L.; Fischer, Barbara L.; Van Hulle, Carol A.; Gooding, Diane C.; Yarasheski, Kevin E.; Wyman, Mary F.; Johnson, Adrienne L.; Lambrou, Nickolas H.; James, Taryn T.; Bouges, Shenikqua; Carter, Fabu P; Salazar, Hector; Norris, Nia; Chin, Nathaniel A.; Ennis, Gilda E.; Jonaitis, Erin M.; Simó, Carola A. Ferrer; Kirmess, Kris; Meyer, Matthew R.; Holubasch, Mary S.; Venkatesh, Venky; West, Tim; Verghese, Philip B.; Carlsson, Cynthia M.; Asthana, Sanjay; Johnson, Sterling C.
• Is Part Of: Alzheimer's & dementia. Volume 18(2022)Supplement 5
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-12-20
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.061055
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: In an emerging biomarker era, scientists, clinicians and patients must trust that Alzheimer's disease (AD) biomarkers are broadly applicable across populations. However, biomarker studies inadequately include minoritized populations – largely because biomarker collection is invasive and centralized, and outreach wholly insufficient. The AA‐FAIM study collected plasma and cognitive data from a Black cohort by leveraging existing resources from the Wisconsin Alzheimer's Disease Research Center and Wisconsin Registry for Alzheimer's Prevention. We compared two easily‐decentralized and low‐burden options for measuring preclinical changes–either alone or combined, testing associations of baseline plasma Aβ42/40 and intra‐individual cognitive variability (IICV) with longitudinal cognition. Methods: Using AA‐FAIM participants' cognitive data, we derived an estimate of baseline IICV from z‐scores, representing standard deviation across five indices from list learning, executive function, and confrontational naming tests. PrecivityAD TM assays quantified baseline plasma Aβ42/40. After comparing baseline IICV associations for factor‐validity (Figure 1), we conducted linear mixed‐effects models examining associations of baseline plasma Aβ42/40, IICV, or the combination of markers with longitudinal cognition on eight outcomes. Fully adjusted models included age‐at‐assessment, gender, education, and random intercepts. Benjamini‐Hochberg (BH) corrections controlling false discovery rate at 5% were applied across outcomes. Using Bayesian Information Criterion (BIC), plasma Aβ42/40 models were assessed for comparative fits with and without IICV interactions and main effects. Results: Models included between 163‐667 observations from 56‐174 participants (Table 1). In models without plasma, IICV * age was weakly significant only for Trails tests with higher baseline IICV being associated with worse trajectories (Table 2; uncorrected/corrected p‐values=0.057/0.227 (log Trails A) and 0.044/0.227 (log Trails B)). Plasma Aβ42/40 was significantly associated with trajectories across multiple cognitive outcomes regardless of IICV inclusion (Table 3). Model estimates suggested lower baseline Aβ42/40 was associated with worse performance over time (Figures 2‐3). All interactions survived correction with IICV in the model. Without IICV, only Trails B survived correction. BIC comparisons suggested removing the IICV * age interaction from plasma models, but retaining IICV main effects. Conclusions: Baseline plasma Aβ42/40 predicted cognitive trajectory in a Black cohort. Although IICV appeared less sensitive than plasma Aβ42/40 in moderating cognitive trajectory, models including both IICV and plasma Aβ42/40 may be better in predicting trajectories.