Data‐driven approach for early detection of pathological pathways in middle‐aged adults with family history of sporadic Alzheimer's disease. (31st December 2021)

• Record Type: Journal Article
• Authors: Peña‐Gomez, Cleofé; Akinci, Muge; Sánchez‐Benavides, Gonzalo; Shekari, Mahnaz; Grau‐Rivera, Oriol; Operto, Grégory; Milà‐Alomà, Marta; Sala‐Vila, Aleix; Salvadó, Gemma; Kollmorgen, Gwendlyn; Suridjan, Ivonne; Blennow, Kaj; Zetterberg, Henrik; Molinuevo, Jose; Gispert, Juan Domingo; Suarez‐Calvet, Marc; Arenaza‐Urquijo, Eider M
• Is Part Of: Alzheimer's & dementia. Volume 17(2021)Supplement 5
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12-31
• 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.057768
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: In a sample of cognitively unimpaired (CU) middle‐aged adults with family history (FH) of sporadic Alzheimer's disease (AD), we (i) classified participants based on similarity patterns of expression in cerebrospinal fluid (CSF) biomarkers of AD, axonal damage, synaptic dysfunction, neuroinflammation, and glial response, and (ii) characterized the participants expressing these patterns using neuroimaging, demographics, clinical, and FH variables. Method: 254 CU participants with FH of sporadic AD from the ALFA study (Table 1) underwent imaging, lumbar puncture, clinical, and neuropsychological evaluations. CSF‐biomarkers Ab42, Ab40, NfL, neurogranin, sTREM2, YKL40 and GFAP were measured by Roche NeuroToolKit immunoassays; p‐tau181 and p‐tau217 were measured with an in‐house Simoa‐based assay, p‐tau231 with an ADx ELISA, and SYT1 and SNAP25 using immunoprecipitation mass spectrometry. Participants were clustered using an agglomerative clustering algorithm according to their similarity of CSF‐biomarker expression–Subsequently, we performed statistical groups comparisons. Result: Four clusters were identified. Higher Aβ burden (lower Aβ42 /40 ) was the hallmark of only 1 cluster (C1: "AD‐like", mean centiloid [CL]: 15), while the other 3 clusters showed lower Aβ burden but higher expression of Glial (C2), Tau (C3, "SNAP‐like "), or synaptic (C4) markers (mean CL: ‐2.6, ‐4.5, ‐2.2, respectively) (Figure 1). The "AD like" cluster comprised the oldest participants with the lowest memory performances (C1<C2<C3<C4), the highest frequency of APOE‐ε4 carriers (C1>C3>C2>C4), participants with the closest age to the parental onset (C1>C2>C4>C3), and low gray matter in the hippocampus and AD‐sensitive regions (C1<C3<C4). The "Glial" cluster showed a higher proportion of hypertensive participants (C2>all) with preserved executive performances and hippocampus (C2>all). The "SNAP‐like" cluster, composed mainly by females (80%), showed the highest proportion of participants with both parents affected with AD and the lowest hippocampus volume (C3<all). The " Synaptic" cluster showed a greater proportion of males, with the best memory performance among the clusters. Conclusion: In CU middle‐aged adults, we identified differential expressions of CSF‐biomarkers conforming to Aβ‐dependent and independent biological pathways, some of which resemble previously described biomarker conditions such as SNAP. These distinct biomarker expressions were associated with specific imaging and clinical features. Longitudinal studies are warranted to investigate the clinical progression of participants showing these distinct biomarker profiles.