A multimodal, longitudinal study of cognitive heterogeneity in early‐onset Alzheimer's disease. (17th September 2021)

• Record Type: Journal Article
• Authors: Pollet, Marianne; Skrobala, Emilie; Lopes, Renaud; Kuchcinski, Grégory; Bordier, Cécile; Rollin‐Sillaire, Adeline; Bombois, Stéphanie; Pasquier, Florence; Delbeuck, Xavier
• Is Part Of: European journal of neurology. Volume 28:Number 12(2021)
• Page Start: 3990
• Page End: 3998
• Publication Date: 2021-09-17
• Subjects: cluster analysis -- cognition -- early‐onset Alzheimer's disease -- follow‐up studies -- phenotype; Neurology -- Periodicals; Nervous system -- Diseases -- Periodicals; 616.8
• DOI: 10.1111/ene.15097
• Languages: English
• ISSNs: 1351-5101

Abstract

Abstract: Background and purpose: Alzheimer's disease (AD) is a heterogeneous pathology. Young patients with AD are particularly likely to have an atypical presentation. The objectives of the present cluster analysis were to determine whether patients with early‐onset AD (EOAD) had several distinct cognitive profiles and to compare the resulting clusters with regard to clinical, neuroimaging, and laboratory characteristics. Methods: We collected cognitive, behavioural, functional, neuroimaging, and laboratory data on 72 patients meeting the criteria for probable mild EOAD. The patients were first classified into clinical phenotype groups by a multidisciplinary board of clinicians. The patients' cognitive and functional decline was monitored for 24 months. A k ‐means clustering analysis was then used to determine clusters on the basis of the patients' neuropsychological test results. Results: Two distinct clusters were identified: the patients in the first cluster (C1, n = 38) had a predominant memory impairment, whereas patients in the second (C2, n = 34) did not. Dyslipidaemia and the presence of ɛ4 apolipoprotein E allele were more frequent in C1, whereas the cognitive and functional decline was faster in the patients in C2. Moreover, posterior brain abnormalities were more severe in patients in C2 than in patients in C1. Conclusions: By applying a k ‐means clustering analysis, we identified two clusters of patients in an EOAD cohort. The clusters differed with regard to certain clinical, imaging, and laboratory characteristics. This clustering procedure might be of value for managing patients with EOAD in general and for identifying those at risk of more rapid decline in particular. Abstract : Based on their cognitive performance, we identified two clusters of patients in an early‐onset Alzheimer's disease cohort. These clusters differed with regard to clinical, imaging, and laboratory characteristics. Differences were also observed concerning the risk of a more rapid decline.