Abstract

AbstractBackgroundAlzheimer’s disease (AD) is known to be heterogeneous in pathobiology and clinical manifestation. Most previous studies demarcating AD subtypes rely on data‐driven clustering methods that are likely confounded by interindividual differences in disease progression within samples. Addressing this problem, we aim to identify and characterize distinct AD atrophy progression patterns in a memory clinic‐based cohort.Method N = 914 subjects (296 cognitively unimpaired, 366 with subjective cognitive impairment, 153 with mild cognitive impairment, and 99 with AD dementia) from the multicenter observational DZNE Longitudinal Cognitive Impairment and Dementia Study (DELCODE) were included. The Subtype and Stage Inference algorithm (Young et al., 2018, Nat. Commun.) was applied to ROI‐based cortical thickness and gray matter volume data obtained from T1w and T2w 3T‐MRI using FreeSurfer. Subtypes were characterized in demographics, ApoE genotype, various fluid biomarkers, and cognition.ResultTwo atrophy subtypes were identified. A limbic‐predominant subtype exhibited initial volume loss in the amygdala and hippocampus, followed by cortical thinning in the medial temporal, temporal, frontal, parietal, and occipital lobes. Contrary, in a hippocampal‐sparing subtype, cortical thinning first occured in the frontal, parietal, and occipital lobes before the temporal and medial temporal lobes, hippocampus, and amygdala were affected. Compared to the hippocampal‐sparing subtype, the limbic‐predominant subtype included subjects who were older, reported fewer years of education, and tended to have more ApoE e4 alleles, higher risk of vascular pathology, and higher plasma neurofilament light levels. The subtypes differed in their CSF biomarker profiles indicating higher β‐amyloid and tau burden in the limbic‐predominant subtype. In the limbic‐predominant subtype, cognitive impairment was accentuated in the learning and memory as well as language domains. Belonging to the hippocampal‐sparing subtype was associated with an impairment in executive functions and processing speed, visuospatial functions, and working memory.ConclusionWe provide evidence on atrophy heterogeneity that is largely in line with previous theoretical and empirical contributions while overcoming key limitations that are commonly faced in earlier studies. Heterogeneity may have substantial impact even in cohort studies designed to primarily include “typical” amnestic AD which highlights the need for optimized and subtype‐specific measures of early detection and disease progression.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call