Alzheimer’s disease (AD) is a heterogeneous disorder with individual variability in clinical and biological features. Mild cognitive impairment (MCI) represents transition phase and provides a critical window for early intervention. Clinically applicable metrics for quantifying individual brain structural variations remain limited. This study aimed to delineate neuroanatomical heterogeneity and identify brain structural subtypes using individual deviation analysis. We constructed normative models of regional gray matter volume across lifespan using T1-weighted MRI data from 1185 healthy controls in the Cam-CAN and OASIS-3 cohorts. Individual deviations from normative trajectories were evaluated for AD and MCI patients in the ADNI dataset. Based on deviation patterns, unsupervised clustering analysis was used to identify distinct AD and MCI subtypes. Clinical features, longitudinal progression, and gene expression profiles were evaluated for each subtype. We found that substantial heterogeneity in the spatial distribution and severity of structural abnormalities across individuals. Two robust subtypes were identified in each group. Subtype 2 of AD and MCI showed more pronounced negative deviation patterns than subtype 1, particularly in the hippocampus, parahippocampal gyrus, and amygdala. These subtypes demonstrated significant differences in cognitive performance, biomarker profiles, disease progression, and gene expression patterns. AD and MCI subtypes exhibited similar deviation patterns, and MCI subtype with more severe negative deviations had a higher risk of converting to AD. This study reveals the neuroanatomical heterogeneity of AD and MCI and demonstrates that individualized deviation mapping can detect clinically valuable subtypes. These findings provide a foundation for personalized diagnosis and monitoring strategies in neurodegenerative diseases.
Wei et al. (Mon,) studied this question.