Independent Predictive Value of Amygdala Atrophy in the Alzheimer's Disease Continuum: A Longitudinal Analysis of Hemispheric Asymmetry and Network Neurodegeneration

This work presents a longitudinal analysis of amygdala and hippocampal atrophy within the Alzheimer’s disease (AD) continuum using data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). A total of 696 participants with baseline diagnoses of cognitively normal (CN) or late mild cognitive impairment (LMCI) were analyzed using standardized 3T MRI processed with FreeSurfer. Structural volumes were normalized to estimated total intracranial volume, and significant atrophy was defined as a ≥5% longitudinal decline from baseline. The study evaluates two key questions: (1) whether amygdala atrophy shows hemispheric asymmetry in its temporal relationship to clinical conversion, and (2) whether amygdala volume provides independent prognostic value beyond hippocampal atrophy. Results show that left amygdala atrophy precedes clinical conversion more frequently than the right (55.0% vs 49.9%), although this difference does not reach global statistical significance (χ² = 4.45, p = 0.217). In univariate survival analysis, amygdala volume is associated with conversion risk; however, this effect is attenuated and loses statistical significance after adjustment for hippocampal volume, while hippocampal volume remains a strong independent predictor. These findings support a network-based model of limbic neurodegeneration, in which amygdala and hippocampal atrophy reflect coordinated system-level processes rather than independent regional drivers. This work is a preprint and has not undergone peer review.