AST/ALT, APOE ε4, and Alzheimer's disease progression: A longitudinal cohort study.

INTRODUCTION: The aspartate aminotransferase-to-alanine aminotransferase (AST/ALT) ratio is a routinely available liver-enzyme index. Whether apolipoprotein E (APOE) ε4 modifies associations of baseline AST/ALT with longitudinal Alzheimer's disease (AD) trajectories is unclear. METHODS: In 2709 Alzheimer's Disease Neuroimaging Initiative participants, linear mixed-effects models estimated AST/ALT-associated longitudinal change in Mini-Mental State Examination, Alzheimer's Disease Assessment Scale 13-item Cognitive subscale, and AD-signature cortical thickness; Cox models evaluated continuous AST/ALT and an extreme-tertile APOE ε4 × AST/ALT comparison for clinical progression. RESULTS: Higher baseline AST/ALT was associated with faster cognitive decline and AD-signature cortical thinning, with larger associations among APOE ε4 carriers (interaction p = .0006-0.040). In clinical-progression analyses, continuous AST/ALT was not significantly associated with risk (hazard ratio [HR] 1.07 per standard deviation, p = 0.12). In the extreme-tertile comparison, APOE ε4 carriers with high AST/ALT had the highest adjusted hazard (HR 3.08, 95% confidence interval 2.18-4.34), but multiplicative and additive interactions were null and incremental discrimination was minimal (ΔC = +0.003). DISCUSSION: Baseline AST/ALT was associated with faster cognitive and cortical decline, particularly among APOE ε4 carriers, but provided limited incremental information for clinical progression beyond APOE ε4 and baseline diagnosis. These findings represent a large longitudinal extension of prior liver-enzyme observations and require external validation.