Evaluating the Predictive Power of Cognitive Assessment Tools for Cortical Hypometabolism in Alzheimer’s Disease

Background: Alzheimer’s disease (AD) is a progressive neurodegenerative disorder characterized by memory loss and cognitive decline. One of the leading theories explaining AD pathology is the emergence of cortical hypometabolism. This study aimed to investigate the association between cortical hypometabolism and various cognitive assessment tools across the dementia spectrum.Methods: This cross-sectional and longitudinal study utilized data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), including 1,048 participants: 291 cognitively normal (CN), 579 with mild cognitive impairment (MCI), and 178 with AD. Fluorodeoxyglucose positron emission tomography (FDG-PET) data (as an indicator of hypometabolism) and cognitive assessment scores—including the Alzheimer’s Disease Assessment Scale (ADAS11 and ADAS13 subtests), Montreal Cognitive Assessment (MoCA), Everyday Cognition Scale (ECog), and Mini-Mental State Exam (MMSE)—were analyzed. Statistical methods included ANOVA, multiple regression, and ROC/AUC analyses.Results: Linear regression revealed that ADAS11, ADAS13, and MMSE significantly predicted PET scores in the MCI group (p=0.002, p=0.002, p=0.017, respectively), while MoCA predicted PET scores in the CN group (β=0.016, p=0.045). ROC analysis showed that ADAS13 had the greatest discriminative capacity (AUC=0.786), followed by ADAS11 (AUC=0.767). Over time, PET scores declined significantly across all groups, with the AD group showing the largest decline. At 24 months, PET scores in the CN and MCI groups were notably higher than those in the AD group (p<0.001).Conclusion: ADAS11 and ADAS13 can effectively differentiate between normal and abnormal cortical hypometabolism. Among all cognitive measures, ADAS13 demonstrated the highest discriminative ability, making it a valuable tool for clinicians and researchers in the early detection and longitudinal monitoring of Alzheimer’s disease.