Alhasan · The journal of nutrition, health & aging 2026 · longitudinal cohort study · n=4,732

Baseline FDG-PET Brain hypometabolism as a predictive biomarker of cognitive decline and Alzheimer's disease risk.

Cited 2 times in the scientific literature.

Level 3 - non-randomized controlled study

Longitudinal cohort study with up to 10 years of follow-up.

PubMed 41819777 · doi:10.1016/j.jnha.2026.100823 · record verified 2026-08-27

What was done

Researchers analyzed 4,732 participants (1,685 with longitudinal cognitive data over up to 10 years) from the Alzheimer's Disease Neuroimaging Initiative (ADNI) who had baseline fluorodeoxyglucose positron emission tomography (FDG-PET), Mini-Mental State Examination (MMSE), and Alzheimer's Disease Assessment Scale (ADAS) scores. Mixed-effects models evaluated time × brain glucose metabolism interactions on cognitive decline, and cross-validation assessed predictive accuracy for mild cognitive impairment and Alzheimer's disease conversion.

What was found

Baseline brain glucose metabolism significantly modulated rates of cognitive decline (time × FDG interaction: MMSE β = 0.746, P < 0.001; ADAS β = -1.595, P < 0.001). High versus low glucose metabolism was associated with a protective difference of 1.49 MMSE points/year and 3.19 ADAS points/year. Among cognitively normal participants, low glucose metabolism increased Alzheimer's disease conversion risk nearly four-fold (incidence rate ratio = 3.79, 95% CI: 2.94–4.88). Predictive models achieved an AUC of 0.826 for Alzheimer's disease conversion (Brier score = 0.092), whereas high-metabolism individuals remained essentially stable over 10 years.

Why it matters

This study provides long-term quantitative evidence that baseline brain glucose hypometabolism predicts 10-year cognitive trajectories and Alzheimer's disease conversion risk, supporting FDG-PET as a prognostic biomarker for risk stratification.

Limits

The study is an observational analysis of the ADNI cohort, which may not fully represent broader community populations. Longitudinal cognitive data were available for only 1,685 of the 4,732 total participants, introducing potential attrition or selection bias. The abstract does not specify regional metabolic patterns or exact thresholds used to classify high versus low metabolism.

Cited by