Brain Glucose Metabolism Is Lower in Down Syndrome Two Decades Before Alzheimer's Symptoms

Down syndrome (DS) and autosomal dominant Alzheimer's disease (ADAD) are the two most common genetic forms of Alzheimer's. Both involve amyloid overproduction, and both allow researchers to estimate years to symptom onset. Brain glucose metabolism, measured by FDG-PET, is a marker of neurodegeneration that declines near symptom onset. Until now the two conditions had been studied separately, so it was unknown whether their metabolic trajectories match. The answer affects how FDG-PET is interpreted in clinical care and trials.

This cross-sectional study pooled data from two consortia. It included 76 adults with DS from the Alzheimer's Biomarker Consortium–Down Syndrome (68 cognitively stable, 6 with mild cognitive impairment, 2 with dementia). It also included 297 ADAD mutation carriers (207 cognitively normal, the rest with dementia ranging from very mild to severe) and 188 neurotypical familial controls from the Dominantly Inherited Alzheimer Network. Expected years to symptom onset (EYO) was calculated as 52.5 minus age in DS, with 47.5 and 57.5 tested in sensitivity analyses. In ADAD it was based on the parent's age at symptom onset. All scans went through a shared processing pipeline with partial volume correction. The regions analyzed were an isthmus cingulate/inferior parietal composite, the precuneus, the middle temporal gyrus and the precentral gyrus. These span early-affected to late- or unaffected areas. Because the best reference region in DS is unsettled, the authors used both brainstem and cerebellum. Within-group linear regression and generalized additive models accounted for age, sex, APOE ε4 status and cortical amyloid (PiB PET, available for 516 of 561 participants).

People with DS had lower FDG uptake than neurotypical controls even about 20 years before expected symptom onset (p<.01). This was true in cognitively stable individuals. In ADAD, baseline FDG was similar to controls until roughly 7 years before expected onset. Symptomatic individuals in both groups had lower FDG than controls (p<.01). Higher amyloid burden was associated with lower FDG in both genetic forms, and amyloid-positive DS and ADAD participants showed similar rates of FDG decline.

The authors conclude that glucose hypometabolism in DS appears much earlier than in ADAD, and that the decline patterns differ. FDG-PET's diagnostic value therefore appears to be specific to the genetic form of Alzheimer's. The comparison also suggests that amyloid's effect on metabolism is similar once amyloid is present. Early differences in DS may reflect additional factors such as neurodevelopmental differences, not only Alzheimer's neurodegeneration.

Several caveats apply. The design is cross-sectional, so within-person trajectories were not tracked. The DS sample is much smaller than the ADAD sample and had few symptomatic individuals. The two cohorts used different FDG acquisition windows and different EYO definitions. The supplied text was truncated before the full results and discussion, so this summary relies on the abstract, introduction and methods.