Longitudinal PET Evaluation of Cerebral Metabolic Decline in Dementia: A Potential Outcome Measure in Alzheimer's Disease Treatment Studies.

Abstract

It is well established that regional cerebral metabolic rates for glucose assessed by [(18)F]fluorodeoxyglucose (FDG) positron emission tomography (PET) in patients with Alzheimer's disease in the mental resting state (eyes and ears covered) provide a sensitive, in vivo metabolic index of Alzheimer's disease dementia. Few studies, however, have evaluated longitudinal declines in regional cerebral glucose metabolism in patients with dementia caused by Alzheimer's disease. In addition, the available studies have not used recently developed brain mapping algorithms to characterize the progression of Alzheimer's disease throughout the brain, and none considered the statistical power of regional cerebral glucose metabolism in testing the ability of treatments to attenuate the progression of dementia. The authors used FDG PET and a brain mapping algorithm to investigate cross-sectional reductions in regional cerebral glucose metabolism, longitudinal decline in regional cerebral glucose metabolism after a 1-year follow-up, and the power of this method to evaluate treatments for Alzheimer's disease in patients with mild to moderate dementia. PET scans were initially acquired in 14 patients with Alzheimer's disease and 34 healthy comparison subjects of similar age and sex. Repeat scans were obtained in the patients 1 year later. Power analyses for voxels showing maximal decline over the 1-year period in regional cerebral glucose metabolism (mg/100 g per minute) were computed to estimate the sample sizes needed to detect a significant treatment response in a 1-year, double-blind, placebo-controlled treatment study. The patients with Alzheimer's disease had significantly lower glucose metabolism than healthy comparison subjects in parietal, temporal, occipital, frontal, and posterior cingulate cortices. One year later, the patients with Alzheimer's disease had significant declines in glucose metabolism in parietal, temporal, frontal, and posterior cingulate cortices. Using maximal glucose metabolism reductions in the left frontal cortex, we estimated that as few as 36 patients per group would be needed to detect a 33% treatment response with one-tailed significance of p</=0.005 and 80% power in a 1-year, double-blind, placebo-controlled treatment study. These findings indicate that brain metabolism as assessed by FDG PET during mental rest is a sensitive marker of disease progression in Alzheimer's disease over a 1-year period. These findings also support the feasibility of using FDG PET as an outcome measure to test the ability of treatments to attenuate the progression of Alzheimer's disease.