Characterization of 7- and 19-month-old Tg2576 mice using multimodal in vivo imaging: limitations as a translatable model of Alzheimer's disease

With 90% of neuroscience clinical trials failing to see efficacy, there is a clear need for the development of disease biomarkers that can improve the ability to predict human Alzheimer’s disease (AD) trial outcomes from animal studies. Several lines of evidence, including genetic susceptibility and disease studies, suggest the utility of fluorodeoxyglucose positron emission tomography (FDG-PET) as a potential biomarker with congruency between humans and animal models. For example, early in AD, patients present with decreased glucose metabolism in the entorhinal cortex and several regions of the brain associated with disease pathology and cognitive decline. While several of the commonly used AD mouse models fail to show all the hallmarks of the disease or the limbic to cortical trajectory, there has not been a systematic evaluation of imaging-derived biomarkers across animal models of AD, contrary to what has been achieved in recent years in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (Miller, 2009). If animal AD models were found to mimic endpoints that correlate with the disease onset, progression, and relapse, then the identification of such markers in animal models could afford the field a translational tool to help bridge the preclinical-clinical gap. Using a combination of FDG-PET and functional magnetic resonance imaging (fMRI), we examined the Tg2576 mouse for global and regional measures of brain glucose metabolism at 7 and 19 months of age. In experiment 1 we observed that at younger ages, when some plaque burden and cognitive deficits have been reported, Tg2576 mice showed hypermetabolism as assessed with FDG-PET. This hypermetabolism decreased with age to levels similar to wild type (WT) counterparts such that the 19-month-old transgenic (Tg) mice did not differ from age matched WTs. In experiment 2, using cerebral blood volume (CBV) fMRI, we demonstrated that the hypermetabolism observed in Tg mice at 7 months could not be explained by changes in hemodynamic parameters as no differences were observed when compared with WTs. Taken together, these data identify brain hypermetabolism in Tg2576 mice which cannot be accounted for by changes in vascular compliance. Instead, the hypermetabolism may reflect a neuronal compensatory mechanism. Our data are discussed in the context of disease biomarker identification and target validation, suggesting little or no utility for