Brain amyloid PET scan in Alzheimer's disease, mild cognitive impairment and normal aging: The first prospective longitudinal study in Thailand

1061 Background: The incidence of dementia in elderly population has been rapidly increasing and causes significant burden on quality of lives, socio-economy and healthcare. Early detection of amyloid deposition in the brain may play important role in accurate diagnosis, identifying patients at risk and planning for patient management. Here we report results from the first prospective longitudinal study in Thailand on the potential use of amyloid PET scan. Objectives: To evaluate the use of [F-18]Florbetapir PET scan in assessing brain amyloid deposition in Alzheimer’s disease (AD), mild cognitive impairment (MCI) and healthy cognitive elderly (HC). The correlation between results obtained from brain [F-18]Florbetapir PET scan and neurocognitive test, MRI, [F-18]FDG PET scan as well as the longitudinal changes in both neurocognitive and imaging results were also determined. Methods: Prospective study was conducted in 86 elderly subjects (21 HCs, 33 MCIs and 32 ADs according to clinical diagnosis and neurocognitive test results). All subjects underwent clinical and neurocognitive assessment (TMSE, CDR-SB, ADAS-COG, brain MRI, [F-18]FDG PET and [F-18]Florbetapir PET scan of the brain within 6-week period, at baseline and repeated at 2-year follow-up. Visual and semi-quantitative analyses of brain scans were performed. The 3D-SSP analysis was used to determine Z-score map patterns and extract numeric data from both [F-18]florbetapir and [F-18]FDG PET scans for further calculation of cerebellar normalized regional SUVRs (Fig. 1&2). MTA scale was used to evaluate mesial temporal atrophy. Chi-square tests and one-way ANOVA were applied for statistical comparison. Correlation between brain amyloid deposition, glucose metabolism, mesial temporal atrophy and neurocognitive scores were assessed. Statistical significance was defined using P < 0.05. Results: From 86 subjects initially enrolled in the study, 72 subjects (19 HCs, 31 MCIs and 22 ADs) also underwent all studies at 2-year follow-up. (Table 1) At baseline, the results from all imaging techniques were highly concordant with clinical diagnosis and significantly correlated with the neurocognitive scores. The ability to differentiate AD from HC by [F-18]Florbetapir was higher than MRI but lower than [F-18]FDG PET. (Table 2) [F-18]florbetapir PET results also showed significant correlation with [F-18]FDG PET and higher than that with MRI. Correlation between baseline [F-18]florbetapir PET results with the change of neurocognitive scores, which varied among clinical subgroups and type of tests, were partly noted. The change in [F-18]florbetapir SUVRs only correlated with change in ADAS-COG scores in HC. (Table 3) The baseline [F-18]florbetapir PET results showed significant correlation with the follow-up [F-18]florbetapir PET in AD>MCI>HC. There was significant negative correlation between baseline Aβ deposition and glucose hypometabolism at follow-up in parietal and occipital lobes in AD but no significant correlation with degree of mesial temporal atrophy was observed. (Table 4) Conclusions: The results from all studied neuroimaging techniques were highly correlated with degree of neurocognitive performance. The ability in differentiating AD from HC in [F-18]FDG PET >[F-18]Florbetapir PET > MRI. Baseline amyloid deposition were spatially well correlated with degree of glucose hypometabolism and mesial temporal atrophy and may be able to project the longitudinal changes of neuropsychological scores and glucose metabolism, especially in AD. Keywords: [F-18]Florbetapir, amyloid, [F-18]FDG, PET scan, dementia, Alzheimer’s disease.