%0 Journal Article %A Tanyaluck Thientunyakit %A Chakmeedaj Sethanandha %A Weerasak Muangpaisan %A Orasa Chawalparit %A Kuntarat Arunrungvichian %A Tossaporn Siriprapa %A Swatabdi Kamal %A Thonnapong Thongpraparn %A Yudthaphon Vichianin %A Rujaporn Chanachai %A Juri Gelovani %T A Novel Multi-Modality Molecular Imaging Index of Amyloid, Glucose Metabolism, and Morphologic Changes in the Brain for Evaluation of Patients with Dementia:A Thailand Perspective. %D 2019 %J Journal of Nuclear Medicine %P 1459-1459 %V 60 %N supplement 1 %X 1459Objectives: To test the utility of a new regional index that is calculated as a ratio of regional SUV of [18F]florbetapir (normalized to cerebellar cortex) over the corresponding regional SUV of [18F]FDG, divided by the structural volume, measured in the co-registered MRI and normalized to the age-matched control group (amyloid/FDG/NVol). Methods: A total of 43 subjects were involved in this study, including 20 age-matched healthy volunteers, 8 patients with mild cognitive impairment (MCI), and 15 patients with Alzheimer''s disease (AD).The neurocognitive performance tests including TMSE, CDR, ADAG-COG, Thai ADLs, and NPI-Q have been performed. Multimodal imaging studies of the brain including MRI, [18F]florbetapir PET/CT and [18F]FDG PET/CT were done in all subjects within 2-month period from neurological assessment. For quantitative analysis, 3D high-resolution T1 weighted FEE images were used to calculate normalized volume (nVol) for each cortical region using Freesurfer image analysis suite version 6.0.0.The PETSurfer workflow provided with Freesurfer were used for both [18F]florbetapir and [18F]FDG scans, which were co-registered to the high-resolution segmented brain images, then average uptake for each subregion were extracted for further calculation of regional SUVR using cerebellar cortex as a reference region. The amyloid/FDG/nVol index for a region was calculated by dividing florbetapir SUVR/FDG SUVR for that region by nVol for the same region.For characterization of demographic data and neurocognitive performance in study groups we used one-way ANOVA for continuous data and Chi-Square for categorical data.Statistical significance was defined as p < 0.05. Results: Comparing the MCI group with control, there were no differences in amyloid/FDG/NVol indices for the majority of brain regions except for right cingulate> right n.accumbens> right insula and left frontal cortex. Comparing the AD group with control, statistically significant differences in amyloid/FDG/NVol indices were observed in all brain regions, especially in n.accumbens, cingulate and parietal cortex and in the white matter of the parietal lobes. Less statistically significant differences were observed in various brain regions when comparing AD and MCI groups. From cluster analyses, the initial median threshold value of amyloid/FDG/NVol index for differential diagnosis of advance stage AD vs early stage AD is >2, or 5 key structures >2 (i.e., parietal cortex, cingulate, insula, hippocampus, amygdala), while the initial median threshold value of amyloid/FDG/NVol index for differential diagnosis of AD vs MCI is >1.4, or 5 key structures >1.4. No significant differences between study groups were observed for in median amyloid/FDG/NVol indices measured in the white matter segments. Discussion: Current study demonstrated that amyloid/FDG/NVol index mapping of the brain can enable a more accurate diagnosis, staging, and prognosis of AD, as well as differential diagnosis of AD from MCI. Additional larger scale clinical studies are required to further evaluate the efficacy of this new quantitative index as a diagnostic and prognostic biomarker of AD as well as for evaluation of safety and efficacy of novel agents undergoing clinical trials for therapy of AD. %U