Abstract
514
Objectives Biomarkers are increasingly employed to supplement clinical diagnosis of Alzheimer’s disease (AD). For that purpose, depending on the concept, pathology/neurodegeneration or diagnosis/progression biomarkers are used1. However, so far it was not possible to derive both biomarker categories within one brain imaging session2. This project investigated the respective potential of PET/MRI to simultaneously image cerebral amyloid load and relative cerebral blood flow (rCBF) by arterial spin labeling (ASL).
Methods 65 subjects (68±10yrs, 31 female) with cognitive deficits (CDR蠅1, MMSE=23±6, MCI: n=38, probable AD dementia (ADD): n=12, possible ADD: n=15) underwent simultaneous [18F]Florbetaben PET/MRI (Siemens 3T mMR). The [18F]Florbetaben PET data were acquired 90-110min p.i. of 300MBq. They were analyzed visually (binary evaluation) and relative quantitatively (PMOD, composite SUVRs). Simultaneously, the ASL MRI data were acquired (PICORE Q2TIPS, 18 slices a 4mm in 64x64 matrix, TE/TI/TR = 16/2400/3400ms). They were analyzed visually (image quality, binary judgment: normal/abnormal, pattern recognition) and using a voxel-based approach (SPM).
Results 32% of the ASL images were visually judged as of inappropriate quality for individual analysis. Of the remaining cases with pathological amyloid PET images, 79% had AD patterns in ASL MRI, while 53/17/30% of the remaining cases with normal amyloid PET images had normal/FTLD/AD patterns in ASL MRI. In the possible/probable ADD patients, 53% were pathologic in amyloid PET only, and 25% both in amyloid PET and ASL MRI. In the MCI subjects, all pathologic amyloid PET cases had an AD pattern in ASL MRI. SPM analysis confirmed the ASL MRI differences between amyloid PET-positives vs. -negatives (see supplementary figure).
Conclusions While ASL MRI provides promising results with typical pathology patterns on a group level, the current data quality is not sufficient for individual AD neurodegeneration/progression biomarker information. The individualization of acquisition parameters may improve the quality of ASL MRI, which then may enable combined amyloid PET/ASL MRI in a one-stop shop fashion.