Related Experiment Video
Updated: Jun 29, 2026
A Dual Tracer PET-MRI Protocol for the Quantitative Measure of Regional Brain Energy Substrates Uptake in the Rat
Published on: December 28, 2013
Utility of Regional Cerebral Blood Flow SPECT and MRI for Predicting Amyloid Deposition in Real-World Clinical
Toshiki Takayama1,2, Takumi Hirose2, Masami Goto3
1The Medical Center for Dementia, Juntendo Hospital, Tokyo, Japan.
Background:
With the advent of disease-modifying therapies for Alzheimer's disease, there is a growing demand for more cost-effective methods that predict amyloid PET positivity. In Japan, the combination of structural brain MRI and regional cerebral blood flow (rCBF) single-photon emission computed tomography (rCBF SPECT) is routinely used in clinical practice for neuroimaging. The present study investigated the utility of this imaging approach for amyloid PET positivity in real-world clinical settings.
Methods:
A retrospective analysis of 101 patients who visited the Memory Clinic at Juntendo University Hospital between April 2019 and September 2023 was performed. All patients had undergone amyloid PET imaging at their own expense, in addition to a neuropsychological assessment, structural brain MRI and rCBF SPECT. Among 58 cases that underwent a voxel-based specific regional analysis system for Alzheimer's disease (VSRAD), volumetric differences between the amyloid-positive and -negative groups were assessed using Statistical Parametric Mapping (SPM) version 12.
Results:
Among the 101 patients analysed, 68 were positive and 33 were negative for amyloid. Twenty-nine patients had a change in clinical diagnosis after amyloid PET imaging, which affected subsequent management. Using VSRAD, VOI severity (continuous value) was set to 0.95 based on Youden's index and AUC (95% CI) was 0.72 (0.572-0.867), with a sensitivity of 90%, specificity of 52% and accuracy of 78%. In the visual assessment, MRI and rCBF SPECT were combined, and AUC (95% CI) was 0.767 (0.629-0.905), with a sensitivity decrease to 21% and specificity increase to 95%. The SPM12 analysis revealed clusters in the right medial temporal lobe at an uncorrected threshold (p < 0.001). However, no voxels survived family-wise error (FWE) correction (p < 0.05).
Conclusion:
A VSRAD analysis of brain MRI has moderate discriminatory ability as a screening tool for predicting amyloid PET positivity. The addition of SPECT to MRI was not associated with a statistically significant improvement compared with MRI alone. SPECT may contribute to the differential diagnosis of non-Alzheimer's disease. MRI alone could be considered an option to increase the pre-test probability of amyloid PET positivity.
