3168
Multimodal Magnetic Resonance Imaging versus 18F-FDG-PET to Identify Mild Cognitive Impairment1Department of Radiology, University of Pennsylvania, Philadelphia, PA, United States, 2Department of Neurology, University of Pennsylvania, Philadelphia, PA, United States
Synopsis
18F-FDG-PET provides a functional neurodegenerative biomarker in the Alzheimer’s continuum, but it is costly and involves exposure to ionizing radiation. Arterial Spin Labeled (ASL) perfusion MRI can be acquired during routine MRI session to measure cerebral blood flow (CBF), which is tightly coupled with cerebral metabolism. We demonstrated that the ASL hypoperfusion pattern was similar to that of FDG-PET-hypometabolism in patients with mild cognitive impairment. Further, ASL-CBF provided complementary information to hippocampal atrophy measured with structural MRI. Multimodal MRI may provide a cost-effective and totally noninvasive substitute for 18F-FDG-PET in clinical and research setting for detecting Alzheimer’s neurodegeneration.
Introduction
18F-Fluorodeoxyglucose Positron Emission Tomography (18F-FDG-PET) provides a measure of cerebral glucose metabolism and serves as a functional neurodegenerative biomarker in the Alzheimer’s disease (AD) continuum, especially in the early stage, when cerebral functional alterations are believed to be more apparent than structural changes.1,2 Glucose metabolism is tightly coupled with cerebral blood flow (CBF),3 which can be measured non-invasively by Arterial Spin Labeled (ASL) perfusion MRI.4 Unlike 18F-FDG-PET, ASL-MRI does not require exposure to ionizing radiation and is less expensive. Additionally it can be acquired as part of routine MRI to provide multimodal data in a single scanning session and thus can potentially be combined with structural MRI, which is also a well-established neurodegenerative biomarker. Here we aimed to compare ASL-MRI and 18F-FDG-PET in discriminating patients with mild cognitive impairment (MCI) from older adult controls and determine if ASL-CBF provides complementary discriminatory value to structural MRI.Methods
Multimodal MRI and 18F-FDG-PET were acquired in close proximity, usually on the same day, from 50 MCI patients (age=73.0±7.0 years, 16 female) and 35 elderly controls (age=70.2±6.9 years, 20 female) recruited from the Penn Memory Center. ASL data were obtained using pseudo-continuous labeling with a labeling time=1.5s and post labeling delay (PLD)=1.5s and acquired with non-background suppressed 2D echo planar imaging with in plane resolution=3.4x3.4mm2 and slice thickness=6mm with a 20% distance factor. T1-weighted MRI for each subject was acquired using a 3D MPRAGE protocol with TR/TE/TI=1.9s/2.89ms/900ms, flip angle=90, bandwidth=170Hz/px, voxel size=1x1x1mm3. CBF maps were determined using advanced signal processing strategies5,6 and absolute and relative (putamen reference) CBF maps were produced. Cortical thickness for each subject was estimated from T1 images using Diffeomorphic Registration based Cortical Thickness (DiReCT).7 The CBF and cortical thickness maps were normalized to the MNI space using ANTs normalization parameters estimated from the T1 images to perform voxel-wise comparison. Additionally, hippocampal volume of each subject was extracted with the automatic segmentation of hippocampal subfields (ASHS) pipeline,8 and normalized with intracranial volume, which was also extracted with ASHS pipeline.
For FDG-PET scans, subjects received an intravenous injection of 5.0±0.5 mCi of 18F-FDG 30 minutes prior to a 30-minute 3D emission scan obtained with 256mm FOV, 128x128 matrix, 2x2x2mm3 voxel size. Line-of-response row-action maximum likelihood algorithm reconstruction using sharp setting was performed followed by CT attenuation correction. The processing of the PET data involved coregistation to the high resolution T1 images and normalization to the MNI space using ANTs as described for ASL-MRI. Standardized Uptake Value Ratio (SUVR) maps were generated by normalizing the raw counts with mean uptake in cerebellum.
Results
The MCI group displayed significantly reduced FDG-PET SUVR in precuneus, middle and posterior cingulate cortex, bilateral parietal, bilateral medial and inferior temporal, insular, caudate and amygdalar regions (Fig. 1A), consistent with the typically observed pattern of prodromal AD.2,9 Similar patterns were observed for ASL absolute (Fig 1B) and relative (Fig. 1C) CBF though the spatial extent of hypoperfusion with absolute CBF was larger suggesting a more diffuse decrease in CBF in the MCI group. Further, differences in temporal lobe between MCI patients and controls were less prominent with ASL-CBF compared to FDG-PET SUVR. Fig. 2 shows the overlap of the regions showing statistically significant group differences between the two groups with ASL, using relative CBF, and FDG-PET. Finally, cortical thickness was significantly reduced in MCI patients in bilateral medial temporal and lateral temporal cortex and in right parietal cortex (Fig. 1D). Notably, in distinction from the PET and ASL-MRI data, midline parietal structures, e.g. precuneus and posterior cingulate cortex (PCC) did not meet statistical significance with the cortical thickness maps.
Relative CBF and 18F-FDG SUVR in a priori selected PCC demonstrated moderate discriminatory power in predicting MCI status (both AUC=0.74), while hippocampal volume from structural MRI demonstrated excellent discriminative power (AUC=0.87±0.09). The combination of PCC relative CBF and hippocampal volume produced the strongest group discrimination (AUC, 0.89 ± 0.09). A step-wise logistic regression model including PCC CBF, PCC FDG-SUVR and hippocampal volume showed that the strongest prediction (χ2=43.5, p<0.001) included both hippocampal volume (β=-2615.6, p<0.001) and PCC relative CBF (β=-2.9, p<0.05) in the model, indicating MRI measures of CBF and hippocampal volume provide complementary predictive power.
Discussion and Conclusion
ASL-MRI measurements of CBF produced considerable overlap with measures of 18F-FDG SUVR in known regions of early AD neurodegeneration. Given the possibility of adding ASL to MRI protocols and its potential complementary role to structural imaging, it may serve as a useful alternative to 18F-FDG-PET for classifying degree of neurodegeneration in individuals with prodromal AD in the clinical and research setting.Acknowledgements
The authors would like to thank Dr. Abass Alavi for his contribution in setting up the PET imaging protocol. This study was supported by NIH grants R01 MH080729, P41 EB015893, R01 AG040271, R01 AG010124 and R01 AG055005.References
1. Jack CR, Jr., Knopman DS, Jagust WJ, et al. Tracking pathophysiological processes in Alzheimer's disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol 2013;12(2):207-216.
2. Landau SM, Harvey D, Madison CM, et al. Associations between cognitive, functional, and FDG-PET measures of decline in AD and MCI. Neurobiology of aging 2011;32(7):1207-1218.
3. Raichle ME. Behind the scenes of functional brain imaging: a historical and physiological perspective. Proceedings of the National Academy of Sciences of the United States of America 1998;95(3):765-772.
4. Alsop DC, Detre JA, Golay X, et al. Recommended implementation of arterial spin-labeled perfusion MRI for clinical applications: A consensus of the ISMRM perfusion study group and the European consortium for ASL in dementia. Magnetic resonance in medicine : official journal of the Society of Magnetic Resonance in Medicine / Society of Magnetic Resonance in Medicine 2015;73(1):102-116.
5. Dolui S, Wang Z, Shinohara RT, Wolk DA, et al. Structural Correlation-based Outlier Rejection (SCORE) algorithm for arterial spin labeling time series. Journal of magnetic resonance imaging : JMRI 2016;45(6):1786-1797.
6. Dolui S, Wolk DA, Detre JA. SCRUB: A Structural Correlation and Empirical Robust Bayesian Method for ASL Data. Proceedings of the International Society of Magnetic Resonance in Medicine. Singapore; 2016.
7. Das SR, Avants BB, Grossman M, Gee JC. Registration based cortical thickness measurement. NeuroImage 2009;45(3):867-879.
8. Yushkevich PA, Pluta JB, Wang H, et al. Automated volumetry and regional thickness analysis of hippocampal subfields and medial temporal cortical structures in mild cognitive impairment. Hum Brain Mapp 2015;36(1):258-287.
9. Mosconi L, Mistur R, Switalski R, et al. FDG-PET changes in brain glucose metabolism from normal cognition to pathologically verified Alzheimer's disease. European journal of nuclear medicine and molecular imaging 2009;36(5):811-822.
Figures
