Doctor of Philosophy
Keith St Lawrence
Frontotemporal dementia (FTD) is a devastating neurodegenerative disease characterized by a rapid decline in behavioural, language, and motor abilities. Advances in the understanding of FTD genetics and pathophysiology, and the subsequent development of novel disease modifying treatments have highlighted the need for tools to assess their efficacy. While structural magnetic resonance imaging (MRI) and functional imaging with 18F-flurodeoxyglucose (FDG) positron emission tomography (PET) are used for clinical diagnosis, structural changes are subtle at the early stages and PET imaging is expensive and access limited. Given the coupling of cerebral blood flow (CBF) to energy metabolism, an attractive alternative is the MRI perfusion technique arterial spin labeling (ASL). Unlike PET, ASL is completely non-invasive, which is ideal for mapping longitudinal changes in brain function. However, inconsistent results across FTD studies highlight the need to optimize ASL, particularly given that the quality of CBF images is sensitive to the imaging parameters. Accordingly, this thesis investigated the potential of ASL for the detection of longitudinal perfusion changes associated with FTD and differential diagnosis of FTD subtypes.
To evaluate the sensitivity of ASL, CBF measurements from ASL were compared to PET with radiolabeled water (15O-water), the gold-standard for imaging CBF. To avoid arterial sampling, in Study I, I developed and validated a non-invasive PET/MR approach (i.e. PMRFlow) to quantify perfusion by 15O-water using a porcine model. Excellent agreement was found when compared to PET-only measurements (R2 = 0.9, slope = 0.88) over a flow range from 30 to 100 ml/100g/min. In Study II, I evaluated the sensitivity of ASL relative to 15O-water for identifying regional hypoperfusion. While 15O-water showed superior sensitivity, ASL was also able to identify regional hypoperfusion specific to FTD subtypes (sensitivity = 70%, specificity = 78%). In Study III, I characterized the longitudinal reproducibility and reliability of ASL using optimized sequence parameters. Good agreement of repeat measures (month-separated) was found in both patients (CV = 16.3%, ICC = 0.62) and controls (CV = 13.9%, ICC = 0.62). Additionally, with a post labeling delay of 2s, transit time errors were not a significant source of error.
In capitalizing on the unique features of PET/MR imaging, most notably the ability to simultaneously acquire PET and MRI-based perfusion, this thesis demonstrates the utility of ASL and supports its use in longitudinal studies of FTD.
Summary for Lay Audience
Frontotemporal dementia (FTD) is a debilitating neurodegenerative disease with no known cure. Recent advancements in potential symptomatic and disease modifying therapies, highlight the need for imaging biomarkers to distinguish FTD subtypes and evaluate the efficacy of novel therapies. Considering that brain perfusion is an early marker of neurodegeneration, perfusion MRI technique arterial spin labeling (ASL) is an attractive approach for studying the natural disease progression to identify clinical endpoints. Ultimately, this could allow for earlier intervention, and provide a means to track changes in response to novel therapies. However, inconsistent results across FTD studies highlight the need to optimize ASL, particularly given that it is well known that the quality of brain perfusion images are sensitive to the chosen imaging parameters.
The objective of this thesis was to explore the role of ASL in differential diagnosis and longitudinal monitoring of FTD by: developing and validating PMRFlow, a non-invasive approach for measuring perfusion by the gold standard method, positron emission tomography (PET) with radiolabeled water (15O-water); optimizing ASL imaging parameters by a direct comparison to the aforementioned non-invasive approach for measuring perfusion by 15O-water PET; evaluating the sensitivity of ASL for detecting disease-related changes in perfusion relative to 15O-water PET; and evaluating its longitudinal reproducibility and reliability. In capitalizing on the unique features of PET/MR imaging, most notably the ability to simultaneously acquire PET- and MRI-based perfusion images, this thesis demonstrates the utility of ASL and supports its use as a biomarker in FTD.
Ssali, Tracy, "Validation of Arterial Spin Labeling for Longitudinal Monitoring and Differential Diagnosis of Frontotemporal Dementia" (2021). Electronic Thesis and Dissertation Repository. 8187.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Available for download on Friday, September 01, 2023