Advancements in Saturation Transfer Imaging of the Spinal Cord
Lawless, Richard Dylan
The spinal cord is a narrow, cylindrical structure responsible for rapidly transmitting electrical signals throughout the nervous system and is the primary sensorimotor pathway for the human body. Therefore, damage to the spinal cord caused by neurodegenerative diseases, such as multiple sclerosis (MS), can lead to severely impaired neurological function. Conventional magnetic resonance imaging (MRI) is sensitive to late-stage inflammatory lesions and tissue atrophy, but are often poor indicators of disease progression and do not report on underlying pathophysiology of MS. Quantitative MRI biomarkers capable of detecting tissue changes earlier in the disease pathology may have significant implications in the diagnosis, prognosis, and treatment of MS. The primary focus of this work is saturation transfer (ST) imaging, including both quantitative magnetization transfer (qMT) and chemical exchange saturation transfer (CEST). In a ST experiment, a pool of exchangeable protons is saturated using an off-resonance RF pulse, and through direct chemical exchange (CEST) and dipolar coupling (MT) the magnetization is transferred to the surrounding water protons, resulting in an observable attenuation of the water signal. The magnitude of signal attenuation provides an indirect measurement of the exchanging species, which is related to its concentration and exchange rate. The ability to observe and measure the biochemical alterations within tissue provides a distinct advantage over T1- and T2-weighted MRI, which measures water content only, and may help to resolve the disconnect between radiological findings and clinical presentation. qMT provides quantitative estimates of the semi-solid concentration within tissue, primarily the pool-size-ratio (PSR) which has been shown to correlate with white matter myelin density. CEST, conversely, is a more spectrally selective saturation transfer method that provides sensitivity to endogenous mobile solutes with exchangeable protons. While Amide Proton Transfer (APT), the most commonly explored CEST effect, can be related to protein concentration and pH in vivo, it also suffers from several confounding factors. Contributions from larger concentration pools, such as direct water saturation (DS) and the macromolecular (MT) component, often obfuscate the lower concentration metabolite pools which is the target of CEST experiments and may offer greater insight into pathology. Additionally, due to the lack of a well-established quantitative model, CEST results in the literature are primarily reported by calculating the asymmetry of the CEST Z-spectrum around the amide resonance frequency. This technique is susceptible to changes in the semi-solid pool size as well as relaxation time differences and is dependent on scan parameters, leading to difficulties in reproducibility and limited multisite comparison. Thus, the studies proposed herein will seek to develop a standardized, quantitative technique for modelling the results of CEST experiments in order to extract more specific, clinically relevant information. Both myelin density and protein/peptide concentration can be aberrant in patients with MS because MS is both a demyelinating and inflammatory process, respectively. Additionally, treatment of neurodegenerative disorders such as MS will alter the chemical composition of the tissue well before any structural changes can be observed with clinical MRI. However, it has not yet been explored how combining the molecular information gained from qMT and CEST experiments can further our understanding of MS pathology, evolution, or therapeutic intervention. Therefore, we performed a longitudinal, cross-cohort study of MS patients with varying disease severity within a low-severity cohort to determine if qMT and CEST parameters can act as biomarkers for disease progression and treatment efficacy. We have investigated our hypothesis in both ordered (spinal cord) and disordered (cerebrospinal fluid) tissues to gain a more complete understanding of MS pathology and, ultimately hope the work herein will improve the diagnostic and prognostic capabilities of MRI in vivo, closing the gap between radiological imaging and neurological disability.