Toward an Olfactory Imaging Biomarker for Early-Stage Parkinson’s Disease

Background and Problem Statement Despite a stubborn 19th-century myth questioning the importance of olfaction in humans, it is clear that the sense of smell plays a fundamental role for human behavior and cognition (1). For instance, olfaction is critical for guiding food search, identifying potential threats in the environment, and for navigating social interactions. Olfaction is among the first senses to be affected by neurodegenerative disorders, including Parkinson’s Disease (PD)(2). In fact, olfactory dysfunction is one of the earliest non-motor features of PD (3, 4), and is present in 90% of early-stage PD patients (5), a prevalence that is higher than many cardinal motor signs (6). Because olfactory deficits precede the onset of motor symptoms in PD, olfaction may serve as a biomarker for early-stage PD (7). Olfactory perceptual performance has received some attention as a potential biomarker for the early and differential diagnosis of PD (7). Beyond perceptual performance, the structural integrity of olfactory pathways, as measured in vivo using neuroimaging methods, could provide additional diagnostic information. Importantly, imaging biomarkers could help to dissociate age-related olfactory decline from early-stage PD-related neurodegeneration. However, imaging biomarkers for PD based on the olfactory system are relatively unexplored. A few imaging studies have focused on olfactory bulb volume, but provided mixed results (8, 9). More recently, diffusion-weighted imaging (DWI) methods have been applied to measure potential microstructure changes in olfactory tissues (10-12). Imaging olfactory brain structures in vivo is technically challenging (13-15), and inconsistencies among studies likely stem from these difficulties. Most importantly, olfactory structures are located in ventral brain areas that are prone to susceptibility artifacts. These artifacts result in substantial image deformations (stretching and shrinking). In addition, conventional single-shot echo planar imaging (SS-EPI) techniques suffer from “T2-blurring”, reducing image contrast (16). Finally, white-matter fibers in olfactory areas are small and sparsely myelinated, resulting in less restricted diffusion of water molecules through this tissue. This leads to weaker diffusion signals compared with larger white-matter fiber bundles (17). Moreover, the conventional tensor model is unable to resolve multiple directions of constrained diffusivity within a single voxel, and tractography based on this model is therefore unable to capture the small, interleaved white matter fibers that connect olfactory regions (18, 19). These challenges are not trivial; in fact, they have prevented a comprehensive characterization of the basic anatomy of the olfactory system in humans (17, 20, 21). The goal of the current project is to resolve these technical issues and to characterize the anatomy of olfactory pathways in humans. Mapping out olfactory structures with high precision will allow us to generate olfactory pathways templates. These templates will then be used in conjunction with microstructure measurements calculated using diffusion scans from PD patients to derive imaging biomarkers for early-stage PD.