Background

Dementia with Lewy bodies (DLB) is the second most common cause of degenerative dementia in the elderly after Alzheimer's disease (AD). While DLB is neuropathologically defined by the accumulation of misfolded α-synuclein into Lewy bodies and neurites, many individuals with DLB frequently show AD copathologies such as amyloid-β and intracellular tau neurofibrillary tangles. However, the precise mechanisms through which they influence the cascade of neurodegenerative processes in DLB remains unclear. Neurite Orientation Dispersion and Density Imaging (NODDI) is a relatively recent development in neuroimaging, allowing for new avenues to probe subtle disease-related processes in DLB. Unlike conventional MRI methods, NODDI employs advanced multi-compartment modeling to selectively map the brain’s intricate network of white matter fiber bundles down to the level of axons and dendrites. As such, NODDI may be well-suited to capture early and potentially more sensitive changes in response to pathological insults, which DTI may overlook.  To date, little is known regarding (i) the spatial distributions of NODDI parameters across different brain regions in individuals with DLB and (ii) the potential associations between these parameters and PET biomarkers of amyloid-β and tau.

Objectives of the study

Our study pursued three primary objectives: First, we used NODDI and DTI to compare white matter microstructure integrity between 45 patients across the DLB spectrum and 45 cognitively unimpaired controls. At the Mayo Clinic Alzheimer’s Disease Research Center, we included patients with clinically probable DLB at mild to moderate clinical stages (n = 32) and those with prodromal DLB (n = 13) to comprise a group of people on the DLB spectrum. Secondly, we delineated the multivariate associations of the APOE ε4 genetic risk factor and PET markers of amyloid-β and tau pathology with white matter injury using Structural Equation Modeling. Thirdly, we examined the relationship of tau-associated neurite injury with clinical measures of disease severity.

Widespread extent of neurite abnormalities in DLB

Compared to cognitively unimpaired controls, DLB patients exhibited widespread abnormalities in the white matter, spanning major fiber pathways across the frontal, parietal, temporal, and limbic regions. DTI comparisons revealed that Fractional Anisotropy, a general measure of fiber coherence and microstructural integrity, was decreased, while Mean Diffusivity, a sensitive marker reflecting overall magnitudes of water diffusivity, was markedly elevated. Moreover, these DTI abnormalities spatially coincided with significantly lower values of Neurite Density Index from NODDI. Although our results broadly align with prior DTI studies documenting widespread white matter injury in DLB, the implementation of NODDI in this study extended the literature by pointing to axonal injury as a likely mechanism underlying white matter injury in DLB.

Influences of Alzheimer’s disease pathologies on neurite microstructure in DLB

Among individuals with DLB, tau deposition within the medial temporal lobe was strongly associated with more severe white matter injury. This relationship was especially pronounced along the temporo-limbic fiber tracts, which are intricately connected to the "cortical signature" of AD - a group of regions particularly susceptible to early tau accumulation and grey matter atrophy. Interestingly, our group at the Mayo Clinic has previously demonstrated that longitudinal increases in tau PET uptake were linked to accelerated atrophy in brain regions spatially proximal to the white matter abnormalities identified in the DLB group (Chen et al., 2022). This observation raises the intriguing possibility that the spread of tau pathology from the affected white matter tracts to the adjacent gray matter regions may trigger neuronal death and ensuing atrophy, as observed on structural MRI in patients with DLB.

 In contrast to the robust associations observed between tau pathology and neurite microstructural injury, amyloid-β initially appeared to have no significant univariate effect on any of the DTI and NODDI markers. While these null findings align with the mixed literature on the influence of amyloid-β in DLB, the Structural Equational Models revealed a significant, indirect pathway in which amyloid-β contributes to white matter injury through its mediating effect on tau accumulation, which then acts as the more proximal driver of axonal degeneration (amyloid-β → tau → neurite density loss). Supporting this interpretation, we found that tau-related white matter injury was associated with worse disease severity, particularly in DLB patients classified as amyloid-β positive on PET.

Future directions

Despite the biological plausibility of our SEMs, longitudinal studies are still needed to establish NODDI change rates and how they may relate to clinical decline, longitudinal accumulation of tau pathology, and amyloid-β over time in patients with DLB.  While our study included participants with prodromal DLB, the statistical analyses lacked sufficient power to discern group differences that may emerge at this early stage. Ongoing data collection and future research would help delineate the regional patterns of NODDI findings in prodromal DLB and investigate the contributions of AD pathologies on the transition from mild cognitive impairment with Lewy bodies to probable DLB.  Our study represents a significant step forward in mapping the correlations between AD co-pathology and white matter injury in DLB. Nevertheless, the complex interrelationships among amyloid-β, tau, and alpha-synuclein – each of which may promote downstream aggregation of the others – and their impact on DTI and NODDI abnormalities remain unclear. Future research integrating emerging biomarkers of α-synuclein would be highly anticipated to distinguish between additive or synergistic pathways of amyloid, tau, and α-synuclein on neurite alterations and clinical decline in patients with DLB.

Conclusion

Our NODDI study has demonstrated widespread white matter microstructural abnormalities in DLB, likely attributable to the cumulative effects of both AD and alpha-synuclein. These insights could be important for optimizing future disease-modifying trials in DLB, indicating that successful treatment strategies for individuals with DLB may adopt a comprehensive approach targeting amyloid and tau in addition to alpha-synuclein.