Multi‐modal MR and synaptic PET changes in dementia with Lewy bodies.

Background: Novel, validated imaging biomarkers for assessing disease severity and progression in dementia with Lewy Bodies (DLB) are needed as standard neuroimaging changes such as atrophy are not pronounced. Both multi‐modal MR and PET offer opportunities, for example cortical microstructure and dendritic imaging can be assessed using advanced MR diffusion weighted imaging (DWI) methods such as neurite orientation dispersion and density imaging (NODDI) and PET with UCB‐J can assess synaptic density. We applied these and other markers to people with DLB and controls. Method: In linked studies, people with probable DLB and similarly aged controls (n = 10‐30 per group depending on study) underwent multimodal 3T‐MRI (including structural, multi‐shell DWI and arterial spin labelling imaging) and PET imaging with AV1451 (for assessment of tau), PiB (for amyloid status) and [11C]UCB‐J PET/MR. For MR, T1‐MRI and DWI datasets were processed using FreeSurfer 7.2 and FSL 6.0.4 respectively with whole‐brain ODI derived from the Microstructure Diffusion Toolbox. The [11C]UCB‐J and [18F]AV1451 binding potentials (BPND) were assessed using a simplified reference tissue model (reference centrum semiovale for UCBJ and cerebellum for AV1451). Group differences in ODI, UCB‐J and AV1451 BPND were examined and correlated with structural changes and clinical and cognitive variables (ACE‐R score). Result: Compared to controls, people with DLB showed significantly thinner neocortex (T = ‐2.8, p = 0.01) in temporo‐parietal regions but more widespread reductions in cortical microstructure (ODI, T = ‐3.9, p < 0.001)(Figure 1). Within the DLB group, hippocampal ODI was positively correlated with cognitive performance (robust linear regression: p < 0.001)(Figure 2). Reduced synaptic density (UCB‐J BPND) was found for the DLB group in posterior regions (p<0.05, FWE corrected)(Figure 3), with a trend for greater reductions in those who were amyloid positive. ACE‐R scores significantly correlated with synaptic density (p<0.05, FWE corrected). There were no significant differences in AV1451 binding between DLB and controls. Conclusion: Extensive reductions in ODI and reduced synaptic density, both associated with cognitive impairment and occurring in regions without significant atrophy or tau deposition, suggests that NODDI and synaptic PET imaging are promising biomarkers for DLB which merit further investigation, including in longitudinal studies. [ABSTRACT FROM AUTHOR]