Abstract

AbstractBackgroundWhile research has focused on gray matter alterations in neurodegeneration, investigating the relationship between white matter and neuropathology allows for additional insight into disease effects on broad brain networks. TAR DNA‐binding protein 43 (TDP‐43), which has shown to be involved in various neurodegenerative disorders involving axonal damage including ALS, FTLD, and LATE, is associated with lower WM integrity. We used sparse canonical correlation analysis (SCCA) to explore the relationships between ante‐mortem WM integrity and post‐mortem TDP‐43.MethodDiffusion weighted images were gathered on a 1.5T scanner and processed using TORTOISE in order to calculate fractional anisotropy (FA) in a sample of 70 older adults from two longitudinal cohort studies, the Religious Orders Study and Rush Memory and Aging Project. We registered the white matter ROIs from the John’s Hopkins University (JHU) DTI atlas to the study‐specific template using ANTS. Average FA was computed for each ROI and was included in the SCCA. A semi‐quantitative rating of TDP‐43 severity was assessed in 5 regions. Disease burden in each region and average FA within 48 ROIs were inputs for the SCCA. A fused lasso penalty based on permutation testing was employed to promote sparsity.ResultThe 70 subjects were 91.10 (SD=6.08) years old at death with a median interval from MRI to death of 3.11 years (SD=1.45). Sixty‐four percent were female, and 36% had dementia at their last study visit prior to death. TDP pathology was present in 47% of individuals. SCCA produced one canonical variable (Cor=0.36) which identified one regional pathology variable (TDP‐43 within the entorhinal cortex; Feature weight =‐1.0) and six FA ROIs: pontine crossing tract (0.597), splenium of corpus callosum (0.531), right anterior limb of internal capsule (0.001), superior corona radiata (0.411), and bilateral anterior corona radiata (Right: 0.115 and Left: 0.422).ConclusionIdentified WM ROIs within the brainstem align with previous research within TDP‐43 related disorders, including those with motor neuron involvement such as FTLD‐U. Further, findings within bundles connecting parietal and temporal areas indicate the importance of TDP‐43 in broad brain networks which could relate to the deficits seen in memory and executive functions within FTLD and related disorders.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call