Chronic traumatic encephalopathy (CTE) is a progressive neurodegenerative disease that has been increasingly linked to traumatic brain injury. The neuropathology that distinguishes CTE from other tauopathies includes hyperphosphorylated tau (pTau) tangles and tau positive astrocytes irregularly distributed in cortical sulcal depths and clustered around perivascular foci. These features are clearly identified using immunohistochemistry, but are undetectable to current clinical imaging methods. Diffusion imaging has been proposed as a noninvasive method to detect the pathognomonic lesion of CTE in vivo because of its high sensitivity to microstructural alterations in tissue structure. While several diffusion imaging approaches, ranging from diffusion tensor imaging (DTI) to more advanced schemes such as generalized q-sampling imaging (GQI) and diffusion kurtosis imaging (DKI) may prove useful, the relationship between changes in diffusion-derived metrics and the underlying pathology remains unknown. We have developed and implemented a method of perform radiological-pathological correlations in tissues with diagnoses of CTE, aimed to determine whether high spatial resolution diffusion imaging is capable of sensitively detecting pTau pathology.

Human ex vivo cortical tissues diagnoses with Stage III/IV CTE, Alzheimer's disease (AD) or frontotemporal lobar dementia (FTLD) were scanned in an 11.74T Agilent MRI scanner using DTI, GQI and DKI acquisition schemes with isotropic in-plane spatial resolution of 250microm and 500microm slice thickness. Following image acquisition, tissues were sectioned and stained for histopathological markers including AT8 (pTau), GFAP (astrocytes) and Myelin Black Gold II (myelinated white matter). A custom script was used to co-register histological to MRI images, allowing for the ability to perform high spatial resolution correlations of histological with diffusion metrics. Using this approach, we found no relationship between pTau in sulcal depths and any of our DTI, GQI and DKI based measures. Interestingly, we found that white matter underlying sulcal depths in CTE tissues showed signs of disruption, a finding that we did not observe in AD or FTLD tissues. Furthermore, white matter integrity in these regions was correlated with fractional anisotropy. These findings demonstrate that high spatial resolution diffusion imaging is capable of detecting white matter disorganization closely related to pTau pathology in CTE, and may provide a more sensitive and specific means of diagnosing CTE.