Multi‐parametric Neuroimaging in a Non‐Human Primate Model of Sporadic Cerebral Amyloid Angiopathy (Saimiri boliviensis boliviensis)

Abstract

AbstractBackgroundSquirrel monkeys (SQMs) are a unique non‐human primate model (NHP), which unlike other NHPs, develops extensive age‐associated cerebral amyloid angiopathy (CAA). The growing recognition of CAA contribution to clinical decline and CAA‐related complications in current clinical trials underscore the utility of the SQM model for studying CAA pathology development. Neuroimaging can play a key role in noninvasively detecting and monitoring disease progression in conjunction with histological methods. We have previously shown the utility of quantitative R2 * to map age‐related differences in SQM neuropathology cross‐sectionally and longitudinally. Here we aim to complement our protocol with advanced MRI approaches sensitive to brain microstructure to enable early and specific detection of alteration in tissue integrity. Our overarching goals aim to establish the in vivo and ex vivo MRI metrics profile of young and geriatric SQM cohorts.Method2D T2‐w (a) Turbo‐spin‐echo (TSE) and (b) FLAIR MRIs were acquired in vivo and ex vivo, while (c) multi‐shell diffusion‐weighted MRI (DWI) was limited to ex vivo imaging. Diffusion tensor imaging (DTI) and Diffusion kurtosis imaging (DKI) metrics were derived from comprehensive DWI datasets. Immunohistochemistry was subsequently performed to characterize the microstructural underpinnings of MRI abnormalities.ResultBothin vivo and ex vivo T2‐w TSE and FLAIR helped identify spontaneous asymmetric hyperintensities (ARIA‐E) in geriatric SQMs (21‐23yrs), predominantly in the parieto‐occipital WM, not seen in the young cohort (5‐yrs). Importantly, DKI‐derived mean kurtosis proved to be the most sensitive of DWI metrics for detecting lesions in the parieto‐occipital region. Widely adopted DTI metrics were far less sensitive in identifying microstructural changes. Histological assessment of region‐matched sections revealed Iba1 microgliosis, as well as immunoreactivity for both extravascular fibrinogen and decreased MBP when compared to the contralateral side. Aß immunohistochemistry (6E10/4G8) confirmed presence of CAA in the affected regions.ConclusionAdvanced DWI models are sensitive and specific neuroimaging tools for identifying abnormalities associated with tissue integrity that are potentially linked to age‐associated cerebrovascular dysfunction. Ex vivo conventional MRI detected similar abnormalities observed in vivo and proved to be a useful step to validate the performance of advanced imaging strategies, which can help bridge the gap between in vivo MRI and histology.