Translational Modeling Of Morphological And Physiological Features Of Vascular Dementia: A Review

Vascular dementia (VaD) is the second-most common form of dementia, accounting for 20% of all dementia cases. Although the pathobiology of VaD is poorly understood and lacks effective treatment, clinically relevant factors such as age and sex are rarely considered in preclinical modeling. This places efforts to develop future therapies at a severe disadvantage, as clinically relevant translation of underlying pathobiology may be lost. We reviewed common morphological and physiological outcomes in six major rodent models of VaD in a total of 258 full-text publications: 1) chronic cerebral hypoperfusion, 2) high fat diet, 3) diabetes, 4) hypertension, 5) carotid arterial calcification, and 6) and CADASIL (Cerebral Autosomal Dominant Arteriopathy with Sub-cortical Infarcts and Leukoencephalopathy), and evaluated translational features relevant to human pathophysiology with an analysis of timepoints of observed pathology and rodent sex and age. We found that all models of VaD shared common features of decreased CBF and upregulation of inflammatory signaling molecules such as TNF-α, IL-1β, and IL-6, and reactive oxygen species (ROS) such as SOD and NOX, consistent with clinical presentations of VaD. Only the carotid artery calcification model showed inconsistent evidence for brain endothelial tight junction loss, astrogliosis, and macrophage reactivity, in disagreement with other models. We conclude that improved translational insight may critically depend on 1) inclusion of a constellation of comorbidities in preclinical modeling instead of modeling each in isolation to better capture human disease pathogenesis, 2) increased use of middle-aged and aged preclinical rodent models which are sensitive to clinically relevant disease presentation in middle-age and elderly individuals, and 3) inclusion of female animals in disease models as sex is a relevant biological factor which may better define future therapeutic strategy in real-world populations.