SIMULATED COGNITIVE TRAJECTORIES IN PATIENTS WITH/WITHOUT HISTORY OF STROKE WHO ARE AT RISK OF DEVELOPING ALZHEIMER’S DISEASE USING AD ACE

Abstract

There is substantial evidence in the literature that cerebrovascular factors such as history of stroke have a cumulative effect on Alzheimer's disease (AD) development and are associated/correlated with an increased risk of degenerative AD pathology and cognition decline. The AD Archimedes Condition-Event simulator (AD ACE) employs a set of AD disease equations to predict the progression of disease over time based on change in AD biomarkers including measures of amyloid PET (AV45) and tau (CSF t-tau) levels and their connections to clinical presentation of AD, including cognition and behavioral scales. History of stroke, however, was not included as a predictor in the AD ACE disease equations. We used AD ACE to compare patients with history of stroke versus those without and studied how potential effect of stroke may be mediated through other clinical AD biomarkers on disease progression over time. We simulated the disease progression of cognitively normal (CN) individuals and those with subjective memory complaint (SMC) in the ADNI1 dataset. We analyzed the change in two cognitive scales (MMSE, CDRSB) over 10 years between two patient subgroups based on their history of stroke at baseline. A total of 149 CN/SMC patients were identified in ADNI1 for inclusion in this study (18 with Stroke History, 131 without Stroke History). The two patient subgroups were very similar in terms of their mean age and other baseline biomarkers and cognition scales. The simulated trajectories showed faster cognitive decline (ΔCDRSB=1.3 and ΔMMSE=2) and an odds ratio of 1.37 for incidence of dementia for the subgroup with vs. without history of stroke over 10 years which is aligned with reported outcomes in the literature including a recent analysis of the patients from the Rotterdam Study. A simulation using the AD ACE equations, which do not explicitly include stroke history, projects faster cognitive decline and greater risk of dementia for patients with history of stroke. This suggests the effects of stroke on cognitive decline may be mediated through other clinical AD biomarkers.