Heterogeneity of factors associated with cognitive decline and cortical atrophy in early- versus late-onset Alzheimer's disease

The objectives of this study were to investigate the variable factors associated with cognitive function and cortical atrophy and estimated variable importance of those factors in affecting cognitive function and cortical atrophy in patients with EOAD and LOAD. Patients with EOAD (n = 40), LOAD (n = 34), and healthy volunteers with normal cognition were included (n = 65). All of them performed 3T MRI, [18F]THK5351 PET (THK), [18F]flutemetamol PET (FLUTE), and detailed neuropsychological tests. To investigate factors associated with neuropsychological test results and cortical thickness in each group, we conducted multivariable linear regression models, including amyloid, tau, cerebral small vessel disease markers on MRI, and vascular risk factors. Then, we estimated variable importance in associating cognitive functions and cortical thickness, using relative importance analysis. In patients with EOAD, global THK retention was the most important contributor to the model variances for most neuropsychological tests, except for memory. However, in patients with LOAD, multiple contributors beyond tau were important in explaining variance of neuropsychological tests. In analyses with mean cortical thickness, global THK retention was the main contributor in patients with EOAD, while in LOAD patients, multiple factors contributed equally to mean cortical thickness. Therefore, EOAD and LOAD may have different pathomechanistic courses.