S4‐02‐03: Longitudinal CSF changes in markers of neurodegeneration and utilization of these markers in predicting conversion from normal cognition to cognitive impairment

Increasing evidence suggests that the pathology of Alzheimer's disease (AD) begins to occur ∼10-15 years prior to onset of any clinically detectable cognitive impairment. By utilizing CSF biomarkers and assessing middle aged and elderly people longitudinally, one may be able to stage the different periods of brain health including the absence of significant AD neuropathology as well as the presence or amount of various aspects of AD pathology and brain injury. We assessed cognitively normal individuals at entry that were 40-90 years of age. A subset of these individuals are participants in the Washington University Adult Children Study (ACS), a longitudinal clinical and biomarker study of cognitively normal middle-aged individuals (age 45-74) that is designed to investigate the time course of AD pathologic biomarker changes during the preclinical period. Levels of CSF Ab42, tau, ptau181, YKL-40, and visinin-like protein-1 (VILIP-1) were assessed at baseline and in some individuals longitudinally. Longitudinal analysis of individuals in the ACS who start with normal CSF biomarkers reveal that the first changes consistent with AD pathology appear to be decreases in Aβ42 beginning in some people in the age range 55-64 with increases in tau occurring in some individuals by later middle age (ages 65-74). These changes are greater in those with a positive family history of AD. Assessment of individuals with a mean age in their 70's is consistent with the changes in Aβ42 being early and tau as well as other markers being later in that the best predictors of conversion from being cognitively normal to cognitive impairment over a 3-5 year window are 1) the ratios of tau/Aβ42 and p-tau/Aβ42 (markers of neurodegeneration and possibly tangles combined with plaques) and 2) the ratios of VILIP-1/Aβ42 and YKL-40/Aβ42 (markers of neurodegeneration and inflammation respectively, combined with plaques). This data supports the hypothesis that Aβ deposition and tauopathy/neurodegerenation begin to occur years before cognitive decline and that assessments of tau, VILIP-1, and YKL-40 may be useful for diagnosis, staging, and prognosis. They may also be markers that respond to treatments that prevent or decrease neurodegeneration and inflammation.