The goal of this project is to develop an early diagnostic test for Alzheimer's disease (AD). Neuropathology studies show, in mild cognitive impairment (MCI) and some normal (NL) elderly, neurofibrillary and plaque pathology that is associated with neuronal and volume losses in the entorhinal cortex (EC) and hippocampus. Our prior work shows that MRI volume reductions in these regions predict future MCI and AD. However, volume losses lack disease specificity. Together, our most recent publication and those of our collaborators show that tangle-related abnormal tau proteins, hyperphosphorylated at threonine 231 (P-tau231), are: 1) elevated in the CSF in MCI and AD; 2) diagnostically specific for AD; 3) useful in predicting further cognitive decline in MCI; and 4) diluted in the CSF by the progressive ventricular enlargement of AD, such that CSF volume adjustment enables detection of longitudinal P-tau231 increases in MCI. Our pilot data demonstrates that P-tau231 significantly adds to the MRI hippocampal volume in the classification of MCI and NL. A-beta40 levels are also elevated in MCI. Overall, these findings suggest that by combining the sensitivity of the MRI hippocampal volume with the specificity of CSF P-tau231, a sensitive and specific in vivo recognition of AD pathology in MCI is possible. We propose the longitudinal study of 30 NL, 80 MCI, 30 mild AD, and 30 frontotemporal dementia (FTD) patients. Over 5-years, we will complete baseline and 24-month follow-up exams. We will use standardized clinical protocols at each observation and collect normreferenced neuropsychological data, high-resolution MRI, and CSF. We hypothesize that elevated Ptau231 is specific for AD, and that it independently contributes to the hippocampal volume in both the differential diagnosis of AD, and in predicting the MCI conversion to AD. All the required clinical components, laboratory, and imaging for this study are in place with established quality control. There are ample numbers of volunteer subjects and patients available. There is adequate statistical power for testing the hypotheses.
Showing the most recent 10 out of 183 publications
