The discovery of biomarkers for Alzheimer's disease (AD) is essential for identifying people in the earliest stages of the disease when therapy will be most effective. A major thrust of biomarker research has been to screen the cerebrospinal fluid (CSF) of patients to measure levels of amyloid and tau proteins associated with hallmark AD brain lesions. These resulting biomarkers have been favorable but not optimal for discriminating early stage AD subjects from controls, thus highlighting the need to augment the CSF biomarker panel with novel proteins to improve its diagnostic accuracy. During our studies of the cell survival substance nerve growth factor (NGF) in AD, we discovered that the cell death-promoting NGF parent molecule, proNGF, is much higher in the brains of people who died with mild cognitive impairment (MCI, a preclinical AD stage) or mild AD compared to those with no cognitive impairment (NCI). This suggests a pathological shift from NGF-mediated cell survival to proNGF-mediated cell death in the brain. Based on these findings, we tested whether CSF proNGF levels mark the onset of cognitive impairment using postmortem CSF samples from subjects who died with NCI, MCI or AD. We found that proNGF levels were increased 55-70% in CSF samples from MCI and AD compared to NCI. Moreover, increased proNGF levels were associated with poorer cognitive performance as measured by tests of global cognition. An additional pilot study of CSF from living control and AD subjects showed a 3-fold increase in proNGF levels in AD. Together, these findings suggest that increased CSF levels of proNGF may act as a biomarker for the progression of AD. This proposal will show that increased proNGF protein levels in CSF is a novel AD biomarker with high predictive value for identifying at-risk subjects who will benefit most from therapeutic intervention.