Assessments of the Utilities of CSF NPTX2 for Disease Progression in Cognitively Normal Individuals Who Progress to Clinical MCI and AD

Abstract

Cerebrospinal fluid (CSF) amyloid and tau biomarkers provide reliable detection of amyloid plaque and, to a lesser extent, tau pathology across the Alzheimer’s disease continuum; however, they predict risk for progression from cognitively normal to mild cognitive impairment (MCI) or to dementia with limited accuracy. Amyloid pathology as detected using CSF Aβ42 occurs substantially earlier than does tau pathology. Reliable detection of amyloid plaque burden alone has limited ability to accurately predict time to cognitive decline and MCI in cognitively normal individuals. Synaptic dysfunction and loss is considered to be one of the earliest pathological mechanisms – and consequences-of Alzheimer disease (AD) and may be a sensitive measure of neurodegeneration. Thus measurements of synaptic biomarkers in biofluids have attracted heightened interest for the potential to improve upon the accuracy in prediction of risk for disease progression, beyond Aβ42 and tau, in cognitively normal individuals. The possibility that a marker of synaptic integrity such as Neuronal pentraxin-2 (NPTX2) becomes abnormal earlier than a CSF measure of tau and adds predictive value in cognitively normal individuals on an AD trajectory is a major aim of the study reported by Soldan et al. Of note, these authors developed and validated an LC/MSMS methodology based upon a composite of 3 NPTX2-specific peptides produced by trypsinization to quantify CSF NPTX2 and found that levels correlate highly with the author’s previously described immunoassay. They included measurements of CSF Aβ42/40, t-tau, and p-tau181 by highly validated fully automated immunoassays. Soldan et al. conducted their study in BIOCARD study participants (269 cognitively normal individuals with average age at baseline of 57.7 years with average follow-up time of 16.3 years; 77 progressed to MCI or to dementia). The average age of this cohort is lower compared to the vast majority of AD biofluid biomarker studies and is a very important feature as it enables disease detection and biomarker transitions at an earlier age. The major findings in this study include: (A) association of lower CSF NPTX2 concentration values with the onset of MCI clinical symptoms, a finding consistent with prior reports across AD disease progression, e.g. from MCI to dementia; (B) the relationship between lower NPTX2 concentration and risk for disease progression was similar for males and females and did not differ between apolipoprotein E (ApoE) ε4 carriers and non-carriers; (C) CSF NPTX2 added predictive value to p-tau181 or t-tau for predicting progression to MCI consistent with the hypothesis that this is as early a marker for cognitive decline as is Aβ42/40, thus adding to predictive power of these biomarkers alone and providing a tauindependent added biomarker measure. These are encouraging findings, although the extent of increased predictive value was relatively small after including amyloid and p-tau181 in the model. A similar set of analyses in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort measured CSF NPTX2 by mass spectrometry using a different method and found that NPTX2 decreased longitudinally by about 10% per year in MCI and predicted MCI to AD progression, but cross-sectional NPTX2 did not predict change from normal cognition to MCI. The difference between the results of ADNI and Soldan et al may lie in ADNI enrolling people at older ages and having shorter follow-up (up to 3 years was examined). There may be differences in neurodegeneration among people who develop MCI at relatively younger ages. A few studies have looked at other synaptic biomarkers in CSF as predictors. A recent report found that higher levels of the presynaptic protein GAP43 in CSF predicted progression in amyloid positive cognitively normal elders, although it is not clear how much additional value there is beyond Aβ42 and P-tau. GAP43 levels are increased in CSF and the mechanisms that contribute to the release of this marker into CSF are likely different