Premium
Longitudinal changes in established and exploratory cerebrospinal fluid biomarkers by PTAU/AΒ 42 status in cognitively unimpaired adults
Author(s) -
Van Hulle Carol A,
Betthauser Tobey J,
Jonaitis Erin M,
Batrla Richard,
Wild Norbert,
Kollmorgen Gwendlyn,
Andreasson Ulf,
Bendlin Barbara B,
Asthana Sanjay,
Carlsson Cynthia M,
Johnson Sterling C,
Zetterberg Henrik,
Blennow Kaj
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.047156
Subject(s) - biomarker , neurogranin , cerebrospinal fluid , receiver operating characteristic , neurodegeneration , medicine , oncology , psychology , pathology , disease , chemistry , biochemistry , protein kinase c , enzyme
Background Characterizing longitudinal changes in AD and neurodegeneration related CSF biomarkers in cognitively unimpaired adults is critical to distinguishing typical aging from disease pathology, Yet longitudinal change in established and especially emerging CSF biomarkers are not well characterized. We examined linear age trends in CSF biomarkers for AD (Ab 42 , pTau), neurodegeneration (NFL, Neurogranin and α‐Synuclein) and glial activation (YKL40, GFAP, sTREM2, S100B) assayed using the Roche NeuroToolKit (NTK), a panel of automated prototype CSF immunoassays. Methods Analyses were based on a subsample of 146 cognitively unimpaired participants who underwent at least two lumbar punctures (total N= 402). Cut‐offs for the ratio of pTau/Ab 42 were determined using receiver operating characteristic curve analysis in a larger sample of impaired and unimpaired middle‐aged adults ( N =167) treating a positive [C‐11]PiB visual rating as the standard of comparison. CSF analytes, plus ratios Ab 42 /Ab 40 and pTau/Ab 42 , were fit to linear mixed models with random intercepts and slopes. Biomarker status (pTau/Ab 42 +/‐) and covariates (gender, APOE 4 status, years education) were entered as fixed effects. We also tested for moderation of age effects by biomarker status. P‐values were adjusted for the false discovery rate. Results Sample characteristics are shown in Table 1. Twenty participants were AD biomarker positive. We observed small but significant subject‐level variation in slopes for most analytes, including the core AD biomarkers, NFL, neurogranin, YKL40, and S100B. Biomarker status (pTau/Ab 42 ) significantly predicted levels of AD related CSF biomarkers (Table 2), but not levels of neurodegeneration or glial activation related analytes. Ab 42 /Ab 40 , but not Ab 42, levels significantly decreased over time; all other analytes increased over time with the exception of S100B (Figure 1). Changes in biomarker values over time was not moderated by baseline biomarker status for any of the CSF analytes (not shown). Conclusion Prior preliminary analyses (recreated in Figure 2) suggested that baseline values for biomarkers of neurodegeneration and glial activation are not related to core AD biomarkers until later in disease pathology. We have added to those findings, showing that among cognitively unimpaired individuals, significant longitudinal changes in both core AD‐related and emerging CSF biomarkers appear to be independent of early signs of joint amyloid and tau pathology.