EFFECTS OF UP TO 14 YEARS OF BIOBANK STORAGE OF CSF BIOMARKERS AB42, TTAU, AND PTAU

Abstract

The combined profile of CSF aβ42, tTau, and pTau aids in diagnosing Alzheimer’s disease (AD). Unfortunately, sensitivity to (pre-)analytical variation has so far hampered universal cut-off values and worldwide clinical implementation of these markers (Fourier et al., 2015; Le Bastard, De Deyn, & Engelborghs, 2015; Willemse & Teunissen, 2015). The effect of long-term storage on CSF quality is an important pre-analytical variation factor, especially in studies using historical cohorts. So far, only two studies have addressed this topic directly, reporting storage at -80°C not influencing CSF aβ42, tau, and p-tau levels during storage periods of two and six years (Bjerke et al., 2010; Schipke et al., 2011). Here, we study the levels of CSF aβ42, tTau, and pTau in relation to storage duration, based on the assumption that mean biomarker levels in a clinical homogenous AD patient cohort remain constant over the years. The pre-analytical CSF biobanking protocol was essentially the same over these years, precluding pre-analytical bias. As such, storage duration is the only dependent variable. We selected an AD cohort (n=150) from the VU University Medical Center matched on age, MMSE score, sex, and ApoE genotype, with inclusion dates equally distributed over the years 2001-2013. Aβ42, tTau, and pTau concentrations were measured with one lot of ELISAs (Fujirebio). Regression analyses were performed to assess the effect of storage time on biomarker levels. We found that levels of CSF biomarkers aβ42, tTau, and pTau were not related to storage time at -80°C. However, we found a difference in biomarker concentrations measured during routine over the years, compared to the concentrations at re-measurement in this experiment. The longer ago the routine analysis was done, the larger the discrepancy in concentrations, resulting in significant correlations with storage time for aβ42 (r2=0.232) and tTau (r2=0.040). Levels of CSF aβ42, tTau, and pTau did not vary due to longer storage at -80°C in a homogenous AD cohort. Differences in aβ42 and tTau levels measured at start versus measured after storage become larger when storage time increases, which will consequently be rather due to changes in the assay than changes due to storage.