Introduction: We tested how tube types (ethylenediaminetetraacetic acid [EDTA], serum, lithium heparin [LiHep], and citrate) and freeze-thaw cycles affect levels of blood biomarkers for Alzheimer's disease (AD) pathophysiology, glial activation, and neuronal injury.

Methods: Amyloid beta (A beta)42, A beta 40, phosphorylated tau181 (p-tau181), glial fibrillary acidic protein, total tau (t-tau), neurofilament light, and phosphorylated neurofilament heavy protein were measured using single molecule arrays.

Results: LiHep demonstrated the highest mean value for all biomarkers. Tube types were highly correlated for most biomarkers (r > 0.95) but gave significantly different absolute concentrations. Weaker correlations between tube types were found for A beta 42/40 (r = 0.63-0.86) and serum t-tau (r = 0.46-0.64). Freeze-thaw cycles highly influenced levels of serum A beta and t-tau (P<.0001), and minor decreases in EDTA A beta 40 and EDTA p-tau181 were found after freeze-thaw cycle 4 (P<.05).

Discussion: The same tube type should be used in research studies on blood biomarkers. Individual concentration cut-offs are needed for each tube type in all tested biomarkers despite being highly correlated. Serum should be avoided for A beta 42, A beta 40, and t-tau. Freeze-thaw cycles > 3 should be avoided for p-tau181.