Pre-analytical pitfalls: How blood collection tubes influence exercise-induced cell-free DNA concentrations

Abstract

Circulating cell-free DNA (cfDNA) is a promising biomarker for physiological stress, including exercise-induced responses. However, the lack of standardization in blood collection tubes (BCTs) for quantification of cfDNA hampers inter-study comparisons. In this study, we assessed the impact of different BCTs on exercise-induced cfDNA dynamics. Eleven participants [25 (SD 2.3) years of age] performed three different treadmill exercise protocols, including an all-out test and combinations of constant and interval load. Blood samples were collected before, 5 min and 30 min post-exercise using EDTA, lithium-heparin (LH) and serum BCTs. Concentrations of cfDNA were quantified using quantitative PCR. The cfDNA increased significantly across all protocols and BCTs. A significant effect of BCT on cfDNA concentrations (P = 0.034) was found, with serum showing higher concentrations than EDTA and LH. Although absolute differences from pre- to post-exercise were comparable across BCTs (P = 0.476), fold changes differed significantly (P = 0.012), with the highest observed in EDTA and the lowest in serum. Bland-Altman analyses demonstrated better agreement between EDTA and LH compared with serum. Significant correlations of cfDNA with energy expenditure and peak oxygen uptake were found. These correlations were stronger in EDTA and LH than in serum. Our findings highlight the crucial influence of BCT choice on cfDNA measurements in exercise settings. Given that EDTA and LH reflected exercise load better, they could be preferred for exercise physiology research. This work underscores the need to account for the choice of BCT to improve data comparability across studies. Additionally, these findings might have broader implications for clinical settings where cfDNA is used as a biomarker.

Keywords: blood biomarker; blood collection tubes; cell‐free DNA; circulating DNA; exercise; pre‐analytical factors; running.

FIGURE 1

Study design. Every participant made two visits, one in week 1 and one in week 2. They underwent different exercise protocols. At T0, T2 and T3 blood was sampled at three time points: before the onset of exercise (pre); within 5 min after cessation of the exercise (post +5); and 30 min after cessation of the exercise (post +30). At every time point, blood was sampled into serum, LH and EDTA BCTs. Abbreviations: BCT, blood collection tube; IAT, individual anaerobic threshold; LH, lithium–heparin.

FIGURE 2

Impact of exercise on the cfDNA concentration in the different BCTs. (a) Absolute concentrations of cfDNA for the three exercise protocols. (b) Absolute differences in cfDNA compared with pre‐exercise concentrations for the three exercise protocols. (c) Fold changes of cfDNA compared with pre‐exercise concentrations for the three exercise protocols. Significant P‐values of post hoc comparisons with Bonferroni correction are labelled. Abbreviations: BCT, blood collection tube; cfDNA, cell‐free DNA.

FIGURE 3

Scatter diagrams and Bland–Altman plots of cfDNA absolute differences +5. (a, d) Comparison between EDTA and LH. (b, e) Comparison between EDTA and serum. (c, f) Comparison between LH and serum. The dotted line in scatterplots is the line of equality. Continuous and dashed lines in Bland–Altman plots are the mean and the mean ± 1.96 SD. Abbreviations: cfDNA, cell‐free DNA; LH, lithium–heparin.

FIGURE 4

Spearman correlation between cfDNA and exercise parameters. (a) All‐out test (T0). (b) Endurance exercises (T2 and T3). Spearman correlation was conducted for the absolute difference +5 (triangles, absolute difference +5) and for the fold changes +5 (circles, fold change +5) in all BCTs. Significant correlations are depicted in black and non‐significant correlations in grey. The duration of endurance tests was constant for all subjects. Abbreviations: BCT, blood collection tube; cfDNA, cell‐free DNA; RPE, rating of perceived exertion; ${\dot{V}}_{{\mathrm{O}}_2}$ , oxygen uptake; ${\dot{V}}_{{\mathrm{O}}_2\mathrm{peak}}$ , peak oxygen uptake.