Quantitative impact of pre-analytical process on plasma uracil when testing for dihydropyrimidine dehydrogenase deficiency

Abstract

Aims: Determining dihydropyrimidine dehydrogenase (DPD) activity by measuring patient's uracil (U) plasma concentration is mandatory before fluoropyrimidine (FP) administration in France. In this study, we aimed to refine the pre-analytical recommendations for determining U and dihydrouracil (UH₂ ) concentrations, as they are essential in reliable DPD-deficiency testing.

Methods: U and UH₂ concentrations were collected from 14 hospital laboratories. Stability in whole blood and plasma after centrifugation, the type of anticoagulant and long-term plasma storage were evaluated. The variation induced by time and temperature was calculated and compared to an acceptability range of ±20%. Inter-occasion variability (IOV) of U and UH₂ was assessed in 573 patients double sampled for DPD-deficiency testing.

Results: Storage of blood samples before centrifugation at room temperature (RT) should not exceed 1 h, whereas cold (+4°C) storage maintains the stability of uracil after 5 hours. For patients correctly double sampled, IOV of U reached 22.4% for U (SD = 17.9%, range = 0-99%). Notably, 17% of them were assigned with a different phenotype (normal or DPD-deficient) based on the analysis of their two samples. For those having at least one non-compliant sample, this percentage increased up to 33.8%. The moment of blood collection did not affect the DPD phenotyping result.

Conclusion: Caution should be taken when interpreting U concentrations if the time before centrifugation exceeds 1 hour at RT, since it rises significantly afterwards. Not respecting the pre-analytical conditions for DPD phenotyping increases the risk of DPD status misclassification.

Keywords: dihydropyrimidine dehydrogenase; dihydrouracil; intra-individual variability; pre-analytical practices; uracil.

FIGURE 1

Mean relative deviations (± standard deviation) of the uracil concentrations in (A) whole blood and (B) plasma samples kept at room temperature (RT, red points) or +4°C (blue points). Vertical dotted lines represent the current recommended maximal time to keep whole blood samples at RT (1.5 h, in red) and +4°C (4 h, in blue). ±20% acceptance area is in grey. *P < .05 (Wilcoxon signed rank paired test)

FIGURE 2

Mean relative deviations (± standard deviation) of UH₂ concentrations and UH₂/U ratio in whole blood (left panel) and plasma samples (right panel) kept at room temperature (RT, red points) or +4°C (blue points). Vertical dotted lines represent the current recommended maximal time to keep whole blood samples at RT (1.5 h, in red) and +4°C (4 h, in blue). ±20% acceptance area is in grey. *P < .05 (Wilcoxon signed rank paired test)

FIGURE 3

Observed (dots) and individual predicted (lines) uracil (U) concentrations as a function of time (in h), stratified by storage conditions. The Gompertz model adequately captures the saturation in U increase for samples stored at room temperature.

FIGURE 4

Bland–Altman plot showing differences between two samples for a given patient (Y axis), against the mean U concentration of the two samples (X axis, log scale). Patients were separated into two categories: (A) those having two compliant samples (n = 495, first sample chosen as the reference, Uref) and (B) those having only one compliant sample (n = 65, compliant sample chosen as the reference, Uref). Difference between the two samples was calculated as follows: U test − Uref. The central dotted blue line represents the mean difference in U of repeated sampling, and the dotted red lines represent the 95% confidence interval.

FIGURE 5

Effect of sampling time on uracil concentrations. (A) The individual values (grey symbols) of U from patients with compliant samples were plotted as a function of their sample timing. Means and standard deviations of the corresponding 1‐hour interval appear in blue. (B) The U concentrations were compared for patients double sampled at two different moments of the day (before or after 12 pm, Wilcoxon paired comparison test).