Last Updated: 17-10-2019 10:58
Within the framework of the HBM4EU project, an Inter-Laboratory Comparison Investigation (ICI) and External Quality Assurance Schemes (EQUAS) was organized and conducted for the analysis of Bisphenol A (BPA), Bisphenol S (BPS) and Bisphenol F (BPF) in human urine. The study was conducted from December 2018 to February 2019.
In total, 34 laboratories from 21 countries participated in this ICI/EQUAS. Among participants, four expert laboratories were included, two from Europe (HBM4EU consortium) and two from outside Europe (USA, Canada). The participation in the ICI/EQUAS was satisfactory, as 32 out of 34 laboratories submitted their results.
Four different test samples consisting of 10 mL urine each were prepared, corresponding to different concentration levels of the targeted biomarkers (incurred and fortified), and sent to the participating laboratories for analysis. These samples were defined as follows:
– 1 sample at very low level (VL), corresponding to a non-fortified pool of individual human urine samples for which the concentration of the targeted markers were expected to be below or as close as possible to expected instrumental detection limits,
– 1 sample at low level (L), corresponding to an incurred pool of individual human urine samples for which the concentration of the targeted markers were expected to be near the p25 of the concentration distribution of a European general population,
– 1 sample at medium level (M), corresponding to a fortified pool of individual human urine samples for which the concentration of the targeted markers were expected to be near the p50 value of a European general population,
– 1 samples at high level (H), corresponding to a fortified pool of individual human urine samples for which the concentration of the targeted markers were expected to be near the p95 value of a European general population.
Homogeneity and stability assessment of the control materials confirmed that the very low, low, medium and high concentration levels samples were adequately homogeneous and stable for BPA, BPS and BPF.
The HBM4EU QAU selected four expert laboratories for the Bisphenols/urine combination.
Selection criteria included:
– Number of years of experience with the biomarker/matrix combination of interest.
– Application of highly sensitive and selective analytical techniques for the analysis.
– Application of isotopically labelled standards for quantification.
– Availability of in-house validation reports, data on on-going intra-laboratory performance, ISO17025 accreditation for the biomarker of interest.
– Success rate in inter-laboratory comparisons, external quality assessment schemes or at least comparative results in application studies
The expert-assigned value is the target value based on analysis results obtained from analysis of the control material by at least 3 expert laboratories (see HBM4EU-SOP-QA-001). The expert assigned values were calculated by averaging the values obtained by the expert labs. When the expert-assigned value couldn’t be used as the assigned value (as described in 5.2), the consensus value calculated from the results submitted by the participants was used to perform the assessment.
Laboratory results were rated using z-scores in accordance with ISO 13528 and ISO 17043. The default standard deviation applied for proficiency assessment (i.e. target standard deviation) was set to FFP = 25 %, as described in 5.3.
Assessments’ scores were calculated for BPA, BPS and BPF for low, medium and high level samples. Only z-scores associated to BPA were calculated for very low sample. As a global overview, the proportion of satisfying results (-2 < Z-score < 2) was from 47% to 94% for BPA, from 61% to 89% for BPS and from 63% to 79% for BPF.
Scores associated to the very low concentration level sample (VL) were not calculated for BPS and BPF. Expert assigned value and consensus value couldn’t be used to calculate z-scores for these compounds (as described in 5.2 and 5.4).