|
Abstract INTRODUCTION:
Certified reference materials (CRMs) are crucial for ensuring method validation and underpinning the reliability of clinical measurements. Their analysis enables a laboratory’s methods to be standardised and their performance to be benchmarked against other measurement providers. However, in many clinical areas, commercially available matrix CRMs are limited. This is the case for the monitoring of phenylketonuria (PKU), an inherited metabolic disease that causes raised levels of phenylalanine in blood. If left untreated, this condition can lead to severe health repercussions such as irreversible brain damage. PKU patients manage their condition by severely restricting dietary phenylalanine intake, which is monitored against target decision limits through regular phenylalanine blood concentration measurements. However, the lack of a matrix-matched amino acid CRM with SI traceable assigned values, impedes the reliable comparability of hospital results obtained over a period of time. Additionally, the quality of this service is known to be inconsistent between providers, with between laboratory variability estimated to be greater than 35%, highlighting the need for measurement harmonisation and standardisation.
METHODS:
The National Measurement Laboratory (NML), hosted by LGC, collaborated with Synnovis at Guy’s and St Thomas’ Hospital to produce four traceable value assigned whole blood materials, to be used in a three-part inter-laboratory study in the absence of suitable CRMs. The aim of the study was to improve quality and comparability of phenylalanine measurements for PKU monitoring across the UK by comparison of results to traceable value assigned materials, with proven stability for the course of the project. The first part of the study assessed the variability in measurements when laboratories (n = 15) used their individual routine clinical methods. In the second part, the effect of using a harmonised method on the main source of variance in the results, as between laboratory variability (n = 5), was investigated. Furthermore, three routine calibration approaches were applied to the data and compared. Finally, the accuracy of the harmonised method and most promising calibration approach was assessed against traceable materials value assigned by the NML.
The amount of phenylalanine in each of the materials was assigned fully traceable to the International System of Units (SI) with an expanded measurement uncertainty below 2%. In order to value assign the materials for phenylalanine mass fraction with the required low associated uncertainty, a double exact matching isotope dilution gas chromatography tandem mass spectrometry (DEM-ID-GC-MS/MS) methodology was used. The use of GC-MS/MS was a key factor in achieving a lower total measurement uncertainty than typically achievable, which increased the reliability and strength of conclusions drawn through evaluation against the materials. GC-MS/MS provided enhanced analyte selectivity in comparison with other mass spectrometry systems, which decreased chromatographic interferences in the natural and labelled channels. This allowed for excellent peak area ratio precision, which is the main contributor to the total measurement uncertainty associated with traceable value assignment.
RESULTS:
The main source of variability using routine methods was the between-laboratory variation. All the laboratories used a similar analytical method, however the specifics of these varied considerably between participants. One of the key parameters that differed was how measurements were calibrated, with three general calibration approaches typically being used. For this reason, the second study aimed to assess how these three approaches affected the measurement variability when a common measurement approach was used. This demonstrated that the agreement between results improved by at least 85% when a harmonised method, including isotopically labelled internal standards, was implemented in combination with external calibration. Therefore, in the final study, this calibration strategy was used alongside the common analytical method for measuring the materials produced by the NML. This final evaluation demonstrated the accuracy of measurements, through comparison with a traceable value assigned matrix material, and showed agreement between every participant’s result and the assigned value at the 95% confidence interval, k = 2.
CONCLUSION:
A harmonised analysis method for PKU monitoring measurements was developed and evaluated using a standardised calibration approach. This significantly improved agreement between laboratories when compared to the varied methods which are routinely used. Supplying traceable value assigned whole blood materials as part of an inter-laboratory study underpinned PKU measurements to a true value. This allowed laboratories to conclusively determine the accuracy of the implemented method and benchmark their performance against other measurement providers, without the additional complexity required for CRM production.
REFERENCES:
1. Carling, R., et al., Investigation of the relationship between phenylalanine in venous plasma and capillary blood using volumetric blood collection devices. Clin Chem Lab Med, 2023. 64(6): p. 468-476.
|
= Discovery stage. (53.14%, 2025)
= Translation stage. (22.33%, 2025)
= Clinically available. (24.53%, 2025)
