Towards the development of a certified reference material for hemoglobin A2

Background: In 2004, a working group on the standardization of hemoglobin A2 (HbA2) was created within the IFCC, with the aim of developing a reference system for this analyte. One goal was to prepare a certified reference material in collaboration with the Institute for Reference Materials and Measurements (IRMM). This paper describes the properties of a first batch of this candidate study material. Methods: Eighty millilitre of fresh whole blood, collected from a healthy blood donor, was treated by removing plasma, white blood cells and platelets. Red cells were hemolyzed to prepare 100 vials of lyophilized material (approximately 155 mg per vial). After reconstitution, the HbA2 content was measured with a total of seven HPLC methods, three electrophoretic techniques, and two capillary electrophoresis (CE) methods. Homogeneity was tested in a subset of five vials. Stability during storage at +4°C and -20°C was tested monthly over a period of 1 year. The commutability of this material was assessed by analysing the study material together with a set of 54 fresh blood samples, with a subgroup of the above mentioned methods, only by one routine HPLC (Bio-Rad Variant II, dual kit) and by a CE method (Beckman PA800, Analis kit), respectively. Results: The chromatographic and electrophoretic patterns obtained by all the HPLC, electrophoretic and CE techniques did not show any difference between those obtained using the first study material and those obtained with fresh blood samples. The lot was found to be homogeneous on the basis of the content of lyophilized powder per vial. The HbA2 concentration in the lyophilized material remained stable at +4°C and -20°C, even after 1 year of storage. After reconstitution, the HbA2 concentration did not change for more than 2 weeks in the refrigerator at +4°C. The normalized residual of the study material, measuring the degree of its commutability was 0.9, similar to that obtained on other home prepared and some commercial controls. Conclusions: Ideally, fresh whole blood is the best reference material in the meterological traceability chain for HbA2 analysis. However, for a number of reasons the preparation of large batches of fresh whole blood to be used as secondary reference material for HbA2 is not practical. In our work, we have proven that lyophilization does not appear to cause any matrix effect or inhomogeneity in the study material, which also confirmed to be commutable for the Bio-Rad Variant II (dual kit) and Beckman PA800 (Analis kit) methods. We conclude that a material similarly prepared as the current study material and value assigned with the candidate reference measurement procedure still under development will be suitable to calibrate various routine methods for HbA2. This will result in improvement of the inter-method variability for this important biochemical marker.