Automated immunoassay methods for ferritin: recovery studies to assess traceability to an international standard

Background: Ferritin standardisation is problematical due to the heterogeneity of ferritin isoforms and the antibodies used in its immunoassay, and the lack of a reference measurement procedure. We investigated the performance of the 1st (liver), 2nd (spleen) and 3rd (recombinant) International Standards (ISs) for ferritin in major assays. Methods: The ferritin in a serum pool 'spiked' with either the 2nd or 3rd IS for ferritin was measured by 52 laboratories using five automated methods and the recovery of the target values calculated. A smaller serum pool was 'spiked' with the 1st IS for a limited recovery exercise. The ferritin values of five serum samples were also measured and recalculated relative to the ISs. Results: Recoveries of each of the 2nd and 3rd ISs were 90%-110% for four of five methods; recoveries of the 1st IS were 104% and 111% for two of three methods claiming traceability to this IS. One method significantly over-recovered each of the IS (124%-155%). Recalculating the ferritin values of the serum samples relative to the IS reduced the overall inter-method agreement, largely because of the anomalous over-recovery of the IS by one method. Conclusions: The use of the 3rd IS to standardise assays will minimise assay drift due to manufacturers adopting a 'harmonisation' approach in which the calibration is adjusted to conform to overall mean values. Standardisation against the current IS also ensures compliance with the European Union In-Vitro Diagnostic Directive which requires traceability of assay calibrators to reference materials of a higher order. Assay drift may result in poor sensitivity and specificity in the diagnosis of iron status, and would require laboratories to continually re-evaluate reference intervals.