Evaluation of commutability of the ERM - DA 470k reference material for two albumin assays using immunochemical principles

Background. To standardize the measurement of human serum proteins, a new reference material (RM)(IRMM ERM-DA 470k/
IFCC) has recently been released. We evaluatedthe commutability of this RM for two immunochemical albumin methods based
on turbidimetry (Roche Cobas Tina-quant Gen. 2) and nephelometry (Delta Radim) principles.
Methods. We measured albumin concentrations on the two systems in20 human serum samples (albumin concentrations, 16.3
to 42.3 g/L) and in the RM in duplicate in two different runs on the same day. Biological samples were collected, aliquoted and
stored(£24 h) at +4°C until their use. Manufacturer’s control materials were used to validate analytical runs. The RM commutability
was estimated from data of regression analysis using the 95% prediction interval (95PI) and multiples of the standard error of
regression (Sy-x), according to the CLSI C53-A document.
Results. The RM results were contained within the 95PIbasedon the results for the native clinical samples. In addition, using an
acceptance criterion for commutability of ±2 times the experimental Sy-x (±3.135), the relative residual for the RM (-2.107) was
within the acceptable range. Albumin values for RM on both analytical systems (34.6 g/L for turbidimetry and 34.2 g/L for nephelometry)
were, however, markedly lower than expected (certified value±uncertainty, 37.2±1.2 g/L).
Conclusions. The new RM is commutable between the evaluated immunochemical assays and can be used as a basis to maintain
their traceability to higher-order references. However, some inconsistency in the assignment of values to the manufacturer’s
working calibrators is likely and should be verified.