Analysis of Comparability Test between Reagent Lots in Immunoassay

ABSTRACT

Lot-to-lot reproducibility is an important issue in immunoassays and reagent lot-to-lot comparability test comparing the results of new reagent lot with those of used lot using patients' samples or controls is usually performed to detect the difference between lots. However, there are no universally used acceptability criteria regarding reagent lot-to-lot comparability test. We performed reagent comparability test between different lots of alpha-fetoprotein (AFP) reagents and tried to determine its comparability by several criteria. Both the commercialized controls and in-house controls for AFP made from pooled patients' sera were measured 10 times using the old and new lots of reagents, whenever a reagent lot was changed. The differences in the mean control values, the percent difference (% difference), and the difference to between-run standard deviation ratio (D:SD ratio) between successive lots were calculated. We compared the results of reagent comparability test to arbitrarily determined acceptability criteria suggested by CLSI C54-A. Although comparability between reagent lots was determined according to how strictly we set the criteria, some lot-to-lot differences between certain pairs of lots exceeded the criteria. We hope that the results of this study might be helpful to perform reagent lot-to-lot comparability test and set the criteria for reagent comparability test between lots in other laboratories.