Twenty-two sera samples were tested in parallel with Gen-Probe (GP; Lot 03022V) and One Lambda (OL; Lot 9) single antigen MHC Class II reagents. GP assay requires addition of 40 ¦Ìl bead mix + 10 ¦Ìl sera, and the OL assay requires 5 ¦Ìl bead mix + 20 ¦Ìl sera, while the absolute number of beads is similar between assays. Murine monoclonal antibodies against HLA-DR, -DQ, and -DP with a concentration capable of saturating reagents were used to compare antigen density of GP and OL reagents. GP Background Corrected Mean (BCM) value and OL Normalized value in mean fluorescence intensity (MFI) from antigen density experiments were analyzed by paired student¡¯s t-test. Correlation analysis (Pearson r) was performed on GP BCM and OL Normalized values between beads with equivalent molecular antigens (50 total, 27 HLA-DR, 13 HLA-DQ, and 10 HLA-DP) as well as on titer analysis (neat, 1:16, 1:256, and 1:1024) of 5 sera samples.
Antigen density was significantly greater on GP reagents versus OL reagents with mean strength of saturating murine monoclonal antibodies being 8,080 ¡À 2,257 MFI versus 2,150 ¡À 516 MFI for HLA-DR, 9,955 ¡À 2,866 MFI versus 5,175 ¡À 1,256 MFI for HLA- DQ, and 11,084 ¡À 1,615 MFI versus 3,291 ¡À 553 MMF for HLA-DP (p < 0.001 for all comparisons). Correlation for all specificities resulted in an R square value of 0.7547 (Fig. 1). Observed correlation was greatest between HLA-DR beads (R square = 0.8898) and lowest between HLA-DQ beads (R square = 0.4839). Correlation of results improved as sera titers were performed with R square values of 0.6916, 0.7184, 0.8633, and 0.8090 observed in neat, 1:16, 1:256, and 1:1024 titers respectively. In these samples, a prozone effect or saturation of reagents was observed predominantly in OL HLA-DQ specificities but was not observed with GP reagents.
Both the significantly higher density of HLA antigens on the GP beads and a dilution effect introduced by the difference in serum to bead-volume ratio may contribute to the lower correlation observed in this study. This is supported in part by the higher prozone effect observed with OL HLA-DQ specificities but not in GP, and the improved correlation observed with titers. Further investigation is warranted to determine correlation between GP and OL reagents.