1.1000
#6
RECOGNITION OF HLA ANTIBODIES USING A FLOW CLASS I BEAD ASSAY.
R. Vorhaben, CHS, J. Crumpton, MT, B. Lavingia, CHS and P. Stastny, MD. Dallas TX, The University of Texas Southwestern Medical Center, Internal Medicine.

In highly sensitized patients defining anti-HLA specificities by serum screening is often problematic because conventional chi square analysis fails. Dilution and absorption studies are labor intensive and only define high titer antibodies. To identify specificities from results of the flow-PRA assay, we have performed a cluster analysis of HLA antigens after sorting on fluorescent intensity (Fi). Daily instrument variability was removed by converting to Equivalent Soluble Fluorescent Molecules (ESFM) and readings were corrected for variable antigen density (AD) by adjusting values to 75th percentile of maximal antigen Fi with W6/32, a monoclonal antibody against HLA class I heavy chains. Results from 30 normal human sera were used to determine normal binding on each bead. Antibody specificity was determined by sorting on ESFM followed by antigen frequency (Af) calculation of beads within 15% of peak Fi. A chi square analysis was performed on the antigen with the highest Af. When a significant chi-square was obtained, a tail analysis was performed, followed by Af determination and chi square on the next set of beads. Results were compared on 15 exchange specimens and yielded 100% concordance with consensus. Correlation was excellent in 10 sera with PRA of 56-95% , tested with class I single-antigen beads. However, the assay with single-antigen beads identified more specificities (214 antigens vs 118), suggesting that assay is more sensitive. Including both antigens and Fi in a cluster analysis allowed identification of both high and low titer antibodies that previously could not be recognized by panel screening.