15-OR

Abstract

Solid phase immunoassays enable us to identify unacceptable donor HLA specific antigens (DSA) which allows for the prediction of compatible donor/recipient combinations. This process is referred to as the virtual crossmatch (vXM). The aim of the study was to evaluate the performance of the vXM against the final flow cytometric crossmatch (FCXM). All FCXM performed in the past three years were included in this study. Flow cytometry antibody screening and Luminex single antigen bead assay (LSA) were performed to determine the presence of HLA antibodies and their specificities. MFI = 1000 was used as cutoff for LSA. A final FCXM either prospective or retrospective for the transplant recipients was performed with potential donors selected upon the vXM. Donor cells used for final crossmatch were treated with pronase. Total 2070 FCMX were recorded and analyzed in the study. Among these crossmatches, 1891 were vXM negative, 768 crossmatches were performed for HLA sensitized recipients with 179 positive vXM due to presence of DSA. There were total 297 positive final FCXM, of them 123 were expected and 174 (9%) were unexpected compared with the vXM results. The overall prediction rate of the final negative FCXM was 91%. The overall unexpected positive final FCXM was 8%. The majority of the unexpected final T cell positive and B cell negative FCXM became negative when the FCXM repeated with non-pronase treated donor cells. Our vXM practice provided a >90% correct predication of negative FCXM. We could expected that the unexpected positive final FCXM mainly caused by non-HLA reactivity, for example, the pronase treatment and/or presence of antibodies not directed to HLA because the current LSA could finely define the specificities of HLA antibodies. However, it could not be excluded that the presence of anti-DP and/or DQA1 antibodies could trigger a positive B cell FCXM in cases where donors’ DP and DQA1 were not typed. This will be further investigated.