VIRTUAL CROSSMATCH AND CLINICAL RELEVANCE OF HLA ANTIBODIES DETECTED BY LUMINEX-SINGLE ANTIGEN BEADS.

makes hard to enable a "virtual crossmatch" (v-XM) for prediction of XM results in donor selection. Methods: We retrospectively analyzed correlation between v-XMs and actual crossmatches (a-XMs) performed within 2007 and 2011. In carrying out v-XM, the donor HLA molecules against which patients showed LSA-DSA with MFI5000 were considered "unacceptable DSA"; all the LSA-DSA showing MFI5000 were defined "acceptable DSA". As donors were typed for HLA-A, B, DR and DQB1, we didn't consider anti HLA-C and -DP DSA. On the basis of a negative v-XM, we performed 284 a-XMs between 177 sensitized renal transplant recipients and 197 donors using both cytotoxic and flow-cytometric methods (CDCXM and FCXM). In some cases we analyzed the cytotoxic activity of DSA by Luminex-C1q Screen. Results: Virtual-XM results showed good correlation with both CDC and FC a-XM (96% and 90% respectively). Eleven CDC/FC a-XMs gave positive results; in these XMs, sera contained "acceptable" DSA. Sixteen a-XMs gave positive results only by FCXM. In all but 2 XMs, sera contained "acceptable" DSA; the remaining 2 sera contained "unacceptable" anti-DP DSA. But 50 a-XMs were negative by both methods, even if DSA were present. In 47 a-XMs, "acceptable" DSA were present. Anti-DPB1 (MFI 11200) or anti-DQA1 (MFI 9700) DSA were only found in 3 cases. Interestingly anti-DQA1 DSA showed high cytotoxic activity by Luminex-C1q Screen. Twenty-seven DSA positive/FCXM negative patients were transplanted. Antibody mediated rejection was observed in 2 patients with "unaceptable" anti-DP DSA. Conclusions: Our v-XM protocol showed high sensitivity in predicting donor-recipient immunologic compatibility. Prediction based on LSA assay feels the amount of antigen on beads and may not be representative of the amount found on cells. Our findinfs highlight the importance to discard weak and likely irrelevant DSA in performing a clinically useful v-XM.