45-P

Abstract

Aim A negative flow crossmatch (FXM) represents the gold standard for successful kidney transplantation, since this highly sensitive test detects low levels of donor specific antibody (DSA). Positive correlation between FXM and luminex antibody assays is vital for survival of kidney transplant recipients and their grafts. We identified patients eliminated from the UNOS kidney waiting list due to a positive virtual crossmatch, (i.e. DSA levels EXPECTED to yield a positive FXM), yet who actually had a negative FXM result. This enabled our laboratory to develop an improved algorithm which more accurately selects patients’ unacceptable antigens (UA) in UNET. Methods FXMs were performed using sera from 87 potential kidney recipients against 51 different pronase treated (15 min/37 °C) donor cells. This sera was also analyzed for HLA class I and II DSA using luminex single antigen assay (One Lambda). We focused on patient sera containing only one DSA to donor cells tested in the FXM. A correlation between results of the luminex assays (MFI = median fluorescence intensity) and FXM was then determined. Results The average luminex MFI levels for sera exhibiting positive virtual crossmatches and resulting in positive actual FXM was 4,638 ± 379 for class I (n = 89) and 6,684 ± 776 for class II (n = 44) DSA. However, in sera containing DSA expected to produce positive virtual crossmatches but which resulted in negative FXM, the average MFI level was 2,609 ± 142 (n = 60) and 4,824 ± 656 (n = 41) for class I (p Conclusions To reduce false positive predictive results (i.e. virtual crossmatches), luminex MFI cutoffs to select patients’ UA in UNET were increased based on results of actual FXMs. Continual monitoring of correlation between FXM and luminex antibody testing should be performed.