Reappraisal of HLA antibody analysis and crossmatching in kidney transplantation.

It has been established that preformed IgG antibodies specific for donor HLA antigens may accelerate graft failure. An increasing number of studies have demonstrated adverse graft survival in patients who have anti-HLA antibodies, whether preformed or developed post-transplant. More recently, ELISA and flow cytometric techniques were introduced to overcome the limited sensitivity and specificity of the CDC assay. These emerging approaches can be reliably used to predict crossmatches in highly sensitized patients and also to monitor the development of clinically relevant anti-HLA antibody after transplantation. This retrospective study used LAT-M screening and Luminex HLA class I and II specificity assay to re-examine: (a), the impact of pre-transplant HLA antibody on long term graft survival; (b), the accuracy with which detection of HLA antibody and specificity by ELISA predicts pretransplant CDC crossmatch; (c), a comparison of Luminex and ELISA methods in detecting HLA antibodies. In this study, pre-transplant sera from 288 renal patients followed up at NCKUH were tested by the ELISA method, LAT-M. The tests showed that 19% had HLA antibodies before transplantation. Among the 234 of the patients who did not have pre-transplant antibodies, 85% enjoyed 5-year functional graft survival, 76% 10-year, and 56% 15-year functional graft survival. The corresponding functional graft survival for the 54 patients who tested HLA antibody-positive was 65%, 53% and 28% (P=0.0021). Sera from 481 patients awaiting kidney transplantation at NCKUH were tested by the ELISA method LAT-M and by CDC to determine how well HLA antibodies detected by ELISA predict the crossmatches shown by CDC. HLA antibodies found by ELISA ranged from 24% weak reactivity (OD "2") to 17% strongly reactive (OD "8"). The positive predictive value (PPV) of ELISA-detected antibodies for positive CDC crossmatch at the time of transplant was found to be 43-54%. The negative predictive value (NPV)-ELISA found no antibodies, CDC no crossmatches- was 88%. The PPV was 55% for sera with HLA class I DSA and 67% with HLA class II sera. On the other hand, NPV was 84% with sera negative for HLA class I DSA and 86% with sera negative for HLA class II DSA. Pretransplant sera from 48 patients with followup data at NCKUH were retested by both ELISA LAT-M and Luminex in order to compare the efficacy of those two methods. ELISA found pre-transplant HLA antibodies in 8 of the 48 (17%). Luminex found HLA antibodies in 27 (56%). Functional graft survival at 5, 10 and 15 years was not significantly different between the 27 patients whom Luminex identified as having pre-transplant HLA antibodies and the 21 patients Luminex found to be free of those antibodies (P=0.7197). For patients shown by Luminex to have pre-transplant class II DSA (N=8), functional graft survival was significantly lower than for those Luminex showed negative for HLA antibodies (P=0.0036). The concept of virtual XM relies on accurate HLA typing and thorough evaluation of HLA antibodies by solid-phase assays. While a negative virtual XM proved to be very reliable to rule out the presence of donor-specific HLA antibodies, it becomes more a concern whether all HLA antibodies detected by flow-beads are in fact clinically relevant. The virtual XM approach-in which antibodies are characterized by solid-phase assays prior to crossmatching-was reported to predict a negative flow XM in greater than 90% of cases. The predictive value for a correct CDC XM, however, was only 75%. A potential disadvantage of the virtual XM approach is that transplants may be excluded based on antibody results with unknown clinical relevance. Based on our results, we believe HLA antibody identification using ELISA still has a role in predicting long term graft survival and negative predictive value for CDC crossmatch before transplantation. Further analysis of HLA antibody, using Luminex, will be done to compare with present data.