ALLOSPECIFIC IFN-??-PRODUCING CELLS IN PERIPHERAL BLOOD AS A FUNCTIONAL MEASURE OF HLA HAPLOTYPE MATCH

Abstract

458 Standard tissue typing methods identify differences in HLA molecules between transplant donors and recipients, with the number of mismatches correlating inversely with the long term outcome of the allograft. It is not clear whether HLA haplotype mismatches correlate with the strength of the alloresponse, however, as the differences may or may not have functional significance. Our laboratory has been interested in identifying a surrogate marker for the strength of the allospecific immune response that would improve upon current matching procedures and thus aid in predicting graft survival. To this end we used a cytokine ELISA spot assay to evaluate the frequency of allospecific, IFNγ-producing cells in the peripheral blood of putative renal transplant recipients and potential living donors. PBLs were obtained from 30 individuals (10 families) at the time of initial crossmatch, and were tested for allospecific IFNγ production using one way MLR ELISA spot assays. The resultant frequencies were correlated with the HLA haplotypes. HLA A and B alleles were defined using microlymphocytotoxicity assays and HLA DR alleles were identified using PCR with sequence specific primers. The PBLs of donor-recipient pairs with identical haplotypes had a significantly lower mean frequency of allospecific IFNγ-producing cells than those with single (3/6 alleles) or full (6/6) haplotype mismatches (p< 0.05 for each), but the number of mismatches did not directly correlate with the number of IFNγ spots detected:Table Many fully mismatched pairs had low frequencies(< 18 spots/300,000 PBLs) of allospecific IFNγ-producing cells, suggesting that the mismatches may not be of functional significance. Furthermore, we often found that the frequency of IFNγ-producing cells was significantly higher in one direction than the other. In one illustrative example, one sibling expressed HLA A2 and a second expressed HLA A30, with the other 5 loci being identical. A2 anti-A30 yielded 68 spots/300,000 PBLs, while A30 anti-A2 yielded only 12. These findings confirm that expression of certain alleles is associated with stronger alloresponses than others despite identical numbers of mismatches. Our studies provide an alternative functional approach for the evaluation of donor-recipient histocompatibility. Long term follow-up of the graft recipients will be required to determine whether the pretransplant frequency of allospecific IFNγ-producing cells provides predictive information regarding graft outcome beyond standard HLA typing.