HLA matching effects.

Abstract

1. Transplants without A,B mismatches that had good early function (grade A at 3 months) had a strikingly lower long-term failure rate (equivalently, longer half-life) than transplants that had poor early function. This was found in HLA-identical sibling donor transplants as well as in cadaver donor grafts. The same superiority was observed in parent donor grafts although the number of zero mismatch grafts in this group was too small to permit a sensitive analysis. In cadaver donor grafts, zero A,B mismatch transplants had a longer half-life than mismatched grafts in both early function groups in first transplants but only in the poorer early function group in regrafts. These results were based on transplants done from 1970 to 1979; there was not enough experience with DR-typed grafts for their inclusion into a similar analysis. This finding does not at present have consequences for recipient selection. The finding would have consequences if an association of early function with matching at another system (e.g., Lewis) or with some assessable non-responsiveness were to be shown. 2. The Excess Match Index (EMI), defined as number of matches less number of mismatches, was shown to be associated with graft survival. It was shown that within groups defined by number of mismatches, survival tended to be greater if the number of matches were more; likewise, within groups defined by number of matches, survival tended to be greater if the number of mismatches was less. The within-match group and within-mismatch group comparisons were based on small numbers of transplants and the corresponding findings are not firmly established. The potential significance of the index, which follows from the finer gradation of the extent of matching, is that it may supply a useful basis for priority schemes for selection of recipients, particularly in donor sharing programs. 3. A recipient selection scheme based on the EMI applied to DR matching was evaluated in a simulation study with the result that most of the gains from DR matching could be obtained with relatively small recipient pools. Nearly all of the gain was achieved by a pool size of 400 (pool size for ABO-identical recipients) and most of the gain could be achieved with smaller pools. The result is encouraging since it indicates that much of the gain from HLA matching can probably be achieved with practically attainable size recipient pools. It was important that the priority scheme not lead to a distribution of phenotypes in the pool that differs substantially from the population.(ABSTRACT TRUNCATED AT 250 WORDS)