50-P

Abstract

Aim The “virtual crossmatch” is advanced as a predictor of HLA reactivity/compatibility between the recipient and the potential donor. HLA laboratories and transplant centers rely on this practice. Antibody detection however is limited by the makeup of the alleles in the test panel and the genetic makeup of the donor pool is endless. A laboratory with limited resources, specimen access or time may need to make an immediate “virtual” call about a donor/recipient pairing with low resolution molecular HLA typing paired with allelic level HLA Antibody information. Knowing that low resolution molecular HLA typing kits are geared towards identifying the most common alleles for each locus, we wanted to survey the actual frequency that these low resolution SSO kits identified the high resolution typing of these loci with a “single pass”. Methods We surveyed the last 7 ASHI high resolutions PT challenges. We compared our initial low resolution typing (and most likely implied high resolution typing) with the graded high resolution typing returned for each of these surveys. We then determined the percentage of matched results between the predicted high resolution typing with the actual high resolution typing at each of the following HLA loci: A, B, C, DRB1, DRB3,4,5 and DQB1. Results Percentage matched, Predicted High Resolution Vs. Actual High resolution typing A locus: 96% matched B locus: 96% matched Cw locus: 91% matched DRB1 locus: 81% matched DRB3, 4, 5 locus: 48% matched DQB1 locus: 89% matched. Conclusions Our survey has resolved that HLA typing based on low resolution SSO kits results in 90% matched HLA typing as ascertained by high resolution HLA-A, B and Cw typing; our study also demonstrates that HLA class II typing by low resolution, especially DRB3,4,5 typing, leaves much to be desired. The advantages and disadvantages unmasked by our study may be of value in informing clinical decisions and associated risks and benefits.