Multilaboratory Evaluation of Serum Analysis for HLA Antibody and Crossmatch Reactivity by Lymphocytotoxicity Methods

Abstract

Abstract Context.—This report presents results of the serum antibody analysis and crossmatch challenges in the proficiency testing program for histocompatibility testing jointly sponsored by the American Society for Histocompatibility and Immunogenetics and the College of American Pathologists. Objective.—To obtain information about consensus rates among participating laboratories that reported antibody screening and crossmatch results by direct complement-dependent lymphocytotoxicity (CDC) and/or anti-human globulin (AHG)–augmentation methods. Design.—We analyzed responses from approximately 165 laboratories participating in 32 surveys during 1993–2000. Most of the testing was done by CDC methods, but increasing proportions of laboratories are using AHG augmentation of these techniques; almost one half of the serum screenings and crossmatches were done by AHG. Results.—A total of 40 serum specimens were screened to determine the percent panel-reactive antibody (PRA) and identify HLA-specific antibodies. Participants often reported very wide ranges of PRA values. Panel-reactive antibody ranges exceeded 60 percentage points for 16 (40%) of the serum screening results by CDC and for 31 (77%) of the results by AHG. The interlaboratory variability of PRA values suggests that in many laboratories, the CDC or AHG procedures were often too insensitive or overly sensitive. The antibody identification results revealed inconsistent patterns among the participants performing CDC or AHG screening. Most participants reported the same primary antibody specificities by both methods. The consensus levels were generally high for the monospecific sera. On the other hand, there was much less agreement among the participants if the sera reacted with 2 or more HLA antigens. Participants using the more sensitive AHG method reported additional antibody specificities in many specimens, but invariably the consensus levels were rather low. A total of 192 serum-cell combinations were used for the crossmatch challenges. There was considerable interlaboratory variability; 21% of the CDC crossmatches and 36% of AHG crossmatches failed to reach the 90% consensus threshold. Conclusions.—This experience demonstrates considerable inconsistencies in serum screening and crossmatching among laboratories participating in the American Society for Histocompatibility and Immunogenetics/College of American Pathologists surveys. A lack of uniformity in test results may limit the efficient application of these methods in a clinical setting. Standardization of crossmatch and antibody screening techniques is highly desirable.