Forty-five years have passed since the seminal studies of Patel and Terasaki left no doubt that pretransplant donor directed antibodies (subsequently shown to be antibodies to human leukocyte antigens [HLA] were associated with hyperacute rejection of kidney allografts (1). Based on those observations, it is no surprise that pretransplant detection of such antibodies became (and still is) a major focal point of histocompatibility testing. Clearly, our early concepts of HLA antibodies and antigens were understandably naïve because, understandably, we were naïve. There was serendipity in how antibodies (and antigens) were described. For example, HLA-A9 became the ninth defined HLA antigen when antisera were identified which reacted with cells not expressing any of the eight other HLA antigens. Subsequently, antisera were identified which only reacted with mutually exclusive subsets of HLA-A9–positive cells. This resulted in naming two new HLA antigens (HLA-A23 and HLA-A24) which were considered to be “splits” of the “parent” HLA-A9 antigen (2). Conversely, HLA-B7, HLA-B27, and HLA-B42 were considered to be distinct antigens when antisera reacting with cells expressing any of those specificities were discovered. In this situation, HLA-B7, HLA-B27, HLA-B42 (and others) were considered to belong to a cross-reactive group of antigens (3). When reviewed with a retrospective eye, whether the target of an antibody was considered a “parent” or a cross-reactive antigen was a matter of timing and chance. Today, both would simply be referred to as an epitope, one that could be mapped by a first year graduate student with access to readily available HLA sequence data.
The pioneers of HLA testing, however, began with variables on both sides of the equation. Neither HLA antibodies nor HLA antigens were well characterized and one was often needed to define the other. Furthermore, even under the best of conditions, testing was limited to relatively insensitive and frequently erratic cell-based agglutination and cytotoxic assays. Investigators compensated for these technical shortcomings with strong pattern recognition skills and creative scientific approaches to resolve the vagaries and ambiguities that were HLA antigen or antibody reactions. For example, after recognizing that HLA-B7, HLA-B27, and HLA-B42 were cross-reactive, absorption, and elution studies revealed that cells expressing any one of those antigens could remove and restore, respectively, reactivity against the other specificities. By immunoprecipitation and polyacrylamide gel electrophoresis, the cross-reactivity was shown to be caused by a single antibody that recognized a shared determinant on each antigen (4). Although such serological experiments were tedious and cumbersome, they were quite informative.
In the current era, new tools and assays are available that, in most circumstances, simplify the analysis of HLA antibodies, even among those patients who are highly sensitized. Fluorescence-based, solid phase antibody detection systems (e.g., Luminex; Luminex Corporation, Austin, TX) are more sensitive, specific and efficient than the cellular assays that had been the gold standard for detection and identification of HLA antibodies (5, 6). Multiplexing technology permits up to 100 different fluorescently (red and infrared) labeled microparticles, each coated with a single HLA allele, to be simultaneously interrogated with a fluorescent reporter (phycoertythrin) after incubation with patient sera. Sophisticated software analyzes the data and generates a histogram displaying each bead, typically ordered from high to low based on its mean fluorescence intensity. When mean fluorescence intensity values exceed preestablished threshold levels, the HLA antigen or allele expressed on the corresponding bead is generally considered to be an unacceptable donor antigen for that recipient. Next, the patient’s own HLA antigens and any HLA antigens deemed acceptable (i.e., negative bead activity with patient serum) can be entered into HLAMatchmaker, a sophisticated matching program that enumerates epitope (actually eplet) mismatches between the patient and nonself HLA antigens (7). Recent retrospective studies by Wiebe et al. (8) suggest that HLA epitope matching is less likely to lead to de novo production of class II donor-specific antibodies than conventional HLA antigen matching, and may therefore be a better approach to achieve long term graft survival.
Collectively, the advances referred to above revolutionized the practice of histocompatibility testing. Virtual crossmatching, acceptable mismatching, and paired donor exchange programs, once only concepts, are now the norm. The future holds promise for further advances. Nonetheless, prospective lymphocyte crossmatches continue to be performed between highly sensitized recipients and donors whose virtual crossmatches are negative. Transplant programs still prefer a physical crossmatch to confirm the absence of antibodies before proceeding to surgery, a subtle acknowledgment that computerized matching algorithms are not infallible.
In this issue of Transplantation, Lutz and his colleagues (9) describe and characterize an HLA-Bw4 antibody produced by an HLA-Bw4 positive patient. Interestingly, HLA-Bw4, one of the first HLA antigens to ever be described (10), is considered to be a “public” epitope, that is, a sequence found within multiple HLA-A and HLA-B locus antigens. Previous studies have reported “variants” of HLA-Bw4, which is simply another way to describe Bw4 alleles (11). In essence, and without referring to it as such, Lutz et al. identified an allele-specific antibody to HLA-Bw4. However, based on the collective data of Luminex single antigen bead testing and HLA Matchmaker, the production of this Bw4 antibody should not have been possible. Clearly, the HLA complex is still laden with secrets not predictable even with state-of-the-art hardware and software. The human elements of pattern recognition, serum adsorption, and perseverance solved a problem that computers could not. This study by Lutz et al. reminds us that we still have much to learn from the past.
1. Patel R, Terasaki PI. Significance of the positive crossmatch test in kidney transplantation. N Engl J Med
1969; 280: 735.
2. Dausset J. The genetics of transplantation antigens. Transplant Proc
1971; 3: 8.
3. Joysey VC, Wolf E. HLA-A, -B, and -C antigens, their serology and cross-reaction. Br Med Bull
1978; 3: 217.
4. Schwartz BD, Luehrman LK, Rodey GE. Public antigenic determinant on a family of HLA-B molecules. J Clin Invest
1979; 64: 938.
5. Fulton RJ, McDade RL, Smith PL, et al. Advanced multiplexed analysis with the FlowMetrix system. Clin Chem
1997; 43: 1749.
6. Gebel HM, Liwski RS, Bray RA. Technical aspects of HLA antibody testing. Curr Opin Organ Transplant
2013; 18: 455.
7. Duquesnoy RJ. Antibody-reactive epitope determination with HLAMatchmaker and its application in transplantation. Tissue Antigens
2011; 77: 525.
8. Wiebe C, Pochinco D, Blydt-Hansen TD, et al. Class II HLA epitope matching—a strategy to minimize de novo donor-specific antibody development and improve outcomes. Am J Transplant
2013; 13: 3114.
9. Lutz CT, Al-Attar A, May JR, Jennings CD. Alloantibody to a Bw4 epitope in a Bw4+ B*27:05 patient. Transplantation
2014; 98: 853.
10. van Rood JJ, van Leeuwen A. Leukocyte grouping. A method and its application. J Clin Invest
11. Muller CA, Engler-Blum G, Gekeler V, et al. Genetic and serological heterogeneity of the supertypic HLA-B locus specificities Bw4 and Bw6. Immunogenetics
1989; 30: 200.