From its first experimental origins, hematopoietic cell transplantation (HCT) evolved into a standard procedure used across a broad array of hematologic conditions, including benign and malignant diseases of the blood, aplastic anemia, and hemoglobinopathies, as well as non-hematologic conditions.1 The discovery and characterization of human leukocyte antigens (HLAs) and the crucial role they play in histocompatibility when transplanting tissues from allogeneic donors were key to this major medical achievement. HLA typing became the most critical component of matching hematopoietic cells between donors and recipients to help avoid acute graft rejection and graft-versus-host disease (GVHD) and improve patient outcomes.1
Modern HLA typing techniques deliver much more detailed information than earlier methods, identifying finer levels of mismatch. Paul M. Armistead, MD, PhD, of the Bone Marrow Transplant Program and the Advanced Cellular Therapeutics Facility at the University of North Carolina at Chapel Hill, underscored something of a paradox, explaining, “With more recent [GVHD] prophylaxis regimens, we actually achieve very similar outcomes even with what used to be considered highly significant HLA mismatches.”
ASH Clinical News spoke with Dr. Armistead and other experts in the field about the evolution of HLA typing, how it is used in hematology, and some key considerations for clinicians.
In the 1930s, scientists first began to identify antigens that seemed to be responsible for the rejection of allogeneic tumor transplantation in mice. Eventually, scientists discovered the genetic locus associated with acceptance versus strong rejection, which ultimately became known as the major histocompatibility complex.2 The equivalent genetic locus in humans became known as the HLA system. In 1958, French immunologist Jean Dausset, PhD, identified and characterized the first of these transplantation antigens in humans.
“They found that serum from pregnant women actually agglutinated leukocytes,” explained Medhat Z. Askar, MD, PhD, director of clinical services at Be The Match/National Marrow Donor Program (NMDP) in Minneapolis and associate vice president of Qatar University. However, when exposed to leukocytes from certain specific individuals – later determined to have matching HLA – the sera did not agglutinate.2
Dr. Dausset’s experiments paralleled early work in HCT, which demonstrated both the potential for a newly engrafted immune system to prevent leukemia relapse and the potential for that engrafted immune system to act against the recipient via GVHD. The first allogeneic HCT (alloHCT) performed in unrelated, HLA-unmatched donors in 1957 resulted in graft failure rejection.3
Dr. Dausset recognized the significance of his HLA finding, for which he ultimately shared the Nobel Prize in 1980. In 1958, he wrote, “The study of leukocyte antigens might become of great importance in tissue transplantation, in particular, bone marrow transplantation.”2
Nature of the HLA Region
Located on chromosome 6, the HLA region contains more than 200 genes. It is the most polymorphic genetic system in humans; more than 36,000 HLA and related alleles have been described, and this number continues to grow, especially with the wider use of high-throughput sequencing methods.4,5
The class I HLA region contains the genes HLA-A, HLA-B, and HLA-C. These genes encode the heavy chains that are one component of the class I molecules on the cell surface. These molecules provide an antigen recognition site to express endogenous peptides on their surface as part of immune cell surveillance.6
In contrast, class II HLA molecules are expressed by antigen-presenting leukocytes. These heterodimer molecules contain binding sites used to exhibit pathogenic peptides for T-lymphocyte examination. Important gene regions here include DP, DQ, and DR.
The HLA region also encodes other genes important in the innate response, as well as additional non-classical HLA genes thought to be less important for tissue compatibility.6
HLA genes, because of their chromosomal proximity, are almost always inherited as a group in a Mendelian fashion, with one haplotype of HLA genes inherited from one parent and another haplotype inherited from the other. Full siblings thus have a 25% chance that they will inherit the same haplotype from both parents (complete match), a 50% chance that they will have one haplotype in common (termed haploidentical), and a 25% chance of having no shared haplotypes.6
Not surprisingly, an identically matched related donor, ideal for HCT, is only available 15% to 30% of the time, and haploidentical donors are more easily found.7
Work in solid organ transplantation underscored the importance of HLA matching in preventing rejection. By the mid to late 1960s, scientists had developed methods to identify and type HLA, such that donors and recipients could be matched as part of HCT, and clinical trials for HCT soon began.8
It took time for scientists to characterize the loci that were most important for donor matching. In 2007, Stephanie Lee, MD, and colleagues demonstrated that the HLA-A, -B, -C, and -DR loci seemed to be the most important ones for unrelated donor transplantation for increased survival, with about a 10% worse overall survival in patients with a 7/8 match compared to an 8/8 match.9
Comprehensive literature reviews have demonstrated increased overall survival, decreased GVHD, and reduced transplant-related mortality when recipients who receive HCT match on both their sets of alleles on HLA-A, -B, -C, and -DRB1, sometimes termed an 8/8 match. Joint guidelines from the NMDP and the Center for International Bone and Marrow Transplant Research (CIBMTR) recommend an 8/8 match, although a 7/8 match can be considered if an 8/8 match is not available.10
Eventually it became clear that mismatches at the HLA-DQ locus also seemed to pose risk, although not as significant as those first four, leading some institutions to move to a 10/10 match model.
The impact of the HLA-DP locus has also become more salient; matching on all six of these loci is sometimes termed a 12/12 match. Eric Weimer, PhD, of the University of North Carolina School of Medicine in Chapel Hill, explained that most unrelated donors who match at DR with an 8/8 match are also likely to match at DQ, making them a 10/10 match; however, because of greater physical distance down the chromosome, 80% of unrelated donors who are 8/8 matches will not match at DP.
To help address this, scientists have developed a framework of permissive matching at DP, based on T-cell reactivity against the actual DP protein. Thus, some DP alleles do not technically match but are considered to carry less immunogenic risk (permissive mismatched), and other allelic combinations carry higher risks (non-permissive mismatched).11
Dr. Askar noted that although DQ and DP mismatches should be minimized where possible, consistent with NMDP/CIBMTR guidelines, he suggested framing the matching as 8/8 versus 10/10 or 12/12, because these additional loci influence outcomes less profoundly than HLA-A, -B, -D, and DRB1.
The first methods of HLA typing, based on serology, were universally used for many years. By examining the combination of sera that led to complement engagement and cell lysis, one could determine the HLA type of the lymphocyte donor. However, Dr. Askar pointed out, “With this method you can only distinguish molecules that are different enough to be discovered by antibody specificity.”
Laboratories could use this method to assign a basic, broad HLA typing in several hours, which led to many early discoveries about HLA. However, this method had poor sensitivity for detecting small differences in amino acid sequence in HLA protein, and it could only identify HLA alleles that had previously been characterized.11
Molecular typing techniques based on DNA sequencing have since replaced serologic typing. Intermediate-resolution molecular techniques came first, using DNA-matched sequence specific primers, based on antigens. Not all alleles could be distinguished with this technique alone, though. It was often used in combination with polymerase chain reaction (PCR) after its invention to help resolve ambiguous typing.12 Such intermediate-resolution tests are still sometimes used, such as when initially screening siblings for HLA compatibility.
Dr. Weimer explained that most HLA typing today involves higher resolution allele-based molecular typing techniques, either Sanger-based sequencing, which was first used to sequence the human genome, or next-generation sequencing (NGS). These tests give information about the specific genetic makeup of the alleles. Guidelines recommend high-resolution testing for HCT matching, including at minimum the HLA antigen recognition site portion of relevant HLA genes.10
Sanger-based typing is slower, more expensive, and more labor intensive compared to lower resolution PCR-based techniques, but it provides more precision. However, Dr. Weimer explained that many HLA labs are now planning to move over to NGS, and many have already made this switch.
“Sanger sequencing at our institution used to take three weeks, but when we switched to NGS, our turnaround time became about four days,” Dr. Weimer noted. After initial implementation costs, NGS is also less expensive to perform than Sanger-based methods.
Some scientists have even begun exploring the use of ultra-high resolution HLA testing, which matches not just for coding nucleotides but for non-coding ones as well. The picture on this is evolving, but at least some data suggest that in patients who have more than one 10/10 donor available on high-resolution matches, using such ultra-high resolution HLA testing to further minimize mismatches might further improve outcomes.13 It’s unclear whether this approach will become widespread.
Dr. Weimer also noted that high-resolution methods such as NGS have exponentially increased the amount of data available. “We have far more information than is useful to my hematology colleagues for matching. That means it is more incumbent on the laboratory to properly identify and report the clinically useful information,” he said.
Although the amount of information to consider for matching can be overwhelming, clinicians are not making such decisions in isolation, and they can lean on members of the transplant team with the most expertise in these topics.
“It’s important for clinicians to know that you can talk to your individual local laboratory, have conversations with them about how the lab can support you, and give you what you feel you need to be successful,” Dr. Weimer noted.
Somewhat paradoxically, our knowledge of HLA matches has become more detailed and nuanced while our treatment regimens for GVHD have allowed clinicians to become somewhat less stringent with matching. Dr. Armistead pointed out that 20 years ago, it was much riskier to perform a transplant with an HLA-A, -B, -C, or -DR mismatch. This was before post-transplant cyclophosphamide GVHD prophylaxis regimens were developed.
“That’s not to say that we’ll get rid of HLA typing at any point, but we are not as concerned about HLA-A, -B, -C, and -DR mismatches as we used to be,” Dr. Armistead added.
With these treatment innovations, Dr. Askar noted, some experts believe that a degree of mismatch might even become desired in some patients receiving HCT in a malignancy setting, to heighten the so-called graft-versus-leukemic effect. If born out, Dr. Askar pointed out, less stringent HCT matching criteria would make more unrelated donor matches available to more people, and patients may be better matched using other desirable criteria, such as younger donor age.
The better ability to treat GVHD may also influence the choice of donor when a fully matched related donor is not available. Dr. Armistead noted the practice at his institution, traditionally, has been to choose a fully matched unrelated donor over a haploidentical (4/4) related donor, which is similar to the preference at many centers. However, perhaps because of COVID-19, it has been taking longer to get access to unrelated donor tissue.
“The outcomes aren’t that different,” Dr. Armistead said. “The haploidentical donor is usually a family member, so it’s easier to have them come in and donate stem cells – you’re decreasing time to transplant.”
“Delaying transplant by waiting for the ideal hypothetical donor is a bad trade-off, because even if you get it, the patient’s [condition] has deteriorated,” Dr. Askar noted.
Dr. Askar is collaborating with colleagues from NMDP as well as other experts to revise the existing NMDP/CIBMTR guidelines to provide more up-to-date guidance on choosing the most favorable donors. They also will provide updated guidance around emerging findings from studies, such as matching the B leader sequence of the HLA-B gene, an area outside the antigen recognition site region commonly used for HLA-B matching, which may also affect clinical outcomes.14
Scientists have explored many other potential areas for matching, such as major histocompatibility complex class I polypeptide-related sequence A and natural killer cell immunoglobulin-like receptors. Results have been mixed, though, and it’s not yet clear how this information might best be incorporated into matching practices.11
However, Dr. Weimer pointed out, “The issue is always that the more we constrict and say you have to match here and match there, the smaller the donor pool becomes. It becomes rate limiting, because you limit opportunities, particularly for underrepresented minorities with rarer HLA types in Be The Match.”
“Twenty years ago, safely performing an [alloHCT] on a Black American who did not have an HLA matched unrelated donor was a significant challenge,” Dr. Armistead said. “Now, you can take racial and ethnic minorities who are not as well represented in the unrelated registry and proceed to transplant and expect your outcomes to be very similar. That’s obviously a very good thing.”
- Fürst D, Neuchel C, Tsamadou C, et al. HLA matching in unrelated stem cell transplantation up to date. Transfus Med Hemother. 2019;46(5):326-336.
- Thorsby E. A short history of HLA. Tissue Antigens. 2009;74(2):101-116.
- Thomas ED, Lochte HL, Lu WC, et al. Intravenous infusion of bone marrow in patients receiving radiation and chemotherapy. N Engl J Med. 1957;257(11):491-496.
- Hurley CK, Kempenich J, Wadsworth K, et al. Common, intermediate and well-documented HLA alleles in world populations: CIWD version 3.0.0. HLA. 2020;95(6):516-531.
- HLA alleles. Dec 1, 2022. Accessed March 28, 2023. https://hla.alleles.org/alleles/index.html.
- Choo SY. The HLA system: genetics, immunology, clinical testing, and clinical implications. Yonsei Med J. 2007;48(1):11-23.
- Gragert L, Eapen M, Williams E, et al. HLA match likelihoods for hematopoietic stem-cell grafts in the U.S. registry. N Engl J Med. 2014;371(4):339-348.
- Henig I, Zuckerman T. Hematopoietic stem cell transplantation-50 years of evolution and future perspectives. Rambam Maimonides Med J. 2014;5(4):e0028.
- Lee SJ, Klein J, Haagenson M, et al. High-resolution donor-recipient HLA matching contributes to the success of unrelated donor marrow transplantation. Blood. 2007;110(13):4576-4583.
- Dehn J, Spellman S, Hurley CK, et al. Selection of unrelated donors and cord blood units for hematopoietic cell transplantation: guidelines from the NMDP/CIBMTR. Blood. 2019;134(12):924-934.
- Edgerly CH, Weimer ET. The past, present, and future of HLA typing in transplantation. Methods Mol Biol. 2018;1802:1-10.
- Cornaby C, Weimer ET. HLA Typing by next-generation sequencing: lessons learned and future applications. Clin Lab Med. 2022;42(4):603-612.
- Mayor NP, Wang T, Lee SJ, et al. Impact of previously unrecognized HLA mismatches using ultrahigh resolution typing in unrelated donor hematopoietic cell transplantation. J Clin Oncol. 2021;39(21):2397-2409.
- Sajulga R, Bolon YT, Maiers MJ, et al. Assessment of HLA-B genetic variation with an HLA-B leader tool and implications in clinical transplantation. Blood Adv. 2022;6(1):270-280.