The human leukocyte antigen (HLA) system, located on the short arm of chromosome 6 (6p21.3), is divided into three regions: class I, class II, and class III. The three primary class I genes, HLA-A, HLA-B, and HLA-C, encode the α chain of the corresponding major histocompatibility complex class I proteins, HLA-A, -B, and -C. These class I molecules present endogenous peptides to CD8+ T cells, while class II molecules, coded by HLA-DR, -DP, and -DQ, present exogenous peptides to CD4+ T cells. HLA is the most polymorphic genetic system in humans and plays a vital role in distinguishing self from nonself [1]. This polymorphism significantly affects graft rejection, graft-versus-host disease (GvHD), and the overall outcomes in patients. Advances in HLA typing technology have made hematopoietic stem cell transplantation a standard treatment for many hematologic diseases and have greatly improved the outcomes of solid organ transplants [2,3]. In kidney transplantation, specifically, the donor-specific anti-HLA antibodies (DSA) negative group and the extended second field high-resolution negative and low-resolution positive DSA groups demonstrated comparable 10-year graft survival rates (82.4% and 93.8%; P=0.27). This indicates that high-resolution HLA typing is crucial for accurately assessing DSA in kidney transplants [4].

HLA typing technology began with serologic typing and subsequently advanced to polymerase chain reaction (PCR)-based methods. Currently, next-generation sequencing (NGS) is widely utilized, as illustrated in Fig. 1. Modern HLA typing technologies offer more precise and comprehensive information than those available previously. In this review, we explore the developmental history of HLA typing technology and introduce the latest advancements, including third-generation sequencing (TGS). TGS, offered by Pacific Biosciences and Oxford Nanopore Technologies, features long-read, single-molecule, real-time sequencing. This technology enables the sequencing of entire HLA genes without the need to assemble short-read sequences. Additionally, we discuss the impact of HLA typing on transplantation outcomes.

Figure 1. Timeline of technological advancements in human leukocyte antigen (HLA) typing methods. This timeline shows the introduction of developing HLA typing methods from serological assays from the 1960s to third-generation sequencing technology. PCR, polymerase chain reaction.

Serologic Typing

The first HLA typing technology began with serological methods in the 1950s. Jean Dausset first described HLA in 1958, noting the leukocyte agglutination reaction patterns in the serum of repeatedly transfused patients. The leukocyte antigen identified in this study was named 'MAC' in honor of the volunteers who participated in the experiment [5]. Additionally, the presence of leukocyte antibodies in the sera of multiparous women was confirmed, further supporting the existence of leukocyte antigens [6]. Since the complement-dependent cytotoxicity (CDC) microassay was introduced by Terasaki and McClelland in the early 1960s [7], it has been the primary method of HLA typing for over 25 years. Other approaches, such as flow cytometry [8], immunoprecipitation/isoelectric focusing techniques [9] or cellular methods such as mixed lymphocyte responses [10] have also been introduced for histocompatibility testing; nonetheless, convenience has made the CDC microassay a universally accepted method for HLA typing technology. The CDC assay starts by incubating viable lymphocytes with HLA-specific antisera. If the antigens on the cells match the antibodies in the serum, the complement system activates, leading to cell death and the uptake of a vital dye. Results are obtained by counting the live cells after the addition of rabbit serum, which serves as a source of complement. While HLA class I typing was already being performed, class II typing only began with the advent of cell separation technology. The discovery of HLA-D, later renamed HLA-DR (DR for D-related), was confirmed when an HLA-A and -B identical sibling pair exhibited a mixed lymphocyte reaction [11]. Unlike HLA class I, which is expressed on all nucleated cells, class II molecules are exclusively expressed on professional antigen-presenting cells, including dendritic cells, B cells, and monocytes. For class II typing, cell separation techniques were essential as they enabled the isolation of B cells, which predominantly express class II antigens. HLA-DR specificity was first reported in 1977 by analyzing the reaction patterns of B lymphocyte-reactive sera at the 7th International Histocompatibility Testing workshop [5]. Since then, cell separation techniques have facilitated the HLA-DR and HLA-DQ typing of B lymphocytes.

The CDC microassay is challenging and labor-intensive, with issues such as cross-reactivity complicating interpretation. Additionally, the limited availability of serological reagents further constrains its utility. This assay requires viable cells and antisera, primarily sourced from individuals sensitized to various HLA types, including multiparous women or patients who have undergone multiple transfusions. Another significant limitation is the resolution. The definition of histocompatibility typing terms, released in 2011, specifies that low resolution corresponds to the first field in DNA-based nomenclature, such as A*01 or A*02 (Fig. 2). High-resolution typing identifies alleles that encode the protein sequence for the antigen binding site and excludes alleles not expressed as cell-surface proteins. High-resolution achieves a resolution level of at least the second field and resolves ambiguities caused by polymorphisms in exons 2 and 3 for class I loci, and exon 2 for class II loci. Serological typing can only detect differences that are sufficiently distinct to be identified based on antibody specificity [13]. Furthermore, HLA antisera may exhibit cross-reactions, complicating the differentiation of minor amino acid sequence variations and leading to mistyping. For instance, one study reported a 25% error rate in HLA-B typing [14]. Therefore, transplantation based on serologic HLA typing results in poor outcomes [15,16]. Due to these limitations, serological techniques are being rapidly replaced by DNA-based methods.

Figure 2. Nomenclature of HLA alleles. Each HLA allele name has a unique number of four fields. The first field describes the allele group, which often corresponds to the serological type. The second field is used to describe the specific HLA protein, and third field provide the information of synonymous mutation within the coding sequence. The fourth field describes the mutations in noncoding regions such as 3’UTR, 5’UTR, and intron. Suffixes may be added to an allele to indicate the expression status of the allele. HLA, human leukocyte antigen. Adapted from Luo et al. Nat Genet 2021;53:1504–16 with permission of Springer Nature [12].

Polymerase Chain Reaction-Based Human Leukocyte Antigen Typing

With the development of PCR technology, DNA-based HLA typing technologies were introduced in the early 1990s [17]. This significantly improved the resolution of HLA typing for both class I and II loci. Concurrently, there has been a rapid increase in reports on allelic sequence diversity. Since most polymorphic regions are located in the peptide-binding regions—exons 2 and 3 for HLA class I genes and exon 2 for HLA class II genes—these are referred to as the core exons and are the primary targets for genotyping (Fig. 3). DNA-based HLA typing typically involves the use of primer pairs designed to amplify alleles at these specific loci. After amplification, the polymorphic sequences between the primer pairs are identified using various methods, including oligonucleotide probes, sequencing reactions, restriction enzymes, and sequence-specific primer binding. The three most commonly employed techniques are sequence-specific oligonucleotide (SSO) probe hybridization, sequence-specific primers (SSP), and sequence-based typing (SBT).

Figure 3. Schematic presentation of HLA class I and II gene organization and protein structure. In humans, separate clusters of HLA class I and II genes are located on chromosome 6. Three main class I genes are HLA-A, HLA-B, and HLA-C. HLA class I molecules consist of one α chain and β2microglobulin (β2m). The class II region includes the α and β chains for HLA-DR, -DP, and -DQ molecules. The most polymorphic regions are considered to reside in peptide-binding regions, which are exons 2 and 3 in HLA class I genes and exon 2 in HLA class II genes. HLA, human leukocyte antigen.

The SSO method employs a panel of oligonucleotide probes, each specific to a different HLA allele. These labeled probes bind exclusively to their complementary sequences in the amplified DNA sample, producing distinct patterns of probe reactivity that correspond to each allele. By analyzing these patterns, the types of HLA alleles can be determined [18]. Originally introduced as the "dot-blot" method, the SSO technique involved immobilized PCR products that were hybridized with labeled SSO probes. This approach evolved into the "reverse-blot" method, which relies on the hybridization of labeled PCR products with immobilized probes [19]. It was further developed into a multiplexed flowmetrix system [20]. Some SSO methods, such as reverse-SSO probes, can yield allele-level typing results [21].

The SSP method is based on the specific binding of primers to each allele [22], and is also known as allele-specific amplification [23]. For each HLA allele, specific primer pairs were designed based on polymorphic sequence motifs [24–27]. A positive PCR result is indicated by the presence of a band on the electrophoresis gel; the absence of a specific band suggests that the target allele is not present, and thus, the sample is considered negative for that allele. The patterns of positive and negative bands on the gel are used to determine the HLA type of the sample. This method offers the advantage of providing typing results relatively quickly, and the level of typing detail can be tailored by varying the number of PCR primer sets used. However, due to the high number of primer sets required for each locus, the SSP method is more practical for processing a smaller number of samples.

SBT is a direct sequencing method using chain-termination sequencing technology that provides allele-level HLA typing results. This has contributed to the rapid increase in the number of HLA alleles [28–30]. Due to the highly polymorphic nature of HLA, SBT has become the gold standard in allele-level HLA typing technology. Although it has been common since the early 1990s to type using only exons 2 and 3 for HLA class I and exon 2 for HLA class II [29,31,32], the scope of sequencing has expanded to include additional exons, thereby providing more comprehensive sequence information [33–36].

In contrast to serological typing, DNA-based typing technologies do not require cell viability or cell surface HLA expression. Moreover, these methods utilize synthetic standardized reagents to ensure more accurate typing results. DNA-based typing approaches facilitate intermediate-resolution HLA typing, which can distinguish between changes at the amino acid level [37]. For instance, the greatest discrepancies between serologic typing and DNA typing results were observed in HLA-B allele typing, which is recognized for having the highest diversity among HLA loci. This underscores the superior accuracy of DNA-based typing technologies [38]. Additionally, this technology has led to the identification of a significantly larger number of antigenically variable HLA alleles than those detectable by serological typing. While serologic typing identified several different HLA-DR specificities, the advent of DNA typing technology revealed more than 200 DRB1 alleles based on the second exon sequence within a decade of its introduction [39].

The introduction of DNA-based typing technologies has led to an increase in the number of new alleles identified in both class I and II (Fig. 4) [40]. However, this addition has also resulted in greater ambiguity, as HLA typing data may now be consistent with multiple alleles, complicating the interpretation of these data (Fig. 5) [41]. Ambiguities typically arise from incomplete sequencing and phase ambiguity. Incomplete sequencing ambiguity occurs when alleles differ only in regions that are not sequenced, such as introns, 3’UTR, and 5’UTR. Phase ambiguity arises when typing technologies fail to distinguish between polymorphisms on the same chromosome (cis) and those on different chromosomes (trans). Indeed, 41% of HLA-A and 24% of HLA-B typing results were found to be ambiguous when using the SBT method [42]. In cases of ambiguous results, HLA alleles are often assigned by predicting the most frequent allele in the population [43]. This issue can be resolved by completely sequencing the two separate alleles in a heterozygous sample. For instance, Voorter et al. [44] successfully demonstrated an unambiguous SBT typing technology by sequencing the full-length HLA genes after performing group-specific full-length amplification, starting from low-resolution typing results. Using this technology reduced ambiguities to 4.4% for HLA-A and-B, and 0% for HLA-C. Table 1 outlines the mechanisms and key features of each HLA typing technology.

Table 1. Overview of different HLA typing techniques

Typing method	Basic mechanism	Advantages	Disadvantages
Serologic method	Detection of HLA molecules using antisera	• Rapid screening • Suitable for deceased donor typing	• Low resolution • Limited availability of serological reagents • Typing limited to known alleles
SSO	Hybridization with short oligonucleotide DNA probes	Multiple sample typing	• Low to intermediate resolution • Typing limited to known alleles • Certain amount of ambiguities
SSP	Amplification of HLA alleles with sequence-specific primers	• Low cost • Applicable for deceased donor • Different resolutions can be obtained depending on the primers	• Unsuitable for large numbers of samples • Low to intermediate resolution • Typing limited to known alleles • Certain amount of ambiguities
SBT	Direct DNA sequencing	• High resolution • Able to sequence novel alleles	• Requires longer time • High cost • Unable to set phase between polymorphisms • Not suitable for deceased donor typing
NGS	Sequencing of small fragments of DNA in parallel	• High resolution • High throughput • Able to sequence novel alleles • Low ambiguity • Whole-gene coverage	• Expensive sequencer • Requires suitable software for analysis • Not suitable for deceased donor typing
TGS	• Individual DNA molecule sequencing in real time (PacBio SMRT sequencing) • Single DNA molecule sequencing through a nanopore (Oxford Nanopore sequencing)	• High resolution • Able to sequence novel alleles • Low ambiguity and phasing • Whole-gene coverage • Applicable for deceased donor typing (Oxford Nanopore sequencing)	• Expensive equipment • Relatively high error rate • Require suitable bioinformatics tools

HLA, human leukocyte antigen; SSO, sequence-specific oligonucleotide; SSP, sequence-specific primers; SBT, sequence-based typing; NGS, next-generation sequencing; TGS, third-generation sequencing, SMRT, Single-Molecule Sequencing in Real Time.

Figure 4. The number of new alleles submissions by year to the IPD-IMGT/HLA database from 2000 to 2023. This data includes small number of human leukocyte antigen (HLA)-related genes such as MICA/B. The predominant methods for HLA typing are indicated above the bars.

Figure 5. Ambiguous human leukocyte antigen typing results. (A) This type of ambiguity arises when the typing method cannot differentiate between the polymorphisms located on the same chromosome (cis) and that located on different chromosomes (trans). (B) Furthermore, ambiguity arises when two or more alleles differ only in the regions that are not sequenced such as introns, 3’UTR, and 5’UTR.

Introduction of Next-Generation Sequencing Typing

Since the early 2010s, NGS has been recognized as a high-resolution HLA typing technology capable of resolving ambiguity [41,45]. This technology encompasses methods such as IonTorrent’s semiconductor technology and Illumina's reversible terminator approach, both of which have become widely used in HLA typing. Various NGS platforms and methodologies have been investigated, including long-range PCR amplification [41,43,46], exome capture [47,48], and whole-genome sequencing [49,50]. The introduction of massively parallel NGS platforms has enabled high-throughput, high-resolution HLA typing.

Early research on NGS-based HLA typing technology primarily focused on demonstrating its efficiency, achieving high accuracy and low ambiguity [51–53]. By integrating two suitable software programs, HLA genotyping accuracy reached 100%, with an ambiguity rate of only 0.8% [54]. Another study demonstrated a 100% concordance between NGS and Sanger sequencing results, along with a 93.5% reduction in ambiguity [55].

NGS also aids in the identification of novel HLA alleles and the completion of partially sequenced alleles [41]. The IPD-IMGT/HLA database serves as the primary repository for HLA allele sequences. As of September 2024, the IMGT/HLA database has cataloged over 40,000 HLA alleles, with the adoption of NGS technology significantly boosting the rate of new HLA allele discoveries due to its high-resolution capabilities. Fig. 4 illustrates the sharp increase in the number of newly registered alleles in the IMGT database following the widespread application of NGS for HLA typing. The ability to type a large number of samples, leveraging the high-throughput benefits of NGS, enabled Baek et al. [56] to identify 13 new types by analyzing 26,202 samples. Furthermore, they highlighted the capability of NGS to rapidly process large sample volumes, extending sequencing to 11 loci, including HLA-A, -B, -C, -DRB1/3/4/5, -DQA1, -DQB1, -DPA1, and -DPB1 [57].

Despite its advantages, NGS technology, which relies on prior PCR amplification and short reads, faces several limitations when used for HLA typing. First, short tandem repeat regions such as HLA-DR can disrupt accurate typing due to replication slippage mutations [58]. Second, the 100–300 bp short reads are unable to span all HLA genes completely, resulting in residual phasing ambiguity [41,59–61]. Additionally, issues such as inappropriate primer binding can lead to allele dropout, and PCR amplification bias during HLA gene enrichment before sequencing may cause allelic imbalance [62–64]. Allelic dropout must be carefully considered when confirming homozygous results. Although the integration of bioinformatics with NGS technologies has improved HLA typing [47,49,59,65], long-read sequencing is necessary to address these limitations and further enhance HLA typing accuracy.

After the introduction of NGS, TGS technologies were developed. TGS differs from NGS in that it features long reads from single-molecule sequencing and real-time sequencing, whereas NGS involves library preparation, amplification, and sequencing to produce short reads [66]. For highly polymorphic or repetitive sequences, such as HLA regions, sequencing long reads from a single molecule is preferable to reconstructing long reads from short sequences [67,68]. TGS technologies have advanced HLA typing by providing high-resolution and unambiguous results that cover the entire HLA gene [2]. TGS technologies such as PacBio’s Single-Molecule Sequencing in Real Time (SMRT) sequencing [69] and Oxford Nanopore technology [70] are capable of sequencing an entire gene from a single DNA template without fragmenting the DNA into shorter pieces (Fig. 6). Since the majority of HLA gene polymorphisms are located within the peptide-binding groove, many previous HLA typing methods focused only on specific regions, such as exons 2 and 3 for HLA class I genes and exon 2 for HLA class II genes. However, the long reads provided by TGS technology enable the direct sequencing of the entire HLA region, yielding fully phased, high-resolution typing results [52,71,72].

Figure 6. The schematic workflow of a PacBio SMRT sequencing and Oxford Nanopore Technologies. (A) PacBio SMRT sequencing. Hairpin adaptors are ligated to both ends of the template DNA to form a single-stranded circular DNA. The addition of complementary fluorescently labeled dNTP during sequencing is recorded. (B) Oxford Nanopore sequencing. The DNA template binds to adaptors with motor protein. Sequencing is initiated by entering a nanopore attached to the motor protein. As the template DNA strand passes through the nanopore, specific change in the ionic current for each nucleotide across the membrane is detected. This specific current is further translated into nucleotide sequence. SMRT, Single-Molecule Sequencing in Real Time.

PacBio Single-Molecule Sequencing in Real Time Sequencing

PacBio introduced the first TGS technology, known as SMRT sequencing [69]. This method involves forming double-stranded DNA into single-stranded circular DNA using hairpin adapters. The circular DNA is then anchored with a DNA polymerase inside a nanowell, referred to as a zero-mode waveguide, for direct sequencing. Sequencing is achieved by synthesizing a complementary DNA strand with fluorescently labeled nucleotides, and the sequence is captured through optical signals. High accuracy is attained by sequencing multiple circles [69,73,74].

Turner et al. [75] utilized SMRT technology to perform HLA typing on 126 B-lymphoblastoid cell lines, achieving a 96% concordance rate with the IPD-IMGT/HLA database. The discrepancies in typing results were attributed to the identification of novel alleles and previously unreported types. Additionally, 64% of the typings were conducted at a higher resolution, and patterns of linkage disequilibrium were elucidated through four-field resolution typing. Another study employed a sequencing strategy that combined shotgun sequencing on an Illumina MiSeq with SMRT sequencing to achieve whole-gene characterization of HLA alleles [76]. By leveraging the strengths of each platform and mitigating their limitations, the researchers were able to submit 1,056 fully characterized HLA alleles to the IMGT/HLA database, nearly doubling the existing number.

Oxford Nanopore Technology

Oxford Nanopore sequencing technology offers a significant advantage in HLA typing due to its ability to cover megabase scales [72,77,78]. This technology utilizes flow cells equipped with nanopores. As single-stranded DNA passes through these nanopores, distinct ionic currents are generated based on the nucleotide sequence, and these currents are recorded in real time. The currents are then translated into DNA sequences using base-calling software. Similar to SMRT sequencing, this method delivers high-resolution typing results with fewer ambiguities.

Comparing the typing results with SSO typing technology, 100% concordance was achieved for all loci and samples, with notably low ambiguity for nanopores at only 9%, compared to 20% ambiguity with SSO [79]. Additionally, both technologies required only two working days. When compared to SBT typing, the concordance rate was also 100% [72]. Using Athlon, a bioinformatics pipeline that maps nanopore reads to the HLA allele database, 100% accuracy was achieved in HLA class I typing at a two-field resolution [80]. Furthermore, another study demonstrated concordance rates of 99.5% and 99.3% at 2-field and 3-field resolutions, respectively, for multiple HLA loci compared to reference typing using R10.3 flow cells [81]. High-resolution HLA typing from Oxford Nanopore sequencing showed 100% concordance with NGS methods, with 93.2% of alleles reported as unambiguous at a two-field typing resolution [82]. Nanopore technology enables rapid and high-resolution HLA typing for deceased donors [82], making it suitable for emergency deceased organ donor typing without disrupting the organ allocation process and potentially improving transplantation outcomes.

Hematopoietic Stem Cell Transplantation

The development of HLA typing technologies has substantially advanced the field of hematopoietic stem cell transplantation. These advancements have not only improved transplantation outcomes but have also broadened the range of donor options, extending beyond HLA-identical siblings to include unrelated donors and umbilical cord blood. Numerous studies have highlighted the crucial role of HLA class I and II matching in the success of hematopoietic stem cell transplants [83–86]. Specifically, a single mismatch at any of the HLA-A, -B, -C, -DRB1, or -DQB1 loci was linked to a notable reduction in survival, with a hazard ratio of 1.41. Furthermore, the presence of multiple mismatches was associated with even poorer survival outcomes, showing a hazard ratio of 1.91 and severe grade III–IV acute GvHD [82]. The advent of high-resolution typing technologies has led to new insights. Pairs previously deemed fully matched using low-resolution typing were discovered to have mismatches when reevaluated with high-resolution methods. For instance, one study found that 26% of what were considered full matches (six out of six matches) for HLA-A, -B, and -DR based on low-resolution typing actually had at least one high-resolution mismatch [87]. Additionally, among the matches for HLA-A, -B, -C, and -DR (eight out of eight matches) determined by low-resolution typing, 19% exhibited a single high-resolution mismatch.

Several studies have shown that discrepancies between high-resolution HLA typing and low-resolution HLA typing can influence transplantation outcomes [88]. For instance, patients who were matched 12/12 by ultra-high-resolution typing exhibited better 5-year overall survival rates compared to those matched 12/12 by conventional methods but were found to be mismatched upon ultra-high-resolution analysis (54.8% vs. 30.1%, P=0.022) [89]. In addition, high-resolution HLA mismatches are also linked to poor transplant outcomes when they are combined [87]. Because low-resolution mismatched donor-recipient pairs are more likely to differ at a larger number of immunogenic epitopes, especially in the antigen recognition domain (ARD) regions, the impact of high-resolution mismatches may be considered more permissive. Nevertheless, these studies have clearly shown that high-resolution mismatches can significantly affect outcomes, highlighting the importance of high-resolution typing in enhancing the success of transplantations [90,91]. Furthermore, high-resolution typing that includes loci such as HLA-DQ and HLA-DP has been associated with improved transplantation outcomes. For example, mismatching of HLA-DQ has been linked to an increased risk of graft loss (hazard ratio, 1.18) and a higher incidence of acute rejection within the first year (odds ratio, 1.14) among living kidney donor recipients [92]. In the context of hematopoietic stem cell transplantation, mismatching of the HLA-DPB1 allele was associated with higher risks of grades II–IV acute GvHD and grades III–IV acute GvHD, with odds ratios of 1.33 and 1.26, respectively [93].

Solid Organ Transplantation

HLA matching is also considered important in solid organ transplantation, with numerous studies demonstrating its impact on clinical outcomes [94–96]. However, unlike in hematopoietic stem cell transplantation, solid organ transplantation often results in mismatches at the HLA antigen level due to the limited availability of donor organs [97]. Therefore, while high-resolution HLA typing is primarily used in hematopoietic stem cell transplantation to find highly matched donors, its role in solid organ transplantation should be considered differently. Beyond assessing donor-recipient histocompatibility, high-resolution HLA typing is essential for risk stratification and supports personalized approaches to immunosuppression and graft rejection treatment regimens. Additionally, several studies have indicated that further high-resolution genotyping can provide valuable information for assessing alloantibody reactivity during solid organ transplantation [3,4,98].

Low-resolution typing technologies have been employed due to the time constraints associated with deceased donor allocation. The MinION device from Oxford Nanopore Technologies has demonstrated its capability to provide high-resolution typing information within 6 hours, establishing its potential utility in deceased donor allocation. High-resolution typing results can enhance clinical decision-making in solid organ transplantation by offering detailed molecular mismatch information, such as eplet or epitope mismatches. An eplet is defined as the minimal amino acid configuration on the HLA molecule that triggers an antibody response, while an HLA epitope consists of the amino acids involved in antibody binding [99]. Regarding molecular mismatch, a recent study highlighted the use of an algorithm designed to predict mismatched HLA peptides that could potentially activate CD4 T cell alloreactivity [100]. When applied to kidney transplantation recipients, higher scores predicted by this algorithm were linked to adverse clinical outcomes, including the formation of de novo donor-specific antibodies, graft rejection, and reduced graft survival [101–104]. However, this algorithm primarily relies on low-resolution HLA typing data. When high-resolution typing data is used instead, patients can be categorized into different risk levels, thereby enhancing both the accuracy and predictive capabilities of the outcomes. These findings collectively highlight the potential to enhance patient outcomes and increase the success of various types of transplantations through the use of high-resolution HLA typing.

With evolving HLA typing technologies, transplantation practices, and the development of immunosuppressive regimens, the concept of "HLA matching" should also evolve. The additional data obtained from high-resolution HLA typing, which now includes introns and other gene regulatory sequences, will ultimately enhance clinical outcomes. Consequently, this expanded typing information should not be used to exclude mismatched donors. Instead, it should be employed for risk stratification and to optimize treatment regimens tailored to individual patients [105].

Among the HLA genes, the highest degree of polymorphism is observed in specific exons of the ARDs, which are responsible for encoding the peptide and T cell receptor-binding regions of the HLA protein. The future of HLA typing will involve expanding beyond traditional sequencing of the HLA region by extending coverage to include non-ARD exons, introns, and regulatory regions. A comprehensive analysis of the immunogenicity of the HLA protein, including its expression levels [65] and its binding to other immunoreceptors such as killer cell immunoglobulin like receptor [106,107], will further increase our understanding of the HLA matching in the field of transplantation.

This review outlines the developments in HLA typing technologies and their impact on transplantation outcomes. Research showing a correlation between HLA matching and improved transplantation results has broadened donor options and enhanced clinical outcomes. As HLA typing technologies continue to evolve, we anticipate further enhancements in transplant outcomes due to more precise donor selection. Additionally, in the realm of solid organ transplantation—where procedures often proceed with minimal mismatches due to donor scarcity—high-resolution HLA typing enables patient risk stratification. This advancement supports personalized approaches to immunosuppression and graft rejection treatment protocols. With ongoing technological advancements, the criteria for optimal HLA matching will continually adapt to fulfill the evolving demands of clinical transplantation.

Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Funding/Support

This research was supported by Basic Medical Science Facilitation Program through the Catholic Medical Center of the Catholic University of Korea and was funded by the National Research Foundation of Korea (NRF) grant funded by the Korean Ministry of Science and ICT (2023R1A2C1002458).

Author Contributions

Conceptualization: HC, TGK, YJC. Resources: EJC, HJK, ICB, AKW, SJP. Writing–original draft: HC. Writing–review & editing: all authors. All authors read and approved the final manuscript.