Unrelated Donor Hematopoietic Cell Transplantation

The Human Leukocyte Antigen and Killer-Cell Immunoglobulin-LIKE Receptor Genetic Systems

Human Leukocyte Antigen Typing Methods

The development of standardized typing methods and nomenclature for HLA genes have been facilitated by a series of 17 international histocompatibility workshops ( Table 106.2 ). The terms “low-resolution,” “intermediate-resolution”, and “high-resolution” are a reflection of the evolution of DNA methods and their equivalency to serologically defined specificities. The term “6/6” match refers to recipients and donors who share the same low-resolution-defined HLA-A, HLA-B, and HLA-DR genes. The term “8/8” refers to high-resolution matching at the four loci HLA-A, HLA-B, HLA-C, and HLA-DRB1; “10/10” to the addition of HLA-DQB1, and “12/12” to the addition of HLA-DPB1.

Next-generation sequencing (NGS) platforms are the most current sequencing methodology which provide not only “ultra-high resolution” of HLA alleles (typing of the gene beyond its coding regions) but also short-range phasing of exons 2, 3, and 4 of class I genes, and of exons 2 and 3 of class II genes. NGS has the capability of linking sequences across exons that substantially reduces ambiguities of allele combinations. NGS approaches for HLA typing are supported by software for automated allele assignment. Because many samples may be tested simultaneously, NGS is a cost-effective typing method for typing donors recruited into registries.

Genetics of the Killer-CellImmunoglobulin-Like Receptors System

Recognition that natural killer (NK) cells engage HLA, leading to the capacity to discern self from non-self, has generated interest in the NK receptors responsible for HLA binding ( Table 106.3 ). Located on chromosome 19, the KIR gene region is similar to the HLA gene complex in that it is also polygenic and polymorphic; however, unlike the HLA genes, whose loci are conserved across all individuals, KIR genes frequently differ from individual to individual. As few as eight and as many as 14 different KIR genes and pseudogenes may be found on one haplotype, with some genes existing more than once. KIR genes encode two broad categories of receptors: mmunoreceptor tyrosine-based inhibition motif (ITIM)-containing inhibitory receptors and mmunoreceptor tyrosine-based activation motif (ITAM)-binding activating receptors. Nomenclature for the KIR genes is structurally inspired: the first two characters (2D or 3D) refer to the number of extracellular Ig-like domains characterizing the receptor; the third character (L or S) refers to the “long” inhibitory or “short” activating cytoplasmic tail of the receptor, and the last character (numbering 1 through 5) differentiates receptors with similar numbers of extracellular domains and cytoplasmic tail length. The inhibitory receptors recognize HLA class I ligands, and interaction between the receptor and its cognate HLA ligand is responsible for establishing the response threshold upon encountering a putative target. It is in this way that a KIR-expressing NK cell becomes educated to behave indifferently toward cells bearing self-HLA ligands and aggressively toward cells that have lost or altered HLA expression.

The KIR genes are arranged neatly in a head-to-tail manner within a highly organized cluster with prototypic gene order, leading to LD between pairs and clusters of KIR genes and alleles. The so-called anchor genes and pseudogenes, recognized for their distinctive positions on nearly all identified KIR haplotypes, demarcate the beginning, middle and end of the KIR gene cluster. KIR3DL3 and KIR3DL2 flank the KIR genetic region, while KIR3DP marks the end of the centromeric portion and KIR2DL4 marks the beginning of the telomeric portion of the haplotype. The centromeric and telomeric portions appear to have evolved and diversified separately due to reproductive and pathogenic pressures, and the gene content distinctive to each region may have clinical relevance in HCT.

Population studies of KIR genotypes identified the presence of two major KIR haplotype groups, the A-haplotype and the B-haplotype, where, apart from the anchor genes, the canonical A-haplotype is comprised of the genes KIR2DL3, -2DL1, -3DL1, and -2DS4. By contrast, B haplotypes are more variable in gene content, exhibiting more than one activating KIR gene—KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS4, KIR2DS5, and KIR3DS1—in the centromeric and/or telomeric regions.

Allelic polymorphism of each KIR locus lends additional diversity to the genetic region, with new alleles reported regularly. The loci encoding inhibitory KIR receptors are the most diverse, with 183 KIR3DL1, 165 KIR3DL3, 164 KIR3DL2, 111 KIR2DL1, 64 KIR2DL3, and 34 KIR2DL2 alleles reported as of March 2020. KIR allele sequences are named in an analogous fashion to HLA. An asterisk separates the KIR gene name (e.g., KIR2DL1) from the numeric allele sequence (KIR2DL1*). The first three digits following the asterisk indicate alleles that encode different protein sequences (KIR2DL1*003), followed by two digits that distinguish alleles whose DNA differences within the coding sequence are synonymous (KIR2DL1*00302). The last 2-digit field identifies alleles that differ by DNA substitutions in non-coding regions (KIR2DL1*0030202). For some KIR loci (e.g., KIR3DL1 and KIR2DL1), alleles can be grouped into subtypes based on shared sequences that distinguish them based on cell surface expression or function.

Killer-Cell Immunoglobulin-Like Receptors Typing Methods

Identification of KIR genotypes can be performed at the resolution of the KIR gene, subtype, and allele by a number of different molecular approaches. Currently, polymerase chain reaction (PCR) sequence-specific primers (SSP) and sequence-specific oligonucleotide probes (SSOP) are employed by clinical laboratories for KIR gene content. For research purposes, PCR-SSP is used to identify subtypes and some alleles. Whole-genome sequencing and RNA-seq databases are rich resources for algorithmic extraction of KIR gene content and copy number. More recently, NGS has emerged as a powerful research tool for KIR gene and allele discrimination, with the anticipation of more widespread use as evidence for the clinical importance of KIR genotyping increases.

Donor Identification and Likelihood of Transplantation

The field of unrelated donor HCT has witnessed rapid growth over the past 30 years. Among all allogeneic transplantations performed in the US, the number of transplants from unrelated donors exceeds that from related donors since 2006. The primary indications for unrelated donor HCT are acute myeloid leukemia (AML), myelodysplastic syndrome/myeloproliferative disorders (MDS/MPD), acute lymphoblastic leukemia (ALL), non-Hodgkin lymphoma (NHL), and other nonmalignant disorders. Starting in 2005, patients between 60 and 69 years of age comprised the fastest-growing group of transplant recipients as a result of the introduction of non-myeloablative and reduced-intensity conditioning regimens. Since 2011, patients at the age of 70 or older are the fastest-growing group of allogeneic transplant recipients.

Unrelated HCT has been made feasible by the establishment of registries of volunteer donors worldwide. The Anthony Nolan Appeal was the first effort to demonstrate the feasibility of donor recruitment. Now known as the Anthony Nolan Research Institute, this registry was the first to promote access to HLA-matched bone marrow (BM) donors for patients around the world. In the United States, early efforts for donor recruitment were spearheaded by individual centers. The growing interest in unrelated donor HCT led the US Congress to authorize the creation of a national registry comprising a network of donor centers, transplant centers, and a national coordinating center through the Transplant Act of 1984. Two years later, a federal contract to establish a national registry was awarded to the National Marrow Donor Program (NMDP). In the Netherlands, Professor Jon J. van Rood led the Europdonor Foundation in the collection of HLA data from donor registries around the world and the formation of a database of HLA phenotypes known as Bone Marrow Donors Worldwide (BMDW). Continued growth in registry size worldwide has increased the chances that well-matched donors can be identified. Today, more than 37 million donors are available through international cooperative agreements with registries around the world. The likelihood of identifying an HLA-A, -B, -C, -DRB1-matched (HLA-8/8) unrelated donor depends on the patient’s race and ethnicity, with higher rates in patients of White European descent than other ethnicities. In the Japanese population, the probability of identifying an HLA-8/8 match increases as registry size increases but is predicted to reach a limit due to population-specific allele and haplotype frequencies. Strategies to estimate the likelihood of any given patient’s success in identifying HLA-matched donors take into account the patient’s race and ethnicity. When criteria are relaxed to permit mismatching, over 98% of patients of European descent and over 80% of patients of non-European origin have HLA-7/8 donors. Use of cord blood can further extend access to transplantation.

The identification of HLA-matched donors is only one potential barrier to transplantation for patients. Donor availability, donor health issues, donor race, delays due to coordination, and deterioration of patient health may also lower the chances that a patient can undergo transplantation ( Box 106.1 ). Identification of primary and backup donors allows patients to reach transplantation and improves the overall survival of patients with good performance status, compared to not having a suitable donor.

Human Leukocyte Antigen Criteria

Selection of unrelated donors includes consideration for the degree of HLA matching and the presence of recipient anti-donor HLA antibodies, which place the patient at high risk for non-engraftment in the setting of HLA mismatching. Evaluation of unrelated donors includes a health screening history, physical examination, and laboratory testing for risks associated with transmissible elements with a special focus on hepatitis B, hepatitis C, human immunodeficiency virus (HIV 1 and 2), human T-cell lymphotrophic virus I and II, malaria, West Nile virus, transmissible spongiform encephalitis (Creutzfeldt-Jacob disease), Chagas disease, Treponema pallidum , and cytomegalovirus (CMV). This core screening is supplemented with an infectious disease health questionnaire screening with novel epidemics including Zika risk areas and Covid-19 exposure. Increased awareness for the health and safety of the unrelated donor has led to the establishment of standards for donation.

A large-scale registry study demonstrates that donor HLA match status and donor age are the two most important determinants of transplant survivorship. Two-year survival was 3% better when a donor 10 years or younger is selected. These data demonstrate that transplant centers have the flexibility in donor selection, with the prioritization of HLA-matched young donors where possible (see Box 106.1 ).

Human Leukocyte Antigen Vector of Mismatching

The “vector” or “direction” of HLA compatibility between a donor and a recipient has biologic relevance in risks of graft failure and GVHD and applies to all classical HLA loci. The presence of donor alleles not shared by the recipient determines HVG allorecognition; the presence of recipient alleles not shared by the donor provides the immunologic basis for GVH allorecognition ( Table 106.4 ). “Bidirectional” mismatching occurs when both HVG and GVH vectors are present at a given HLA locus. A recent analysis of the vector of mismatching confirms that outcome is associated with a GVH vector mismatch, and not when a GVH mismatch is absent.