The epitopes of human MHC were first found on white blood cells and, therefore, were called human leukocyte antigen (HLA). Later these Ags were identified on virtually all nucleated cells throughout the body including cells in all three layers of the cornea.³ In the 1960s, the use of serologic reagents to identify transplantation Ags brought the HLA system into focus.⁴ By the end of the 1977 HLA workshop, it was known that three loci HLA-A, HLA-B and HLA-C controlled the production of a group of proteins that were called Class I Ag.⁵ Similar testing identified loci of the MHC in the mouse (H-2), rat (RT-I), chimpanzee (ChLA), rhesus monkey (RhLA), dog (DLAI), and numerous other species. The MHC is highly conserved throughout many mammals, and both animal and human studies aided in increasing the success of tissue and organ transplants. Microcytotoxicity testing to determine individuals' complete HLA genotype and computer codification of the reaction patterns of thousands of anti-HLA alloantisera have made possible the delineation of the entire HLA system. The Ags are reported centrally and updated monthly to the HLA sequence database held at the Anthony Nolan Research Institute.¹

In 1973, HLA-B 27 was found to be associated with ankylosing spondylitis in a large proportion of cases, providing a second impetus for further study of the MHC. The motivation for defining HLA disease and association can help to determine at risk populations and genetic susceptibility factors. These may shed light on the mechanisms of disease, which include autoimmune, immune complex, and nonimmune, allowing for more precise strategies in disease prevention and treatment.

In this chapter, nomenclature of the HLA system, genetic organization, biochemistry, and structure are reviewed. The historic and modern techniques of HLA typing are discussed, as well as the ophthalmic diseases associated with the HLA system.

NOMENCLATURE

The HLA region occupies a segment of DNA approximately 2 centimorgans (4 million base pairs) in length on the short arm of human chromosome 6.⁶ The centimorgan is a standard unit of physical map distance on a chromosome equivalent to a 1% frequency of genetic recombination between linked genes. The MHC contains numerous loci where individual genes may be found (Fig. 1). These intrinsic MHC loci are categorized into three classes, according to function (Table 1). With more sensitive serologic determination, many specificities have been refined and renamed. For example, DR5 was split into DR11 and DR12; B15 splits into B62, B63, B75, B76 and B77. The HLA Nomenclature Committee, governed by the World Health Organization, has met every 2 or 3 years to address the evolving knowledge and to come to agreements on the criteria used for defining serologic determinants. The problems with the complexity of the HLA nomenclature system has been evident since the first HLA workshop in 1964.⁷ New discoveries within the HLA system make it difficult to maintain organization of the nomenclature. A number of alleles named in error were renamed or abandoned.^1,8 Currently identified alleles of each type include: 151 HLA-A, 301 HLA-B, 83 HLA-C (which are Class Ia—classical Class I Ags), 5 HLA-E, 1 HLA-F, 14 HLA-G (which are Class Ib—nonclassical Class I), 2 HLA-DRA, 281 HLA-DRB (D-related), 20 HLA-DQA, 43 HLA-DQB1 (formerly DS, MB, or DC), 18 HLA-DPA1, 87 HLA-DPB1 (formerly SB), 4 HLA-DMA, 6 HLA-DMB, 8 HLA-DOA, and 3 HLA DOB (which are Class II Ags). As of September 1999, a total of 1027 HLA alleles have been determined, and new alleles are updated monthly at a rate of about 150 alleles per year.¹ The complete, updated database can be accessed on the Internet (www.anthonynolan.org). Additional loci continue to be recognized, and several are apparently silent (i.e., no gene product is produced). Despite its complexity, the HLA nomenclature system has been a prototype for naming genes in the Human Genome Project.⁷

Table 1. Components of the MHC

ALLELIC POLYMORPHISM

Within the MHC complex at any given locus, one of several alternative forms or alleles of a gene can be found. The high level of polymorphism has been attributed to the rate of mutation of HLA.⁹ There are three methods for the generation of new classical MHC gene sequences: point mutation, recombination, and microrecombination. One example of this is the HLA-A69 coding allele, which has one exon that is identical to the HLA-A68 coding allele and the rest of the gene from the HLA-A2 coding allele, clearly resulting from a large-scale recombination event.¹⁰ There are also poorly defined specificities that lack official WHO nomenclature designations; they may be designated with an X, with a local name; or treated as a known specificity with a similar serologic pattern.¹¹

PUBLIC AND PRIVATE ANTIGENS

Public Ags were determined from early studies with epitopes that appear on several different molecules. For example, Bw4 is also found on B-13, B-27, B-37, and many others. Private Ags, on the other hand, were subsequently determined with more sensitive serologic discrimination. This resulted in revision to the initial nomenclature.⁴

CROSS-REACTIVE GROUPS (CREGs)

There is a limited polymorphism at any given amino acid position and common sequence motifs are often shared by closely related HLA alleles and specificities. This interrelationship leads to amino acid residue and epitope matching as CREGs.¹² CREGs may interfere with accurate Ag assignment. One example is the family of HLA-B specificities that has serologic cross reactivity with HLA-B27 (the B7-CREG family) that is associated with African-American patients with ankylosing spondylitis.^13,14 It has been postulated that matching CREG specificities may aid in the prevention of graft rejection with corneal transplantation.

LINKAGE DISEQUILIBRIUM

Linkage disequilibrium is a difference between the expected association of alleles with random matings in a population at equilibrium and the observed frequency of finding a given allele at one HLA locus. Certain combinations of alleles, however, are found with a frequency far exceeding the expected.¹⁵ For example, the HLA-A1 allele and the HLA-B8 allele have been demonstrated in American Caucasians at frequencies of 27.5% and 15.7%, respectively. The expected frequency with which the HLA-A1-B-8 haplotype should be found is 0.27 × 0.15 = 4.3%. However, this haplotype is actually found with a frequency of about 9.8%. Thus, the linkage disequilibrium is 9.8 - 4.3 = 5.5%. This indicates a low frequency of recombination between HLA-A and HLA-B. Recombination is not observed between HLA-DQ and -DR, suggesting hotspots of recombination.¹⁸ There are numerous examples of linkage disequilibrium. Etiologies of linkage disequilibrium in human populations include inbreeding, migration and mixing of two populations, random drift, and selective advantage of certain haplotypes.¹⁹ Clinically, this can lead to an association of both HLA alleles with a disease process. Several diseases have been demonstrated to have an association with HLA Ags in one race but not another. For example, Sjögren's syndrome (SS) has been associated with HLA-Dw3 and, therefore, with HLA-A8 (which is in linkage disequilibrium with Dw3). However, one study showed no association of HLA-A8 with SS in Eastern European Jews, whereas another showed an association among Caucasians in the United States.²⁰ A similar situation has been demonstrated in Grave's disease and diabetes mellitus.^20–22

PARALAGOUS CHROMOSOMES

Collections of loci similar to some of those near the MHC have been identified on chromosomes 1, 9, and 19. For example, NOTCH4, PBX2, and TN-X, which are found on chromosome 6, are related to NOTCH1, PBX3, and TN-C on chromosome 9. One theory to explain this is that primordial chromosomal duplication took place. More comparisons continue to be made, as well as evaluation of the function of these paralogous chromosomes.¹

CLASS I ANTIGENS

Both Class I and Class II MHC Ags are heterodimeric (two polypeptide chains) glycoprotein cell-surface molecules (see Fig. 2). Class I Ags consist of a polymorphic glycoprotein heavy (alpha) chain on the cell surface, and a noncovalently associated, nonpolymorphic protein called β₂-microglobulin (which is coded on chromosome 15).²⁸ The detailed structures were determined by x-ray crystallography.²⁹ The entire Class I Ag is anchored into the cell membrane lipid bilayer and uses the heavy chain molecule to circulate.³⁰ The peptides bound to Class I molecules are predominantly short, usually octamers and nonamers. It is bound by both ends, with interactions between the MHC molecule and the N- and C-termini of the peptide. The most important MHC interactions are those with side chains of the second amino acid of the peptide (referred to as P2) and the final amino acid.^1,2

More information on the structure and function of Class I has been gained by evaluating the large-scale chromatin organization of chromosome 6.³¹ The gene encoding Class I Ags consists of seven exons (translated regions of DNA), separated by introns (untranslated regions of DNA). The second, third, and fourth exons correspond to the first, second, and third alpha domains, respectively, in the heavy chain.¹⁰ Therefore, complex mechanisms for gene splicing, assembly, and ribonucleic acid (RNA) translation are required to produce intact HLA molecules.

CLASS II ANTIGENS

The HLA Class II region encodes three main glycoproteins: HLA-DP, -DQ, and -DR.³² Each of the Class II glycoproteins are noncovalently linked heterodimers with an alpha chain and a beta chain in noncovalent association (see Fig. 2). The alpha chain has two glycoprotein residues, and the beta chain has one residue. These chains each consist of three regions including an extracellular hydrophilic region, a transmembrane hydrophobic region, and an intracellular hydrophilic region similar to the other members of the immunoglobulin family. The intracellular portion of the alpha chain can be phosphorylated for signal transduction. The alpha chain extracellular region contains two domains (homology units) each of which has an attachment site for oligosaccharide, and the alpha 2 domain has an intrachain disulfide bond. The beta chain extracellular hydrophilic region also contains two domains, the first of which bears the oligosaccharide and both of which contain intrachain disulfide bonds. The second extracellular domain of both the alpha and beta chains located most proximal to the cell membrane shows significant sequence homology to immunoglobulin constant region domains. The HLA-DP, -DQ, and -DR are expressed on the surfaces of the cells. HLA-DO is closely related to the classical Class II molecules but shows only limited polymorphism. HLA-DMA and DMB are only distantly related and have relatively little polymorphism.^33,34

The map of the MHC-D region (see Fig. 1) shows that each subregion contains several genes.¹⁸ The DR subregion, for example, contains three beta chain genes, one of which is a pseudogene. Each of the beta chain genes can combine with the product of the DR alpha chain locus to form a complete HLA-DR molecule.³⁵ Similarly, the DP and DQ subregions contain two alpha and two beta chain genes, which combine to form complete molecules. The DP, DQ, and DR genes, like Class I genes, are composed of introns and exons, with the exons corresponding roughly to the protein domains.¹

CLASS III ANTIGENS

In addition to the histocompatibility loci, the Class III region contains several clusters containing related genes determining the second (C2) and fourth (C4) components of the complement pathway and two members of the heat shock protein family; several cytokines also localize to the MHC on chromosome 6.³⁶ The C4 locus has been duplicated, so there are two C4 genes on each chromosome, which are more polymorphic than those of C2. There are also proteins without known roles in the immune system coded here including NOTCH4, PPT1, and steroid 21-hydroxylase.⁶

CLASS I ANTIGENS

HLA Class I Ag (HLA-A, -B, and -C) are expressed on virtually all nucleated cells and are often referred to as classical transplantation Ags. Class I Ag are located on cells within all layers of the cornea including epithelium, keratocytes in the stroma, and, albeit in low levels, on the endothelial cells.^37,38 The low level of MHC-Ia on the endothelium is an index of the immune privilege attributed to the anterior chamber.³⁹ These Ags, present on the cell surface, are scrutinized by T cells and are the principal Ags recognized by the host during tissue transplant rejection (see Fig. 2). In cell-mediated cytolysis, which is the experimental correlate of graft rejection, Class I Ags are the target molecules recognized by killer T lymphocytes.²⁹ Both private and public HLA Ags are recognized independently by T cells.^40,41

The natural role of Class I Ags, however, is related to histocompatibility restriction of cell-mediated lysis (CML) by CD8+ cytotoxic T lymphocytes of virus-infected cells.⁴² When T lymphocytes are exposed to the viral Ag that is expressed on the surface of an infected cell, lymphocytes recognize the viral epitope only when associated with a Class I molecule.³⁰ The destructive lymphocytes are activated by a response to viral Ag and are limited to killing only the target cells that bear the same viral Ag associated with the same Class I Ag present on the original, sensitizing infected cell.⁴³ Interestingly, the cytotoxic T cells will not destroy target cells bearing the same viral Ag but a different Class I Ag, nor will they kill cells with the same Class I Ag but a different viral Ag.⁴⁴

CLASS II ANTIGENS

Class II Distribution

The Class II Ags, unlike the Class I Ags, are usually expressed only on immunocompetent cells, including tissue macrophages and circulating monocytes, resting T cells in low amounts, activated T cells in high amounts, and B lymphocytes. Both Class I and Class II Ags are readily detected on the endothelium of rejected corneal buttons but not on the endothelium of normal corneal endothelium or on corneal buttons removed for herpetic disease. Cells in the ciliary body and choroid have been found to express Class II Ags.⁴⁵

Class II Ags are expressed on the Langerhans' cells that migrate in the conjunctival and corneal epithelium. These wandering Ag presenting cells have a turnover rate of every 3 days,⁴⁶ are critical in the corneal and ocular surface immune response,⁴⁷ and are much more densely distributed in the peripheral rather than central cornea. In the setting of ocular inflammation, Langerhans' cells proliferate on the ocular surface; this proliferation is reversed with the administration of topical and systemic corticosteroids. Class II Ags have not been detected in high concentrations on corneal epithelial, stromal, or endothelial cells from normal donors⁴⁸ but can be found scattered throughout the epithelium in leukocytes and occasional epithelial cells. Class II Ags are absent on the blood vessel endothelium, reflective of the immunologically privileged status of the eye.⁴⁵

Antigen Presentation

Class II molecules are designed for the capture of peptides derived from exogenous Ag because they present peptides created in the endosomal compartment of the antigen-presenting cells. The mechanisms for uptake of Ag for class-II restricted presentation includes endocytosis, phagocytosis, and macropinocytosis.¹ Before the loading into the cleft of Class II molecules, protein Ag need to be cleaved to peptides of 12 to 18 amino acids in length. Once the cytosolic Ag is contained in vacuoles, the Class II molecules may present them on the cell surface where they remain for 1 to 2 days.

In physiologic situations in which cells bearing identical HLA haplotypes interact productively, the Class II Ags are actively involved in Ag presentation by macrophages, B lymphocytes, and other antigen-presenting cells to T lymphocytes.⁴⁹ The interaction between immunocompetent cells is essential to normal immune function. TCRs have two distinct regions that aid in the corecognition of fragments of protein Ags in association with self Class II molecules.¹ These conformational changes are transduced through the T-cell membrane via the TCR molecule, leading to a cascade of intracellular events leading to T-cell activation and proliferation.²⁷ TCRs (see Fig. 2) exist only on the cell surface, unlike antibody molecules, which can also circulate freely and bind intact, free Ag. The TCR can only recognize processed Ag when it is embedded in Class II molecules.¹ Intact exogenous Ags, such as free tetanus toxin, enter the antigen-presenting cell by means of macropinocytosis after binding Class II molecules.^50,51 Once inside the cell, the foreign protein undergoes proteolysis into shorter amino acid sequences, leading to a highly immunogenic protein fragment. After further processing in the Golgi apparatus, the immunogenic protein is recycled on the cell surface, probably within the same endosome, and presented in close association with a Class II molecule, ready for T-cell recognition. Endogenous Ags, such as viral glycoproteins, are synthesized via a similar mechanism within the endoplasmic reticulum and Golgi apparatus under the virally encoded regulatory gene products.⁵² These immunogenic peptides exit the Golgi to reach the surface of the infected cell where they associate with Class I Ags. Thus, Ag internalization, processing, and presentation on the cell surface in conjunction with MHC products are essential to T-cell recognition and activation by both Class I and Class II molecules.⁵³

T-Cell Interactions

The CD 4 and CD 8 molecules (see Fig. 2) are intimately involved in T-cell recognition of the MHC molecule/processed Ag complex. The CD 8-bearing, cytotoxic T-cell subset is correlated with recognition of Class I Ags and contains the main fraction of cytotoxic T cells. CD 4-bearing helper T cells specifically recognize Class II MHC Ags. Monoclonal antibodies specific for CD 4 block Ag induced proliferation and production of interleukin 2 (IL-2) by CD 4+ cells, and anti-CD 8 antibodies inhibit cytotoxicity by interfering with the interaction between cytotoxic T cells and target cells bearing foreign Class I Ags.⁵³ The glycoproteins on these cells function as cell adhesion molecules, enabling nonspecific binding interactions between T cells and target cells bearing appropriate MHC Ags. This promotes binding of processed Ag by the TCR, thereby enhancing subsequent T cell activation.

COMPLEMENT-DEPENDENT CYTOTOXICITY (CDC) ASSAY

HLA typing began in the early 1960s as the CDC assay. Over the past 25 years, the technique has been modified and refined, but the underlying concept remains the same. In defining Class I Ags, typing sera (hence the term serology) are obtained from multiparous women who have been sensitized to paternal haplotypes during gestation. Their sera have high antibody titers directed against a limited number of HLA epitopes.^38,54 Monoclonal antibodies have also been produced to provide serotyping reagents. The standard method for HLA Class I typing is the lymphocyte microcytotoxicity assay in which multiple antisera against HLA-A, -B, and -C Ags are placed in microtiter plates and the plates are frozen until needed. About 2000 peripheral blood lymphocytes (PBLs) are added to each well in the plate, and, after a short period of incubation, complement is added and the incubation continued. Next, a vital dye, usually eosin, is added. Under phase contrast microscopy, cells lysed by the antisera and complement take up the dye and appear red, whereas intact living cells do not stain. The percentage of cells lysed by each antiserum is calculated and the HLA-A, -B, and -C phenotype can be determined on the basis of the reaction pattern.¹ The disadvantages of these techniques, when used to type in corneal transplantation, include the need to establish a cell culture requiring several weeks and the need for retina pigment epithelial cells. Retina pigment epithelial cells, where ocular MHC expression is high, are used when tissue typing is performed only for corneal transplantation. This requires the whole globe, which is not as readily available as a corneoscleral rim.^55,56

LYMPHOCYTE MICROTOXICITY ASSAY

The CDC test was never as robust for the Class II HLA-DR and HLA-DQ Ags, and these have been historically typed by the lymphocyte microlymphotoxicity assay (with the exception that the typing is performed on purified populations of B lymphocytes).⁵⁵ Banks of alloantisera and monoclonal antibodies are used to define tertiary epitopes and CREGs of HLA glycoproteins on the cell surface. Double-staining procedures can also be employed to distinguish between T and B lymphocytes without purification and separation of B cells. Typing sera are pretested to ensure that they are not detecting Class I Ags.¹ The typing on cadaveric blood must be performed within a few hours after death because of the lack of viable lymphocytes.

MIXED LEUKOCYTE REACTION (MLR)

The HLA-D Ags have historically been typed by the MLR.^11,57 Leukocytes from individuals with different HLA-D genotypes are mixed in cell culture and undergo blast transformation, DNA synthesis, and then proliferation in response to the presence of foreign HLA-D Ags on cells from the other individual. This method of HLA-D typing employs a panel of HLA-D homozygous typing cells (HTCs) displaying all of the known HLA-D surface molecules. These stimulator cells are exposed to radiation or treated with mitomycin-C to prevent DNA synthesis and proliferation, allowing for a one-way MLR when the unknown leukocytes are added. The responding lymphocytes are incubated with the HTCs for 5 days. Next, tritiated thymidine (³H-thymidine) is then added for an additional 12 to 18 hours (the allogenic MLR). After terminating the culture and isolating the cells, the amount of thymidine incorporated into the DNA measures the responder lymphocyte DNA synthesis. If the unknown responder cells synthesize DNA after incubation with a known HTC specificity, then the unknown cells are determined to not possess the same HLA-D phenotype as the HTC. If there is no DNA synthesis by unknown responder cells, then the unknown cells are determined to possess the same HLA-D Ag as the HTC. If only one HLA-D type can be found, reversing the one-way MLR allows the laboratory to determine whether the unknown cell is heterozygous or homozygous for that particular HLA-D type. In this reaction, the unknowns from a patient to be typed are irradiated (or given mitomycin-C) and used as stimulators.¹ Thus, a lack of DNA synthesis indicates that the unknown cell is HLA-D homozygous. On the other hand, DNA synthesis by the HTC indicates that the unknown cell is HLA-D heterozygous and is, therefore, recognizing a foreign HLA-D type that is not available in the HTC panel.⁵⁸

The development of immunoprecipitation, flow cytometry, and isoelectric focusing techniques in some laboratories has provided higher resolution and has enabled the definition of HLA specificities previously unidentified by CDC.⁵⁶ This leads to fewer blanks in the HLA evaluation. As molecular biology typing techniques arose, DNA amplification using polymerase chain reaction (PCR) and sequence-specific oligonucleotide probes (SSOPs) allowed for precise definition of HLA gene polymorphisms even when only a small amount of DNA was available.

ANTIGEN CROSS MATCHING

Previous transplantation, blood transfusions, or pregnancy can lead to sensitization and the production of cytotoxic antibodies against certain HLA Ags. The crossmatch test is useful in determining the presence of preformed antibodies to HLA Ags. Crossmatching is performed using the patient's most recent serum and donor lymphocytes, either PBLs or isolated T cells.

If the patient's serum kills a potential donor's cells, a positive cross match indicates the presence of preformed antibodies. A positive test is an absolute contraindication to transplantation of major organs, because a positive cross match is associated with early, rapid, uncontrollable rejection and irreversible graft loss. In addition, the patient's own cells can be cross matched with their own serum to detect the presence of any nonspecific autolymphocytotoxic antibodies. These autoantibodies may produce a false-positive crossmatch test and erroneously eliminate a potential donor.

MOLECULAR HLA TYPING

Molecular typing of HLA can be performed using DNA or RNA. Restriction fragment length polymorphic DNA probes (RFLPs) have been proven to be a useful method of typing HLA Class II DNA.⁵⁹ The pattern of DNA fragments for a given HLA gene depends on the gene sequence, the restriction endonuclease employed, and the probe used. Several studies have demonstrated the rapidity and reliability of PCR techniques in the evaluation of HLA Class I and Class II, and they are often useful when the sample of DNA is too degraded or in insufficient amounts for RFLP typing.^55,60 The use of PCR has aided in identifying about 200 DRB1 alleles, when serology could identify only 15. Currently, our institution, the University of Texas Medical Branch in Galveston, uses PCR only if there is a discrepancy in the initial serology. Widespread use of genomic typing will require a great deal of collaboration and standardization but with tremendous potential for improved accuracy, decreased cost, and increased rapidity in HLA typing.

In ophthalmology, post-mortem typing can be performed via culture of the donor retina pigment epithelium by double fluorescence⁶¹ but many HLA-A, -B, and -C specificities were declared blank.⁶² A technique in which RPE cells are cultured and stimulated to express HLA by interferon gamma, using cytotoxicity and one-dimensional isoelectric focusing, allows for HLA typing in about 14 days.⁶²

The use of PCR has revolutionized the speed for HLA type I matching, reducing the time needed from several weeks to around 5 hours.

GENETIC SUSCEPTIBILITY TO HLA-ASSOCIATED DISEASE

The first step to identification of disease susceptibility is to ensure that there is a major genetic component in the disease. This is usually estimated by studying concordance rates in identical twins compared with nonidentical twins.

Once a significant genetic component is established and the chromosomal region is determined, relative risk (RR) can be established. RR is the chance that an individual with a given disease-associated HLA Ag has of developing the associated disease compared with an individual who lacks that same Ag. RR is calculated with the formula:

RR = (p+)(c-)/(p-)(c+)

where p+ = the number of patients possessing the associated HLA Ag; c- = the number of control patients in a similar population lacking the particular HLA Ag; p- = the number of patients lacking the particular HLA Ag; and c+ = the number of controls possessing the particular HLA Ag. The higher the RR, the more frequent an Ag is seen in a given patient population of affected individuals.

The absolute risk (AR) is the chance that an individual who possesses the disease-associated HLA Ag will actually develop the disease during his or her lifetime. AR is calculated with the formula where P = the prevalence of the disease in the general population:

AR = (p+/c+) P

Thus, it is possible to quantify these concepts with data on the prototypic association between ankylosing spondylitis and HLA-B27. Eighty-nine percent of Caucasians with ankylosing spondylitis are B27 positive, compared with 9% of Caucasian controls. A rough estimate of the RR is, therefore, (p+c-)/(p-c+) = (89 × 91)/(11 × 9) = 82; Which means that an HLA-B27 positive individual is calculated to be 82 times more likely to develop ankylosing spondylitis than a B27 negative person. However, studies of the association with HLA-B27 and ankylosing spondylitis have determined the RR to be 69.1. Thus, the total number of patients and total number of controls tested dictate a more accurate RR derived directly from the data and not from estimates, taking into account the other unknown factors.

In populations in which the prevalence of clinically apparent severe ankylosing spondylitis is about 0.4%, the AR is then (89/9) × 0.004 = 0.04. Ultimately, this means that only four out of every 100 HLA-B27 positive people will actually develop clinically severe ankylosing spondylitis during their lifetime. In populations with higher prevalence rates, the AR increases accordingly.

MECHANISMS OF HLA DISEASE ASSOCIATION

There are a number of hypotheses to explain the propensity of syndromes like ankylosing spondylitis and birdshot chorioretinopathy to have such high associations with particular Ags coded by the MHC. A great deal of research has shed light on potential mechanisms and a variety of theories, which can be categorized into two main groups. Either HLA Ags are directly involved in disease pathogenesis, or genes closely linked to the HLA loci are the cause.²⁹

Direct Involvement of HLA Antigens

Receptors.

The receptor hypothesis suggests that microbes recognize specific MHC molecules and use them as a vehicle to enter the cell. Evidence that HLA-B27 may alter the uptake of Salmonella support this theory.⁶³ HLA Ags may act as receptors for pathogenic organisms, such as the case of HLA-B27 on the lymphocyte surface acting as a receptor for those organisms responsible for Reiter's syndrome, ankylosing spondylitis, or reactive arthritis.

Molecular Mimicry.

The molecular (or antigenic) mimicry hypothesis postulates that the molecular or antigenic structures of infectious agents have similarities with a particular HLA Ag on the cell surface.⁶⁴ When infected with the pathogen, the host has an elevated immune response and has antibodies against the pathogen, even without prior exposure to the pathogen.⁶⁵ One striking example of this is the amino acid homology between Klebsiella pneumoniae nitrogenase and HLA-B27.^65,66 Certain strains of Klebsiella organisms isolated from patients with ankylosing spondylitis will elicit the production of antibodies in rabbits that cross react with HLA-B27. Interestingly, these cells retain their reactivity even after multiple subcultures, suggesting that the information for the modulation of the B27 Ag is incorporated into the genome of the cell.^5,67 Another example is in multiple sclerosis in which many microbial agents have regions of sequences that may serve as binding motifs for HLA-DR2.⁶⁴

Interaction with Superantigens.

Superantigens are proteins produced by bacteria and viruses that are able to stimulate large numbers of T cells by cross linking the TCRs with the MHC Class II molecules of presenting cells. Superantigens can induce MHC Class II expression on a variety of cells and can lead to reactivation or exacerbation of autoimmune disease.⁴⁴ It is possible that there is an HLA-specific superantigen that could cause disease in this manner.¹

HLA-Derived Peptides May Be Presented by HLA Class I or II Molecules.

The most abundant naturally processed peptides from the HLA Class II molecules are those derived from other HLA molecules. It is possible that certain HLA Class II restricted T lymphocytes, stimulated by bacterial infection, may have a cross-reactive specificity for an HLA-derived peptide.¹

Involvement of Genes Closely Linked to the MHC

Disease Susceptibility Genes.

The hypothesis states that certain genes that confer susceptibility are linked to the HLA markers. This type of susceptibility is clear in cases in which formal genetic analysis has demonstrated the linkage (i.e., complement deficiency [see Fig. 1] and hemochromatosis).^5,36 Linkage disequilibrium between HLA genes and genes that actually produce disease susceptibility would explain the association between certain HLA phenotypes and specific disease states.⁶⁸ Thus, the OR 2 gene would simply be a marker for the gene responsible for multiple sclerosis.⁶⁹ Numerous other factors explain the lack of the HLA type in all persons with the associated disease, such as disease heterogeneity, environmental factors, and multigene pathogenesis. Some disease susceptibility genes may have resulted from a single mutation in the gene of an ancestor.

Immune Response (IR) Genes.

Ir genes control antibody and delayed-type hypersensitivity responses to protein Ags. There are three major mechanisms by which Ir gene products can influence immune reactivity: (1) they can dictate potential autoimmunity by controlling clonal deletion of Ag reactive cells, thereby effectively selecting the animal's immunologic repertoire; (2) they influence the antigen-presenting cells and can induce T-cell activation; and (3) they can control the predominant regulatory interaction, thereby favoring either an enhancement or a suppression of immune reactivity.⁵ Ir genes have been well mapped in the I region of the mouse, which corresponds to the D/DR loci in humans. Many diseases in people have stronger associations with Class II Ags than with Class I Ags (Table 2), which supports this hypothesis. Furthermore, some diseases are associated with more than one allele of the same locus, such as Graves' disease with B8 in Caucasians, and Bw35 in the Japanese. This type of disease association can then be explained by the different patterns of linkage disequilibria with the disease susceptibility gene in different populations.⁶⁹

Table 2. HLA Disease Associations

DISEASES ASSOCIATED WITH SPECIFIC HLA ANTIGENS

Numerous diseases seen by ophthalmologists are associated with HLA Ags. Meaningful use of HLA typing in the clinical evaluation of the uveitis patient is very much dependent on the judgment of the physician.⁴⁹ In certain circumstances, the clinical presentation is not classic or fully expressed, leading to the consideration of several diagnoses. It is in this context that typing for specific HLA Ags is most likely to be useful, adding one bit more of information to the selection of a likely diagnosis.^70,71 Complete HLA typing is never indicated in the uveitis workup, unless, of course, a study looking for HLA associations is being undertaken in a distinct population.

Because HLA-B27 occurs in one of every 13 or 14 healthy individuals and the overwhelming majority of HLA-B27 positive patients do not have the associated disease,¹⁴ testing for HLA-B27 may be of only limited usefulness in the evaluation of arthritis patients by the internal medicine specialist.^69,72 Likewise, routine testing for HLA-B27 in all patients with uveitis is certainly not indicated; only those patients in whom the characteristic disease pattern of acute, alternating, recurrent nongranulomatous iridocyclitis is present are likely to test positive, and a positive test may not change the management of the patient. Careful evaluation of the patient's history may reveal back problems, dysentery, nongonococcal urethritis, or colitis, all of which should prompt consideration of ordering a B27 test and discussing the findings with the patient's internist. HLA typing has been used to help with the diagnosis of Behçet's disease, but it is not included in the current International Study Group Criteria.⁷³ Occasionally, HLA-B27-related uveitis may present with vitreitis or as an acute fibrinous panuveitis with hypopyon, in which case a positive B27 test can be reassuring in the absence of diagnostic information implicating a more devastating disease such as endophthalmitis. Similarly, HLA-DR53 testing in non-Japanese patients with suspected Vogt-Koyanagi-Harada syndrome is likely to be of low yield, because this association has been made in only one racial group.

HLA B27-Associated Diseases

HLA-B27 represents a family of 25 different alleles that encode 23 different proteins. Since 1972, HLA-B27 has been identified as associated with ankylosing spondylitis.¹³ The initial observations of HLA-B27 associated with ankylosing spondylitis was made in Caucasians from Europe and North America. A twin survey, however, suggests that HLA-B27 provides only about 16% of the overall genetic susceptibility to ankylosing spondylitis.¹⁴ HLA-B27 has also been found to be significantly associated with acute iridocyclitis, regardless of the presence of a seronegative spondyloarthropathy such as ankylosing spondylitis.⁷⁴ This association includes studies of patient populations with a diagnosis of acute anterior uveitis (AAU);^75,76 unspecified uveitis;⁷⁷ adult iridocyclitis; and pauciarticular, late-onset juvenile rheumatoid arthritis (JRA).⁷⁸ Interestingly, adult iridocyclitis and AAU in American black patients are associated with the HLA-B8 alloantigen, and not HLA-B or HLA-D.⁷⁶

HLA-B27 is also associated with seronegative arthropathies, whether or not uveitis is present. The seronegative arthropathies (in which the rheumatoid factor is consistently negative) include Reiter's syndrome, psoriatic arthropathy, inflammatory bowel disease (ulcerative colitis), Crohn's disease, postdiarrheal reactive arthritis, sexually acquired postchlamydial and postgonorrheal reactive arthritis, late-onset pauciarticular JRA, and ankylosing spondylitis.

A number of other diseases display somewhat weaker associations with HLA-B27. These include asbestosis, chronic brucellosis, granulocytic leukemia, meningococcal meningitis, rheumatoid arthritis, keratoconus and Whipple's disease.⁷⁰

Many published reports have associated HLA-B27-positive patients with AAU¹³ as more likely to be younger at the age of onset and male and to have frequent unilateral alternating eye involvement, and a higher incidence of ocular complications.^13,79 Clinically, AAU patients lack mutton-fat keratic precipitates, have associated seronegative spondyloarthropathies, and tend to have hyperacute presentations including fibrinoid anterior chamber reactions and severe vitreitis.^70,80 Ordering the HLA-B27 typing as a diagnostic test may help in predicting prognosis.^71,75

B27-related iridocyclitis must be treated aggressively and early with topical, systemic, or injectable anti-inflammatory medications. Inadequately treated B27 disease can easily deteriorate into chronic, recalcitrant, bilateral uveitis.

Ankylosing Spondylitis

The 1973 discovery that ankylosing spondylitis was associated with the B27 locus dramatically increased the enthusiasm for pursuing the study of the histocompatibility Ags and immunologic epidemiology.¹³ Population prevalence rates indicate that a B27-positive person of either sex has about a 20% probability of developing ankylosing spondylitis during his or her lifetime. In addition, HLA-B60 has been shown to increase the risk for ankylosing spondylitis threefold regardless of HLA-B27 status. Thus, the prevalence is as high as 10 to 15 cases per 1000 in some populations.⁷² Approximately 25% of patients with ankylosing spondylitis will develop iridocyclitis (Fig. 3). The characteristic changes involving the vertebral column and sacroiliac joints in these patients are also found in the other seronegative spondyloarthropathies.^67,75,79

Reiter's Syndrome

Reiter's syndrome is a postinfectious autoimmune disease that has strong associations with the HLA-B27 locus. The classic triad of polyarthritis, papillary conjunctivitis, and urethritis, as well as highly characteristic mucocutaneous lesions, are the most specific signs of Reiter's syndrome.⁶⁶ Uveitis and iritis are also prominent ocular findings of this disease.

Sexually transmitted chlamydial infection, or acute dysentery resulting from gram-negative organisms such as Klebsiella, Salmonella, Shigella, or Yersinia species often precedes the development of joint inflammation, hence the term “reactive arthritis.” Thirty-five percent of patients with sexually acquired reactive arthritis (SARA) following nongonococcal urethritis have conjunctivitis and other lesions characteristic of the full complex of Reiter's syndrome. There are a variety of theories to explain the role of HLA-B27 in the pathogenesis of reactive arthritis. It is likely that the features of HLA-B27 as an antigen-presenting molecule; the possible role of the free B27 heavy chains; and the ability to bind long extracellular peptides, as well as the single cysteine at position 67 in the peptide chain, may all play a role in the pathogenesis.^113,79

Birdshot Chorioretinopathy

Birdshot chorioretinopathy is a clinically distinct form of chronic posterior uveitis with bilateral, multiple, hypopigmented postequatorial retinal pigment epithelial (RPE) and choroidal inflammatory lesions (Fig. 4). This disease is often complicated by cystoid macular edema, posterior synechiae, cataract, and vitreitis. The association between birdshot chorioretinopathy and HLA-A29, which has been documented as high as 95.6% with a RR factor of 224.35, is probably the strongest established to date for any disease.^81,82

Presumed Ocular Histoplasmosis Syndrome (POHS)

POHS is a triad of peripheral, punched-out chorioretinal atrophic lesions; peripapillary atrophy; and macular subretinal neovascular membrane formation specifically in the absence of vitreitis. It is strongly associated with skin test responsiveness to Ag from the filamentous fungus Histoplasma capsulatum, which is endemic to the Mississippi River valley. Certain individuals in endemic areas may have an increased susceptibility to POHS following exposure to the fungus, as well as an additional propensity to develop visually disabling macular lesions. HLA-B7 is strongly associated with disciform macular lesions⁸³ but not with peripheral atrophic scars seen in POHS. HLA-DR2, on the other hand, is associated with both macular and peripheral scarring,⁸⁴ which incriminates a genetic predisposition to macular neovascularization in this syndrome and apparently has a differing genetic risk than multifocal choroiditis and panuveitis.⁸⁴

Vogt-Koyanagi-Harada (VKH) Disease

VKH disease is a rare multisystem disease affecting the eye, inner ear, the skin, and the central nervous system.⁸⁵ It is believed to be due to an autoimmune condition in which cytotoxic T cells target the melanocytes leading to uveal depigmentation, poliosis, and vitiligo. Patients present initially with iridocyclitis and choroiditis, occasionally severe enough to produce an exudative retinal detachment. Many of the initial studies on this syndrome demonstrating significant HLA associations have been performed on Japanese populations. The syndrome has been associated with HLA-Bw54 (a split of HLA-Bw22 also known as Bw22J), HLA-DwWa (a new D locus Ag), and HLA-DR4. This syndrome has been described in Caucasians, in whom these Ags do not appear to be associated with the disease process.⁸⁴ VKH has been demonstrated to have combined allelic predisposition in which DQA1*0301 acts as the primary factor and HLA-DR4 acts as an additive factor in the development of the disease. There is also a negative association of DQB1*0604; it has been proposed that DQB1*0604 may provide protection.⁸⁷ There have been some case reports of VKH syndrome following cutaneous injury, which suggests that this disorder may result from systemic sensitization to shared melanocytic Ags.⁸⁸ Ethnicity is also a complicating factor with HLA association, because VKH is more common in persons with higher levels of skin pigmentation. In non-Japanese studies, HLA-DRB1*0405 has been associated with Brazilian patients,⁸⁹ and DRB1*0101was associated with Mexican populations.⁸⁹

Sympathetic Ophthalmia

Sympathetic ophthalmia is a rare uveitic syndrome seen in patients who had a prior history of penetrating trauma. A Japanese study found these patients to have HLA associations similar to those found in VKH^85,90,91 (see Table 2). Experimental autoimmune uveitis is an animal model of disease that has similar findings to sympathetic ophthalmia.⁹² Studies of this disease have demonstrated the possibility of retina S-Ag as a factor in this disease.⁹³

Glaucomatocyclitic Crisis

Also known as the Posner-Schlossman syndrome,⁹⁴ glaucomatocyclitic crises are heralded by acute, unilateral attacks of often severe ocular hypertension associated with mild iridocyclitis. A strong association with HLA-Bw54, determined to not be related to linkage disequilibrium, has been described in Japanese patients with this syndrome.⁹⁵

Pseudoexfoliation

Patients with glaucoma resulting from capsular pseudoexfoliation were more likely found to be HLABw35 positive than the other study patients with chronic, open-angle glaucoma in a Scandinavian study (Fig. 5).⁹⁶ This conclusion was not confirmed in a subsequent study on a similar population.⁹⁷ In an Irish study, an HLA association with pseudoexfoliation is identified for 14 Ags. Eleven Ags (HLA-A1, -A33, -B8, -B47, -B51, -B53, -B57, -B62, -DR3, -DR12, and -DR13) are significantly more common in the pseudoexfoliation group, whereas 3 Ags (HLA-B12, -B17, and -DR2) are significantly less common.⁹⁸ Once again, more extensive studies are needed to confirm the association.

Behçet's Disease

Behçet's disease is a multisystem vasculitic disorder with widespread clinical manifestations, including iritis, recurrent oral and genital ulcers, occlusive retinal vasculitis, arthritis, central nervous system involvement, and skin lesions (Fig. 6). An association of HLA-B5 was first described in Japanese patients in 1973.⁹⁹ Later this was confirmed and further specified as an association with HLA-B51 and not of an associated with HLA-B52 (the two split products of HLA-B5). This association was found in Turkish, Jewish, English, and Mediterranean populations.^100,101 Many of these populations are found along the ancient east and west passage known as the Silk Route that was used in medieval times to bring the material to Europe.¹⁰² The HLA association is less clear in Northwestern Europeans and Americans. Early studies in the 1970s did not show an association, but later findings found an increased frequency of HLA-B51 associated with those of Irish, Scottish, and German descent (far from the silk route).¹⁰⁰ A separate study failed to find an association with those in an English clinic.¹⁰³

Some authorities classify Behçet's disease into four subtypes: mucocutaneous, arthritic, necrologic, and ocular. HLA-B12 appears to be elevated in the mucocutaneous and arthritic groups, whereas HLA-B5 is increased in the ocular subgroup.¹⁰⁴ About 70% of patients with Behçet's have ocular involvement,¹⁰¹ but there are no immunologic predictors indicating which patients will develop ocular disease. Ocular involvement may be the presenting sign in 20%. Refractory Behçet's disease may be associated with HLA-DQ genomic variations as detected by restriction endonuclease studies.¹⁰³

Juvenile Rheumatoid Arthritis (JRA)

Patients with pauciarticular JRA are particularly prone to uveitis, whereas those with Still'sdisease (systemic onset polyarthritis) and polyarticular rheumatoid factor positive JRA are usually spared.¹⁰⁵ Still's disease is associated with HLABw35, whereas rheumatoid factor positive polyarticular JRA is associated with DR4. Further associations with DRB1*04 were observed in adults. DPB1*0201 was increased in the persistent pauciarticular group. DPB1*0402 was decreased in all pauciarticular groups with or without conversion.¹⁰⁶ HLA-DR5 is associated with the early onset form of pauciarticular JRA seen most frequently in girls. Fifty-three percent of these children are likely to develop chronic bilateral ocular disease, band keratopathy, synechiae, and cataract and show positive antinuclear antibody (ANA) titers.⁶⁷ Later onset pauciarticular JRA is more frequently seen in males and is a subset of HLA-B27 disease. Twenty-five percent of these patients develop uveitis. In poorly characterized JRA, the use of HLA typing for specific Ags may suggest the prognosis for ocular involvement.

Nonuveitic Corneal and Conjunctival Disease with HLA Associations

The first reports of association of keratoconus with HLA was by Gasset and colleagues¹⁰⁷ who found one family in which a mother and her two children with severe keratoconus had HLA-B27 positivity; HLA-B27 negativity was found in the family members with no evidence of keratoconus. Keratoconus has been found to have HLA association in familial cases in Finland.¹⁰⁸ More recently, the HLA-B27 Ag was demonstrated in the phenotype and increased incidence of haplotypes A2-B27 and A3-B35 were typical of patients with the progressive course of keratoconus, whereas in the acute form of keratoconus, there was an increased incidence of HLA-A1, -A19, -B7, and complete absence of HLA-A2.¹⁰⁹ Other studies failed to demonstrate an association. More extensive and reliable studies are needed to determine the association of keratoconus and HLA.

In Cogan's early descriptions of what has ultimately been called Cogan's syndrome, three of the four patients were found to have W-17 histocompatability Ag positivity.¹¹⁰ In a later study, HLA typing of 10 patients with well-documented Cogan's syndrome failed to support previous reports of an increased incidence of HLA-B17. Although HLA-A9 (Aw24), Bw35, and Cw4 appeared increased in frequency among patients, the small number of cases precluded attaching importance to the significance of these increased frequencies.¹¹¹

Mooren's ulcer has been found to have an increased frequency of HLA-DR17 and -DQ2.¹¹² An aberrant expression of HLA-DR Ag was found in keratoconjunctival epithelial cells and keratocytes in the corneal stroma biopsied from patients with Mooren's ulcer, implicating an excessive autoimmune reactivity as possibly the direct cause of Mooren's ulcer.¹¹³

Ocular cicatricial pemphigoid, an autoimmune mediated disorder (Fig. 7), was initially reported to be associated with HLA-B12, whereas later studies did not confirm this association. Other studies found that HLA-DR-4, DR-5, DQw3, B8, B35, and B49 have associations. With the use of RFLPs, HLA-DQB1*030 was confirmed in several studies to have an association.¹¹⁴ This association has led to further studies demonstrating that cell-mediated immunity, rather than antibody-mediated immunity may be a critical cause of the scarring in this disease.¹¹⁴

Stevens-Johnson syndrome has been associated with the DQB*0601 allele, with a RR of 7.2.¹¹⁵ HLA-Bw44 (a split from HLA-B12, which is of interest because of the association with ocular cicatricial pemphigoid) was also found to have an association with Stevens-Johnson syndrome.¹¹⁶ The HLA association may be one of several factors contributing to this autoimmune disease, because immune complexes have been demonstrated in the blood and immune globulins are deposited at the dermal-epidermal junction.¹¹⁵

Thygeson's superficial punctate keratopathy has been associated with HLA-DR3.¹¹⁷

Ophthalmic Diseases with Conflicting Reports of HLA Association

As more information is gathered about HLA and their associations with disease, it becomes clearer that HLA polymorphisms appear to be responsible for genetic variations in the extent of the immune response seen in different individuals in response to different Ags. There have been many studies in which no MHC Ags could be linked to ophthalmic diseases. In the field of cataract, senile cataract has not been found to have any HLA association.¹¹⁸ Diabetes, mitochondrial myopathies, and uveitis are all associated with the formation of cataracts, but there has been no demonstrable HLA association with the cataract.¹¹⁹

Primary open-angle glaucoma (POAG) is familial in 20% of the cases; however, even with the considerable evidence of genetic factors in POAG, there have been inconsistent reports.¹²⁰ Various studies demonstrate association with increased and decreased association with various HLA types; however, the associations with HLA-D and -DR have not yet been studied in detail.^119,120 There have been no HLA associations on the A or B loci found for acute angle closure glaucoma.^120–122 No association was found with pigment dispersion glaucoma,^123,124 oral steroid-induced ocular hypertension and steroid-induced cataract, or chronic uveitis.

There have been inconsistent reports of HLA associations with herpetic keratitis.

HLA Antigens and Fundus Diseases

Retinoblastoma (Rb) was found to have a possible HLA association in a small study in the 70s,¹²⁵ whereas another study of 16 Rb patients did not confirm this.¹²⁶ Interestingly, when a patient has Rb, it has been shown that the expression of Rb leads to modification of the chromatin environment of the HLA-DRA promoter independent of transcription.¹²⁷

Eales's disease was recently demonstrated to have a positive disequilibrium of haplotypes A3-B44 and A11-B12;¹²⁸ however, there was no association demonstrated in other studies.

Choroidal malignant melanoma (Fig. 8) has been associated with HLA-Aw32¹²⁹ and a loss of heterozygosity on chromosome 6,¹³⁰ but a separate study of 514 patients with melanoma in other parts of the body and a study of 64 patients with uveal melanoma failed to reveal any significant association.¹³¹ No HLA association was demonstrated with central serous retinopathy,¹²⁹ chorioretinitis, toxoplasmic retinochoroiditis, chronic juvenile iridocyclitis, age-related disciform macular degeneration, retinitis pigmentosa, or rhegmatogenous retinal detachment. Proliferative vitreoretinopathy (PVR) was found to have increased HLA-DR expression on RPE cells; this could play a role in triggering a local immune response in PVR.¹³² Sympathetic ophthalmia was found in an early study to have no HLA association,⁹⁰ whereas in a Japanese study, a significant association was found with HLA-DR4 and DR53⁹¹; and in North American patients, an association with DR-4 was noted.⁸⁵

Ophthalmic Diseases with Systemic Components and HLA

Sarcoidosis has not been found to have a consistent HLA association.¹³³

Diabetes mellitus has been evaluated closely since 1972 for HLA associations.¹³⁴ There has been an increased risk of insulin-dependent diabetes mellitus (IDDM) in HLA-Dw3 or DRw3 positive patients. In retinopathy, a variety of HLA associations have been suggested including the risk for advanced retinopathy in HLA-B15 positive patients and more common severe retinopathy in patients with DR3-DQ2/DR4-DQ8 alleles.¹³⁵

Temporal arteritis has been found to have an association with HLA-DRB1 *04 alleles.^136–138

There is considerable evidence that Graves' disease is of autoimmune etiology, and there have been several studies suggesting a significant association between HLA-B8 and Graves' disease in Caucasian populations.^{21,22,120,139} Molecular mimicry appears to play a role in the pathogenesis.

It is not entirely clear whether HLA matching is beneficial to the survival of uncomplicated, nonvascularized corneal transplants for conditions such as keratoconus or central corneal scarring. Very favorable rejection rates have also been attained in high-risk patients by obtaining donor tissue with HLA Ags to which the recipient was not sensitized. Previous sensitization is determined by a lymphocyte crossmatch study. Two disincentives to HLA matching is the high degree of polymorphism and the cost and the delay in time with traditional serology typing techniques.

Studies evaluating HLA-matching for stem cell transplantation have demonstrated a possible benefit with HLA-matching. One patient with a complete HLA mismatch had poor epithelial healing and a trophic ulcer. Because the conjunctiva is a vascularized tissue, some groups have decided to use donors with at least a 50% HLA match.¹⁵⁰

HLA expression is thought to be altered in the face of herpes infection, autoimmune reactions, and vascularization.¹⁵¹ This elevated levels of HLA may make these corneas more susceptible to attack by cytotoxic cells and hence graft rejection.¹⁵² However, the genetic susceptibility to corneal herpetic disease appears to be dependent on non-MHC loci.⁴² Graft rejection in patients with previous herpes infection has been shown to be associated with induction of Class I HLA-ABC and Class II HLA-DR Ags on cells in the donor button by a factor associated with cellular inflammation.¹⁵³

COLLABORATIVE CORNEAL TRANSPLANTATION STUDY

The CCTS is addressing the value of donor-recipient histocompatibility matching by pooling high-risk recipients and donor tissue through a nationwide network for donor selection and tissue distribution.¹⁵⁴ This multicenter, randomized, double-masked clinical trial is assessing recipients for detectable lymphocytotoxic antibodies. Those who have antibodies will be assigned randomly to receive either a negatively or positively crossmatched cornea. Patients without detectable antibodies are assigned donor tissue with varying degrees of HLA matching. They found that there was no difference in highly matched (zero or one HLA-A, -B mismatch) or low matched (two or more mismatched transplants).^3,40,148 In fact, an inverse correlation for HLA-DR was found, in which incompatible grafts had a superior survival rate compared with matched grafts.¹⁵⁵ One confounding factor for the CCTS is the aggressive use of immunosuppressants leading to a high overall survival of the high-risk grafts (65% at 3 years).¹⁴⁸ The CCTS group did find a significant survival advantage with blood group compatibility, which was previously believed to have little or no relevance. A non-CCTS, retrospective analysis found rejection in 1681 patients to be almost 14 times greater in recipients with HLA-AB and HLA-DR mismatched donor corneas.¹⁴⁹ The diversity of experimental results and complexity of the HLA system confirms the need for more study of the benefit of HLA matching for corneal transplantation.

1. Lechler R, Warrens AN. HLA in Health and Disease. 2nd ed. San Diego, CA: Academic, 2000:xvii, 472

2. Bach FH. Immunogenetics of HLA. Wien Klin Wochenschr 1991;103:141

3. Taylor CJ, Dyer PA. Histocompatibility antigens. Eye 1995;9:173

4. Wordsworth P. HLA nomenclature: A user-friendly system? Ann Rheum Dis 1994;53:153

5. Zmijewski CM. HLA and disease. Crit Rev Clin Lab Sci 1984;20:285

6. Trowsdale J, Campbell RD. Complexity in the major histocompatibility complex. Eur J Immunogenet 1992;19:45

7. Bodmer WF. HLA: What's in a name? A commentary on HLA nomenclature development over the years. Tissue Antigens 1997;49:293

8. Bodmer JG, Marsh SG, Albert ED et al. Nomenclature for factors of the HLA system, 1994. Eur J Immunogenet 1994;21:485

9. Kasahara M. Major Histocompatibility Complex: Evolution, Structure, and Function. Tokyo; New York: Springer, 2000:xxiii, 561

10. Holmes N, Parham P. Exon shuffling in vivo can generate novel HLA Class I molecules. EMBO J 1985;4:2849

11. Schreuder GM, Hurley CK, Marsh SG et al. The HLA dictionary 1999: A summary of HLA-A, -B, -C, -DRB1/3/4/5, -DQB1 alleles and their association with serologically defined HLA-A, -B, -C, -DR and -DQ antigens. Tissue Antigens 1999;54:409

12. Wujciak T, Opelz G. Evaluation of HLA matching for CREG antigens in Europe. Transplantation 1999;68:1097

13. Reveille, JD. HLA-B27 and the seronegative spondyloarthropathies. Am J Med Sci 1998;316:239

14. Reveille JD, Ball EJ, Khan MA. HLA-B27 and genetic predisposing factors in spondyloarthropathies. Curr Opin Rheumatol 2001;13:265

15. Carrington M. Recombination within the human MHC. Immunol Rev 1999;167:245

16. Alfonso C, Karlsson L. Nonclassical MHC Class II molecules. Annu Rev Immunol 2000;18:113

17. Fischer GF. Molecular genetics of HLA. Vox Sang 2000;78(Suppl 2):261

18. Davidson JA, Kippax RL, Dyer PA. A study of HLA-A, B, DR and Bf bearing haplotypes derived from 304 families resident in the north west of England. J Immunogenet 1988;15:227

19. Mittal KK. Immunobiology of the human major histocompatibility complex: Association of HLA antigens with disease. Acta Anthropogenet 1984;8:245

20. Schreuder GM, Hurley CK, Marsh SG et al. HL-A antigen in Sjögren's syndrome. Ann Ophthalmol 1980;12:1084

21. Marga M, Denisova A, Sochnev A et al. Two HLA DRB 1 alleles confer independent genetic susceptibility to Graves disease: Relevance of cross-population studies. Am J Med Genet 2001;102:188

22. Philippou G, Krimitzas A, Kaltsas G et al. HLA DQA1*0501 and DRB1*0301 antigens do not independently convey susceptibility to Graves' disease. J Endocrinol Invest 2001;24:88

23. Browning MJ, McMichael AJ. HLA and MHC: Genes, Molecules, and Function. Human Molecular Genetics Series. Oxford, UK; Herndon, VA: BIOS Scientific, 1996:xvii, 438

24. Hood L, Campbell JH, Elgin SC. The organization, expression, and evolution of antibody genes and other multigene families. Annu Rev Genet 1975;9:305

25. Huang SK, Yi M, Kumai M et al. Immunogenetic aspects of IgE-mediated responses. Adv Exp Med Biol 1994;347:11

26. Le Bouteiller P, Blaschitz A. The functionality of HLA-G is emerging. Immunol Rev 1999;167:233

27. Nelson CA, Fremont DH. Structural principles of MHC Class II antigen presentation. Rev Immunogenet 1999;1:47

28. O'Callaghan CA, Bell JI. Structure and function of the human MHC Class Ib molecules HLA-E, HLA-F and HLA-G. Immunol Rev 1998;163:129

29. Natarajan K, Li H, Mariuzza RA et al. MHC Class I molecules, structure and function. Rev Immunogenet 1999;1:32

30. Le Bouteiller P. HLA Class I chromosomal region, genes, and products: Facts and questions. Crit Rev Immunol 1994;14:89

31. Volpi EV, Chevret E, Jones T et al. Large-scale chromatin organization of the major histocompatibility complex and other regions of human chromosome 6 and its response to interferon in interphase nuclei. J Cell Sci 2000;113:1565

32. Morita N, Munkhbat B, Gansuvd B et al. Effect of HLA-A and -DPB1 matching in corneal transplantation. Transplant Proc 1998;30:3491

33. Urban RG, Chicz RM. MHC molecules: Expression, Assembly, and Function. Medical intelligence unit. New York; Austin: Chapman & Hall; RG Landes, 1996:298

34. Woolfrey AE, Nepom GT. Differential transcription elements direct expression of HLA-DQ genes. Clin Immunol Immunopathol 1995;74:119

35. Sonderstrup G, McDevitt HO. DR, DQ, and you: MHC alleles and autoimmunity. J Clin Invest 2001;107:795

36. Truedsson L, Alper CA, Awdeh ZL et al. Characterization of type I complement C2 deficiency MHC haplotypes. Strong conservation of the complotype/HLA-B-region and absence of disease association due to linked Class II genes. J Immunol 1993:151:5856

37. Treseler PA, Foulks GN, Sanfilippo F. The expression of HLA antigens by cells in the human cornea. Am J Ophthalmol 1984;98:763

38. Carpenter CB. HLA Class I DNA typing in organ transplantation. Tissue Antigens 1997;50:322

39. Niederkorn JY, Chiang EY, Ungchusri T et al. Expression of a nonclassical MHC Class Ib molecule in the eye. Transplantation 1999;68:1790

40. Batchelor JR, Casey TA, Werb A et al. HLA matching and corneal grafting. Lancet 1976;1:551

41. Nelson KA. Human leukocyte antigen system and the immune response to it. Clin Orthop 1996;326:35

42. Streilein JW, Dana MR, Ksander BR. Immunity causing blindness: Five different paths to herpes stromal keratitis. Immunol Today 1997;18:443

43. Sompayrac L. How the Immune System Works. Malden, MA: Blackwell Science, 1999:vii, 111

44. Davey MP, Rosenbaum JT. The human leukocyte antigen complex and chronic ocular inflammatory disorders. Am J Ophthalmol 2000;129:235

45. Abi-Hanna D, Wakefield D, Watkins S. HLA antigens in ocular tissues. I. In vivo expression in human eyes. Transplantation 1988;45:610

46. Steptoe RJ, Holt PG, McMenamin PG. Origin and steady-state turnover of major histocompatibility complex Class II-positive dendritic cells and resident-tissue macrophages in the iris of the rat eye. J Neuroimmunol 1996;68:67

47. Steptoe RJ, Holt PG, McMenamin PG. Functional studies of major histocompatibility Class II-positive dendritic cells and resident tissue macrophages isolated from the rat iris. Immunology 1995;85:630

48. Diaz-Araya CM, Madigan MC, Provis JM et al. Immunohistochemical and topographic studies of dendritic cells and macrophages in human fetal cornea. Invest Ophthalmol Vis Sci 1995;36:644

49. McCluskey J, Peh CA. The human leucocyte antigens and clinical medicine: An overview. Rev Immunogenet 1999;1:3

50. Fernandez N, Butcher G. MHC: A Practical Approach. Practical approach series. Oxford; New York: IRL Press at Oxford University Press, 1997:v, 180

51. Whitton JL. Antigen Presentation. Current Topics in Microbiology and Immunology, 232. Berlin; New York: Springer, 1998:vi, 244.

52. Singh N, Agrawal S, Rastogi AK. Infectious diseases and immunity: Special reference to major histocompatibility complex. Emerg Infect Dis 1997;3:41

53. Viret C, Janeway CA Jr. MHC and T cell development. Rev Immunogenet 1999;1:91

54. Begovich AB, Erlich HA. HLA typing for bone marrow transplantation. New polymerase chain reaction-based methods. JAMA 1995;273:586

55. Wernet P, Kogler G, Enczmam J et al. Rapid method for successful HLA Class I and II typing from cadaveric blood for direct matching in cornea transplantation. Graefes Arch Clin Exp Ophthalmol 1998;236:507

56. Bray RA. Flow cytometry in human leukocyte antigen testing. Semin Hematol 2001;38:194

57. Erlich HA, Opelz G, Hansen J. HLA DNA typing and transplantation. Immunity 2001;14:347

58. Odum N. Cellular typing for HLA-DF: Bulk-expansion of primed lymphocyte typing (PLT) cells. Technical and immunogenetic aspects. Dan Med Bull 1989;36:551

59. Munkhbat B, Hagihara M, Sato T et al. HLA Class II DNA typing using ocular tissue and its usefulness in corneal transplantation. Transplant Proc 1996;28:1257

60. Blake E, Mihalovich J, Higuchi R et al. Polymerase chain reaction (PCR) amplification and human leukocyte antigen (HLA)-DQ alpha oligonucleotide typing on biological evidence samples: Casework experience. J Forensic Sci 1992;37:700

61. Baumgartner I, Huber-Spitzy V, Grabner G et al. HLA typing from human donor eyes. Graefes Arch Clin Exp Ophthalmol 1989;227:541

62. Zavazava N, Halene M, Westphal E et al. Expression of MHC Class I and II molecules by cadaver retinal pigment epithelium cells: Optimization of post-mortem HLA typing. Clin Exp Immunol 1991;84:163

63. Kapasi K, Inman RD. ME1 epitope of HLA-B27 confers Class I-mediated modulation of gram- negative bacterial invasion. J Immunol 1994;153:833

64. Liblau R, Gautam AM. HLA, molecular mimicry and multiple sclerosis. Rev Immunogenet 2000;2:95

65. Ebringer A, Ahmadi K, Fielder M et al. Molecular mimicry: The geographical distribution of immune responses to Klebsiella in ankylosing spondylitis and its relevance to therapy. Clin Rheumatol 1996;15(Suppl 1):57

66. Schwimmbeck PL, Yu DT, Oldstone MB. Autoantibodies to HLA B27 in the sera of HLA B27 patients with ankylosing spondylitis and Reiter's syndrome. Molecular mimicry with Klebsiella pneumoniae as potential mechanism of autoimmune disease. J Exp Med 1987;166:173

67. Yu DT, Choo SY, Schaack T. Molecular mimicry in HLA-B27-related arthritis. Ann Intern Med 1989;111:581

68. Bittencourt PL, Goldberg AC, Cancado EL et al. Different HLA profiles confer susceptibility to autoimmune hepatitis type 1 and 2. Am J Gastroenterol 1998;93:1394

69. Thomson G. HLA disease associations: Models for the study of complex human genetic disorders. Crit Rev Clin Lab Sci 1995;32:183

70. Huhtinen M, Karma A. HLA-B27 typing in the categorisation of uveitis in a HLA-B27 rich population. Br J Ophthalmol 2000;84:413

71. Kijlstra A. The value of laboratory testing in uveitis. Eye 1990;4:732

72. Allen RL, Bowness P, McMichael A. The role of HLA-B27 in spondyloarthritis. Immunogenetics 1999;50:220

73. Criteria for diagnosis of Behçet's disease. International Study Group for Behçet's Disease. Lancet 1990;335:1078

74. Zierhut M, Feltkamp B, Forrester J et al. Immunology of the eye and the joint. Immunol Today 1994;15:249

75. Dodds EM, Lowder CY, Meisler DM. Posterior segment inflammation in HLA-B27+ acute anterior uveitis: Clinical characteristics. Ocul Immunol Inflamm 1999;7:85

76. Saari KM, Hakli J, Solja J et al. HLA-B27 frequency and MLC reactions in acute anterior uveitis. Albrecht Von Graefes Arch Klin Exp Ophthalmol 1981;216:23

77. Feltkamp TE. HLA and uveitis. Int Ophthalmol 1990;14:327

78. Shinzato M, Yamamoto J, Hirata CE et al. Eye disease in a patient with rheumatoid arthritis. Postgrad Med J 1999;75:676

79. Lyons JL, Rosenbaum JT. Uveitis associated with inflammatory bowel disease compared with uveitis associated with spondyloarthropathy. Arch Ophthalmol 1997;115:61

80. Schlosstein L, Terasaki PI, Bluestone R et al. High association of an HL-A antigen, W27, with ankylosing spondylitis. N Engl J Med 1973;288:704

81. Jabs DA, Rosenbaum JT, Foster CS et al. Guidelines for the use of immunosuppressive drugs in patients with ocular inflammatory disorders: Recommendations of an expert panel. Am J Ophthalmol 2000;130:492

82. Baarsma GS, Priem HA, Kijlstra A. Association of birdshot retinochoroidopathy and HLA-A29 antigen. Curr Eye Res 1990;9(Suppl):63

83. Meredith TA, Smith RE, Duquesnoy RJ. Association of HLA-DRw2 antigen with presumed ocular histoplasmosis. Am J Ophthalmol 1980;89:70

84. Spaide RF, Skerry JE, Yannuzzi LA et al. Lack of the HLA-DR2 specificity in multifocal choroiditis and panuveitis. Br J Ophthalmol 1990;74:536

85. Davis JL, Mittal KK, Freidlin V et al. HLA associations and ancestry in Vogt-Koyanagi-Harada disease and sympathetic ophthalmia. Ophthalmology 1990;97:1137

86. Richard G, Dieckhues B, Junemann G. [Immunological examinations in Vogt-Koyanagi-Harada disease (author's transl)]. Albrecht Von Graefes Arch Klin Exp Ophthalmol 1981;217:325

87. Islam SM, Numaga J, Fujino Y et al. HLA Class II genes in Vogt-Koyanagi-Harada disease. Invest Ophthalmol Vis Sci 1994;35:3890

88. Rathinam SR, Nampcrumalsamy P, Nozik RA et al. Vogt-Koyanagi-Harada syndrome after cutaneous injury. Ophthalmology 1999;106:635

89. Goldberg AC, Yamamoto JH, Chiarella JM et al. HLA-DRB1*0405 is the predominant allele in Brazilian patients with Vogt-Koyanagi-Harada disease. Hum Immunol 1998:59:183

90. Reynard M, Shulman IA, Azen SP et al. Histocompatibility antigens in sympathetic ophthalmia. Am J Ophthalmol 1983;95:216

91. Shindo Y, Ohno S, Usui M et al. Immunogenetic study of sympathetic ophthalmia. Tissue Antigens 1997;49:111

92. Biswas J, Raizada S. Immunopathology of uveitis. Indian J Lepr 1998;70:11

93. Nussenblatt RB. Proctor Lecture. Experimental autoimmune uveitis: Mechanisms of disease and clinical therapeutic indications. Invest Ophthalmol Vis Sci 1991;32:3131

94. Malinovsky VE. Glaucomatocyclitic crisis. J Am Optom Assoc 1983;54:1069

95. Hirose S, Ohno S, Matsuda H. HLA-Bw54 and glaucomatocyclitic crisis. Arch Ophthalmol 1985;103:1837

96. Olivius E, Polland W. Histocompatibility (HLA) antigens in capsular glaucoma and simplex glaucoma. Acta Ophthalmol Copenh 1980;58:406

97. Slagsvold JE, Nordhagen R. The HLA system in primary open angle glaucoma and in patients with pseudoexfoliation of the lens capsule (exfoliation or fibrillopathia epitheliocapsularis). Acta Ophthalmol Copenh 1980;58:188

98. FitzSimon JS, Mulvihill A, Kennedy S et al. Association of HLA type with pseudoexfoliation of the lens capsule. Br J Ophthalmol 1996;80:402

99. Ohno S, Kimura SJ, O'Connor GR et al. HLA antigens and uveitis. Br J Ophthalmol 1977;61:62

100. Kilmartin DJ, Finch A, Acheson RW. Primary association of HLA-B51 with Behçet's disease in Ireland. Br J Ophthalmol 1997;81:649

101. Yazici H, Kimura SJ, O'Connor GR et al. HLA antigens in Behçet's disease: A reappraisal by a comparative study of Turkish and British patients. Ann Rheum Dis 1980;39:344

102. Mizuki M, Ohno S, Ando H et al. Major histocompatibility complex Class II alleles in Kazak and Han populations in the Silk Route of northwestern China. Tissue Antigens 1997:50:527

103. Gul A, Uyar FA, Inanc M et al. Lack of association of HLA-B*51 with a severe disease course in Behçet's disease. Rheumatology Oxford 2001;40:668

104. Kotter I, Gunaydin I, Stubiger N et al. Comparative analysis of the association of HLA-B*51 suballeles with Behçet's disease in patients of German and Turkish origin. Tissue Antigens 2001;58:166

105. Murray KJ, Moroldo MB, Donnelly P et al. Age-specific effects of juvenile rheumatoid arthritis-associated HLA alleles. Arthritis Rheum 1999;42:1843

106. Cerna M, Vavrincova P, Havelka S. HLA and juvenile rheumatoid arthritis. Acta Univ Carol Med 1994;40:69

107. Gasset AR, Richman AV, Frias JL. HLA antigens and keratoconus. Ann Ophthalmol 1977;9:767

108. Ihalainen A. Clinical and epidemiological features of keratoconus genetic and external factors in the pathogenesis of the disease. Acta Ophthalmol 1986;178(Suppl):1

109. Gorskova EN, Sevost'ianov EN. [Associations of HLA Class I haplotype antigens with various patterns of keratoconus]. Vestn Oftalmol 1997;113:31

110. Cogan DG, Sullivan WR Jr. Immunologic study of nonsyphilitic interstitial keratitis with vestibuloauditory symptoms. Am J Ophthalmol 1975;80:491

111. Kaiser-Kupfer MI, Del Valle LA, Mittal KK et al. HLA and Cogan's syndrome. N Engl J Med 1978;298:1094

112. Taylor CJ, Smith SI, Morgan CH et al. HLA and Mooren's ulceration. Br J Ophthalmol 2000;84:72

113. Wang Z, Chen J, Zheng H. Changes in local immune functions in Mooren's ulcer. Yan Ke Xue Bao 1996;12:33

114. Chan LS, Hammerberg C, Cooper KD. Significantly increased occurrence of HLA-DQB1*0301 allele in patients with ocular cicatricial pemphigoid. J Invest Dermatol 1997;108:129

115. Power WJ, Saidman SL, Zhang DS et al. HLA typing in patients with ocular manifestations of Stevens-Johnson syndrome. Ophthalmology 1996;103:1406

116. Mondino BJ, Brown SI, Biglan AW. HLA antigens in Stevens-Johnson syndrome with ocular involvement. Arch Ophthalmol 1982;100:1453

117. Darrell, RW. Thygeson's superficial punctate keratitis: Natural history and association with HLA DR3. Trans Am Ophthalmol Soc 1981;79:486

118. Hida M, Kuwahara K. [Cataract and HLA antigen (author's transl)]. Nippon Ganka Gakkai Zasshi 1978;82:751

119. Saari KM. HLA antigens in ophthalmology. Acta Ophthalmol 1984;165(Suppl):18

120. Ladas ID. Histocompatibility (HLA) antigens and eye diseases other than uveitis. Surv Ophthalmol 1983;27:233

121. Ritch R, Podos SM, Henley W et al. Lack of association of histocompatibility antigens with primary open-angle glaucoma. Arch Ophthalmol 1978;96:2204

122. Salmon JF, Martell R. The role of ethnicity in primary angle-closure glaucoma. S Afr Med J 1994;84:623

123. Shin DH, Sugar HS. HLA in pigment dispersion and glaucoma. Tissue Antigens 1982;19:301

124. Kaiser-Kupfer MI, Mittal KK. The HLA and ABO antigens in pigment dispersion syndrome. Am J Ophthalmol 1978;85:368

125. Bertrams J, Schildberg P, Hopping W et al. HL-A antigens in retinoblastoma. Tissue Antigens 1973;3:78

126. Roberts DF, Duggan-Keen M, Aherne GE et al. Immunogenetic studies in retinoblastoma. Br J Ophthalmol 1986;70:686

127. Osborne A, Tschickardt M, Blanck G. Retinoblastoma protein expression facilitates chromatin remodeling at the HLA-DRA promoter. Nucleic Acids Res 1997;25:5095

128. Biswas J, Mukesh BN, Narain S et al. Profiling of human leukocyte antigens in Eales' disease. Int Ophthalmol 1997;21:277

129. Bertrams J, Spitznas M, Rommelfanger M. Missing evidence for HLA antigen association with Eales' disease, chorioretinitis, central serous retinopathy, and malignant choroidal melanoma. Invest Ophthalmol Vis Sci 1978;17:918

130. Metzelaar-Blok JA, Jager MJ, Moghaddam PH et al. Frequent loss of heterozygosity on chromosome 6p in uveal melanoma. Hum Immunol 1999;60:962

131. Ericsson C, Seregard S, Bartolazzi A et al. Association of HLA Class I and Class II antigen expression and mortality in uveal melanoma. Invest Ophthalmol Vis Sci 2001;42:2153

132. Proenca R, Carvalho M, Proenca D et al. HLA antigens and lymphocytes in proliferative vitreoretinopathy. Graefes Arch Clin Exp Ophthalmol 1994;232:25

133. Ohno S. HLA antigens in ophthalmology. Acta Ophthalmol 1984;165(Suppl):7

134. Bertrams J, Dewald G, Spitznas M et al. HLA-A, B, C, DR, Bf, and C2 alleles in insulin-dependent diabetes mellitus with proliferative retinopathy. Immunobiology 1980;158:113

135. Agardh D, Gaur LK, Agardh E et al. HLA-DQB1*0201/0302 is associated with severe retinopathy in patients with IDDM. Diabetologia 1996;39:1313

136. Gonzalez-Gay MA. Genetic epidemiology. Giant cell arteritis and polymyalgia rheumatica. Arthritis Res 2001;3:154

137. Rauzy O, Fort M, Nourhashemi F et al. Relation between HLA DRB1 alleles and corticosteroid resistance in giant cell arteritis. Ann Rheum Dis 1998;57:380

138. Salvarani C, Boiardi L, Mantovani V et al. HLA-DRB1, DQA1, and DQB1 alleles associated with giant cell arteritis in northern Italy. J Rheumatol 1999;26:2395

139. Hunt PJ, Marshall SE, Weetman AP et al. Histocompatibility leucocyte antigens and closely linked immunomodulatory genes in autoimmune thyroid disease. Clin Endocrinol Oxf 2001;55:491

140. Streilein JW, Yamada J, Dana MR et al. Anterior chamber-associated immune deviation, ocular immune privilege, and orthotopic corneal allografts. Transplant Proc 1999;31:1472

141. Creemers PC, Kahn D, Hill JC. HLA-A and -B alleles in cornea donors as risk factors for graft rejection. Transpl Immunol 1999;7:15

142. Katami M. Corneal transplantation—immunologically privileged status. Eye 1991;5:528

143. Sonoda Y, Streilein JW. Orthotopic corneal transplantation in mice—evidence that the immunogenetic rules of rejection do not apply. Transplantation 1992;54:694

144. Goulmy E, Pool J, Van Lochem E et al. The role of human minor histocompatibility antigens in graft failure: A mini-review. Eye 1995;9:180

145. Boisjoly HM, Roy R, Dube I et al. HLA-A,B and DR matching in corneal transplantation. Ophthalmology 1986;93:1290

146. Beekhuis WH. Current clinician's opinions on risk factors in corneal grafting. Results of a survey among surgeons in the eurotransplant area. Cornea 1995;14:39

147. Stulting RD, Waring GO, 3rd, Bridges WZ et al. Effect of donor epithelium on corneal transplant survival. Ophthalmology 1988;95:803

148. The collaborative corneal transplantation studies (CCTS). Effectiveness of histocompatibility matching in high-risk corneal transplantation. The Collaborative Corneal Transplantation Studies Research Group. Arch Ophthalmol 1992;110:1392

149. Volker-Dieben HJ, Claas FH, Schreuder GM et al. Beneficial effect of HLA-DR matching on the survival of corneal allografts. Transplantation 2000;70:640

150. Holland EJ. Epithelial transplantation for the management of severe ocular surface disease. Trans Am Ophthalmol Soc 1996;94:677

151. Donnelly JJ, Orlin SE, Wei ZG et al. Class II alloantigen induced on corneal endothelium. Role in corneal allograft rejection. Invest Ophthalmol Vis Sci 1990;31:1315

152. McBride BW, McGill JI, Smith JL. MHC Class I and Class II antigen expression in normal human corneas and in corneas from cases of herpetic keratitis. Immunology 1988;65:583

153. Pepose JS, Gardner KM, Nestor MS et al. Detection of HLA Class I and II antigens in rejected human corneal allografts. Ophthalmology 1985;92:1480

154. Hopkins KA, Maguire MG, Fink NE et al. Reproducibility of HLA-A, B, and DR typing using peripheral blood samples: Results of retyping in the collaborative corneal transplantation studies. Collaborative Corneal Transplantation Studies Group (corrected). Hum Immunol 1992;33:122

155. Vail A, Gore SM, Bradley BA et al. Influence of donor and histocompatibility factors on corneal graft outcome. Transplantation 1994;58:1210