HLA class I serotypes and cytotoxic T-lymphocyte responses among human immunodeficiency virus-1-uninfected Thai volunteers immunized with ALVAC-HIV in combination with monomeric gp120 or oligomeric gp160 protein boosting.

Antigen-induced cellular immunogenicity may vary between populations due to differences in human leukocyte antigen (HLA) diversity and, hence, may play a critical role in the protection afforded by vaccines. In the setting of two, phase I/II human immunodeficiency virus-1 vaccine trials of a recombinant canarypox prime, and boosting with either recombinant monomeric gp120 or oligomeric gp160, we assessed the association between specific human leukocyte antigen (HLA) class I serotypes and the presence of cytotoxic T-lymphocyte response measured by 51Cr-release assay. HLA class I serotypes A11, A24, A33, B46, and B75 were the most common, present in 10% or more of 245 individuals studied. Forty of 187 (21.4%) Thai adults who received either ALVAC-HIV with gp120 or oligomeric gp160 or ALVAC alone had a precursor cytolytic CD8 T-cell response (pCTL). HLA-B44 was positively and significantly associated with a pCTL response (odds ratio 7.6, 95% CI: 2.7-21.2), whereas B46 was negatively associated but not robust when adjusted for multiple comparisons. Responses to Env proteins accounted for the majority (nine of 11) of pCTL activity among those persons with B44. This HLA class I serotype occurred in 9.4% of participants overall (including the placebo group), less commonly than what is reported from populations of European ancestry. These results strengthen the importance of assessing HLA class I distributions in conjunction with studies of vaccines designed to elicit cellular immunity in different populations.

Association of Fcgamma receptor IIb and IIIb polymorphisms with susceptibility to systemic lupus erythematosus in Thais.

We recently reported association of a newly identified polymorphism of Fcgamma receptor (FcgammaR) IIb, I232T, with systemic lupus erythematosus (SLE) in Japanese. To date, information on FcgammaR genotypes and their association with SLE is limited in South-east Asian populations. To gain further insight into the role of FcgammaR polymorphisms in the genetic predisposition of SLE, association of FcgammaRIIa-H131R, IIb-I232T, IIIa-F176V and IIIb-NA1/NA2 (HNA-1a/1b) polymorphisms with SLE was analyzed in the Thai population, using case-control association analysis. FcgammaRIIb-232T/T and IIIb-NA2/NA2 genotypes were associated with SLE with the odds ratio of 2.55. Genotype relative risk analysis revealed significant association of IIb-232T/T and IIIb-NA2/NA2, and a tendency of association of the IIIa-176F/F genotype. Moreover, carriers of FcgammaRIIa-131R were significantly increased in patients with lupus nephritis. Significant linkage disequilibrium was present among FcgammaRIIb, IIIa and IIIb, and two-locus analyses suggested that the tendency of association of FcgammaRIIIa could derive from linkage disequilibrium with IIb and IIIb. These results provided evidence that FcgammaR polymorphisms may be an important predisposing factor also in Thais in a complex manner.

HLA-A and -B allele associations with secondary dengue virus infections correlate with disease severity and the infecting viral serotype in ethnic Thais.

Little is known of the role of classical HLA-A and -B class I alleles in determining resistance, susceptibility, or the severity of acute viral infections. Appropriate paradigms for immunogenetic studies of acute viral infections are dengue fever (DF) and dengue hemorrhagic fever (DHF). Both primary and secondary infections with dengue virus (DEN) serotypes 1, 2, 3 or 4, can result in either clinically less severe DF or the more severe DHF. In secondary exposures, a memory response is induced in immunologically primed individuals, which can both clear the infecting dengue virus and contribute to its pathology. In a case-control study of 263 ethnic Thai patients infected with either DEN-1, -2, -3 or -4, we detected HLA class I associations with secondary infections, but not in immunologically naive patients with primary infections. HLA-A*0203 was associated with the less severe DF, regardless of the secondary infecting virus serotype. By contrast, HLA-A*0207 was associated with susceptibility to the more severe DHF in patients with secondary DEN-1 and DEN-2 infections only. Conversely, HLA-B*51 was associated with the development of DHF in patients with secondary infections, and HLA-B*52 was associated with DF in patients with secondary DEN-1 and DEN-2 infections. Moreover, HLA-B44, B62, B76 and B77 also appeared to be protective against developing clinical disease after secondary dengue virus infection. These results confirm that classical HLA class I alleles are associated with the clinical outcome of exposure to dengue virus, in previously exposed and immunologically primed individuals.

Natural killer cell immunoglobulin-like receptor (KIR) locus profiles in African and South Asian populations.

Natural killer (NK) and some T cells express killer cell immunoglobulin-like receptors (KIRs), which interact with HLA class I expressed by target cells and consequently regulate cytolytic activity. The number of KIR loci can vary and so a range of genetic profiles is observed. We have determined the KIR genetic profiles from one African (n = 62) and two South Asian (n = 108, n = 78) populations. Several of the KIRs are present at significantly different frequencies between the two major ethnic groups (eg KIR2DS4 gene frequency 0.82 African, 0.47 S Asian. Pc < 1 x 10(-6)) and this is due to uneven distribution of two KIR haplotype families 'A' and 'B'. All three populations described here displayed a greater degree of diversity of KIR genetic profiles than other populations investigated, which indicates further complexity of underlying haplotypes; in this respect we describe two individuals who appear homozygous for a large deletion including the previously ubiquitous 2DL4. We have also reanalysed three populations that we studied previously, for the presence of a KIR which is now known to be an indicator of the 'B' haplotype. South Asians had the highest overall frequencies of all KIR loci characteristic of 'B' haplotypes (Pc < 0.0001 to < 0.004). Furthermore, gene frequency independent deviances in the linkage disequilibrium were apparent between populations.

Natural killer cell immunoglobulin-like receptor (KIR) locus profiles in African and South Asian populations

Natural killer (NK) and some T cells express killer cell immunoglobulin-like receptors (KIRs), which interact with HLA class I expressed by target cells and consequently regulate cytolytic activity. The number of KIR loci can vary and so a range of genetic profiles is observed. We have determined the KIR genetic profiles from one African (n = 62) and two South Asian (n = 108, n = 78) populations. Several of the KIRs are present at significantly different frequencies between the two major ethnic groups (eg KIR2DS4 gene frequency 0.82 African, 0.47 S Asian. Pc < 1 x 10(-6)) and this is due to uneven distribution of two KIR haplotype families 'A' and 'B'. All three populations described here displayed a greater degree of diversity of KIR genetic profiles than other populations investigated, which indicates further complexity of underlying haplotypes; in this respect we describe two individuals who appear homozygous for a large deletion including the previously ubiquitous 2DL4. We have also reanalysed three populations that we studied previously, for the presence of a KIR which is now known to be an indicator of the 'B' haplotype. South Asians had the highest overall frequencies of all KIR loci characteristic of 'B' haplotypes (Pc < 0.0001 to < 0.004). Furthermore, gene frequency independent deviances in the linkage disequilibrium were apparent between populations.

HIV-specific cytotoxic T lymphocytes, HLA-A11, and chemokine-related factors may act synergistically to determine HIV resistance in CCR5 delta32-negative female sex workers in Chiang Rai, northern Thailand.

Understanding how highly HIV-exposed individuals remain HIV uninfected may be useful for HIV vaccine design and development of new HIV prevention strategies. To elucidate mechanisms associated with resistance to HIV infection, immunologic and genetic factors were examined in 14 HIV-exposed but persistently seronegative (HEPS) female sex workers from Chiang Rai, northern Thailand and in ethnically matched, HIV-positive (n = 9) and HIV-negative women (n = 9). The HEPS women were identified in a study of commercial sex workers who had an HIV-1 incidence of 20.3 per 100 person-years. A high frequency of HLA-A11 was observed in HEPS women (86%) compared with northern Thai controls (56%). HIV-specific cytotoxic T lymphocyte (CTL) lytic responses were detected in cryopreserved peripheral blood mononuclear cells (PBMCs), using HLA-A-matched subtype E HIV-1 peptides in four of seven (57%) HEPS women, eight of eight HIV-positive women, and zero of nine HIV-negative unexposed controls (p = 0.019 HEPS women vs. HIV-negative controls). CTL lysis levels were low, but responses were detected to peptides from Nef, Pol, Gag, and Env. Nef responses predominated in HEPS women. Compared with controls, HEPS women tended to have higher frequencies of CCR5 promotor 59402GG and SDF-1 3'UTR 801A genotypes known to influence HIV transmission or course of disease. HEPS women also had higher levels of spontaneous RANTES production by PBMCs than other groups. Each of these factors could potentially contribute to HIV resistance. As most HEPS women had one or more of these factors, they may prevent HIV infection synergistically by blocking HIV cell entry, delaying its dissemination, or killing HIV-infected cells.

Distribution of natural killer cell immunoglobulin-like receptor sequences in three ethnic groups.

Killer cell immunoglobulin-like receptors (KIRs) are members of a group of molecules that specifically recognize HLA class I ligands and are found on subsets of human lymphopoetic cells. The number of KIR loci can vary between individuals, resulting in a heterogeneous array of possible KIR genes. The range of observed profiles has been explained by the occurrence of two haplotype families termed A and B which can be distinguished on the basis of certain KIR sequences. Here we attempted to determine whether the frequencies of putative KIR loci and the two haplotype groups vary in three ethnically defined, healthy, and unrelated control populations, namely UK Caucasoid (n=136), Palestinian (n=105) and Thai (n=119). We molecularly typed genomic DNA for the presence of 12 putative KIR loci, KIR2DL1, KIR2DL2, KIR2DL3, KIR2DL4, KIR3DL1, KIR3DL2, KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS4, KIR2DS5, and KIR3DS1, using modified PCR sequence-specific primers. The patterns of KIR locus frequencies combined with the similar linkage disequilibrium values suggest that there was a distinction in the distribution of the two broad haplotype groups between the populations studied. The A haplotype was always the most prevalent, but the ratio of A to B varied between populations. The frequency of B haplotype was highest in the Palestinians and lowest in the Thais (Pc<0.0001).

A comparison of molecular HLA-DR and DQ allele profiles forming DR51-, DR52-, and DR53-related haplotypes in five ethnic Thai populations from mainland southeast Asia.

Using PCR-SSOP typing we have deduced the composition and frequency of HLA-DRB1, -DRB3, -DRB4, -DRB5, -DQA1, and -DQB1 alleles present in DR51-, DR52-, and DR53-related haplotypes, in 519 individuals representative of five ethnic Thai populations recruited in central, northeastern and northern Thailand. In total, we have unequivocally detected at varying frequencies, 17 DR51-related haplotypes, 24 DR52 haplotypes, and 12 DR53 haplotypes in the study groups. We document evidence of north-south gradients of DR51-related haplotypes, whereby the overall frequency of DR51-containing haplotypes is relatively more common in the northern Thai groups. Similarly, within DR53-related haplotypes the frequency of DRB1*0901-containing haplotypes increases in the more northerly groups, and an inverse effect was observed with DRB1*0701-containing haplotypes that were relatively more common in the northeastern and central Thais. We have also compared the class II haplotype profiles of the Thais with the equivalent profiles reported in other non-Thai ethnic groups from mainland and insular SE Asia. One DR51-related haplotype DRB1*1502x, DRB5*0102x, DQA1*0101/4, DQB1*0501, would appear to be characteristic of Thai populations, as it was the most common DR2 haplotype in all five study groups and is also prevalent in other mainland southeast Asians, but is much less evident in the more northern populations of eastern Asia or China.

Allelic variation in the intron 2 and 3 of the MICA gene.

We have determined by sequencing the allelic variation in intron 2 and 3 of hte MICA gene for a total of 22 different alleles. Sequencing of introns was performed in two directions, using DNA from homozygous cell lines from families and from unrelated individuals. Intron 2 is 273 bp long and did in the alleles investigated not reveal any length polymorphism. We found a total of eight polymorphic positions which exhibit a strict biallelism, as it is also found in the polymorphisms for exon 2, 3 and 4 of MICA. Intron 3 is 586 bp long an required an additional set of primers placed near the middle of this intron in order to allow a complete bidirectional sequence. In intron 3, a total of 10 polymorphic positions were identified. Interestingly, we found two variants of the allele MICA*002 which are distinguished only by one basepair difference in intron 3. The variant MICA*002A is associated with HLA-B35 and B58, while the allele MICA*002B is associated with B38 and B39.

Definition of new alleles of MIC-A using sequencing-based typing.

We have sequenced exons 2, 3 and 4 of MIC-A in 23 homozygous cell lines, 22 families and 54 unrelated individuals. This has led to the definition of seven polymorphic positions in exon 2, 13 in exon 3 and 12 in exon 4, yielding a total of 33 different MIC-A allelic specificities, of which 16 have not been described before. The newly defined sequences and those of the alleles defined before were entered into a database of the SCORE program (Helmberg et al., 1998, Tissue Antigens, 51, 587) for comprehensive genotyping analysis. In the tested sample, only one genotype present in two individuals gave rise to an ambiguous genotype. If all possible combinations of the 33 alleles are considered, 10 of 636 combinations are ambiguous. The MIC-A exon 2, 3 and 4 polymorphism is characterized by diallelic single base exchanges and by a considerable degree of exon shuffling. The majority of heterozygote positions identified are non-synonymous, i.e. five of seven in exon 2, 13 of 13 in exon 3 and eight of 12 in exon 4, suggesting an important function for the MIC-A polymorphism.

Association of HLA and T-cell receptor gene polymorphisms with Semple rabies vaccine-induced autoimmune encephalomyelitis.

Semple rabies vaccine is derived from brain tissue infected with rabies virus that is subsequently inactivated with phenol. Semple rabies vaccine-induced autoimmune encephalomyelitis (SAE) occurs in 1 in 220 immunized individuals. The immune response to myelin basic protein and pathological changes of demyelination in SAE suggest that this disease is the human homologue of experimental autoimmune encephalomyelitis (EAE). SAE and EAE are frequently studied as models for the human demyelinating disease multiple sclerosis. Major histocompatibility complex (MHC) class II and T-cell receptor (TCR) gene polymorphisms play important roles in rodent susceptibility to EAE and were analyzed to determine if the same was true in humans with SAE. HLA-DRB1, HLA-DQB1, and TCRBV gene polymorphisms were studied in Thai individuals with SAE (n = 18), with vaccination without neurological complications (n = 43), and without vaccination (n = 140). The allele frequencies of HLA-DR9 (DRB1*0901) and HLA-DR17 (DRB1*0301) were increased in SAE patients (DR9 = 22%, DR17 = 14%) compared with vaccinated controls (DR9 = 13%, DR17 = 6%) and with unvaccinated controls (DR9 = 9%, DR17 = 4%). The allele frequency of HLA-DQ7 (DQB1*0301) was decreased in SAE patients (8%) compared with vaccinated controls (15%) and with unvaccinated controls (25%). These susceptibilities are distinct from those associated with multiple sclerosis. The frequencies of TCRBV alleles and haplotypes were similar in SAE patients and vaccinated controls. These data suggest that genetic susceptibility associated with MHC class II alleles may have a role in the pathogenesis of SAE and its mechanism may be different from those involved in multiple sclerosis.

Serological and molecular analysis of HLA class I and II alleles in Thai patients with psoriasis vulgaris.

The HLA class I and class II alleles in 67 patients with type I psoriasis vulgaris, 23 patients with type II psoriasis vulgaris and 140 healthy individuals were analyzed. The frequencies of HLA-A2, -B46, -B57 and DQB1*0303 were significantly increased in type I psoriasis compared to the controls (Pc<0.05). Molecular analysis of HLA-A2 alleles showed an increase in HLA-A*0207 and a decrease in HLA-A*0203 in type I psoriasis. HLA-DQB1*0301 was significantly decreased in type I psoriasis compared to the normal controls (Pc<0.05). No association of any alleles with type II psoriasis was observed. This data demonstrated two susceptible haplotypes: HLA-A1-B57-DRB1*0701-DQA1*0201-DQB1*0303 (AH57.1) and HLA-A2-B46-DRB1*0901-DQA1*0301-DQB1*0303 (AH46.1) for type I psoriasis in the Thai population. Besides, the haplotype AH46.1 was also associated with type II psoriasis.

Predominance of HLA-restricted cytotoxic T-lymphocyte responses to serotype-cross-reactive epitopes on nonstructural proteins following natural secondary dengue virus infection.

We examined the memory cytotoxic T-lymphocytic (CTL) responses of peripheral blood mononuclear cells (PBMC) obtained from patients in Thailand 12 months after natural symptomatic secondary dengue virus infection. In all four patients analyzed, CTLs were detected in bulk culture PBMC against nonstructural dengue virus proteins. Numerous CD4+ and CD8+ CTL lines were generated from the bulk cultures of two patients, KPP94-037 and KPP94-024, which were specific for NS1.2a (NS1 and NS2a collectively) and NS3 proteins, respectively. All CTL lines derived from both patients were cross-reactive with other serotypes of dengue virus. The CD8+ NS1.2a-specific lines from patient KPP94-037 were HLA B57 restricted, and the CD8+ NS3-specific lines from patient KPP94-024 were HLA B7 restricted. The CD4+ CTL lines from patient KPP94-037 were HLA DR7 restricted. A majority of the CD8+ CTLs isolated from patient KPP94-024 were found to recognize amino acids 221 to 232 on NS3. These results demonstrate that in Thai patients after symptomatic secondary natural dengue infections, CTLs are mainly directed against nonstructural proteins and are broadly cross-reactive.

Bone marrow, peripheral blood and cord blood stem cell transplantation in children: ten years' experience at Siriraj Hospital.

Stem cell transplantations were performed in 69 children at Siriraj Hospital over a ten year period. The source of stem cells was bone marrow (60), peripheral blood (3), or cord blood (6). The diseases treated included 35 thalassemias, 11 Burkitt's lymphoma, five non-Hodgkin's lymphoma, five aplastic anemia, eight acute leukemia, and one each of neuroblastoma, severe combined immunodeficiency, Wiskott-Aldrich syndrome, myelodysplastic syndrome, and pyruvate kinase deficiency. The success rate of stem cell transplantation in Thai children varied according to the underlying diseases of the patients, ranging from 50% in acute leukemia to 100% in aplastic anemia. The outcome of stem cell transplantation in 35 thalassemic children revealed 23 (79.4%) were cured, whereas three (10.3%) remain alive with disease and the other three (10.3%) died. The incidence of graft-versus-host disease was low hen compared with that of Western countries. It is concluded that bone marrow, peripheral blood and cord blood stem cell transplantation will be the treatment of choice and will be widely used in the future to cure many hematologic and malignant disorders in children.

HLA class II allele frequencies in northern Thais (Kamphaeng Phet).

The polymorphism of HLA class II genes (HLA-DRB1, -DQA1, -DQB1 and -DPB1) was investigated in 97 normal Northern Thais (NT) from Kamphaeng Phet province using PCR-SSO typing. Allele frequencies (AF) have been determined. DRB1*1202 (17.5%), DRB1*1502 (16.5%), DQA1*0101 (25.8%), DQA1*0102 (21.7%), DQB1*0502 (22.7%), DPB1*0501 (23.2%) and DPB1*1301 (22.7%) showed the highest frequencies in each locus. These results were more similar to those observed in Present-day Thais (PDT) and Central Thais (CT) than Northern Thais from Chiang Mai (CM) and Dai Lue (DL). However, the data presented in this population study should be useful in many fields, such as anthropology, organ transplantation, disease susceptibility and evolutionary genetics.

HLA-A, -B, -DRB1, -DQA1, and -DQB1 polymorphism in Thais.

In this study we examined HLA-A, -B, -DRB1, -DQA1, and -DQB1, gene allele, and haplotype frequencies in two ethnic Thai populations. We compared these frequencies to the known HLA class I and II allele profiles of non-Thai mainland and insular Southeast (SE) Asians. HLA-A locus gene and allele frequencies, are comparatively homogeneous in both Thai and non-Thai SE Asians. In contrast, HLA-B; -DRB1, -DQA1, and -DQB1 gene and allele frequencies, show more ethnic and geographic variation in SE Asians. Conserved haplotypes, or combinations of linked HLA class I and II alleles were detected in Thais, but at relatively low frequencies. It would appear that ethnic Thais, reflect an admixture of peoples from both the northern mainland and southern island groups of SE Asia.

HLA class I typing by one-dimensional isoelectric focusing and identification of the new variants in Thai population.

One-dimensional isoelectric focusing (1D-IEF) is the technique to define HLA class I antigens based on difference in isoelectric point of HLA molecules. Different IEF subtypes are shown in different populations. In this study, 1D-IEF was employed to study HLA-A and -B subtypes in Thai population. A panel of 117 samples including all serologically defined HLA-A and -B antigens in Thai population were typed by 1D-IEF. Serological specificities and subtypes correlated well with IEF results and some antigens with unclear serological specificities could be confirmed by IEF. In addition, more subtypes could be obtained by IEF than by serology. A total of 17 IEF subtypes from HLA-A and 31 IEF subtypes from HLA-B could be identified. The subtypes predominantly found in Thai population were A2.3, A24.2, A11.1, A33.2, B15.2, B7.1 and B13.1. In addition, new IEF variants were identified in HLA-B35, B5, B56 and B48. The band positions of these variants were different from those previously described. These IEF subtypes are HLA gene products which may be important in transplantation. The combination of IEF and serology for HLA typing can provide a better definition of each allelic product of HLA-A and -B.

Estimated rate of HIV-1-infectious but seronegative blood donations in Bangkok, Thailand.

OBJECTIVES: To determine HIV seroprevalence and incidence among various blood donor types, and to estimate the rate of window-period blood donations. DESIGN: Retrospective cohort from computerized donor records. METHODS: Records were analysed from all 60,483 donors (contributing 97,464 donor units) at a public university teaching hospital blood bank in Bangkok, Thailand, from 1 January 1990 to 30 June 1993. Annual HIV incidence among 14,482 repeat donors who were HIV-seronegative on their first donation was calculated assuming equal probability of seroconversion between last seronegative and first seropositive donations. To estimate the probability of window-period donations, we assumed that the time from HIV infectivity to onset of detectable antibody was 45 days. RESULTS: In 1990, HIV incidence calculated for all repeat donors was 307 per 100,000 person-years; the probability of a window-period donation was 38 in 100,000 donations or one in 2644 donations. During 1991-1993, this probability decreased by one-half. However, one-time donors were more than twice as likely as repeat donors to be HIV-1-seropositive. CONCLUSIONS: The rate of HIV window-period blood donations among Thai repeat donors was relatively high compared with that in developed countries and was probably even higher among one-time donors. Improved donor deferral criteria are needed in Thailand.

PIP: In countries with a high incidence of human immunodeficiency virus (HIV), blood donations made during the "window period" (time between infection and the development of detectable antibody) pose a serious risk. To assess the extent of this risk in Thailand, records of all 60,483 blood donors at Bangkok's Mahidol University from January 1, 1990, to June 30, 1993, were reviewed. A retrospective cohort of 14,482 repeat donors was created by identifying those who were HIV-negative at their first screening and made one or more subsequent donations (mean, 3.6). Among these repeat donors, 40 (0.3%) seroconverted during the study period. The time from HIV infectivity to onset of detectable antibody was assumed to be 45 days. The HIV transmission rate among repeat donors resulting from donations during the window period was estimated to be 1 in 2644 units transfused in 1990 and 1 in 5000 units transfused during 1991-93. Extrapolation of these estimates suggests that, during the 1991-93 period, 57 units in Bangkok (194 units nationwide) were from donors in the HIV window period. Voluntary 1-time donors were 1.9 times more likely to be HIV-infected than voluntary repeat donors, while paid one-time donors were 13.7 times more likely to be seropositive than paid repeat donors. In 1994, the hospital initiated HIV p24 antigen screening of all donated blood; although this technique shortens the window period and should improve blood safety, improved donor referral criteria are needed.

Structural definition of the A*74 group: implications for matching in bone marrow transplantation with alternative donors.

We have identified two new A*74 alleles (A*7402 and 7403) in two unrelated individuals. A*7402 differs from A*7401 by a single amino acid substitution in the signal peptide and may be the result of a gene conversion event at the 3' end of exon 1. A*7403 differs from A*7401 by a single amino acid exchange in the alpha 1 domain and is most likely due to a point mutation in exon 2, since no HLA class I donor allele has been found. Since A*7402 appears to be the ancestor of the other two A*74 alleles, it is possible that A*7401 and 7403 have been created by successive point mutations. The sequences of the expressed proteins of A*7401 and 7402 are identical. The heavy chain sequence of A*7403 differs from these alleles at the crucial residue 79 which is located in the sequence stretch of the alpha 1 alpha-Helix where the Bw4/Bw6 determinants have been identified and which probably affects TCR interaction. This variation can therefore be expected to stimulate alloreactive T cells, graft rejection and graft versus host disease emphasizing the relevance for matching in bone marrow transplantation with alternative donors.