Analysis of 250 HLA‐B44 genotypes in European Caucasoids: high diversity and preferential ABCDRB1 associations in B*4402, B*4403, and B*4405 haplotypes

Abstract

Based on high-resolution DNA typing within 235 pedigrees, a total of 250 HLA-A/B/C/DRB1/DRB3 genotypes have been characterized. These comprise 129 different B44 haplotypes, of which 73.6% occurred only once. Only four different B*44 alleles were identified: B*4402-4405, with B*4402 and B*4403 haplotypes accounting for 57.6 and 36.8%, respectively, of all haplotypes. Although the relative numbers of different A/B/C/DRB1/B3 haplotypic associations were similar in both B*4402 and B*4403 haplotypes, the genotypic profiles were quite different in the two groups. When associated with the A*0101, A*0201, A*2402, A*3201, and A*6801 alleles, a much more extensive polymorphism of B*4402 haplotypes with respect to HLA-C and DRB1 associations was disclosed. On the other hand, B*4403 haplotypes were more diverse in the A23-B44 and A29-B44 groups with respect to DRB1 associations. Considering B-C linkage, B*4402-Cw*0501, B*4402-Cw*0704, B*4402-Cw*1604, B*4403-Cw*0401, B*4403-Cw*1601, B*4404-Cw*1601, and B*4405-Cw*0202 accounted for 98% of all genotypes. Eight A/B/C/DRB1 haplotypes occurred at a relative genotypic frequency of >0.015, with A*2902-B*4403-Cw*1601-DRB1*0701 (11.2%) and A*0201-B*4402-Cw*0501-DRB1*0401 (8.4%) as the two most frequent genotypes. Some A and DRB1 alleles were predominantly, if not exclusively, associated with specific B-C pairs: A*0301 with B*4402-Cw*0501 and B*4403-Cw*0401; A*2301 with B*4403-Cw*0401; A*2608 with B*4402-Cw*0501; A*2902 with B*4403-Cw*1601; DRB1*0101/0401/0403/0404/1101/1104/0801/1301/1302 with B*4402-Cw*0501; and DRB1*0701 with B*4403-Cw*1601. On the basis of this dataset and our experience with searches for phenotypically matched unrelated stem cell donors, several ABDR haplotypes were identified that would confer a higher probability of B44- and C-incompatibility. The analysis of 112 consecutive unrelated stem cell donor searches revealed that 24% of the 400 tested donors were B44-mismatched, and that no single B44 allele- matched donor could be identified for only 7% of the patients. HLA-C incompatibility rate was 22.2% for the patients with > or =1 B44 allele-matched donor(s). This dataset can therefore be used as a predictive tool for B44- and C-disparities in unrelated stem cell transplantation.

The probability of identifying a 10/10 HLA allele-matched unrelated donor is highly predictable

“…This estimation was based on the patients HLA haplotypes (rare alleles, uncommon linkages and frequent haplotypes), on local data sets of preferential haplotypic associations, 11 on available information in the BMDW data file, as well as on local experience in unrelated donor searches for the past 15 years. 4 The present data confirm the feasibility and accuracy of such an approach through high positive and negative predictive value of this estimate compared to the search end result.…”

“…20 For B18, B27, B44 and B51 serotypes, B-Cw associations in the patient were evaluated with respect to data available in the BMDW registry and with published and local databases. [9][10][11][12][13] For example, a low-probability estimate would be assigned for a patient with the A2-B51-Cw*02-DR8 haplotype, since a very strong linkage with Cw*1402 on this haplotype has been observed (Tiercy J-M, unpublished data). When the two haplotypes of the patient did not belong to the most frequent ones, the search qualified as high when the most common alleles are identified by four-digit typing for each locus based on the frequency distribution described by Hurley et al 21 The BMDW search program, which lists the potential donors that are allele mismatched for each locus, provides useful Table 1 The 10 most frequent haplotypes in Caucasoids based on the phenotypes computed by Boucher et al 19 …”

“…The probability estimates were based on sound knowledge of HLA haplotype and allele frequencies in Caucasoid populations. [12][13][14][15][16][17][18] Published and local data sets with allele frequencies and multilocus haplotypes, such as described for the B44 serotype, 11 were useful for the estimations.…”

“…As an example, the A2-B44-DR11 haplotype may comprise any of the three following five-loci combinations: A*0201-B*4402-Cw*0501-DRB1*1101-DQB1*0301, A*0201-B*4402-Cw*0501-DRB1*1104-DQB1*0301 and A*0201-B*4402-Cw*0704-DRB1*1101-DQB1*0301. 11 An additional difficulty with unrelated donors resides in the fact that HLA-allelic combinations do not necessarily segregate on the same chromosome, as is obviously the case in patients with proven genotypes. This may influence compatibility assessments since allele frequency distribution within a given serotype can significantly differ from one ABDR haplotype to another.…”

HLA Haplotypes in Singapore: A Study of Mothers and Their Cord Blood Units

“…This is complicated by differences in the frequencies of alleles and haplotypes among population groups [4 -6]. In population studies, HLA haplotypes have been defined by allele segregation in families [7][8][9] or, in large populations of unrelated individuals, predicted with the use of mathematical algorithms such as the Expectation Maximization algorithm [10 -12].…”