HLA-DR and -DQ typings were performed by a combination of RFLP and PCR
-SSP techniques in 234 Danish women with at least three consecutive un
explained fetal losses (recurrent fetal losses) and 360 controls and t
he DRB1, DQA1 and DQB1 alleles were deduced. In the total group of pat
ients, the frequency of no DRB1-DQA1-DQB1 haplotype was significantly
increased compared with controls. In the subgroup of 97 women with fou
r or more fetal losses (multiple fetal loss group), the frequency of w
omen carrying the DRB10101, DQA1*0101, DQB1*0501; DRB1*0102, DQA1*010
1, DQB10501 and DRB1*0103, DQA1*0101, DQB1*0501 haplotypes or the DRB
10301, DQA1*0501, DQB1*0201 haplotype were significantly increased co
mpared with controls (RR=2.1; p(c)<0.05 with regard to former three ha
plotypes combined and RR=2.2; p(c)<0.05 for the latter). The frequency
of women with at least one of the four haplotypes was significantly (
p<0.002) increased with the number of previous fetal losses in the wom
en's history. Analysis of the DQA1 and DQB1 phenotypes in women with a
t least four fetal losses showed that DQA10501 and DQB1*0501 were inc
reased compared with controls (RR=1.9; p(c)<0.05 and RR=2.2; p(c)<0.02
5, respectively). Analysis of DRB1-DQA1-DQB1/DRB1-DQA1-DQB1 genotypes
suggested that genotypes comprising both DQA10501 and DQB1*0501 allel
es (in trans) exhibited a higher RR for experiencing at least four fet
al losses (RR=3.4, p=0.002) than each of the alleles did alone. Genoty
pes comprising a DQB1 allele encoding a beta chain negative for aspart
ate in position 57 were associated with an increased RR (2.3; p<0.01)
for multiple fetal losses and the etiologic fraction was high (0.49).
The results suggest that genes or gene-sequences, linked to the DRB10
101, DQA10101, DQB1*0501; DRB1*0102, DQA1*0101, DQB1*0501; DRB1*0103,
DQA10101, DQB1*0501 and DRB1*0301, DQA1*0501, DQB1*0201 haplotypes,
confer susceptibility to multiple fetal losses. These alleles/gene-seq
uences are presumably located in the DQA1 and DQB1 loci and may elicit
their effect by mediating an autoimmune reaction against one or more
trophoblast antigens.