A combination of the HLA-DRB1*03 phenotype and low plasma mannose-binding lectin predisposes to autoantibody formation in women with recurrent pregnancy loss.

Abstract

It is documented that a series of autoantibodies can be detected with increased frequency in women with recurrent pregnancy loss (RPL) and they may impact the pregnancy prognosis negatively. It is unknown whether the autoantibodies per se or the basic immune disturbances underlying autoantibody production, are the reason for this association. Our group has previously found that some genetically determined immunological biomarkers are associated with RPL and the same biomarkers are also in various degrees known to predispose to autoantibody production. The aim of this study was to clarify whether the RPL-associated immunogenetic biomarkers are associated with positivity for three major classes of autoantibodies associated with RPL. In 663 patients with RPL in whom we had results for HLA-DRB1 typing and plasma mannose-binding lectin (p-MBL) measurement, it was investigated whether there is a correlation between positivity for the autoantibodies: anticardiolipin antibodies, β2 glycoprotein I antibodies, and lupus anticoagulant (jointly called antiphospholipid antibodies), thyroid-peroxidase antibodies, and antinuclear antibodies and each of the HLA-DRB1 alleles HLA-DRB1*03 or HLA-DRB1*07 either alone or in combination with low p-MBL defined as ≤500 µg/l. Although slightly higher frequencies of positivity of two or more autoantibodies were seen in patients with either p-MBL ≤500 µg/l or being positive for HLA-DRB1*03, none were significantly associated. However, in patients with the combination of low p-MBL and HLA-DRB1*03, presence of at least one autoantibody was significantly more frequent than in patients with no such combination (OR= 2.4; 95% CI 1.2-5.0, p = 0.01). In an analysis of which autoantibodies were most strongly associated with the low p-MBL/HLA-DRB1*03 combination, antinuclear antibodies were significantly more frequent in these patients (OR 2.0; 95% CI 1.0-3.9, p=0.05) whereas the other autoantibodies were also positively but more weakly associated with this combination. In conclusion, to clarify the pathogenetic background, underlying immunogenetic factors should be examined in autoantibody positive RPL patients (as well as other patients with autoimmune diseases) but the genetic background may be complex.