Characterising effects of autoimmune disease liability and targets of their treatment on adverse pregnancy outcomes using Mendelian Randomization

Background:
Autoimmune diseases are often more common in women than men, diagnosed in early adulthood, and associated with adverse pregnancy outcomes. However, whether these diseases or medications used to manage them cause adverse pregnancy outcomes is unknown.

Methods:
We used two-sample Mendelian randomization (MR) to explore the effect of liability to seven autoimmune diseases on nine pre-selected adverse pregnancy outcomes. Genetic instruments were selected from published GWAS and outcome genetic association data included up to 400,000 women within the MR-PREG collaboration. Main analyses used the inverse variance weighted (IVW) method, and sensitivity analyses included MR-Egger and weighted-median.

Results:
We found evidence that liability to rheumatoid arthritis increased risks of low birthweight (OR=1.12, 95% CI:1.06-1.18), neonatal intensive care unit admission (OR=1.08, 95% CI:1.03-1.12), and gestational diabetes (OR=1.06, 95% CI:1.03-1.09), while systemic lupus erythematosus increased risks of low birthweight (OR=1.04, 95% CI:1.01-1.08) and preeclampsia (OR=1.03, 95% CI:1.00-1.05). Multiple sclerosis decreased risks of several outcomes, including preeclampsia (OR=0.95, 95% CI:0.93-0.98) and low birthweight (OR=0.94, 95% CI=0.90-0.98). Sensitivity analyses supported these IVW results. For psoriasis vulgaris, psoriatic arthritis, axial spondyloarthritis and inflammatory bowel disease there was limited evidence supporting any effects, though several estimates were imprecise.

Conclusion:
Our preliminary findings suggest an effect of some autoimmune diseases on fetal growth and preeclampsia. In ongoing analyses, we are exploring the role of the HLA locus in these effects, and using drug target MR to investigate the effects of perturbing targets of monoclonal antibody drugs used to manage these autoimmune conditions on the same pregnancy outcomes.