Persistence of a Proteomic Signature After a Hypertensive Disorder of Pregnancy

Abstract

Background: A hypertensive disorder of pregnancy is associated with a higher risk of cardiovascular disease later in life, but the potential mechanistic links are unknown.

Methods: We recruited 2 groups of women, 1 during pregnancy and another at least 2 years after delivery. Cases had a hypertensive disorder of pregnancy, and controls had a normotensive pregnancy. The pregnancy cohort had study visits antepartum and postpartum; the mid-life group made a single study visit. We assayed 7288 plasma proteins, applied machine learning to identify proteomics signatures at each time point, and performed enrichment analyses to identify relevant biological pathways.

Results: The pregnancy cohort (58 cases and 46 controls) had a mean age of 33.8 years, and the mid-life group (71 cases and 74 controls) had a mean age of 40.8 years. Protein levels differed significantly between cases and controls at each time point: 6233 antepartum, 189 postpartum, and 224 in mid-life. The postpartum protein signature discriminated well between cases and controls (c-index=0.78), and it also discriminated well in the independent mid-life samples (c-index=0.72). Pathway analyses identified differences in the complement and coagulation cascades that persisted across the antepartum, postpartum, and mid-life samples. The 28 proteins present in both the postpartum and mid-life signatures included 5 complement factors (3, B, H, H-related-1, and C1r-subcomponent-like) and coagulation factor IX.

Conclusions: Differences in protein expression persist for years after a hypertensive disorder of pregnancy. The consistent differences in the complement and coagulation pathways may contribute to the increased risk of later life cardiovascular disease.

Keywords: case-control studies; hypertension, pregnancy-induced; machine learning; pre-eclampsia; proteomics.