Exploration of novel mechanisms and biomarkers for venous thromboembolism : a genetics and proteomics study

Abstract

Background: Venous thromboembolism (VTE) contributes to a large health burden and incidence increases exponentially with age. ~25% patients will experience a recurrent event and there is a 2-fold risk for death in the following years. Risk prediction remains a challenge.

Aims to: Investigate a presumed overlap between cardiovascular disease and VTE. Expand current knowledge of established pathways in VTE risk by combining clinical and genetic epidemiology. Apply affinity proteomics to identify novel plasma susceptibility biomarkers.

Methods and results: 39 single nucleotide polymorphisms (SNPs) associated with coronary artery disease (CAD) and 18 risk VTE-SNPs from candidate gene approach studies or genome-wide association studies (GWAS) were identified from a literature search. The SNPs were genotyped in 2,835 women from the ThromboEmbolism Hormone study (TEHS), a Swedish nationwide case-control study in women (2002-2009). Association was assessed with logistic regression. Clinical and genetic predictors that contributed significantly to the fit of the logistic regression model were included in the prediction models. The genetic predictors that contributed to first VTE were assessed in a cohort of 1010 women with VTE from TEHS, followed up until a recurrence, death or November 2011 (TEHS follow-up). The SNP r579459 in the ABO locus was the only CAD-SNP that was significant associated with VTE (OR 1.57 (95% CI: 1.39-1.78, p= 6.4 10-13)). Seven VTE risk-SNPs (F5 rs6025, F2 rs1799963, ABO rs514659, FGG rs2066865, F11 rs2289252, PROC rs1799810 and KNG1 rs710446) with 4 SNP-SNP interactions contributed to the genetic risk score for VTE with an AUC of 0.66 (95% CI: 0.64-0.68). After adding clinical risk factors the AUC attained 0.84 (95% CI: 0.82-0.85). The goodness of fit of the genetic and combined scores improved when significant SNP-SNP interaction terms were included. In TEHS follow-up study, the overall recurrence rate was 20 per 1000 person-years (95% CI; 16-24). Carriers of the risk alleles of F5 rs6025 (FVL) (HR=1.7 (95% CI; 1.1-2.6)) and F11 rs2289252 (HR=1.8 (95% CI; 1.1-3.0)) had significantly higher rates of recurrence compared to non-carriers. The cumulative recurrence was 2.5-fold larger in carriers of both F5 rs6025 and F11 rs2289252 than in non-carriers at 5 years follow-up.

Plasma samples from 88 VTE cases and 85 controls, part of the Venous thromboembolism Biomarker Study (VEBIOS), were screened against 408 candidate proteins targeted by 755 antibodies using multiplex bead arrays. Proteins that significantly associated with VTE after Bonferroni correction were tested for replication in plasma samples of 580 cases and 589 controls from the FARIVE study. In VEBIOS, plasma levels of four proteins, HIVEP1, VWF, GPX3 and PDGFB were significantly associated with VTE after Bonferroni correction. VWF and PDGFB successfully replicated in FARIVE with increased plasma levels in VTE cases compared to controls as initially observed in VEBIOS.

Conclusion: ABO locus was the only shared genetic risk factor between CVD and VTE. A limited set of genetic predictors improved prediction of incident VTE in women at high risk. Our data indicated that interactions among SNPs increase the goodness of fit when predicting VTE. A combination of genotypes i.e. F11 rs2289252 and FVL, may be of potential clinical relevance for risk prediction for recurrence. Affinity plasma proteomics proved to be a valuable research strategy to discover novel biomarkers