Venous thromboembolism and ovarian cancer risk: a Mendelian randomized study

Abstract

Abstract Observational studies have reported an association between venous thromboembolism (VTE) and ovarian cancer risk (OC), but the evidence is not consistent. In this study, we will investigate the relationship between VTE and OC risk from a genetic perspective. Objective: This study explored the causal relationship between VTE and OC risk. Methods : The single nucleotide polymorphism ( SNP ) data associated with VTE and OC were obtained from the Finn and UK Biobank databases, respectively. All the pooled data based on genome-wide association studies ( GWAS ) were subjected to secondary data analysis, and the genetic loci closely associated with venous thromboembolism were selected as the instrumental variables, and were used as the instrumental variables by Mendelian randomised Egger regression, median weighting, IVW random-effects model, IVW random-effects model, and IVW random-effects model, respectively., IVW random-effects model, IVW radial method, IVW fixed-effects model five methods omodel are five evaluate the causal relationship between venous thromboembolism and ovarian cancer risk by OR value. METHODS: VTE patients (n = 9,176) and OC patients (n = 1,218) who met the same ethnicity and came from different regions were selected. Five methods, namely, Mendelian randomised Egger regression, median weighted method, IVW random effects model, IVW radial method, and IVW fixed effects model, were used to evaluate the causality between VTE and OC risk by OR, respectively. Sensitivity analyses were performed using Cochran's Q test, MR ⁃Egger regression intercept term, MR ⁃PRESSO, and leave-one-out method to assess the stability and reliability of the results. RESULTS: The GWAS for VTE and OC were screened for European ethnicity. In the MR analysis, we found that genetically predicted VTE was associated with an increased risk of OC. The results of Weighted median, Simple mode, Weighted mode, and MR Egger all showed similar trends (OR = 1.0006, 95% CI: 1.00007–1.0013). There was no heterogeneity of results ( P = 0.18) and no horizontal pleiotropy (P = 0.77). The instrumental variables selected for venous thromboembolism were all strong instrumental variables ( F = 669.7). The sensitivity analyses remained consistent, indicating that VTE was associated with a high prevalence of OC. CONCLUSION: The results of this study suggest that patients with VTE are at an increased risk of OC.