A Mendelian randomization study of insulin therapy for type 1 diabetes increasing the potential risk of ovarian cancer

BackgroundType 1 diabetes (T1D) has been associated with a higher risk of Ovarian cancer (OC), albeit the mechanisms underlying this association remain elusive. A better understanding of the relationship between T1D and OC may contribute to improved primary prevention of OC. We aimed to investigate the putative causal role of T1D on OC, and to identify the potentially mediatory effects of the usage of insulin product underlying this relationship. MethodsWe performed a two-sample Mendelian randomization (MR) analysis using genetic variants associated with T1D and OC from genome-wide association studies. Then, a multivariable MR analysis was conducted to investigate whether T1DM has an independent effect on OC after adjusting for potential confounders. Finally, the mediating role of insulin product was subsequently explored using mediation analysis via two-step MR. Resultsthe MR estimated based on IVW method indicated a causal association between genetically determined T1D and Ovarian cancer (OC) (OR: 1.0006, 95% CI 1.0001-1.0011; P = 0.0164). After adjusting for body mass index, Smoking, physical activity, age at menopause and age at menarche, respectively, we found that a causal relationship between T1DM and OC was still statistically significant (OR>1, P <0.05). The two-step MR analysis revealed that insulin product acted as a mediating moderator between the T1D and OC (mediated proportion, 1.07%). ConclusionsOur findings suggest that T1D may confer a risk effect to OC, mediated in part by therapeutic insulin product. Therefore, precise dosage of insulin product or an alternative to insulin in T1D patients have a profound significance in terms of the prevention of OC.