Sex hormones in the risk of breast cancer: a two-sample Mendelian randomization study

Abstract

Multiple evidence has suggested the essential role of sex hormones in the susceptibility of breast cancer. However, whether there exists a causal association and the effect direction remains controversial. To examine the causative role of hormones in the risk of breast cancer, we first estimated their genetic correlation, and then conducted two-sample and multivariable Mendelian randomization analyses using summary statistics from genome-wide association studies of major sex hormones including testosterone (N=230,454), estradiol (N=163,985) and progesterone (N=1,261), together with breast cancer (N=228,951). We further performed subtype analysis focusing on estrogen receptor (ER)+ breast cancer (N=175,475) and ER- breast cancer (N=127,442), and conducted extensive sensitivity analyses. We identified significant positive genetic correlation between testosterone level and risk of breast cancer (genetic correlation: 0.09, P=1.10E-03). Genetically determined higher total testosterone level was associated with an increased risk of breast cancer (OR: 1.11, 95% CI: 1.06-1.16, P=4.55E-06). In the subtype analysis, higher total testosterone was associated with an increased risk of ER+ breast cancer (OR: 1.18, 95% CI: 1.11-1.26, P=6.00E-08). In contrast, no association was identified between estradiol, progesterone and the risk of breast cancer. These results elucidated the causal role of major sex hormones in the risk of breast cancer, especially in ER+ breast cancer. Future development of preventive or therapeutic interventions in clinical trials could attach importance to this.

Keywords: Mendelian randomization; Sex hormone; breast cancer.

Figure 1

Genetic correlation between sex hormones and breast cancer. (A-C) Genetic correlation estimated using the LDSC method for (A) breast cancer, (B) ER+ breast cancer, and (C) ER- breast cancer. (D-F) Genetic correlation estimated using the GNOVA method for (D) breast cancer, (E) ER+ breast cancer, and (F) ER- breast cancer. Error bars indicate 95% confidence intervals. ER, Estrogen Receptor.

Figure 2

Forest plot showing results from the Mendelian randomization analysis. (A-C) Results from the Mendelian randomization analysis to evaluate causal role of sex hormones in (A) breast cancer, (B) ER+ breast cancer, (C) and ER- breast cancer using the inverse variance weighted method. (D-F) Results from the Mendelian randomization analysis to evaluate causal role of sex hormones in (D) breast cancer, (E) ER+ breast cancer, and (F) ER- breast cancer using the weighted median method. Estimates are per 1 standard deviation (SD) increase in the trait. ER, Estrogen Receptor.