Poster Presentation 43rd Lorne Genome Conference 2022

Mendelian randomization identifies a possible causal effect of steroid hormones on colorectal cancer risk (#181)

Dulari K. Jayarathna 1 2 , Miguel E. Rentería 2 3 , Emilie Sauret 4 , Jyotsna Batra 1 3 5 , Neha S. Gandhi 1 5
  1. Centre for Genomics and Personalised Health, School of Chemistry and Physics, Queensland University of Technology, Brisbane, QLD 4000, Australia
  2. Department of Genetics and Computational Biology, QIMR Berghofer Medical Research Institute, Brisbane, QLD 4006, Australia
  3. School of Biomedical Sciences, Queensland University of Technology, Brisbane, QLD 4059, Australia
  4. School of Mechanical, Medical & Process Engineering, Queensland University of Technology, Brisbane, QLD 4000, Australia
  5. Translational Research Institute, Brisbane, QLD 4102, Australia

Colorectal cancer is the third most common cancer in both men and women in Australia. Although colorectal cancer is generally not considered a hormone-dependent malignancy, growing evidence suggests that sex hormones are relevant to its development. Sex hormones are a sub-group of steroid hormones that are synthesized by the ovaries and testes. Nevertheless, previous studies have not considered the causal effect of other steroid hormones (except sex hormones) on colorectal cancer. Herein, we used four Mendelian Randomization (MR)-based estimators; IVW (inverse variance weighted), IVW-radial, IVW-fixed effect, and IVW-multiplicative random effect, to identify risk associated steroid hormones in colorectal cancer. The steroid hormones and colorectal cancer GWAS (genome-wide association studies) summary statistics were downloaded from the Leipzig Health Atlas [1] and the MRC IEU (Integrative Epidemiology Unit) GWAS database [2], respectively. All GWAS summary statistics are from GWAS in European ancestry individuals. We used the Two-Sample MR R package to conduct MR analyses [3].

According to the IVW-MR models, we found evidence for potential causal effects of oestradiol (effect size=-0.1327, p-value=0.0016), progesterone (effect size=0.1694, p-value=2.66E-15), androstenedione (effect size=0.1780, p-value=1.37E-10), and 17-hydroxyprogesterone (effect size=0.0999, p-value=5.19E-7) on colorectal cancer risk. Sasso et al. [4] have revealed that activation of oestradiol and progesterone is necessary to reduce cell proliferation and increase apoptosis in colon tumours through estrogen receptor beta activation. Our study identified for the first time evidence for a possible causal relationship between androstenedione, 17-hydroxyprogesterone and colorectal cancer. These hormone-cancer risk associations add to the understanding of colorectal cancer biology. Future studies are warranted to identify colorectal cancer-hormone causal effects across different populations/ancestries. A similar approach can be applied to identify causal associations of steroid hormones and other cancer/diseases of interest.

  1. Pott, J., et al., Journal of Clinical Endocrinology & Metabolism, 2019. 104(11): p. 5008-5023.
  2. Hemani, G., et al., eLife, 2018. 30(7):e34408
  3. Hemani, G., et al., PLOS Genetics, 2017. 13(11):e1007081.
  4. Sasso, C.V., et al., Endocrine Connections, 2019. 8(3):p. 217-229.