Mendelian confusion when key assumption is in doubt
The authors are to be commended for their effort in investigating an important question using large databases and a Mendelian randomisation approach (1). However, there are several concerns with their report which impact on the conclusions drawn.
As noted by Gill (2), variants in the JMJD1C gene region have no clearly defined physiological pathway to regulate testosterone production and may influence health in other ways, providing multiple potential confounders. Therefore, JMJD1C does not fulfil the criteria for an appropriate instrumental variable, namely one that is associated with the risk factor of interest (testosterone), that is not related to confounders, and that affects the outcome only through the risk factor (3). It is reasonable to conclude that JMJD1C can either affect the outcome through another trait or pathway other than testosterone (horizontal pleiotropy): a violation of the instrumental variable assumptions because the effects of the genetic variant on the outcome are not mediated exclusively through the risk factor.
It is also worth noting that the results in the initial Mendelian randomisation in UK Biobank were not confirmed in the validation cohort. In men from UK Biobank, JMJD1C variants were associated with increased risk of heart failure and thromboembolism, but not with myocardial infarction. In men from CARDIoGRAMplus C4D 1000 Genomes based GWAS, JMJD1C variants were associated with increased risk of myocardial infarction, but outcomes for thromboembolism and heart failure were not reported. Therefore, the results from the two analyses are discordant.
Finally, sex hormone-binding globulin (SHBG) is the major carrier protein for sex hormones in the circulation, serum SHBG correlates strongly with serum testosterone and variants in the SHBG gene are known to modulate testosterone concentrations (4). The fact that in UK Biobank variants in the SHBG gene that were associated with testosterone, were not associated with heart failure or thromboembolism, and in UK Biobank and CARDIoGRAMplus were not associated with myocardial infarction, are important negative findings which argue against a role for genetically determined sex hormone exposure on these outcomes.
We suggest that the findings of Luo et al (1) be interpreted with care, as exploratory results highlighting the need for a better understanding of the physiological role of JMJD1C. Whether testosterone has beneficial, neutral or adverse effects on the heart and vasculature remains uncertain, pending randomised clinical trials powered for cardiovascular outcomes (5).
Bu B. Yeap MBBS, FRACP, PhD, Professor, Medical School, University of Western Australia, and Endocrinologist, Department of Endocrinology and Diabetes, Fiona Stanley Hospital, Perth, Western Australia, Australia.
Gary A. Wittert MBBch, MD, FRACP, FRCP, FAHMS, Professor and Head, Discipline of Medicine, Director, Freemasons Foundation Centre for Men's Health, University of Adelaide and Senior Consultant Endocrinologist, Royal Adelaide Hospital, North Terrace, Adelaide, South Australia, Australia.
Mathis Grossmann MD PhD FRACP, Professor, Department of Medicine (Austin Health), The University of Melbourne and Endocrinologist, Department of Endocrinology, Austin Health, Heidelberg, Victoria, Australia.
1. Luo S, Au Yeung SL, Zhao JV, Burgess S, Schooling CM. Association of genetically predicted testosterone with thromboembolism, heart failure, and myocardial infarction: mendelian randomisation study in UK Biobank. BMJ 2019; 364: j476.
2. Gill D. Plausible genetic instruments. https://www.bmj.com/content/364/bmj.l476/rr
3. Davies NM, Holmes MV, Davey Smith G. Reading Mendelian randomisation studies: a guide, glossary and checklist for clinicians. BMJ 2018; 362: k601.
4. Huhtaniemi I, Pye SR, Holliday KL, et al. Effect of polymorphisms in selected genes involved in pituitary-testicular function on reproductive hormones and phenotype in aging men. J Clin Endocrinol Metab 2010; 95: 1898-1908.
5. Yeap BB, Dwivedi G, Chih HJ, Reid C. Androgens and cardiovascular disease in men. In: Endotext [Internet] Editors: Feingold KR, Anawalt B, Boyce A, et al. South Dartmouth (MA): MDText.com, Inc.; 2000-. Section: Endocrinology of Male Reproduction, Chapter 15, 2019 Feb 2. https://www.endotext.org/
Competing interests: No competing interests