The male predominance in oesophageal adenocarcinoma and the role of sex hormones

Abstract

Oesophageal adenocarcinoma is a subtype of oesophageal cancer with rapidly rising incidence over the last decades. This rise in incidence is strongest in western countries and more among men than in women, with a male predominance of up to 9:1 in some countries. Despite several risk factors for oesophageal adenocarcinoma being uncovered, the male predominance is largely unexplained. The aim of this thesis was to investigate the aetiology behind oesophageal adenocarcinoma and its male predominance, with focus on sex hormones.

Study I was a systematic review and meta-analysis investigating the association between obesity-related serum biomarkers and the risk of oesophageal adenocarcinoma and its precursor condition, Barrett's oesophagus. A literature search in MEDLINE and EMBASE was conducted, covering studies published until October 2018. Out of 7,641 studies, 19 were included (12 cross-sectional, 2 nested case-control, and 5 cohort studies). Random-effects meta-analysis pooled the odds ratios comparing the highest and lowest categories of biomarker levels. No associations were found for adiponectin, ghrelin, insulin-like growth factor 1, insulin-like growth factor-binding protein 3, triglycerides, Interleukin 8, or tumour necrosis factor alpha. Higher levels of leptin, glucose, insulin, C-reactive protein, interleukin 6, and soluble tumour necrosis factor receptor 2 were potentially linked with an increased risk of oesophageal adenocarcinoma or Barrett's oesophagus

Study II was a nested case-control study in 1972-2016 examining how circulating sex hormone levels influence the risk of oesophageal adenocarcinoma. The source population was the Norwegian Janus Serum Bank Cohort, and this study included 244 male patients with oesophageal adenocarcinoma and 244 male age-matched control participants. Associations between prediagnostic circulating levels of 12 sex hormones and risk of oesophageal adenocarcinoma were analysed using conditional logistic regression. Associations were observed for testosterone, testosterone:oestradiol ratio and luteinizing hormone, but not for sex hormone–binding globulin, dehydroepiandrosterone sulfate, follicle-stimulating hormone, prolactin, 17-OH progesterone, progesterone, androstenedione, or free testosterone index. In addition, a random-effects meta-analysis, combining the results of this study with a similar prospective study, revealed an inverse association between testosterone levels and the risk of oesophageal adenocarcinoma (pooled OR = 0.60, 95% CI 0.38–0.97). No associations were found for androstenedione, sex hormone–binding globulin, oestradiol, or testosterone:oestradiol ratio in the meta-analysis.

Study III was a nationwide Swedish population-based cohort study between 2005 and 2018 comparing the risk of oesophageal and gastric tumours among 191,156 users and 1,911,560 (10 times as many) non-users of 5α-reductase inhibitors (5-ARIs). Multivariable Cox regression provided hazard ratios (HR) with 95% confidence intervals (CI) adjusted for age, calendar year, smoking, non-steroidal anti-inflammatory drugs/aspirin use, and statins use. Further adjustments were made depending on the tumour analysed. Users of 5-ARIs had a substantially decreased risk of developing oesophageal or cardia adenocarcinoma among those obese or diabetic (adjusted HR 0.55, 95% CI 0.39-0.80), and a decreased risk of oesophageal squamous cell carcinoma (adjusted HR 0.49, 95% CI 0.37-0.65).

Study IV was a nationwide Swedish population-based cohort study hypothesizing that the use of 5-ARIs improves survival in patients with oesophago-gastric cancer. The study included men who underwent surgery for oesophageal or gastric cancer between 2006 and 2015, with follow-up until the end of 2020. Out of the 1769 patients diagnosed with oesophago-gastric cancer, 64 (3.6%) had a history of using 5-ARIs. Multivariable Cox regression estimated HRs for associations between 5-ARIs use and 5-year all-cause mortality (main outcome) and 5-year disease-specific mortality (secondary outcome) with adjustment for confounders. Compared to non-users, users of 5-ARIs were not at any decreased risk of 5-year all-cause mortality (adjusted HR 1.13, 95% CI 0.79-1.63) or 5-year disease-specific mortality (adjusted HR 1.10, 95% CI 0.79-1.52) in gastro-oesophageal cancer.