Background: Sex hormones may influence the development of gastrointestinal cancer, but evidence is inconsistent. Methods: We systematically searched MEDLINE and Embase databases to identify prospective studies examining associations between prediagnostic circulating levels of sex hormones and risk of five gastrointestinal cancers: esophageal, gastric, liver, pancreatic, and colorectal cancer. Pooled ORs and 95% confidence intervals (95% CI) were calculated using random-effects models. Results: Among 16,879 identified studies, 29 were included (11 cohort, 15 nested case–control, and three case–cohort studies). Comparing the highest versus lowest tertiles, levels of most sex hormones were not associated with the studied tumors. Higher levels of sex hormone binding globulin (SHBG) were associated with increased risk of gastric cancer (OR = 1.35; 95% CI, 1.06–1.72), but such associations were restricted in men only (OR = 1.43; 95% CI, 1.10–1.85) when stratified by sex. Higher SHBG levels were associated with increased risk of liver cancer (OR = 2.07; 95% CI, 1.40–3.06). Higher testosterone levels were associated with increased risk of liver cancer overall (OR = 2.10; 95% CI, 1.48–2.96), particularly in men (OR = 2.63; 95% CI, 1.65–4.18), Asian populations (OR = 3.27; 95% CI, 1.57–6.83), and in hepatitis B surface antigen-positive individuals (OR = 3.90; 95% CI, 1.43–10.64). Higher levels of SHBG and testosterone were associated with decreased risk of colorectal cancer in men (OR = 0.89; 95% CI, 0.80–0.98 and OR = 0.88; 95% CI, 0.80–0.97, respectively) but not in women. Conclusions: Circulating levels of SHBG and testosterone may influence the risk of gastric, liver, and colorectal cancer. Impact: Further clarifying the role of sex hormones in the development of gastrointestinal cancer may unravel future novel targets for prevention and treatment.

We conducted a case-control study (532 cases and 532 control) in Chinese adults to investigate the independent and interactive effects of dietary nutrients (pro- or anti-inflammation) on Esophageal Squamous Cell Carcinoma (ESCC) risk. Dietary data were collected using a food questionnaire survey that included 171 items. Two algorithms, the Least Absolute Shrinkage and Selector Operation (LASSO) and Bayesian Kernel Machine Regression (BKMR) were employed to select indicators and evaluate the interactive effect of nutrients’ mixture on ESCC risk. Thirteen nutrients were selected, including three pro-inflammatory nutrients (protein, fat and carbohydrate) and ten anti-inflammatory nutrients (fiber, Vitamin A, riboflavin, niacin, Vitamin C, Fe, Se, MUFA, n-3 PUFA and n-6 PUFA). Single-exposure effects of fat, carbohydrate and fiber significantly contributed to ESCC risk. The pro-inflammatory nutrients’ submodel discovered that the combined effect was statistically associated with increased ESCC risk. In addition, a higher fat level was significantly associated with ESCC risk. On the contrary, for fiber and riboflavin, the anti-inflammatory nutrients’ submodel delineated a significant negative effect on the risk of ESCC. Our result implies that dietary nutrients and their inflammatory traits significantly impacted ESCC occurrence. Additional studies are warranted to verify our findings.

Supplementary Figure S6 shows forest plots (1), sensitivity analysis (2) and funnel plots (3) of the associations between prediagnostic circulating sex hormone levels and colon cancer risk.

Supplementary Table S1 shows characteristics of studies included in the meta-analysis.

Supplementary Figure S4 shows forest plots, sensitivity analysis and funnel plot of the associations between prediagnostic circulating sex hormone levels and pancreatic cancer risk.

Supplementary Figure S5 shows forest plots (1), sensitivity analysis (2) and funnel plots (3) of the associations between prediagnostic circulating sex hormone levels and colorectal cancer risk.