Circulating Sex Hormones and Risk of Colorectal Adenomas and Serrated Lesions in Men

Understanding the role of circulating sex hormones in colorectal carcinogenesis has been hampered by several factors. Data are largely focused on estrogens in women. Only a handful of small studies have examined the role of hormones and colorectal neoplasia in men. Studies examining circulating hormones and colorectal cancer risk (1, 2) and those conducted after androgen deprivation therapy (3, 4) suggest that androgens are protective against colorectal carcinogenesis. Other studies reported no association (5, 6). There has been little consideration for the possibility that these factors may have different associations with colorectal neoplasia in the traditional adenoma versus serrated pathways.

To address this gap, we studied the relationships between testosterone, androstenedione, dehydroepiandrosterone (DHEAS), sex hormone binding globulin (SHBG) levels and preinvasive colorectal neoplasia (adenomas and serrated lesions) in a cohort of men from a large randomized adenoma prevention clinical trial.

The methods for the study have been described in detail previously (7). In brief, participants were enrolled in the Vitamin D/Calcium Polyp Prevention Study. Eligible participants were age 45 to 75 years with a prior history of polyps and a follow-up colonoscopy 3 or 5 years after baseline, scheduled as per gastroenterologist recommendation.

We conducted a nested case–control study within the male study cohort who had a follow-up colonoscopy at least 1 year after randomization. Cases were men who had at least one incident colorectal lesion during the follow-up period. A total of 619 cases were randomly selected from within two groups: “low-risk” participants with a single tubular adenoma < 1 cm during at the end of treatment examination or during an interim colonoscopy (n = 309) and “high-risk” participants with an advanced adenoma (> = 25% villous or >1 cm) or sessile serrated adenoma (SSA) or right-sided serrated polyp or >2 adenomas of any size (n = 310). Controls were defined as men who did not have any precursor lesion during follow-up; 306 were randomly selected for this analysis. In total, there were 925 men (306 no lesion, 309 low risk, 310 high risk) with hormone data included.

At baseline, participants completed questionnaires regarding demographic data, medical history, and lifestyle habits. Serum specimens were drawn at baseline and stored at −80°C. Approximately, 90% of the specimens were obtained >8 hours after last food intake.

Sex hormone assays were performed by ELISA/EIA. All assays were conducted in duplicate and analyzed using BioTek Instruments' Gen5 software. Mean intra-plate CVs were: 6.09% (albumin), 6.82% (androstenedione), 6.62% (DHEAS), 16.81% (SHBG), and 4.01% (testosterone). Mean inter-plates coefficients of variability (CVs) were: 4.33% (albumin), 7.80% (androstenedione), 4.87% (DHEAS), 10.27% (SHBG), and 4.13% (testosterone).

Total free testosterone level was calculated using the Vermuelen formula (8). Logistic regression models were fit to estimate ORs and 95% confidence limits (CI) of hormone quartiles for incident low- and high-risk precursor lesions, adjusting for potential confounders: age, body mass index (BMI), race/ethnicity, and smoking status. We also examined risk for advanced adenomas, SSA, right-sided SSA, and >2 adenomas. In sensitivity analyses among non-Hispanic Whites only (N = 817), we observed no significant differences compared with the entire study population and therefore we present findings for the entire population. We also tested for interactions betwee BMI and hormone levels. Significant interactions were deemed present if P values for interaction were less than 0.05, otherwise interaction terms were omitted from models. Statistical analyses were conducted using Stata version 15.

Data used in this analysis are available upon request.

The mean age of study participants was 59 (SD = 6.9) years with a mean BMI of 28.8 (SD = 4.5) kg/m² (Table 1). There were no statistically significant trends of free testosterone, total testosterone, androstenedione, DHEAS, or SHBG for low- or high-risk lesions or any of the individual lesion endpoints. However, ORs for SSAs in fourth quartile of free testosterone were substantially reduced with borderline statistical significance (Table 2).

In this large prospective study, we observed only limited suggestion for an association between sex hormones or binding globulin levels on risk of early “low-risk” or “high-risk” lesions in the colorectum.

The role for testosterone in the etiology of colorectal cancinogenesis in men is not clear and surprisingly understudied. Conflicting results have been reported for testosterone and colorectal cancer (1–6). Inconsistent findings have also been reported for DHEA and SHBG and colorectal cancer (1, 2, 5, 6). There have been no other studies on the role of free testosterone and early precursor lesions in the colorectum.

Our study has limitations. Estrogens could be mediators of androgens, which we did not measure. Second, this study did not include women, limiting the generalizability of our findings. Strengths of this study include highly sensitive assays of several circulating sex steroid hormones in a large number of men. We were able to examine precursor lesions relevant to both adenoma-carcinoma and serrated pathways, implicated in colorectal carcinogenesis. Men were followed prospectively for the occurrence of incident precursor lesions and a centralized pathology review was conducted. We had good lifestyle data, allowing for appropriate adjustment of potential confounders.

In summary, our results do not support the hypothesis that testosterone and other adrenal hormones are associated with reduction in risk of early precursor lesions.

E.L. Barry reports grants from NIH/NCI during the conduct of the study. L.A. Mott reports grants from NIH/NCI during the conduct of the study. M.N. Passarelli reports grants from NIH outside the submitted work. J.A. Baron reports grants from NCI during the conduct of the study; and Together with Dartmouth College, J.A. Baron Baron holds a use patent for the chemopreventive use of calcium supplementation. No disclosures were reported by the other authors.

J.C. Figueiredo: Supervision, investigation, methodology, writing–original draft, project administration, writing–review and editing. G. Gresham: Formal analysis, writing–original draft, writing–review and editing. E.L. Barry: Resources, investigation, writing–review and editing. L.A. Mott: Formal analysis, writing–original draft, writing–review and editing. M.N. Passarelli: Investigation, methodology, writing–review and editing. P.T. Bradshaw: Methodology, writing–review and editing. C.W. Anderson: Formal analysis, methodology, writing–review and editing. J.A. Baron: Conceptualization, resources, supervision, investigation, methodology, writing–original draft, writing–review and editing.

We thank all study participants in the PPSG. This research was partly supported by NCI R01CA098286.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.