Abstract
Epidemiologic studies evaluating associations between sex steroid hormones and colorectal cancer risk have yielded inconsistent results. To elucidate the role of circulating levels of testosterone, and sex hormone-binding globulin (SHBG) in colorectal cancer risk, we conducted observational and Mendelian randomization (MR) analyses.
The observational analyses included 333,530 participants enrolled in the UK Biobank with testosterone and SHBG measured. HRs and 95% confidence intervals (CI) were estimated using multivariable Cox proportional hazards models. For MR analyses, genetic variants robustly associated with hormone levels were identified and their association with colorectal cancer (42,866 cases/42,752 controls) was examined using two-sample MR.
In the observational analysis, there was little evidence that circulating levels of total testosterone were associated with colorectal cancer risk; the MR analyses showed a greater risk for women (OR per 1-SD = 1.09; 95% CI, 1.01–1.17), although pleiotropy may have biased this result. Higher SHBG concentrations were associated with greater colorectal cancer risk for women (HR per 1-SD = 1.16; 95% CI, 1.05–1.29), but was unsupported by the MR analysis. There was little evidence of associations between free testosterone and colorectal cancer in observational and MR analyses.
Circulating concentrations of sex hormones are unlikely to be causally associated with colorectal cancer. Additional experimental studies are required to better understand the possible role of androgens in colorectal cancer development.
Our results from large-scale analyses provide little evidence for sex hormone pathways playing a causal role in colorectal cancer development.
See related commentary by Hang and Shen, p. 1302
Introduction
Colorectal cancer is one of the most common cancers worldwide with lower incidence rates in women compared with men (1). It has been proposed that differing exposures to endogenous and exogenous sex steroid hormones may contribute to this sex disparity (2). Higher concentrations of endogenous or exogenous estrogens in women may confer a protective role against colorectal cancer development (2, 3), whereas longer-term use of androgen deprivation therapy has been associated with elevated colorectal cancer risk in men (4).
Inconsistent results have been found in the few relatively small epidemiologic studies that examined the association between circulating testosterone concentrations and colorectal cancer risk. In a pooled analysis of four US based studies, an inverse association was found between testosterone levels and colorectal cancer among men, but not women (5), whereas a recent Japanese prospective study of postmenopausal women, reported a positive association between testosterone and colorectal cancer risk (6).
Sex hormone-binding globulin (SHBG) is a hepatically-derived glycoprotein and principal transport protein of estrogens and testosterone, and is therefore an important regulator of their bioactivity. In an analysis nested within the Women's Health Initiative Clinical Trial (WHI-CT), we reported a more than twofold higher colorectal cancer risk when the highest and lowest SHBG concentrations exposure groups were compared (7). However, inconsistent results have been found in other smaller studies that have examined the relationship between circulating SHBG concentrations and colorectal cancer (5, 6, 8).
To further examine associations of circulating testosterone and SHBG concentrations with colorectal cancer risk, we conducted complementary observational and Mendelian randomization (MR) analyses. First, we investigated associations of prediagnostic circulating concentrations of total testosterone, free testosterone, and SHBG with colorectal cancer risk in the UK Biobank study, a large prospective cohort of more than 500,000 participants. We then employed MR to help strengthen causal inference by using genetic variants robustly related to circulating sex steroid hormone concentrations from a recent genome-wide association study (GWAS) in UK Biobank (9), and then assessed the relation of these variants with colorectal cancer from large genetic consortia including 42,886 colorectal cancer cases and 42,752 controls (10).
Materials and Methods
UK biobank: observational analysis
Study participants
The UK Biobank is a prospective cohort study of 502,656 adults ages 40 to 69 years who were recruited between 2006 and 2010 (11). The UK Biobank is approved by the North West Multi-centre Research Ethics Committee, the National Information Governance Board for Health and Social Care in England and Wales, and the Community Health Index Advisory Group in Scotland. Since 2004, an independent Ethics and Governance Council additionally oversees UK Biobank's continuous adherence to the Ethics and Governance Framework (http://www.ukbiobank.ac.uk/ethics/). This research has been conducted using the UK Biobank Resource under application number 25897.
At baseline, participants completed a self-administered touchscreen questionnaire, with questions on sociodemographics (such as age, sex, educational level, and postcode, which were used to calculate the Townsend deprivation score; ref. 12), health/medical history, and lifestyle exposures (including smoking related phenotypes, physical activity, dietary intakes, and alcohol consumption). Several anthropometric measurements were also collected, such as body weight, height, and waist circumference. At baseline, blood samples were collected from all participants, and from a subset of ∼20,000 participants repeat blood samples were also collected during a follow-up visit between 2012 and 2013. Blood samples were centrifuged, and serum stored at −80°C (13).
We excluded the following participants: those who reported having had a diagnosis of cancer at recruitment to help reduce reverse causality as an explanation for any observed associations (n = 27,264 prevalent cases, self-reported and cancer registry identified); participants with missing data on body-size measurements (n = 3,032); self-reported prevalent type-2 diabetes (T2D) or unknown diabetes status at recruitment (as diabetes medications can affect the concentrations of sex steroid hormones (14–16); n = 26,698); women who reported oral contraceptive or menopausal hormone use (as our focus was on endogenous circulating hormone levels; n = 19,802); and participants without a total testosterone, SHBG, or albumin (required to estimate free testosterone concentration) measurement (n = 92,330). Our analysis therefore included 333,530 participants (160,650 women and 172,880 men; Fig. 1).
Flowchart of the exclusion criteria of the study participants in UK Biobank.
Laboratory methods
As part of the UK Biobank Biomarker Project, serum concentrations of testosterone, SHBG, and insulin-like growth factor-1 (IGF-1) were determined by a chemiluminescent immunoassay. Serum high sensitivity C-reactive protein (CRP) concentrations were assayed by the immuno-turbidimetric method. For glycated hemoglobin (HbA1c), the HPLC Variant II Turbo 2.0 system was used. A detailed description of assay performance can be found elsewhere (17). The average within-laboratory (total) coefficient of variation (CV) for low, medium, and high internal quality control level samples for each biomarker ranged from 3.7% to 8.3% for total testosterone and 5.2% to 5.7% for SHBG (17). Free testosterone concentrations were calculated with the Vermeulen equation using measured albumin concentration available for each participant (18, 19). A total of 10,573 and 11,519 participants had SHBG and testosterone concentrations measured, respectively, in blood samples collected both at recruitment and at the repeat assessment visit (median of 4 years apart).
Assessment of outcome
The UK Biobank cohort is linked to national cancer and death registries used to determine incident colorectal cancer cases and cancer cases recorded first in death certificates. Complete follow-up was available through March 31, 2016, for England and Wales and October 31, 2015, for Scotland. The 10th Revision of the International Classification of Diseases (ICD10) was used to code incidence cancer data. We classified as proximal colon cancers those found within the caecum, appendix, ascending colon, hepatic flexure, transverse colon, and splenic flexure (C18.0–18.5). Distal colon cancers were considered those found within the descending (C18.6) and sigmoid (C18.7) colon. Overlapping (C18.8) and unspecified (C18.9) lesions of the colon were included in colon cancers only. Rectal cancers were classified those at the recto-sigmoid junction (C19) and rectum (C20).
Statistical analysis
Intraclass correlation coefficients (ICC) were used to estimate the reproducibility between the two measurements of SHBG and testosterone available in a subsample of participants. These were obtained dividing the between-person variance by the sum of the between-person and within-person variances.
Cox proportional hazards models were used to estimate HRs and 95% confidence intervals (CI). We used age was the primary time variable. In particular, time at entry was age at recruitment and exit time was age at whichever of the following came first: colorectal cancer diagnosis, death, or the last date at which follow-up was considered complete. Stratification by age at recruitment in 5-year categories, Townsend deprivation index (quintiles), and region of the recruitment assessment center was used in all models. Analyses were conducted separately for men and women, and also according to anatomical subsite (colon, proximal colon, distal colon, and rectal cancer). Total testosterone, free testosterone, and SHBG were modelled with participants grouped into sex-specific quintiles of circulating concentrations and on the continuous scale. To allow us to compare the continuous model results with the MR estimates we used the following transformations: for total testosterone concentration, an inverse normal transformation of the rank was used for women and men; for free testosterone, a natural logarithmic transformation was used for women and an inverse normal transformation of the rank for men; and for SHBG, an inverse normal transformation of the rank was used for women and a natural logarithmic transformation for men (Supplementary Fig. S1).
Statistical tests for trend were calculated using the ordinal quintiles of sex steroid hormones entered into the model as a continuous variable. Continuous scale HRs were additionally corrected for regression dilution using regression dilution ratios obtained from the subsample of participants with repeated testosterone and SHBG measurements (20, 21). Regression dilution ratios are calculated as the ratio of the difference between the means of the follow-up measurements of sex steroid hormones of participants in the highest and lowest and quintiles divided by the respective estimates at baseline. To obtain the corrected continuous HRs, the log HRs and their standard errors were divided by the regression dilution ratio for total testosterone (i.e., 0.65 in women and 0.68 in men), free testosterone (i.e., 0.71 in women and 0.57 in men), and SHBG (i.e., 0.82 in women and 0.83 in men), and then exponentiated (22). All models met the proportional hazards assumption, assessed through analyses of Schoenfeld residuals (23).
Our primary multivariable model 1 was adjusted for a set of a priori-determined colorectal cancer risk factors. In particular we adjusted for waist circumference, total physical activity, height, alcohol consumption frequency, smoking status and intensity, frequency of red and processed meat consumption, family history of colorectal cancer, educational level, regular aspirin/ibuprofen use, and ever use of hormone replacement therapy. We also considered models additionally adjusted for inflammation markers and glycemic pathways that correlate with sex steroid hormone concentrations, and are associated with colorectal cancer risk, namely CRP, IGF-1, and HbA1c (6, 7, 24, 25). The testosterone and SHBG multivariable model were mutually adjusted.
Sensitivity analyses excluding colorectal cancer cases occurring in the first 2 years of follow-up were performed. We also performed sensitivity analyses excluding women who were ever menopausal hormone users (N = 50,948) or those with polycystic ovary syndrome (PCOS; N = 329). Analyses for sex steroid hormones on the continuous scale were repeated excluding possible outliers (defined as sex hormone concentrations more than 1.5 times the interquartile range above the third quartile or below the first quartile). We further assessed associations of circulating total testosterone, free testosterone and SHBG with colorectal cancer across subgroups of body mass index (BMI; <25, ≥25 kg/m2), waist-to-hip ratio (WHR; <median, ≥above median), age at recruitment (<60, ≥60 years), follow-up time (<5, ≥5 years), and menopausal status (pre-, post-). The likelihood ratio test was used to evaluate interactions between these variables and circulating sex steroid hormones concentrations.
MR analysis
Data on total testosterone, free testosterone, and SHBG
We selected genetic variants associated with circulating total testosterone, free testosterone, and SHBG concentrations at the genome-wide significant level (i.e., P value threshold for inclusion at <5 × 10−8) from the largest GWAS conducted to date (9). We used data from 230,454 women and 194,453 men of European ancestry for total testosterone, 188,507 women and 178,782 men for free testosterone, and 189,473 for women and 180,726 for men for SHBG. Genotyping chip, age at baseline and 10 genetically derived principal components to account for population stratification were included as covariates in the analysis. For SHBG, BMI was also included as a covariate. However, in the MR analysis for SHBG, genetic loci from the BMI-adjusted analyses were used with corresponding effect estimates from the BMI-unadjusted analyses to mitigate possible collider bias (26).
Data on colorectal cancer
Summary data for associations of the hormone-related variants with colorectal cancer were obtained from a meta-analysis of GWAS involving 85,638 participants (42,886 colorectal cancer cases and 42,752 controls) within the ColoRectal Transdisciplinary Study (CORECT), the Colon Cancer Family Registry (CCFR), and the Genetics and Epidemiology of Colorectal Cancer (GECCO) consortia (10). The GWAS was adjusted for age, sex, genotyping platform, and genomic principal components.
Statistical analysis
We conducted two-sample MR analyses to appraise the potential causal nature of the associations between total testosterone, free testosterone, and SHBG with colorectal cancer risk. Where a variant used as an instrument for one of the hormones of interest was not present in the colorectal cancer GWAS, we identified a 1,000 Genomes proxy with r2 > 0.8. For our main analysis, we used a random-effects inverse-variance weighted (IVW) method (27, 28).
Sensitivity analyses
We conducted sensitivity analyses to mitigate against any pleiotropic effects. We undertook MR-Egger regression (29) and computed the estimator from the weighted median approach (30) to assess the possible influence of horizontal pleiotropy on the effect estimates. We calculated the Cochran Q statistic that quantifies the heterogeneity in effect sizes attributed to the selected genetic variants (31). We also estimated the intercept term from the MR-Egger regression, with a deviation from zero being indicative of directional (nonbalanced horizontal) pleiotropy (29). We excluded genetic variants having larger effects (based on standardized beta) on any one of 11 metabolic traits available in the UK Biobank (fasting glucose, T2D, coronary artery disease, high density lipoprotein cholesterol, low-density lipoprotein cholesterol, triglycerides, total-cholesterol, and diastolic and systolic blood pressure, BMI and waist-to-hip ratio adjusted for BMI), an approach known as Steiger filtering (32). A list of pleiotropic variants for total testosterone, free testosterone, and SHBG can be found in the published GWAS (9). Finally, we used only cis variants at the SHBG gene locus (rs1799941, rs6258). Variant rs1799941 is common, whereas rs6258 is rare and alters SHBG's binding affinity for testosterone (33, 34).
All the observational analyses were implemented in Stata 13.1, whereas for the MR analyses, we used the Mendelian randomization R package (35).
Results
UK biobank: observational analysis
After a median follow-up time of 7.1 years (interquartile range = 6.4–7.7), 2,258 colorectal cancer cases were recorded (833 in women and 1,425 in men). In both women and men, compared with noncases, individuals with colorectal cancer were older, had higher BMI, were more likely to have a family history of colorectal cancer and eat red and processed meat more frequently, and were less likely to be current smokers (Table 1). Participant characteristics according to quintiles of total testosterone, free testosterone, and SHBG are presented in Supplementary Table S1.
Characteristics of UK Biobank study participants (n = 333,530 participants).
. | Women (n = 160,650) . | Men (n = 172,880) . | ||
---|---|---|---|---|
. | Colorectal cancer cases (N = 833) . | Noncases (N = 159,817) . | Colorectal cancer cases (N = 1,425) . | Noncases (N = 171,455) . |
Age at recruitment (years)a | 59.5 (7.1) | 55.8 (8.1) | 61.1 (6.3) | 56.3 (8.2) |
Body mass index (kg/m2)a | 27.2 (4.9) | 27.0 (5.1) | 28.1 (4.0) | 27.6 (4.0) |
Waist circumference (cm)a | 85.5 (12.3) | 84.4 (12.2) | 98.6 (10.6) | 96.2 (10.9) |
Height (cm)a | 162.2 (6.2) | 162.6 (6.3) | 175.4 (6.6) | 175.8 (6.8) |
Total physical activity (MET hours per week) | ||||
<10 | 25.0% | 23.0% | 19.6% | 20.5% |
≥60 | 19.4% | 20.0% | 26.0% | 24.8% |
Smoking status | ||||
Never | 57.6% | 60.4% | 39.8% | 49.9% |
Current | 8.3% | 9.1% | 11.8% | 12.6% |
Alcohol consumption | ||||
Never | 9.2% | 8.7% | 4.4% | 5.8% |
daily/almost daily | 19.3% | 16.3% | 31.7% | 25.8% |
Socio-economic status (Townsend deprivation index) | ||||
Highest quintile | 20.9% | 19.6% | 18.9% | 20.3% |
Family history (first degree relative) of colorectal cancer | ||||
Yes | 12.2% | 10.4% | 15.0% | 11.0% |
Regular aspirin/ibuprofen use | ||||
Yes | 23.5% | 24.4% | 27.6% | 26.3% |
Red and processed meat | ||||
<2 occasions per week | 17.2% | 18.9% | 6.0% | 9.0% |
≥4 occasions per week | 32.3% | 30.8% | 57.1% | 51.7% |
Ever menopausal hormone useb | ||||
Yes | 41.9% | 31.7% | ||
Menopausal statusb | ||||
Postmenopausal | 82.0% | 66.3% | ||
C-reactive protein (CRP; mg/L)a | 2.9 (4.5) | 2.5 (4.0) | 3.0 (4.6) | 2.4 (4.2) |
Total testosterone (nmol/L)a | 1.1 (0.9) | 1.1 (0.6) | 11.7 (3.5) | 12.1 (3.7) |
Free testosterone (pmol/L)a | 14.7 (13.8) | 14.6 (10.5) | 199.0 (53.8) | 210.7 (60.6) |
Sex hormone binding globulin (SHBG; nmol/L)a | 61.5 (28.4) | 60.8 (27.9) | 41.5 (17.7) | 39.8 (16.6) |
IGF-1 (nmol/L)a | 20.9 (5.5) | 21.3 (5.6) | 21.5 (5.8) | 22.0 (5.4) |
Glycated hemoglobin (HbA1c; mmol/mol)a | 35.8 (4.3) | 35.1 (4.3) | 35.7 (4.5) | 35.2 (5.0) |
. | Women (n = 160,650) . | Men (n = 172,880) . | ||
---|---|---|---|---|
. | Colorectal cancer cases (N = 833) . | Noncases (N = 159,817) . | Colorectal cancer cases (N = 1,425) . | Noncases (N = 171,455) . |
Age at recruitment (years)a | 59.5 (7.1) | 55.8 (8.1) | 61.1 (6.3) | 56.3 (8.2) |
Body mass index (kg/m2)a | 27.2 (4.9) | 27.0 (5.1) | 28.1 (4.0) | 27.6 (4.0) |
Waist circumference (cm)a | 85.5 (12.3) | 84.4 (12.2) | 98.6 (10.6) | 96.2 (10.9) |
Height (cm)a | 162.2 (6.2) | 162.6 (6.3) | 175.4 (6.6) | 175.8 (6.8) |
Total physical activity (MET hours per week) | ||||
<10 | 25.0% | 23.0% | 19.6% | 20.5% |
≥60 | 19.4% | 20.0% | 26.0% | 24.8% |
Smoking status | ||||
Never | 57.6% | 60.4% | 39.8% | 49.9% |
Current | 8.3% | 9.1% | 11.8% | 12.6% |
Alcohol consumption | ||||
Never | 9.2% | 8.7% | 4.4% | 5.8% |
daily/almost daily | 19.3% | 16.3% | 31.7% | 25.8% |
Socio-economic status (Townsend deprivation index) | ||||
Highest quintile | 20.9% | 19.6% | 18.9% | 20.3% |
Family history (first degree relative) of colorectal cancer | ||||
Yes | 12.2% | 10.4% | 15.0% | 11.0% |
Regular aspirin/ibuprofen use | ||||
Yes | 23.5% | 24.4% | 27.6% | 26.3% |
Red and processed meat | ||||
<2 occasions per week | 17.2% | 18.9% | 6.0% | 9.0% |
≥4 occasions per week | 32.3% | 30.8% | 57.1% | 51.7% |
Ever menopausal hormone useb | ||||
Yes | 41.9% | 31.7% | ||
Menopausal statusb | ||||
Postmenopausal | 82.0% | 66.3% | ||
C-reactive protein (CRP; mg/L)a | 2.9 (4.5) | 2.5 (4.0) | 3.0 (4.6) | 2.4 (4.2) |
Total testosterone (nmol/L)a | 1.1 (0.9) | 1.1 (0.6) | 11.7 (3.5) | 12.1 (3.7) |
Free testosterone (pmol/L)a | 14.7 (13.8) | 14.6 (10.5) | 199.0 (53.8) | 210.7 (60.6) |
Sex hormone binding globulin (SHBG; nmol/L)a | 61.5 (28.4) | 60.8 (27.9) | 41.5 (17.7) | 39.8 (16.6) |
IGF-1 (nmol/L)a | 20.9 (5.5) | 21.3 (5.6) | 21.5 (5.8) | 22.0 (5.4) |
Glycated hemoglobin (HbA1c; mmol/mol)a | 35.8 (4.3) | 35.1 (4.3) | 35.7 (4.5) | 35.2 (5.0) |
Abbreviation: MET, metabolic equivalents.
aMean and SD.
bAmong women only.
The reproducibility (ICC) of testosterone (n = 11,519 participants; 4,669 women and 6,850 men; median of 4 years apart) was 0.59 (95% CI, 0.58–0.61) for women and 0.66 (95% CI, 0.64–0.67) for men. The ICC of SHBG concentrations measured at both the recruitment and repeat assessment visit (n = 10,573 participants; 4,459 women and 6,114 men) was 0.77 (95% CI, 0.76–0.79) for women and 0.82 (95% CI, 0.75–0.87) for men.
Association of circulating total testosterone and free testosterone concentrations with colorectal cancer risk
In the multivariable model 2 additionally adjusted for circulating concentrations of CRP, HbA1c, SHBG, and IGF-1, there was little evidence that a 1-SD increment of total testosterone concentration was associated with colorectal cancer risk for women (HR = 1.00; 95% CI, 0.90–1.11) and men (HR = 0.97; 95% CI, 0.88–1.07; Table 2; Supplementary Table S2). When stratifying by anatomical subsite, no association between circulating total testosterone concentration and colon cancer was found for women (HR per 1-SD increment = 1.04; 95% CI, 0.92–1.18) and men (HR per 1-SD increment = 0.94; 95% CI, 0.83–1.07); a similar pattern of associations were found for proximal and distal colon cancers (Table 2; Supplementary Table S2).
Risk (HRs) of colorectal cancer associated with circulating total testosterone, free testosterone, and sex hormone binding globulin (SHBG) levels in the UK Biobank.
. | Colorectal cancer . | Colon cancer . | Proximal colon cancer . | Distal colon cancer . | Rectal cancer . |
---|---|---|---|---|---|
Total testosteronea | |||||
Women | |||||
Q1 | 1 | 1 | 1 | 1 | 1 |
Q2 | 0.95 (0.77–1.17) | 1.01 (0.79–1.29) | 1.05 (0.77–1.42) | 0.97 (0.64–1.48) | 0.79 (0.52–1.19) |
Q3 | 1.01 (0.82–1.25) | 1.03 (0.81–1.32) | 0.70 (0.49–1.00) | 1.48 (1.01–2.16) | 0.95 (0.64–1.42) |
Q4 | 0.95 (0.77–1.18) | 1.00 (0.78–1.29) | 0.90 (0.64–1.26) | 1.17 (0.78–1.75) | 0.82 (0.54–1.25) |
Q5 | 0.98 (0.79–1.22) | 1.03 (0.80–1.33) | 0.99 (0.71–1.38) | 1.03 (0.68–1.57) | 0.86 (0.56–1.30) |
P-trend | 0.89 | 0.86 | 0.59 | 0.59 | 0.56 |
HR per 1-SD increment | 1.00 (0.93–1.07) | 1.03 (0.95–1.11) | 0.99 (0.89–1.11) | 1.04 (0.92–1.18) | 0.92 (0.80–1.05) |
HR per 1-SD increment (adjusted)b | 1.00 (0.90–1.11) | 1.04 (0.92–1.18) | 0.99 (0.84–1.17) | 1.07 (0.88–1.30) | 0.88 (0.71–1.08) |
Men | |||||
Q1 | 1 | 1 | 1 | 1 | 1 |
Q2 | 1.12 (0.95–1.31) | 1.13 (0.92–1.38) | 0.95 (0.71–1.27) | 1.30 (0.97–1.78) | 1.11 (0.86–1.44) |
Q3 | 1.01 (0.85–1.21) | 1.08 (0.87–1.34) | 0.82 (0.60–1.12) | 1.36 (0.99–1.88) | 0.92 (0.69–1.23) |
Q4 | 0.96 (0.80–1.16) | 0.94 (0.74–1.19) | 0.85 (0.61–1.19) | 1.01 (0.70–1.45) | 1.00 (0.75–1.35) |
Q5 | 0.91 (0.74–1.13) | 0.87 (0.66–1.13) | 0.69 (0.47–1.01) | 1.14 (0.78–1.69) | 0.99 (0.72–1.37) |
P-trend | 0.19 | 0.16 | 0.06 | 0.97 | 0.75 |
HR per 1-SD increment | 0.98 (0.92–1.05) | 0.96 (0.88–1.05) | 0.92 (0.82–1.04) | 1.01 (0.89–1.14) | 1.01 (0.91–1.12) |
HR per 1-SD increment (adjusted)b | 0.97 (0.88–1.07) | 0.94 (0.83–1.07) | 0.89 (0.74–1.06) | 1.01 (0.84–1.22) | 1.01 (0.87–1.18) |
Free testosterone | |||||
Women | |||||
Q1 | 1 | 1 | 1 | 1 | 1 |
Q2 | 0.80 (0.65–0.99) | 0.90 (0.70–1.16) | 0.90 (0.65–1.24) | 0.86 (0.57–1.29) | 0.58 (0.38–0.88) |
Q3 | 0.82 (0.66–1.01) | 0.90 (0.70–1.16) | 0.76 (0.54–1.07) | 1.02 (0.69–1.51) | 0.64 (0.43–0.97) |
Q4 | 0.87 (0.70–1.08) | 0.94 (0.73–1.21) | 0.92 (0.66–1.28) | 0.88 (0.58–1.33) | 0.73 (0.49–1.10) |
Q5 | 0.83 (0.66–1.04) | 0.91 (0.69–1.18) | 0.77 (0.53–1.10) | 0.94 (0.62–1.43) | 0.66 (0.43–1.02) |
P-trend | 0.23 | 0.59 | 0.21 | 0.85 | 0.16 |
HR per 1-unit increment (log scale) | 0.93 (0.81–1.06) | 0.98 (0.84–1.14) | 0.89 (0.72–1.09) | 1.01 (0.80–1.29) | 0.79 (0.61–1.03) |
HR per 1-unit increment (log scale-adjusted)c | 0.90 (0.75–1.08) | 0.97 (0.78–1.20) | 0.85 (0.63–1.13) | 1.02 (0.73–1.43) | 0.72 (0.51–1.04) |
Men | |||||
Q1 | 1 | 1 | 1 | 1 | 1 |
Q2 | 0.95 (0.82–1.11) | 0.96 (0.80–1.17) | 0.75 (0.57–0.99) | 1.23 (0.93–1.61) | 0.94 (0.73–1.20) |
Q3 | 0.95 (0.81–1.11) | 0.83 (0.68–1.02) | 0.74 (0.56–0.99) | 0.91 (0.67–1.24) | 1.15 (0.90–1.48) |
Q4 | 1.01 (0.86–1.19) | 0.98 (0.80–1.21) | 0.87 (0.65–1.16) | 1.15 (0.85–1.56) | 1.08 (0.83–1.40) |
Q5 | 0.91 (0.76–1.09) | 0.87 (0.69–1.10) | 0.77 (0.56–1.07) | 1.02 (0.72–1.44) | 0.99 (0.74–1.32) |
P-trend | 0.59 | 0.29 | 0.20 | 0.98 | 0.62 |
HR per 1-SD increment | 0.99 (0.93–1.05) | 0.98 (0.91–1.05) | 0.94 (0.85–1.04) | 1.03 (0.92–1.14) | 1.01 (0.92–1.10) |
HR per 1-SD increment (adjusted)c | 0.98 (0.89–1.08) | 0.96 (0.85–1.09) | 0.90 (0.75–1.08) | 1.04 (0.87–1.26) | 1.01 (0.86–1.18) |
SHBGd | |||||
Women | |||||
Q1 | 1 | 1 | 1 | 1 | 1 |
Q2 | 1.12 (0.90–1.40) | 1.07 (0.83–1.37) | 1.17 (0.84–1.64) | 1.00 (0.68–1.49) | 1.31 (0.83–2.06) |
Q3 | 1.02 (0.81–1.29) | 0.93 (0.71–1.22) | 0.91 (0.62–1.32) | 0.94 (0.61–1.43) | 1.33 (0.83–2.14) |
Q4 | 1.39 (1.10–1.76) | 1.26 (0.96–1.65) | 1.37 (0.95–1.98) | 1.25 (0.82–1.91) | 1.86 (1.16–2.98) |
Q5 | 1.40 (1.09–1.81) | 1.30 (0.97–1.74) | 1.48 (1.00–2.19) | 1.28 (0.81–2.02) | 1.76 (1.05–2.94) |
P-trend | 0.002 | 0.045 | 0.044 | 0.18 | 0.012 |
HR per 1-SD increment | 1.13 (1.04–1.23) | 1.13 (1.02–1.24) | 1.19 (1.04–1.36) | 1.11 (0.96–1.30) | 1.15 (0.97–1.35) |
HR per 1-SD increment (adjusted)e | 1.16 (1.05–1.29) | 1.15 (1.03–1.30) | 1.24 (1.05–1.45) | 1.14 (0.95–1.37) | 1.18 (0.97–1.44) |
Men | |||||
Q1 | 1 | 1 | 1 | 1 | 1 |
Q2 | 0.92 (0.77–1.10) | 0.94 (0.75–1.18) | 0.84 (0.61–1.17) | 0.92 (0.66–1.30) | 0.89 (0.68–1.18) |
Q3 | 0.81 (0.67–0.97) | 0.91 (0.72–1.15) | 0.91 (0.65–1.27) | 0.90 (0.64–1.28) | 0.66 (0.49–0.90) |
Q4 | 0.91 (0.75–1.11) | 0.97 (0.76–1.24) | 0.87 (0.61–1.23) | 0.96 (0.67–1.37) | 0.83 (0.61–1.12) |
Q5 | 1.01 (0.82–1.25) | 1.09 (0.83–1.42) | 1.12 (0.77–1.64) | 0.99 (0.66–1.47) | 0.90 (0.65–1.26) |
P-trend | 0.86 | 0.51 | 0.60 | 0.93 | 0.56 |
HR per 1-unit increment (log scale) | 1.03 (0.86–1.23) | 1.08 (0.86–1.36) | 1.13 (0.82–1.57) | 0.95 (0.68–1.32) | 0.95 (0.72–1.26) |
HR per 1-unit increment (log scale-adjusted)e | 1.04 (0.84–1.28) | 1.10 (0.84–1.45) | 1.16 (0.79–1.72) | 0.94 (0.63–1.40) | 0.94 (0.67–1.32) |
. | Colorectal cancer . | Colon cancer . | Proximal colon cancer . | Distal colon cancer . | Rectal cancer . |
---|---|---|---|---|---|
Total testosteronea | |||||
Women | |||||
Q1 | 1 | 1 | 1 | 1 | 1 |
Q2 | 0.95 (0.77–1.17) | 1.01 (0.79–1.29) | 1.05 (0.77–1.42) | 0.97 (0.64–1.48) | 0.79 (0.52–1.19) |
Q3 | 1.01 (0.82–1.25) | 1.03 (0.81–1.32) | 0.70 (0.49–1.00) | 1.48 (1.01–2.16) | 0.95 (0.64–1.42) |
Q4 | 0.95 (0.77–1.18) | 1.00 (0.78–1.29) | 0.90 (0.64–1.26) | 1.17 (0.78–1.75) | 0.82 (0.54–1.25) |
Q5 | 0.98 (0.79–1.22) | 1.03 (0.80–1.33) | 0.99 (0.71–1.38) | 1.03 (0.68–1.57) | 0.86 (0.56–1.30) |
P-trend | 0.89 | 0.86 | 0.59 | 0.59 | 0.56 |
HR per 1-SD increment | 1.00 (0.93–1.07) | 1.03 (0.95–1.11) | 0.99 (0.89–1.11) | 1.04 (0.92–1.18) | 0.92 (0.80–1.05) |
HR per 1-SD increment (adjusted)b | 1.00 (0.90–1.11) | 1.04 (0.92–1.18) | 0.99 (0.84–1.17) | 1.07 (0.88–1.30) | 0.88 (0.71–1.08) |
Men | |||||
Q1 | 1 | 1 | 1 | 1 | 1 |
Q2 | 1.12 (0.95–1.31) | 1.13 (0.92–1.38) | 0.95 (0.71–1.27) | 1.30 (0.97–1.78) | 1.11 (0.86–1.44) |
Q3 | 1.01 (0.85–1.21) | 1.08 (0.87–1.34) | 0.82 (0.60–1.12) | 1.36 (0.99–1.88) | 0.92 (0.69–1.23) |
Q4 | 0.96 (0.80–1.16) | 0.94 (0.74–1.19) | 0.85 (0.61–1.19) | 1.01 (0.70–1.45) | 1.00 (0.75–1.35) |
Q5 | 0.91 (0.74–1.13) | 0.87 (0.66–1.13) | 0.69 (0.47–1.01) | 1.14 (0.78–1.69) | 0.99 (0.72–1.37) |
P-trend | 0.19 | 0.16 | 0.06 | 0.97 | 0.75 |
HR per 1-SD increment | 0.98 (0.92–1.05) | 0.96 (0.88–1.05) | 0.92 (0.82–1.04) | 1.01 (0.89–1.14) | 1.01 (0.91–1.12) |
HR per 1-SD increment (adjusted)b | 0.97 (0.88–1.07) | 0.94 (0.83–1.07) | 0.89 (0.74–1.06) | 1.01 (0.84–1.22) | 1.01 (0.87–1.18) |
Free testosterone | |||||
Women | |||||
Q1 | 1 | 1 | 1 | 1 | 1 |
Q2 | 0.80 (0.65–0.99) | 0.90 (0.70–1.16) | 0.90 (0.65–1.24) | 0.86 (0.57–1.29) | 0.58 (0.38–0.88) |
Q3 | 0.82 (0.66–1.01) | 0.90 (0.70–1.16) | 0.76 (0.54–1.07) | 1.02 (0.69–1.51) | 0.64 (0.43–0.97) |
Q4 | 0.87 (0.70–1.08) | 0.94 (0.73–1.21) | 0.92 (0.66–1.28) | 0.88 (0.58–1.33) | 0.73 (0.49–1.10) |
Q5 | 0.83 (0.66–1.04) | 0.91 (0.69–1.18) | 0.77 (0.53–1.10) | 0.94 (0.62–1.43) | 0.66 (0.43–1.02) |
P-trend | 0.23 | 0.59 | 0.21 | 0.85 | 0.16 |
HR per 1-unit increment (log scale) | 0.93 (0.81–1.06) | 0.98 (0.84–1.14) | 0.89 (0.72–1.09) | 1.01 (0.80–1.29) | 0.79 (0.61–1.03) |
HR per 1-unit increment (log scale-adjusted)c | 0.90 (0.75–1.08) | 0.97 (0.78–1.20) | 0.85 (0.63–1.13) | 1.02 (0.73–1.43) | 0.72 (0.51–1.04) |
Men | |||||
Q1 | 1 | 1 | 1 | 1 | 1 |
Q2 | 0.95 (0.82–1.11) | 0.96 (0.80–1.17) | 0.75 (0.57–0.99) | 1.23 (0.93–1.61) | 0.94 (0.73–1.20) |
Q3 | 0.95 (0.81–1.11) | 0.83 (0.68–1.02) | 0.74 (0.56–0.99) | 0.91 (0.67–1.24) | 1.15 (0.90–1.48) |
Q4 | 1.01 (0.86–1.19) | 0.98 (0.80–1.21) | 0.87 (0.65–1.16) | 1.15 (0.85–1.56) | 1.08 (0.83–1.40) |
Q5 | 0.91 (0.76–1.09) | 0.87 (0.69–1.10) | 0.77 (0.56–1.07) | 1.02 (0.72–1.44) | 0.99 (0.74–1.32) |
P-trend | 0.59 | 0.29 | 0.20 | 0.98 | 0.62 |
HR per 1-SD increment | 0.99 (0.93–1.05) | 0.98 (0.91–1.05) | 0.94 (0.85–1.04) | 1.03 (0.92–1.14) | 1.01 (0.92–1.10) |
HR per 1-SD increment (adjusted)c | 0.98 (0.89–1.08) | 0.96 (0.85–1.09) | 0.90 (0.75–1.08) | 1.04 (0.87–1.26) | 1.01 (0.86–1.18) |
SHBGd | |||||
Women | |||||
Q1 | 1 | 1 | 1 | 1 | 1 |
Q2 | 1.12 (0.90–1.40) | 1.07 (0.83–1.37) | 1.17 (0.84–1.64) | 1.00 (0.68–1.49) | 1.31 (0.83–2.06) |
Q3 | 1.02 (0.81–1.29) | 0.93 (0.71–1.22) | 0.91 (0.62–1.32) | 0.94 (0.61–1.43) | 1.33 (0.83–2.14) |
Q4 | 1.39 (1.10–1.76) | 1.26 (0.96–1.65) | 1.37 (0.95–1.98) | 1.25 (0.82–1.91) | 1.86 (1.16–2.98) |
Q5 | 1.40 (1.09–1.81) | 1.30 (0.97–1.74) | 1.48 (1.00–2.19) | 1.28 (0.81–2.02) | 1.76 (1.05–2.94) |
P-trend | 0.002 | 0.045 | 0.044 | 0.18 | 0.012 |
HR per 1-SD increment | 1.13 (1.04–1.23) | 1.13 (1.02–1.24) | 1.19 (1.04–1.36) | 1.11 (0.96–1.30) | 1.15 (0.97–1.35) |
HR per 1-SD increment (adjusted)e | 1.16 (1.05–1.29) | 1.15 (1.03–1.30) | 1.24 (1.05–1.45) | 1.14 (0.95–1.37) | 1.18 (0.97–1.44) |
Men | |||||
Q1 | 1 | 1 | 1 | 1 | 1 |
Q2 | 0.92 (0.77–1.10) | 0.94 (0.75–1.18) | 0.84 (0.61–1.17) | 0.92 (0.66–1.30) | 0.89 (0.68–1.18) |
Q3 | 0.81 (0.67–0.97) | 0.91 (0.72–1.15) | 0.91 (0.65–1.27) | 0.90 (0.64–1.28) | 0.66 (0.49–0.90) |
Q4 | 0.91 (0.75–1.11) | 0.97 (0.76–1.24) | 0.87 (0.61–1.23) | 0.96 (0.67–1.37) | 0.83 (0.61–1.12) |
Q5 | 1.01 (0.82–1.25) | 1.09 (0.83–1.42) | 1.12 (0.77–1.64) | 0.99 (0.66–1.47) | 0.90 (0.65–1.26) |
P-trend | 0.86 | 0.51 | 0.60 | 0.93 | 0.56 |
HR per 1-unit increment (log scale) | 1.03 (0.86–1.23) | 1.08 (0.86–1.36) | 1.13 (0.82–1.57) | 0.95 (0.68–1.32) | 0.95 (0.72–1.26) |
HR per 1-unit increment (log scale-adjusted)e | 1.04 (0.84–1.28) | 1.10 (0.84–1.45) | 1.16 (0.79–1.72) | 0.94 (0.63–1.40) | 0.94 (0.67–1.32) |
Note: Multivariable Cox regression model using age as the underlying time variable and stratified by sex, Townsend deprivation index (quintiles), region of the recruitment assessment center, and age at recruitment. Models adjusted for waist circumference (per 5 cm), total physical activity (<10, 10–<20, 20–<40, 40–<60, ≥60 MET hours per week, unknown), height (per 10 cm), alcohol consumption frequency (never, special occasions only, 1 to 3 times per month, 1 to 2 times per week, 3 to 4 times per week, daily/almost daily, unknown), smoking status and intensity (never, former, current- <15 per day, current- ≥15 per day, current- intensity unknown, unknown), frequency of red and processed meat consumption (<2, 2–<3, 3–<4, ≥4 occasions per week, unknown), family history of colorectal cancer (no, yes, unknown), educational level (CSEs/O-levels/GCSEs or equivalent, NVQ/HND/HNC/A-levels/AS-levels or equivalent, other professional qualifications, college/university degree, none of the above, unknown), regular aspirin/ibuprofen use (no, yes, unknown), ever use of hormone replacement therapy (no, yes, unknown), circulating levels (sex-specific quintiles, missing/unknown) of C-reactive protein (CRP; mg/L), glycated hemoglobin (HbA1c; mmol/mol), and IGF-1 (nmol/L).
aPlus additional adjustment for SHBG (nmol/L).
bHRs were additionally corrected for regression dilution using a regression dilution ratio (0.65 in women and 0.68 in men) obtained from the subsample of participants with repeat total testosterone measurements.
cHRs were additionally corrected for regression dilution using a regression dilution ratio (0.71 in women and 0.57 in men) obtained from the subsample of participants with repeat free testosterone measurements.
dPlus additional adjustment for total testosterone (nmol/L).
eHRs were additionally corrected for regression dilution using a regression dilution ratio (0.82 in women and 0.83 in men) obtained from the subsample of participants with repeat SHBG measurements.
There was little evidence that circulating concentrations of free testosterone were associated with colorectal cancer risk for women (HR per 1 unit increment in log concentration = 0.90; 95% CI, 0.75–1.08) and men (HR per 1-SD increment = 0.98; 95% CI, 0.89–1.08; Table 2; Supplementary Table S2). There was little evidence for an association between circulating levels of free testosterone and colorectal cancer across anatomical subsites for both men and women. Heterogeneity for the circulating free testosterone concentrations and colorectal cancer association was found for men by follow-up time (Pinteraction = 0.01; Table 3).
Subgroup analyses of the association between circulating total testosterone, free testosterone, and sex hormone binding globulin (SHBG) levels and colorectal cancer risk in the UK Biobank.
. | Total testosterone . | Free testosterone . | SHBG . | ||||||
---|---|---|---|---|---|---|---|---|---|
. | HRa . | HR (adjusted)b . | Pinteraction . | HRc . | HR (adjusted)d . | Pinteraction . | HRe . | HR (adjusted)f . | Pinteraction . |
Women | |||||||||
Body mass index (kg/m2) | |||||||||
<25 | 0.95 (0.85–1.07) | 0.93 (0.78–1.11) | 0.33 | 0.87 (0.70–1.08) | 0.82 (0.61–1.11) | 0.46 | 1.12 (0.98–1.29) | 1.15 (0.98–1.36) | 0.90 |
≥25 | 1.02 (0.94–1.11) | 1.03 (0.91–1.18) | 0.96 (0.82–1.13) | 0.94 (0.75–1.18) | 1.14 (1.03–1.26) | 1.17 (1.03–1.32) | |||
Waist to hip ratio | |||||||||
below median (<0.81) | 1.05 (0.95–1.17) | 1.08 (0.93–1.27) | 0.16 | 0.99 (0.82–1.20) | 0.99 (0.76–1.29) | 0.31 | 1.15 (1.02–1.30) | 1.19 (1.03–1.37) | 0.75 |
above median (≥0.81) | 0.96 (0.87–1.05) | 0.93 (0.81–1.07) | 0.87 (0.74–1.04) | 0.83 (0.65–1.05) | 1.12 (1.01–1.25) | 1.15 (1.01–1.31) | |||
Age at recruitment (years) | |||||||||
<60 | 0.99 (0.89–1.11) | 0.99 (0.84–1.17) | 0.94 | 0.94 (0.77–1.14) | 0.91 (0.69–1.20) | 0.84 | 1.10 (0.99–1.24) | 1.13 (0.98–1.30) | 0.47 |
≥60 | 1.00 (0.91–1.09) | 1.00 (0.87–1.14) | 0.91 (0.77–1.08) | 0.88 (0.70–1.11) | 1.16 (1.05–1.29) | 1.20 (1.06–1.37) | |||
Follow-up time (years) | |||||||||
<5 | 0.94 (0.86–1.03) | 0.92 (0.80–1.05) | 0.28 | 0.84 (0.71–1.00) | 0.79 (0.62–1.00) | 0.16 | 1.13 (1.02–1.25) | 1.16 (1.02–1.31) | 0.33 |
≥5 | 1.02 (0.91–1.15) | 1.04 (0.87–1.24) | 1.02 (0.82–1.25) | 1.02 (0.76–1.38) | 1.05 (0.92–1.19) | 1.06 (0.90–1.24) | |||
Menopausal status | |||||||||
Premenopausal | 1.08 (0.89–1.31) | 1.12 (0.84–1.51) | 0.32 | 0.92 (0.66–1.27) | 0.88 (0.56–1.40) | 1.00 | 1.25 (1.04–1.50) | 1.31 (1.04–1.64) | 0.20 |
Postmenopausal | 0.97 (0.90–1.05) | 0.96 (0.85–1.07) | 0.92 (0.79–1.06) | 0.88 (0.72–1.08) | 1.10 (1.00–1.20) | 1.12 (1.00–1.25) | |||
Men | |||||||||
Body mass index (kg/m2) | |||||||||
<25 | 0.96 (0.85–1.09) | 0.94 (0.79–1.13) | 0.73 | 0.91 (0.81–1.02) | 0.85 (0.69–1.04) | 0.11 | 1.37 (0.98–1.91) | 1.46 (0.98–2.18) | 0.05 |
≥25 | 0.98 (0.91–1.06) | 0.98 (0.88–1.09) | 1.01 (0.95–1.08) | 1.02 (0.91–1.14) | 0.96 (0.80–1.16) | 0.96 (0.76–1.20) | |||
Waist-to-hip ratio | |||||||||
below median (<0.93) | 0.91 (0.83–1.00) | 0.87 (0.76–1.00) | 0.05 | 0.94 (0.86–1.02) | 0.89 (0.76–1.04) | 0.16 | 1.00 (0.78–1.28) | 1.00 (0.74–1.35) | 0.61 |
above median (≥0.93) | 1.02 (0.94–1.10) | 1.02 (0.91–1.15) | 1.02 (0.95–1.09) | 1.03 (0.91–1.16) | 1.07 (0.87–1.32) | 1.09 (0.85–1.39) | |||
Age at recruitment (years) | |||||||||
<60 | 1.04 (0.94–1.14) | 1.05 (0.91–1.22) | 0.13 | 1.03 (0.93–1.13) | 1.05 (0.89–1.24) | 0.29 | 1.07 (0.83–1.38) | 1.09 (0.80–1.47) | 0.80 |
≥60 | 0.95 (0.88–1.02) | 0.93 (0.83–1.04) | 0.97 (0.90–1.03) | 0.94 (0.83–1.06) | 1.03 (0.84–1.27) | 1.04 (0.81–1.33) | |||
Follow-up time (years) | |||||||||
<5 | 1.00 (0.93–1.08) | 1.00 (0.90–1.12) | 0.46 | 1.05 (0.98–1.13) | 1.10 (0.97–1.24) | 0.01 | 0.85 (0.69–1.04) | 0.82 (0.64–1.05) | 0.27 |
≥5 | 0.96 (0.87–1.06) | 0.94 (0.81–1.09) | 0.91 (0.82–1.00) | 0.84 (0.71–1.00) | 1.00 (0.77–1.30) | 1.00 (0.73–1.38) |
. | Total testosterone . | Free testosterone . | SHBG . | ||||||
---|---|---|---|---|---|---|---|---|---|
. | HRa . | HR (adjusted)b . | Pinteraction . | HRc . | HR (adjusted)d . | Pinteraction . | HRe . | HR (adjusted)f . | Pinteraction . |
Women | |||||||||
Body mass index (kg/m2) | |||||||||
<25 | 0.95 (0.85–1.07) | 0.93 (0.78–1.11) | 0.33 | 0.87 (0.70–1.08) | 0.82 (0.61–1.11) | 0.46 | 1.12 (0.98–1.29) | 1.15 (0.98–1.36) | 0.90 |
≥25 | 1.02 (0.94–1.11) | 1.03 (0.91–1.18) | 0.96 (0.82–1.13) | 0.94 (0.75–1.18) | 1.14 (1.03–1.26) | 1.17 (1.03–1.32) | |||
Waist to hip ratio | |||||||||
below median (<0.81) | 1.05 (0.95–1.17) | 1.08 (0.93–1.27) | 0.16 | 0.99 (0.82–1.20) | 0.99 (0.76–1.29) | 0.31 | 1.15 (1.02–1.30) | 1.19 (1.03–1.37) | 0.75 |
above median (≥0.81) | 0.96 (0.87–1.05) | 0.93 (0.81–1.07) | 0.87 (0.74–1.04) | 0.83 (0.65–1.05) | 1.12 (1.01–1.25) | 1.15 (1.01–1.31) | |||
Age at recruitment (years) | |||||||||
<60 | 0.99 (0.89–1.11) | 0.99 (0.84–1.17) | 0.94 | 0.94 (0.77–1.14) | 0.91 (0.69–1.20) | 0.84 | 1.10 (0.99–1.24) | 1.13 (0.98–1.30) | 0.47 |
≥60 | 1.00 (0.91–1.09) | 1.00 (0.87–1.14) | 0.91 (0.77–1.08) | 0.88 (0.70–1.11) | 1.16 (1.05–1.29) | 1.20 (1.06–1.37) | |||
Follow-up time (years) | |||||||||
<5 | 0.94 (0.86–1.03) | 0.92 (0.80–1.05) | 0.28 | 0.84 (0.71–1.00) | 0.79 (0.62–1.00) | 0.16 | 1.13 (1.02–1.25) | 1.16 (1.02–1.31) | 0.33 |
≥5 | 1.02 (0.91–1.15) | 1.04 (0.87–1.24) | 1.02 (0.82–1.25) | 1.02 (0.76–1.38) | 1.05 (0.92–1.19) | 1.06 (0.90–1.24) | |||
Menopausal status | |||||||||
Premenopausal | 1.08 (0.89–1.31) | 1.12 (0.84–1.51) | 0.32 | 0.92 (0.66–1.27) | 0.88 (0.56–1.40) | 1.00 | 1.25 (1.04–1.50) | 1.31 (1.04–1.64) | 0.20 |
Postmenopausal | 0.97 (0.90–1.05) | 0.96 (0.85–1.07) | 0.92 (0.79–1.06) | 0.88 (0.72–1.08) | 1.10 (1.00–1.20) | 1.12 (1.00–1.25) | |||
Men | |||||||||
Body mass index (kg/m2) | |||||||||
<25 | 0.96 (0.85–1.09) | 0.94 (0.79–1.13) | 0.73 | 0.91 (0.81–1.02) | 0.85 (0.69–1.04) | 0.11 | 1.37 (0.98–1.91) | 1.46 (0.98–2.18) | 0.05 |
≥25 | 0.98 (0.91–1.06) | 0.98 (0.88–1.09) | 1.01 (0.95–1.08) | 1.02 (0.91–1.14) | 0.96 (0.80–1.16) | 0.96 (0.76–1.20) | |||
Waist-to-hip ratio | |||||||||
below median (<0.93) | 0.91 (0.83–1.00) | 0.87 (0.76–1.00) | 0.05 | 0.94 (0.86–1.02) | 0.89 (0.76–1.04) | 0.16 | 1.00 (0.78–1.28) | 1.00 (0.74–1.35) | 0.61 |
above median (≥0.93) | 1.02 (0.94–1.10) | 1.02 (0.91–1.15) | 1.02 (0.95–1.09) | 1.03 (0.91–1.16) | 1.07 (0.87–1.32) | 1.09 (0.85–1.39) | |||
Age at recruitment (years) | |||||||||
<60 | 1.04 (0.94–1.14) | 1.05 (0.91–1.22) | 0.13 | 1.03 (0.93–1.13) | 1.05 (0.89–1.24) | 0.29 | 1.07 (0.83–1.38) | 1.09 (0.80–1.47) | 0.80 |
≥60 | 0.95 (0.88–1.02) | 0.93 (0.83–1.04) | 0.97 (0.90–1.03) | 0.94 (0.83–1.06) | 1.03 (0.84–1.27) | 1.04 (0.81–1.33) | |||
Follow-up time (years) | |||||||||
<5 | 1.00 (0.93–1.08) | 1.00 (0.90–1.12) | 0.46 | 1.05 (0.98–1.13) | 1.10 (0.97–1.24) | 0.01 | 0.85 (0.69–1.04) | 0.82 (0.64–1.05) | 0.27 |
≥5 | 0.96 (0.87–1.06) | 0.94 (0.81–1.09) | 0.91 (0.82–1.00) | 0.84 (0.71–1.00) | 1.00 (0.77–1.30) | 1.00 (0.73–1.38) |
Note: Multivariable Cox regression model using age as the underlying time variable and stratified by sex, Townsend deprivation index (quintiles), region of the recruitment assessment center, and age at recruitment. Models adjusted for waist circumference (per 5 cm), total physical activity (<10, 10–<20, 20–<40, 40–<60, ≥60 MET hours per week, unknown), height (per 10 cm), alcohol consumption frequency (never, special occasions only, 1 to 3 times per month, 1 to 2 times per week, 3 to 4 times per week, daily/almost daily, unknown), smoking status and intensity (never, former, current- <15 per day, current- ≥15 per day, current- intensity unknown, unknown), frequency of red and processed meat consumption (<2, 2–<3, 3–<4, ≥4 occasions per week, unknown), family history of colorectal cancer (no, yes, unknown), educational level (CSEs/O-levels/GCSEs or equivalent, NVQ/HND/HNC/A-levels/AS-levels or equivalent, other professional qualifications, college/university degree, none of the above, unknown), regular aspirin/ibuprofen use (no, yes, unknown), ever use of hormone replacement therapy (no, yes, unknown), circulating levels (sex-specific quintiles, missing/unknown) of C-reactive protein (CRP; mg/L), glycated hemoglobin (HbA1c; mmol/mol), and IGF-1 (nmol/L).
aHRs per 1-SD increment in both women and men additional adjusted for sex hormone binding globulin (SHBG; nmol/L);.
bHRs were additionally corrected for regression dilution using a regression dilution ratio (0.65 in women and 0.68 in men) obtained from the subsample of participants with repeat total testosterone measurements;.
cHRs per 1-unit increment (log scale) in women and per 1-SD increment in men;.
dHRs were additionally corrected for regression dilution using a regression dilution ratio (0.71 in women and 0.57 in men) obtained from the subsample of participants with repeat free testosterone measurements;.
eHRs per 1-SD increment in women and per 1-unit increment (log scale) in men additionally adjusted for total testosterone (nmol/L);.
fHRs were additionally corrected for regression dilution using a regression dilution ratio (0.82 in women and 0.83 in men) obtained from the subsample of participants with repeat SHBG measurements.
Association between circulating SHBG concentrations and colorectal cancer risk
In the multivariable model 2 additionally adjusted for circulating concentrations of CRP, HbA1c, testosterone, and IGF-1, a 1-SD increment of SHBG concentrations was associated with a higher colorectal cancer risk amongst women (HR = 1.16; 95% CI, 1.05–1.29; Table 2; Supplementary Table S2). No association between SHBG concentrations and colorectal cancer risk was found for men (HR per 1 unit increment in log concentration = 1.04; 95% CI, 0.84–1.28). Associations of similar magnitude between SHBG concentrations and colorectal cancer risk were found in the quintile models, by anatomical subsite, and according to subgroups of BMI, WHR, age at recruitment, follow-up time, and menopausal status (Table 3; all Pinteractions ≥ 0.05).
Sensitivity analyses
Similar results for total testosterone, free testosterone, and SHBG with colorectal cancer were found when: participants with outlier concentrations were excluded (Supplementary Table S3); cases occurring in the first 2 years of follow-up were excluded (n = 564 colorectal cancer cases excluded; Supplementary Table S4); and ever users of menopausal hormones or women with PCOS were excluded (Supplementary Table S5).
Mendelian randomization analyses
Effect estimates for the association between circulating total testosterone and free testosterone concentrations and colorectal cancer risk
In the random-effects IVW models, higher genetically predicted circulating total testosterone concentration was associated with greater risk of colorectal cancer for women (OR per 1 SD increment in testosterone concentrations = 1.09; 95% CI, 1.01–1.17), but not for men (OR = 0.99; 95% CI, 0.91–1.07); although heterogeneity was observed (P-value for heterogeneity was 0.01 for women and <0.01 men). Positive associations were also found for distal colon cancer and rectal cancer for women (distal colon cancer, OR = 1.15; 95% CI, 1.03–1.28; rectal cancer, OR = 1.13; 95% CI, 1.00–1.28), but not for men (distal colon cancer, OR = 1.06; 95% CI, 0.93–1.20; rectal cancer, OR = 1.02; 95% CI, 0.91–1.15; Table 4; Supplementary Table S6). However, these positive associations were slightly attenuated for the weighted median and Steiger filtered analyses, and were null in the lower powered MR-Egger models (Supplementary Table S6).
MR estimates for the effect of total testosterone, free testosterone, and sex hormone binding globulin (SHBG) on colorectal cancer risk.
. | Women . | Men . | ||||||
---|---|---|---|---|---|---|---|---|
Methods . | ORa . | 95% CI . | P value . | P value for pleiotropy or heterogeneity . | ORa . | 95% CI . | P value . | P value for pleiotropy or heterogeneity . |
Total testosterone | ||||||||
Colorectal cancer | ||||||||
IVW (random effects) | 1.09 | (1.01–1.17) | 0.04 | 0.01 | 0.99 | (0.91–1.07) | 0.76 | <0.01 |
MR-Egger (slope) | 1.01 | (0.88–1.17) | 0.85 | 0.28 | 0.95 | (0.84–1.09) | 0.49 | 0.52 |
Weighted median | 1.08 | (0.94–1.25) | 0.27 | NA | 1.02 | (0.91–1.14) | 0.75 | NA |
Colon cancer | ||||||||
IVW (random effects) | 1.06 | (0.97–1.16) | 0.17 | 0.14 | 1.03 | (0.94–1.12) | 0.58 | <0.01 |
MR-Egger (slope) | 0.94 | (0.80–1.11) | 0.47 | 0.09 | 1.00 | (0.86–1.16) | 0.99 | 0.65 |
Weighted median | 1.05 | (0.90–1.24) | 0.53 | NA | 1.08 | (0.94–1.25) | 0.27 | NA |
Distal colon cancer | ||||||||
IVW (random effects) | 1.15 | (1.03–1.28) | 0.01 | 0.52 | 1.06 | (0.93–1.20) | 0.37 | <0.01 |
MR-Egger (slope) | 1.02 | (0.83–1.25) | 0.87 | 0.18 | 1.14 | (0.92–1.40) | 0.23 | 0.40 |
Weighted median | 1.13 | (0.91–1.40) | 0.26 | NA | 1.27 | (1.06–1.53) | 0.01 | NA |
Proximal colon cancer | ||||||||
IVW (random effects) | 1.02 | (0.91–1.14) | 0.75 | 0.02 | 1.00 | (0.90–1.12) | 0.99 | 0.07 |
MR-Egger (slope) | 0.91 | (0.74–1.13) | 0.40 | 0.24 | 0.90 | (0.75–1.08) | 0.25 | 0.14 |
Weighted median | 1.05 | (0.85–1.28) | 0.67 | NA | 0.94 | (0.79–1.13) | 0.53 | NA |
Rectal cancer | ||||||||
IVW (random effects) | 1.13 | (1.00–1.28) | 0.05 | 0.06 | 1.02 | (0.91–1.15) | 0.68 | <0.01 |
MR-Egger (slope) | 1.16 | (0.91–1.46) | 0.23 | 0.82 | 0.97 | (0.81–1.17) | 0.77 | 0.48 |
Weighted median | 1.25 | (1.00–1.57) | 0.05 | NA | 1.05 | (0.88–1.24) | 0.61 | NA |
Free testosterone | ||||||||
Colorectal cancer | ||||||||
IVW (random effects) | 1.05 | (0.93–1.18) | 0.42 | <0.01 | 1.00 | (0.89–1.13) | 0.98 | <0.01 |
MR-Egger (slope) | 1.14 | (0.93–1.40) | 0.19 | 0.31 | 1.00 | (0.79–1.28) | 0.98 | 0.98 |
Weighted median | 1.06 | (0.89–1.26) | 0.49 | NA | 1.06 | (0.90–1.24) | 0.49 | NA |
Colon cancer | ||||||||
IVW (random effects) | 1.01 | (0.89–1.16) | 0.85 | 0.01 | 1.02 | (0.89–1.17) | 0.78 | 0.06 |
MR-Egger (slope) | 1.07 | (0.85–1.35) | 0.55 | 0.55 | 0.90 | (0.68–1.19) | 0.45 | 0.31 |
Weighted median | 1.07 | (0.87–1.32) | 0.50 | NA | 1.00 | (0.83–1.22) | 0.97 | NA |
Distal colon cancer | ||||||||
IVW (random effects) | 1.08 | (0.91–1.27) | 0.37 | 0.17 | 0.92 | (0.77–1.08) | 0.30 | 0.10 |
MR-Egger (slope) | 0.99 | (0.74–1.33) | 0.96 | 0.50 | 0.80 | (0.57–1.12) | 0.19 | 0.36 |
Weighted median | 1.04 | (0.80–1.35) | 0.78 | NA | 0.88 | (0.69–1.12) | 0.30 | NA |
Proximal colon cancer | ||||||||
IVW (random effects) | 0.98 | (0.83–1.14) | 0.76 | 0.01 | 1.09 | (0.92–1.29) | 0.31 | 0.12 |
MR-Egger (slope) | 1.18 | (0.89–1.55) | 0.24 | 0.10 | 1.05 | (0.74–1.49) | 0.78 | 0.81 |
Weighted median | 1.11 | (0.85–1.45) | 0.43 | NA | 1.09 | (0.85–1.39) | 0.51 | NA |
Rectal cancer | ||||||||
IVW (random effects) | 1.13 | (0.94–1.36) | 0.19 | 0.02 | 1.00 | (0.83–1.20) | 0.99 | <0.01 |
MR-Egger (slope) | 1.51 | (1.10–2.07) | 0.01 | 0.03 | 1.09 | (0.75–1.58) | 0.65 | 0.60 |
Weighted median | 1.10 | (0.83–1.45) | 0.51 | NA | 1.01 | (0.78–1.30) | 0.95 | NA |
SHBG | ||||||||
Colorectal cancer | ||||||||
IVW (random effects) | 1.07 | (0.94–1.23) | 0.32 | <0.01 | 1.06 | (0.92–1.21) | 0.42 | <0.01 |
MR-Egger (slope) | 1.02 | (0.84–1.23) | 0.88 | 0.45 | 1.03 | (0.86–1.24) | 0.71 | 0.74 |
Weighted median | 1.05 | (0.84–1.31) | 0.67 | NA | 0.98 | (0.82–1.16) | 0.80 | NA |
Colon cancer | ||||||||
IVW (random effects) | 1.04 | (0.89–1.21) | 0.64 | <0.01 | 1.06 | (0.91–1.24) | 0.44 | <0.01 |
MR-Egger (slope) | 1.00 | (0.80–1.25) | 0.99 | 0.64 | 1.07 | (0.87–1.32) | 0.52 | 0.91 |
Weighted median | 0.93 | (0.73–1.18) | 0.53 | NA | 1.09 | (0.87–1.37) | 0.43 | NA |
Distal colon cancer | ||||||||
IVW (random effects) | 1.01 | (0.83–1.23) | 0.93 | <0.01 | 1.16 | (0.96–1.40) | 0.12 | <0.01 |
MR-Egger (slope) | 0.95 | (0.71–1.27) | 0.71 | 0.55 | 1.15 | (0.89–1.49) | 0.27 | 0.97 |
Weighted median | 0.98 | (0.74–1.29) | 0.88 | NA | 1.41 | (1.01–1.96) | 0.04 | NA |
Proximal colon cancer | ||||||||
IVW (random effects) | 1.07 | (0.89–1.29) | 0.45 | <0.01 | 0.99 | (0.82–1.20) | 0.91 | <0.01 |
MR-Egger (slope) | 1.05 | (0.81–1.36) | 0.72 | 0.81 | 1.00 | (0.77–1.29) | 0.98 | 0.93 |
Weighted median | 0.93 | (0.68–1.28) | 0.65 | NA | 0.90 | (0.68–1.19) | 0.47 | NA |
Rectal cancer | ||||||||
IVW (random effects) | 1.06 | (0.86–1.30) | 0.58 | <0.01 | 1.09 | (0.92–1.31) | 0.32 | <0.01 |
MR-Egger (slope) | 0.84 | (0.63–1.13) | 0.25 | 0.03 | 1.06 | (0.84–1.36) | 0.61 | 0.74 |
Weighted median | 0.89 | (0.63–1.27) | 0.53 | NA | 0.92 | (0.70–1.21) | 0.55 | NA |
. | Women . | Men . | ||||||
---|---|---|---|---|---|---|---|---|
Methods . | ORa . | 95% CI . | P value . | P value for pleiotropy or heterogeneity . | ORa . | 95% CI . | P value . | P value for pleiotropy or heterogeneity . |
Total testosterone | ||||||||
Colorectal cancer | ||||||||
IVW (random effects) | 1.09 | (1.01–1.17) | 0.04 | 0.01 | 0.99 | (0.91–1.07) | 0.76 | <0.01 |
MR-Egger (slope) | 1.01 | (0.88–1.17) | 0.85 | 0.28 | 0.95 | (0.84–1.09) | 0.49 | 0.52 |
Weighted median | 1.08 | (0.94–1.25) | 0.27 | NA | 1.02 | (0.91–1.14) | 0.75 | NA |
Colon cancer | ||||||||
IVW (random effects) | 1.06 | (0.97–1.16) | 0.17 | 0.14 | 1.03 | (0.94–1.12) | 0.58 | <0.01 |
MR-Egger (slope) | 0.94 | (0.80–1.11) | 0.47 | 0.09 | 1.00 | (0.86–1.16) | 0.99 | 0.65 |
Weighted median | 1.05 | (0.90–1.24) | 0.53 | NA | 1.08 | (0.94–1.25) | 0.27 | NA |
Distal colon cancer | ||||||||
IVW (random effects) | 1.15 | (1.03–1.28) | 0.01 | 0.52 | 1.06 | (0.93–1.20) | 0.37 | <0.01 |
MR-Egger (slope) | 1.02 | (0.83–1.25) | 0.87 | 0.18 | 1.14 | (0.92–1.40) | 0.23 | 0.40 |
Weighted median | 1.13 | (0.91–1.40) | 0.26 | NA | 1.27 | (1.06–1.53) | 0.01 | NA |
Proximal colon cancer | ||||||||
IVW (random effects) | 1.02 | (0.91–1.14) | 0.75 | 0.02 | 1.00 | (0.90–1.12) | 0.99 | 0.07 |
MR-Egger (slope) | 0.91 | (0.74–1.13) | 0.40 | 0.24 | 0.90 | (0.75–1.08) | 0.25 | 0.14 |
Weighted median | 1.05 | (0.85–1.28) | 0.67 | NA | 0.94 | (0.79–1.13) | 0.53 | NA |
Rectal cancer | ||||||||
IVW (random effects) | 1.13 | (1.00–1.28) | 0.05 | 0.06 | 1.02 | (0.91–1.15) | 0.68 | <0.01 |
MR-Egger (slope) | 1.16 | (0.91–1.46) | 0.23 | 0.82 | 0.97 | (0.81–1.17) | 0.77 | 0.48 |
Weighted median | 1.25 | (1.00–1.57) | 0.05 | NA | 1.05 | (0.88–1.24) | 0.61 | NA |
Free testosterone | ||||||||
Colorectal cancer | ||||||||
IVW (random effects) | 1.05 | (0.93–1.18) | 0.42 | <0.01 | 1.00 | (0.89–1.13) | 0.98 | <0.01 |
MR-Egger (slope) | 1.14 | (0.93–1.40) | 0.19 | 0.31 | 1.00 | (0.79–1.28) | 0.98 | 0.98 |
Weighted median | 1.06 | (0.89–1.26) | 0.49 | NA | 1.06 | (0.90–1.24) | 0.49 | NA |
Colon cancer | ||||||||
IVW (random effects) | 1.01 | (0.89–1.16) | 0.85 | 0.01 | 1.02 | (0.89–1.17) | 0.78 | 0.06 |
MR-Egger (slope) | 1.07 | (0.85–1.35) | 0.55 | 0.55 | 0.90 | (0.68–1.19) | 0.45 | 0.31 |
Weighted median | 1.07 | (0.87–1.32) | 0.50 | NA | 1.00 | (0.83–1.22) | 0.97 | NA |
Distal colon cancer | ||||||||
IVW (random effects) | 1.08 | (0.91–1.27) | 0.37 | 0.17 | 0.92 | (0.77–1.08) | 0.30 | 0.10 |
MR-Egger (slope) | 0.99 | (0.74–1.33) | 0.96 | 0.50 | 0.80 | (0.57–1.12) | 0.19 | 0.36 |
Weighted median | 1.04 | (0.80–1.35) | 0.78 | NA | 0.88 | (0.69–1.12) | 0.30 | NA |
Proximal colon cancer | ||||||||
IVW (random effects) | 0.98 | (0.83–1.14) | 0.76 | 0.01 | 1.09 | (0.92–1.29) | 0.31 | 0.12 |
MR-Egger (slope) | 1.18 | (0.89–1.55) | 0.24 | 0.10 | 1.05 | (0.74–1.49) | 0.78 | 0.81 |
Weighted median | 1.11 | (0.85–1.45) | 0.43 | NA | 1.09 | (0.85–1.39) | 0.51 | NA |
Rectal cancer | ||||||||
IVW (random effects) | 1.13 | (0.94–1.36) | 0.19 | 0.02 | 1.00 | (0.83–1.20) | 0.99 | <0.01 |
MR-Egger (slope) | 1.51 | (1.10–2.07) | 0.01 | 0.03 | 1.09 | (0.75–1.58) | 0.65 | 0.60 |
Weighted median | 1.10 | (0.83–1.45) | 0.51 | NA | 1.01 | (0.78–1.30) | 0.95 | NA |
SHBG | ||||||||
Colorectal cancer | ||||||||
IVW (random effects) | 1.07 | (0.94–1.23) | 0.32 | <0.01 | 1.06 | (0.92–1.21) | 0.42 | <0.01 |
MR-Egger (slope) | 1.02 | (0.84–1.23) | 0.88 | 0.45 | 1.03 | (0.86–1.24) | 0.71 | 0.74 |
Weighted median | 1.05 | (0.84–1.31) | 0.67 | NA | 0.98 | (0.82–1.16) | 0.80 | NA |
Colon cancer | ||||||||
IVW (random effects) | 1.04 | (0.89–1.21) | 0.64 | <0.01 | 1.06 | (0.91–1.24) | 0.44 | <0.01 |
MR-Egger (slope) | 1.00 | (0.80–1.25) | 0.99 | 0.64 | 1.07 | (0.87–1.32) | 0.52 | 0.91 |
Weighted median | 0.93 | (0.73–1.18) | 0.53 | NA | 1.09 | (0.87–1.37) | 0.43 | NA |
Distal colon cancer | ||||||||
IVW (random effects) | 1.01 | (0.83–1.23) | 0.93 | <0.01 | 1.16 | (0.96–1.40) | 0.12 | <0.01 |
MR-Egger (slope) | 0.95 | (0.71–1.27) | 0.71 | 0.55 | 1.15 | (0.89–1.49) | 0.27 | 0.97 |
Weighted median | 0.98 | (0.74–1.29) | 0.88 | NA | 1.41 | (1.01–1.96) | 0.04 | NA |
Proximal colon cancer | ||||||||
IVW (random effects) | 1.07 | (0.89–1.29) | 0.45 | <0.01 | 0.99 | (0.82–1.20) | 0.91 | <0.01 |
MR-Egger (slope) | 1.05 | (0.81–1.36) | 0.72 | 0.81 | 1.00 | (0.77–1.29) | 0.98 | 0.93 |
Weighted median | 0.93 | (0.68–1.28) | 0.65 | NA | 0.90 | (0.68–1.19) | 0.47 | NA |
Rectal cancer | ||||||||
IVW (random effects) | 1.06 | (0.86–1.30) | 0.58 | <0.01 | 1.09 | (0.92–1.31) | 0.32 | <0.01 |
MR-Egger (slope) | 0.84 | (0.63–1.13) | 0.25 | 0.03 | 1.06 | (0.84–1.36) | 0.61 | 0.74 |
Weighted median | 0.89 | (0.63–1.27) | 0.53 | NA | 0.92 | (0.70–1.21) | 0.55 | NA |
Note: P value for pleiotropy in MR-Egger regression; P value for heterogeneity in IVW analysis.
aORs per 1-SD increment in total testosterone concentrations in both women and men, per 1-unit increment (log scale) in free testosterone concentrations in women and per 1-SD increment in men, and per 1-SD increment in SHBG concentrations in women and per 1-unit increment (log scale) in men.
No association was estimated between genetically predicted circulating free testosterone concentrations and risk of colorectal cancer for both women (OR per 1 unit increment in log-concentrations = 1.05; 95% CI, 0.93–1.18) and men (OR per 1-SD increment = 1.00; 95% CI, 0.89–1.13; Table 4; Supplementary Table S6). Associations of similar magnitude were estimated for all anatomic subsites in both men and women. The MR-Egger test showed evidence of directional pleiotropy for rectal cancer in women (MR-Egger intercept P value = 0.03). The weighted median approach showed effect estimates of similar magnitude with wider CIs in all models. Steiger filtered analysis showed nearly identical null associations with risk of colorectal cancer in both women and men (Table 4; Supplementary Table S6).
Effect estimates for the association between circulating SHBG concentrations and colorectal cancer risk
In the random-effects IVW models, we found no association between genetically predicted circulating SHBG concentrations and risk of colorectal cancer for women (OR per 1 SD increment = 1.07; 95% CI, 0.94–1.23) and men (OR per 1 unit increment in log-concentrations = 1.06; 95% CI, 0.92–1.21), with evidence for heterogeneity in all analyses (Cochran Q P values <0.001; Table 4; Supplementary Table S6). Similar magnitude effect estimates were found for all anatomic subsites in both men and women. The MR-Egger test showed evidence of directional pleiotropy for rectal cancer in women (MR-Egger intercept P value = 0.03). The weighted median approach showed effect estimates of similar magnitude in all models. Almost identical null associations were estimated for circulating SHBG concentrations and colorectal cancer in both women and men excluding pleiotropic variants indicated by Steiger filtering. No associations were observed using cis variants in the SHBG gene as the genetic instrument (Supplementary Table S7).
Discussion
In our MR analysis, we found a positive effect estimate for circulating total testosterone levels with colorectal cancer risk among women; however, we cannot rule out the possibility of pleiotropy biasing this finding (i.e., the effect is explained by an independent biological pathway). There was little evidence that circulating testosterone levels were associated with risk of colorectal cancer for men in the observational and MR analyses. In observational analyses of UK Biobank data, we found that higher prediagnostic concentrations of circulating SHBG were associated with a greater risk of colorectal cancer, with this relationship limited to women only. These findings were not, however, corroborated by our MR analyses, which showed little evidence of an association between genetically predicted SHBG concentrations and colorectal cancer risk in women.
In our observational analyses in UK Biobank, there was little evidence that circulating testosterone levels were associated with colorectal cancer risk for women. Our findings for total testosterone and free testosterone concentrations are generally similar to those published in other recent UK Biobank studies (36, 37). In the MR analyses, we found positive effect estimates between total testosterone concentrations and colorectal, distal colon, and rectal cancer risk for women. However, the effect estimates were null in the MR-Egger models, indicating that there may be alternative pathways explaining these associations (pleiotropy). Possible biological pathways linking testosterone with colorectal cancer development for women are unclear. In women, testosterone is mainly produced by the ovaries, suprarenal glands, and adipose tissue, with its secretion regulated by aromatase activity. After menopause, testosterone becomes the main source of estradiol when ovarian production of estrogens ceases. Thus, the positive association found between total testosterone and colorectal cancer for women may be an indicator of estrogenic pathways. However, epidemiological studies examining the associations between prediagnostic levels of estrogens and colorectal cancer have reported mixed results (5–8), and stronger genetic instruments for circulating estrogen concentrations are required to undertake suitably powered MR analyses with colorectal cancer. Overall, further studies are needed to better understand the biological pathways through which testosterone may influence colorectal cancer risk for women.
The positive association we found between SHBG concentrations and colorectal cancer for women in our UK Biobank observational analysis was consistent with a prior analysis in the WHI-CT study (7). However, other previous observational studies have reported no association between circulating SHBG levels and colorectal cancer risk (5, 6, 8). For men, the null association we found in our observational analysis was inconsistent with a prior Health Professional Follow-up Study/Physicians' Health Study II analysis (5) and a recently published study in UK Biobank (37) that reported an inverse association. This prior UK Biobank study, however, did not statistically adjust for markers of inflammation and glycemic pathways that are known to be correlated with sex steroid hormone concentrations, and have been linked to colorectal cancer risk, namely CRP, IGF-1, and HbA1c (6, 7, 24, 25). After we adjusted our multivariable models for these serologic factors the inverse SHBG risk estimate attenuated to the null. For our MR analysis, we found little evidence of an association between SHBG concentrations and colorectal cancer risk for both men and women. It is possible that this inconsistency in results between observational and MR evidence is a consequence of measurement error, residual confounding, and/or reverse causality, characteristic of observational epidemiology. MR is an increasingly used method that uses genetic variants robustly associated with the exposure of interest in an instrumental variable analysis to appraise the causal nature of the effects of the exposure on an outcome (38). The random and fixed allocation of alleles at conception makes confounding and reverse causation less likely explanations for associations identified in MR studies (39).
This study is the most comprehensive investigation of the associations between circulating sex steroid hormone concentrations and colorectal cancer incorporating complementary observational and MR analyses. Our observational study, using data from the UK Biobank, was the largest to date (including >2,000 incident cases) which meant we were able to examine circulating sex steroid hormones levels and colorectal cancer association by anatomical subsite and by subgroups of colorectal cancer risk factors. We were also able to control statistically for other factors that are related to the sex hormone pathway, and have been linked to colorectal cancer incidence in some studies, namely CRP, IGF-1, and HbA1c (6, 7, 24, 25). A limitation of our analysis was that single hormones measures were available for most participants and it is possible that these measurements may not reflect longer term exposures. However, in our reproducibility analysis, we estimated a within-person ICC of ∼0.6 and ∼0.8 for SHBG for testosterone over a four-year period, indicating that a single measurement provided moderate to good estimates of longer-term exposures of testosterone and SHBG. Uniquely, the availability of second SHBG measurements in a subset of cohort participants also allowed us to correct for regression dilution bias, resulting in HRs of greater magnitude in all models. A further limitation was that we were unable to estimate the association of circulating concentrations of estrogens with colorectal cancer risk as the assay used in the UK Biobank to assess estradiol levels was not sufficiently sensitive to measure low concentrations commonly found in postmenopausal women and men. For our MR analyses, the summary level data that we used meant we were unable to conduct subgroup analyses by other colorectal cancer risk factors (e.g., age, BMI, smoking, menopausal status). In addition, our two-sample MR analyses using summary-level data assumed a linear relationship between sex steroid hormones and risk of colorectal cancer; consequently, potential nonlinear effects could not be examined.
In conclusion, our complementary observational and MR analyses did not support causal associations of circulating SHBG and free testosterone concentrations with colorectal cancer risk. For total testosterone, our MR analyses found positive associations with colorectal cancer among women only; however, we identified some evidence of pleiotropy that may have biased this result indicating the influence of independent biological pathways. Additional experimental studies are required to better understand the possible role of androgens in colorectal cancer development.
Authors' Disclosures
R.M. Martin reports grants from Cancer Research UK during the conduct of the study. G. Casey reports grants from NIH during the conduct of the study. M. Cotterchio reports grants from NCI (NIH) during the conduct of the study. No disclosures were reported by the other authors.
Disclaimer
Where authors are identified as personnel of the International Agency for Research on Cancer/World Health Organization, the authors alone are responsible for the views expressed in this article and they do not necessarily represent the decisions, policy or views of the International Agency for Research on Cancer/World Health Organization.
The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care.
Authors' Contributions
N. Dimou: Formal analysis, validation, methodology, writing–original draft. N. Mori: Writing–review and editing. S. Harlid: Writing–review and editing. J. Harbs: Writing–review and editing. R.M. Martin: Writing–review and editing. K. Smith-Byrne: Writing–review and editing. N. Papadimitriou: Writing–review and editing. D.T. Bishop: Writing–review and editing. G. Casey: Writing–review and editing. S.M. Colorado-Yohar: Writing–review and editing. M. Cotterchio: Writing–review and editing. A.J. Cross: Writing–review and editing. L. Le Marchand: Writing–review and editing. Y. Lin: Writing–review and editing. K. Offit: Writing–review and editing. N.C. Onland-Moret: Writing–review and editing. U. Peters: Writing–review and editing. J.D. Potter: Writing–review and editing. T.E. Rohan: Writing–review and editing. E. Weiderpass: Writing–review and editing. M.J. Gunter: Conceptualization, resources, supervision, funding acquisition, validation, methodology, writing–original draft. N. Murphy: Conceptualization, resources, formal analysis, supervision, funding acquisition, validation, methodology, writing–original draft.
Acknowledgments
This work was supported by the French National Cancer Institute (INCa SHSESP17, grant no. 2017-127) and by Cancer Research UK (C18281/A29019).
Genetics and Epidemiology of Colorectal Cancer Consortium (GECCO): NCI, NIH, U.S. Department of Health and Human Services (U01 CA137088, R01 CA059045, R01CA201407). Genotyping/sequencing services were provided by the Center for Inherited Disease Research (CIDR), contract numbers HHSN268201700006I and HHSN268201200008I. This research was funded in part through the NIH/NCI Cancer Center Support Grant No. P30 CA015704. Scientific Computing Infrastructure at Fred Hutch is funded by ORIP grant no. S10OD028685.
ASTERISK: A Hospital Clinical Research Program (PHRC-BRD09/C) from the University Hospital Center of Nantes (CHU de Nantes) and supported by the Regional Council of Pays de la Loire, the Groupement des Entreprises Françaises dans la Luttecontre le Cancer (GEFLUC), the Association Anne de Bretagne Génétique and the Ligue RégionaleContre le Cancer (LRCC).
The ATBC Study is supported by the Intramural Research Program of the U.S. NCI, NIH, and by U.S. Public Health Service contract HHSN261201500005C from the NCI, Department of Health and Human Services.
CLUE II funding was from the National Cancer Institute (U01 CA86308, Early Detection Research Network; P30 CA006973), National Institute on Aging (U01 AG18033), and the American Institute for Cancer Research. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government. COLO2&3: NIH (R01 CA60987).
Maryland Cancer Registry (MCR) Cancer data was provided by the Maryland Cancer Registry, Center for Cancer Prevention and Control, Maryland Department of Health, with funding from the State of Maryland and the Maryland Cigarette Restitution Fund. The collection and availability of cancer registry data is also supported by the Cooperative Agreement NU58DP006333, funded by the Centers for Disease Control and Prevention. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the Centers for Disease Control and Prevention or the Department of Health and Human Services.
ColoCare: This work was supported by the NIH (grant no. R01 CA189184 (Li/Ulrich), U01 CA206110 (Ulrich/Li/Siegel/Figueireido/Colditz, 2P30CA015704-40 (Gilliland), R01 CA207371 (Ulrich/Li)), the Matthias Lackas-Foundation, the German Consortium for Translational Cancer Research, and the EU TRANSCAN initiative.
The Colon Cancer Family Registry (CCFR, www.coloncfr.org) was supported in part by funding from the NCI, NIH (award U01 CA167551). The CCFR Set-1 (Illumina 1M/1M-Duo) and Set-2 (Illumina Omni1-Quad) scans were supported by NIH awards U01 CA122839 and R01 CA143247 (to G. Casey). The CCFR Set-3 (Affymetrix Axiom CORECT Set array) was supported by NIH award U19 CA148107 and R01 CA81488 (SBG). The CCFR Set-4 (Illumina OncoArray 600K SNP array) was supported by NIH award U19 CA148107 (SBG) and by the Center for Inherited Disease Research (CIDR), which is funded by the NIH to the Johns Hopkins University, contract number HHSN268201200008I. The content of this manuscript does not necessarily reflect the views or policies of the NCI, NIH, or any of the collaborating centers in the Colon Cancer Family Registry (CCFR), nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government, any cancer registry, or the CCFR.
COLON: The COLON study was sponsored by WereldKankerOnderzoek Fonds, including funds from grant 2014/1179 as part of the World Cancer Research Fund International Regular Grant Programme, by Alped'Huzes and the Dutch Cancer Society (UM 2012-5653, UW 2013-5927, UW2015-7946), and by TRANSCAN (JTC2012-MetaboCCC, JTC2013-FOCUS). The Nqplus study is sponsored by a ZonMW investment grant (98-10030); by PREVIEW, the project PREVention of diabetes through lifestyle intervention and population studies in Europe and around the World (PREVIEW) project which received funding from the European Union Seventh Framework Programme (FP7/2007–2013) under grant no. 312057; by funds from TI Food and Nutrition (cardiovascular health theme), a public–private partnership on precompetitive research in food and nutrition; and by FOODBALL, the Food Biomarker Alliance, a project from JPI Healthy Diet for a Healthy Life.
Colorectal Cancer Transdisciplinary (CORECT) Study: The CORECT Study was supported by the National Cancer Institute, National Institutes of Health (NCI/NIH), U.S. Department of Health and Human Services (grant nos. U19 CA148107, R01 CA81488, P30 CA014089, R01 CA197350, P01 CA196569; R01 CA201407) and National Institutes of Environmental Health Sciences, National Institutes of Health (grant no. T32 ES013678).
CORSA: “Österreichische Nationalbank Jubiläumsfondsprojekt” (12511) and Austrian Research Funding Agency (FFG) grant 829675.
CPS-II: The American Cancer Society funds the creation, maintenance, and updating of the Cancer Prevention Study-II (CPS-II) cohort. This study was conducted with Institutional Review Board approval.
CRCGEN: Colorectal Cancer Genetics & Genomics, Spanish study was supported by Instituto de Salud Carlos III, cofunded by FEDER funds—a way to build Europe—(grants PI14-613 and PI09-1286), Agency for Management of University and Research Grants (AGAUR) of the Catalan Government (grant 2017SGR723), and Junta de Castilla y León (grant LE22A10-2). Sample collection of this work was supported by the Xarxa de Bancs de Tumors de Catalunya sponsored by Pla Director d'Oncología de Catalunya (XBTC), Plataforma Biobancos PT13/0010/0013 and ICOBIOBANC, sponsored by the Catalan Institute of Oncology.
Czech Republic CCS: This work was supported by the Grant Agency of the Czech Republic (grants CZ GA CR: GAP304/10/1286 and 1585) and by the Grant Agency of the Ministry of Health of the Czech Republic (grants AZV 15–27580A and AZV 17–30920A).
DACHS: This work was supported by the German Research Council (BR 1704/6–1, BR 1704/6–3, BR 1704/6–4, CH 117/1–1, HO 5117/2–1, HE 5998/2–1, KL 2354/3–1, RO 2270/8–1, and BR1704/17–1), the Interdisciplinary Research Program of the National Center for Tumor Diseases (NCT), Germany, and the German Federal Ministry of Education and Research (01KH0404, 01ER0814, 01ER0815, 01ER1505A, and 01ER1505B).
DALS: NIH (R01 CA48998 to M. L. Slattery).
EDRN: This work is funded and supported by the NCI, EDRN Grant (U01 CA 84968–06).
EPIC: The coordination of EPIC is financially supported by International Agency for Research on Cancer (IARC) and also by the Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London which has additional infrastructure support provided by the NIHR Imperial Biomedical Research Centre (BRC).
The national cohorts are supported by: Danish Cancer Society (Denmark); Ligue Contre le Cancer, Institut Gustave Roussy, Mutuelle Générale de l'Education Nationale, Institut National de la Santé et de la Recherche Médicale (INSERM; France); German Cancer Aid, German Cancer Research Center (DKFZ), German Institute of Human Nutrition PotsdamRehbruecke (DIfE), Federal Ministry of Education and Research (BMBF; Germany); Associazione Italiana per la Ricerca sul Cancro-AIRC-Italy, Compagnia di SanPaolo and National Research Council (Italy); Dutch Ministry of Public Health, Welfare and Sports (VWS), Netherlands Cancer Registry (NKR), LK Research Funds, Dutch Prevention Funds, Dutch ZON (Zorg Onderzoek Nederland), World Cancer Research Fund (WCRF), Statistics Netherlands (The Netherlands); Health Research Fund (FIS) - Instituto de Salud Carlos III (ISCIII), Regional Governments of Andalucía, Asturias, Basque Country, Murcia and Navarra, and the Catalan Institute of Oncology - ICO (Spain); Swedish Cancer Society, Swedish Research Council and County Councils of Skåne and Västerbotten (Sweden); Cancer Research UK (14136 to EPIC-Norfolk; C8221/A29017 to EPIC-Oxford), Medical Research Council (1000143 to EPIC-Norfolk; MR/M012190/1 to EPIC-Oxford; UK).
EPICOLON: This work was supported by grants from Fondo de Investigación Sanitaria/FEDER (PI08/0024, PI08/1276, PS09/02368, P111/00219, PI11/00681, PI14/00173, PI14/00230, PI17/00509, 17/00878, Acción Transversal de Cáncer), Xunta de Galicia (PGIDIT07PXIB9101209PR), Ministerio de Economia y Competitividad (SAF07–64873, SAF 2010–19273, SAF2014–54453R), Fundación Científica de la Asociación Española contra el Cáncer (GCB13131592CAST), Beca Grupo de Trabajo “Oncología” AEG (Asociación Española de Gastroenterología), Fundación Privada Olga Torres, FP7 CHIBCHA Consortium, Agència de Gestiód'AjutsUniversitaris i de Recerca (AGAUR, Generalitat de Catalunya, 2014SGR135, 2014SGR255, 2017SGR21, 2017SGR653), Catalan Tumour Bank Network (Pla Director d'Oncologia, Generalitat de Catalunya), PERIS (SLT002/16/00398, Generalitat de Catalunya), CERCA Programme (Generalitat de Catalunya) and COST Action BM1206 and CA17118. CIBERehd is funded by the Instituto de Salud Carlos III.
ESTHER/VERDI. This work was supported by grants from the Baden-Württemberg Ministry of Science, Research and Arts and the German Cancer Aid.
Harvard cohorts (HPFS, NHS, PHS): HPFS is supported by the NIH (P01 CA055075, UM1 CA167552, U01 CA167552, R01 CA137178, R01 CA151993, R35CA197735, K07 CA190673, and P50 CA127003), NHS by the NIH (R01 CA137178, P01 CA087969, UM1 CA186107, R01 CA151993, R35 CA197735, K07CA190673, and P50 CA127003) and PHS by NIH (R01 CA042182).
Hawaii Adenoma Study: NCI grants R01 CA72520.
HCES-CRC: the Hwasun Cancer Epidemiology Study–Colon and Rectum Cancer (HCES-CRC; grants from Chonnam National University Hwasun Hospital, HCRI15011–1).
Kentucky: This work was supported by the following grant support: Clinical Investigator Award from Damon Runyon Cancer Research Foundation (CI-8); NCI R01CA136726.
LCCS: The Leeds Colorectal Cancer Study was funded by the Food Standards Agency and Cancer Research UK Programme Award (C588/A19167).
MCCS cohort recruitment was funded by VicHealth and Cancer Council Victoria. The MCCS was further supported by Australian NHMRC grants 509348, 209057, 251553, and 504711 and by infrastructure provided by Cancer Council Victoria. Cases and their vital status were ascertained through the Victorian Cancer Registry (VCR) and the Australian Institute of Health and Welfare (AIHW), including the National Death Index and the Australian Cancer Database.
MEC: NIH (R37 CA54281, P01 CA033619, and R01 CA063464).
MECC: This work was supported by the NIH, U.S. Department of Health and Human Services (R01 CA81488 to SBG and GR).
MSKCC: The work at Sloan Kettering in New York was supported by the Robert and Kate Niehaus Center for Inherited Cancer Genomics and the Romeo Milio Foundation. Moffitt: This work was supported by funding from the NIH (grant no. R01 CA189184, P30 CA076292), Florida Department of Health Bankhead-Coley Grant 09BN-13, and the University of South Florida Oehler Foundation. Moffitt contributions were supported in part by the Total Cancer Care Initiative, Collaborative Data Services Core, and Tissue Core at the H. Lee Moffitt Cancer Center & Research Institute, a NCI-designated Comprehensive Cancer Center (grant no. P30 CA076292).
NCCCS I and II: We acknowledge funding support for this project from the NIH, R01 CA66635 and P30 DK034987.
NFCCR: This work was supported by an Interdisciplinary Health Research Team award from the Canadian Institutes of Health Research (CRT 43821); the NIH, U.S. Department of Health and Human Serivces (U01 CA74783); and NCI of Canada grants (18223 and 18226). The authors wish to acknowledge the contribution of Alexandre Belisle and the genotyping team of the McGill University and Génome Québec Innovation Centre, Montréal, Canada, for genotyping the Sequenom panel in the NFCCR samples. Funding was provided to Michael O. Woods by the Canadian Cancer Society Research Institute.
NSHDS: Swedish Research Council; Swedish Cancer Society; Cutting-Edge Research Grant and other grants from Region Västerbotten; Knut and Alice Wallenberg Foundation; Lion's Cancer Research Foundation at Umeå University; the Cancer Research Foundation in Northern Sweden; and the Faculty of Medicine, Umeå University, Umeå, Sweden.
OFCCR: The Ontario Familial Colorectal Cancer Registry was supported in part by the NCI of the NIH under award U01 CA167551 and award U01/U24 CA074783 (to SG). Additional funding for the OFCCR and ARCTIC testing and genetic analysis was through and a Canadian Cancer Society CaRE (Cancer Risk Evaluation) program grant and Ontario Research Fund award GL201–043 (to BWZ), through the Canadian Institutes of Health Research award 112746 (to TJH), and through generous support from the Ontario Ministry of Research and Innovation.
OSUMC: OCCPI funding was provided by Pelotonia and HNPCC funding was provided by the NCI (CA16058 and CA67941).
PLCO: Intramural Research Program of the Division of Cancer Epidemiology and Genetics and supported by contracts from the Division of Cancer Prevention, National Cancer Institute, NIH, DHHS. Funding was provided by National Institutes of Health (NIH), Genes, Environment and Health Initiative (GEI) Z01 CP 010200, NIH U01 HG004446, and NIH GEI U01 HG 004438.
SFCCR: The Seattle Colon Cancer Family Registry was supported in part by the National Cancer Institute (NCI) of the National Institutes of Health (NIH) under awards U01 CA167551, U01 CA074794 (to JDP), and awards U24 CA074794 and R01 CA076366 (to PAN).
SEARCH: The University of Cambridge has received salary support in respect of PDPP from the NHS in the East of England through the Clinical Academic Reserve. Cancer Research UK (C490/A16561); the UK National Institute for Health Research Biomedical Research Centres at the University of Cambridge.
SELECT: Research reported in this publication was supported in part by the National Cancer Institute of the National Institutes of Health under Award Numbers U10 CA37429 (C.D. Blanke), and UM1 CA182883 (C.M. Tangen/I.M. Thompson). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
SMS and REACH: This work was supported by the National Cancer Institute (grant P01 CA074184 to J.D.P. and P.A.N., grants R01 CA097325, R03 CA153323, and K05 CA152715 to P.A.N., and the National Center for Advancing Translational Sciences at the NIH (grant KL2 TR000421 to A.N.B.-H.)
The Swedish Low-risk Colorectal Cancer Study: The study was supported by grants from the Swedish research council; K2015–55X-22674–01–4, K2008–55X-20157–03–3, K2006–72X-20157–01–2 and the Stockholm County Council (ALF project).
Swedish Mammography Cohort and Cohort of Swedish Men: This work was supported by the Swedish Research Council/Infrastructure grant, the Swedish Cancer Foundation, and the Karolinska Institute's Distinguished Professor Award to Alicja Wolk.
UK Biobank: This research has been conducted using the UK Biobank Resource under Application Number 8614.
VITAL: NIH (K05 CA154337).
WHI: The WHI program is funded by the National Heart, Lung, and Blood Institute, NIH, U.S. Department of Health and Human Services through contracts HHSN268201100046C, HHSN268201100001C, HHSN268201100002C, HHSN268201100003C, HHSN268201100004C, and HHSN271201100004C.
RMM was supported by the NIHR Biomedical Research Centre at University Hospitals Bristol and Weston NHS Foundation Trust and the University of Bristol. RMM was supported by a Cancer Research UK (C18281/A29019) programme grant (the Integrative Cancer Epidemiology Programme).
This work was supported by Cancer Research UK (C18281/A29019).
We are very grateful to Dr. Bruno Buecher without whom this project would not have existed. We also thank all those who agreed to participate in this study, including the patients and the healthy control persons, as well as all the physicians, technicians, and students.
CLUE II: We thank the participants of Clue II and appreciate the continued efforts of the staff at the Johns Hopkins George W. Comstock Center for Public Health Research and Prevention in the conduct of the Clue II Cohort Study.
COLON and NQplus: the authors would like to thank the COLON and NQplus investigators at Wageningen University & Research and the involved clinicians in the participating hospitals.
CORSA: We kindly thank all those who contributed to the screening project Burgenland against CRC. Furthermore, we are grateful to Doris Mejri and Monika Hunjadi for laboratory assistance.
CPS-II: The authors thank the CPS-II participants and Study Management Group for their invaluable contributions to this research. The authors would also like to acknowledge the contribution to this study from central cancer registries supported through the Centers for Disease Control and Prevention National Program of Cancer Registries, and cancer registries supported by the National Cancer Institute Surveillance Epidemiology and End Results program.
Czech Republic CCS: We are thankful to all clinicians in major hospitals in the Czech Republic, without whom the study would not be practicable. We are also sincerely grateful to all patients participating in this study.
DACHS: We thank all participants and cooperating clinicians, and Ute Handte-Daub, Utz Benscheid, MuhabbetCelik and Ursula Eilber for excellent technical assistance.
EDRN: We acknowledge all the following contributors to the development of the resource: University of Pittsburgh School of Medicine, Department of Gastroenterology, Hepatology and Nutrition: Lynda Dzubinski; University of Pittsburgh School of Medicine, Department of Pathology: Michelle Bisceglia; and University of Pittsburgh School of Medicine, Department of Biomedical Informatics.
EPIC: Where authors are identified as personnel of the International Agency for Research on Cancer/World Health Organization, the authors alone are responsible for the views expressed in this article and they do not necessarily represent the decisions, policy or views of the International Agency for Research on Cancer/World Health Organization.
EPICOLON: We are sincerely grateful to all patients participating in this study who were recruited as part of the EPICOLON project. We acknowledge the Spanish National DNA Bank, Biobank of Hospital Clínic–IDIBAPS and Biobanco Vasco for the availability of the samples. The work was carried out (in part) at the Esther Koplowitz Centre, Barcelona.
Harvard cohorts (HPFS, NHS, PHS): The study protocol was approved by the institutional review boards of the Brigham and Women's Hospital and Harvard T.H. Chan School of Public Health, and those of participating registries as required.We acknowledge Channing Division of Network Medicine, Department of Medicine, Brigham and Women's Hospital as home of the NHS.We would like to thank the participants and staff of the HPFS, NHS and PHS for their valuable contributions as well as the following state cancer registries for their help: AL, AZ, AR, CA, CO, CT, DE, FL, GA, ID, IL, IN, IA, KY, LA, ME, MD, MA, MI, NE, NH, NJ, NY, NC, ND, OH, OK, OR, PA, RI, SC, TN, TX, VA, WA, WY.The authors assume full responsibility for analyses and interpretation of these data.
Kentucky: We would like to acknowledge the staff at the Kentucky Cancer Registry.
LCCS: We acknowledge the contributions of Jennifer Barrett, Robin Waxman, Gillian Smith and Emma Northwood in conducting this study.
NCCCS I & II: We would like to thank the study participants, and the NC Colorectal Cancer Study staff.
NSHDS investigators thank the Biobank Research Unit at Umeå University, the Västerbotten Intervention Programme, the Northern Sweden MONICA study and Region Västerbotten for providing data and samples and acknowledge the contribution from Biobank Sweden, supported by the Swedish Research Council (VR 2017–00650).
PLCO: The authors thank the PLCO Cancer Screening Trial screening center investigators and the staff from Information Management Services Inc and Westat Inc. Most importantly, we thank the study participants for their contributions that made this study possible.
SFCCR: The authors would like to thank the study participants and staff of the Seattle Colon Cancer Family Registry and the Hormones and Colon Cancer study (CORE Studies).
SEARCH: We thank the SEARCH team.
SELECT: We thank the research and clinical staff at the sites that participated on SELECT study, without whom the trial would not have been successful. We are also grateful to the 35,533 dedicated men who participated in SELECT.
WHI: The authors thank the WHI investigators and staff for their dedication, and the study participants for making the program possible. A full listing of WHI investigators can be found at: http://www.whi.org/researchers/Documents%20%20Write%20a%20Paper/WHI%20Investigator%20Short%20List.pdf
We acknowledge the National Institute for Public Health and the Environment (RIVM), Bilthoven, the Netherlands, for their contribution and ongoing support to the EPIC Study.
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.