Abstract
Objectives
Several lines of evidence have suggested that female hormones may lower risk for developing colorectal cancer. However, the mechanisms by which sex hormones affect colorectal cancer development remain unknown. We sought to determine whether the association may be under genetic control by evaluating genetic variation in estrogen receptors (ESR1 and ESR2), progesterone receptor (PGR), aromatase cytochrome 450 enzyme (CYP19A1) and 17 beta-hydroxysteroid dehydrogenase type 2 gene (HSD17B2).
Methods
We included 158 incident cases of colorectal cancer and 563 randomly chosen control subjects from 28,345 women in the Women's Health Study aged 45 years or older who provided blood samples and had no history of cancer or cardiovascular disease at baseline in 1993. All cases and controls were Caucasians of European descent. A total of 63 tagging and putative functional SNPs in the 5 genes were included for analysis. Unconditional logistic regression was used to estimate odds ratio (ORs) and 95% confidence intervals (CIs).
Results
There was no association between variation in ESR1, ESR2, PGR, CYP19A1 and HSD17B2 and colorectal cancer risk after correction for multiple comparisons (p values after correction ≥0.25). There was also no association with any of the haplotypes examined (p ≥0.15) and no evidence of joint effects of variants in the 5 genes (p ≥0.51).
Conclusion
Our data offer insufficient support for an association between variation in ESR1, ESR2, PGR, CYP19A1 and HSD17B2 and risk for developing colorectal cancer.
INTRODUCTION
Men tend to have a higher incidence rate of colorectal cancer than women of similar age in the US (1). In families with hereditary nonpolyposis colorectal cancer (HNPCC), the lifetime risk of developing colon cancer is much lower in females (30%) than in males (74%) (2). Male rats experimentally exposed to the carcinogen dimethylhydrazine also have twice the risk of developing colon cancer and shorter survival time than their female counterparts (3-5). Epidemologic studies have further shown that an increase in female hormones as a result of pregnancy and use of exogenous hormones such as oral contraceptives (OC) and postmenopausal hormone therapy (HT) are associated with a lower risk for developing colorectal cancer (6-9). It has been suggested that female hormones such as estrogen reduce synthesis or secretion of bile acids, which are believed to be carcinogenic and trophic to colonic epithelium (10-12).
In agreement with the observational findings, the Women's Health Initiative (WHI) estrogen plus progestin (E+P) clinical trial reported a 40% lower risk for colorectal cancer in the treatment group compared with the placebo group (13). However, the WHI estrogen-alone (E-alone) trial among hysterectomized women did not find a lower risk of colorectal cancer in the treatment group (14). Two recent observational studies also found no reduced risk for colorectal cancer incidence among postmenopausal women with higher circulating levels of estradiol and estrone (15, 16). Findings from these studies seemingly suggest that progesterone, but not estrogen, may be the key candidate for risk reduction in colorectal cancer. Alternatively, estrogen effects on colorectal cancer prevention may be modified by other risk factors. In the WHI E-alone trial, different risk estimates were observed in specific age groups; a decrease in colorectal cancer incidence was seen in women aged <70 (p value for age interaction=0.048) (14). Excess body weight in women tends to be associated with increased estrogen levels and it has been suggested that the beneficial effects associated with elevated estrogen levels on colorectal cancer may be offset by the adverse effects of obesity (17). Likewise, the risk associated with estrogen or progesterone may be under genetic control, as these hormones bind to their respective receptors to exert biological actions in target tissues such as the colorectum. Genes responsible for sex-hormone synthesis also affect changes in sex hormone concentrations. Given that the transcriptional activity of genes in target tissues is determined by genetic variation, variation in these hormone-related genes may affect risk for disease development.
Two estrogen receptors, estrogen receptor α (ESR1) and estrogen receptor β (ESR2) have been identified for mediating estrogen action (18). Both ESR1 and ESR2 were found to be expressed, with a greater expression of ESR2, in the gastrointestinal tract, (19, 20) and loss of both receptors have been detected in colon tumors (19-22). Progesterone receptor (PGR), activated mostly through binding to progesterone, has been detected at low to moderate levels in both normal and malignant colorectal tissues (23-27), and a lower expression of PGR has been reported in tumors than in normal colorectal mucosa (23). Two other estrogen-synthesizers, the aromatase cytochrome 450 enzyme (encoded by CYP19A1) and 17 beta-hydroxysteroid dehydrogenase type 2 gene (encoded by HSD17B2), have also been linked to growth development in colon cancer cells (28). The CYP19A1 gene is responsible for catalyzing the conversion of testosterone to estradiol and of androstenedione to estrone (29). The HSD17B2 gene plays a key role in the conversion of estradiol to estrone (30). To date, variation in these genes and colorectal cancer risk has been evaluated in three candidate-gene studies (31-33) and six genome-wide association studies (GWAS) (34-39), which have yielded different findings.
In this case-control study nested in a large female health professional cohort, we evaluated variation in ESR1, ESR2, PGR, CYP19A1, and HSD17B2 in relation to risk for colorectal cancer development. We undertook a comprehensive evaluation of common variants and putative functional variants in these 5 genes. We additionally evaluated effect modification by age, BMI, and hormone therapy (HT) use in relation to variation in these 5 genes and colorectal cancer risk.
MATERIALS AND METHODS
Study Population
We conducted a case-control study nested in the Women's Health Study (WHS), a completed randomized trial of aspirin and vitamin E in the primary prevention of cancer and cardiovascular disease. Beginning in 1993, 39,876 US female health professionals aged ≥45 years and free of cancer or cardiovascular disease enrolled in the study and completed a baseline questionnaire about their medical history and potential risk factors for colorectal cancer. Blood samples were collected in both EDTA and citrate tubes from 28,345 women before randomization. Baseline characteristics of women who provided blood were largely similar to those who did not (40).
In the present analysis, we included 158 women who had a confirmed diagnosis of incident colorectal cancer as of December 2005. A total of 563 healthy controls were randomly selected from the cohort. All cases and controls were Caucasians of European descent.
Single nucleotide polymorphism (SNP) selection
We selected a set of tagging SNPs that capture common variation and linkage disequilibrium (LD) structure across the ESR1 and HSD17B2 genes using the Tagger program implemented in Haploview software (41). The data source for tagging SNP selection was from the CEPH Utah residents with European ancestry in the International Hapmap Project (Release number #18) on the National Center for Biotechnology Information Build 34 assembly available in September 2005 (http://www.hapmap.org). Selection of tagging SNPs was based on a pairwise correlation coefficient (r2) of 0.8 or greater between tagging SNPs and untyped SNPs and a minor allele frequency (MAF) of 5% or greater in the CEPH population. A total of 15 and 3 tagging SNPs in the ESR1 and HSD17B2 genes, respectively, were selected for subsequent genotyping.
Selection of tagging SNPs in the ESR2, PGR and CYP19A1 genes was based on the studies from the Breast and Prostate Cohort Consortium project (BPC3), which has performed SNP discovery and dense genotyping to capture most common haplotype diversity among multi-ethnic groups including 70 American Caucasians (http://www.uscnorris.com/MECGenetics). The tagging SNPs, chosen from common SNPs with a MAF of 1% or greater among whites, predicted an Rh2 of 0.70 or greater between observed haplotypes and those predicted based on tagging SNP genotypes (42-44). Altogether, we included 4, 20, and 11 tagging SNPs in ESR2, PGR, and CYP19A1, respectively.
We also undertook a literature search in identifying putative functional SNPs associated with risk of cancers including colorectal cancer. Most of these chosen SNPs were either at coding (synonymous or nonsynonymous) or promoter regions. As a result, an additional 10 SNPs were selected for the 5 genes. In the present analysis of 63 SNPs with a MAF of ≥12%, we had ≥60% power to detect a relative risk of 1.85 or greater using a 2-sided test with a p value of 0.001 (Supplementary Table 1).
Genotyping
Genomic DNA was extracted from the whole blood samples of all subjects by the MagNa Pure LC instrument with the MagNA Pure LC DNA isolation kit (Roche Applied Science, Penzberg, Germany). Genotyping determination was performed with either an ABI fluorescence-based allelic discrimination method (Applied Biosystems, Foster City, CA) on an ABI 7900HT Sequence Detection System or a Sequenom genotyping method (Sequenom, San Diego, CA) according to the respective manufacture's specifications. Lab personnel were blinded to case-control status. Overall, the average call rate for all SNPs was 96%, ranged from 90% to 98%.
Statistical Analysis
Hardy-Weinberg equilibrium was first confirmed for each SNP among control subjects using a chi-square test. Genotyping data were then analyzed with unconditional logistic regression to calculate odds ratios (ORs) and 95% confidence intervals (CIs) as risk estimates for colorectal cancer in subjects with a linear (log-odds additive) scoring for 0, 1, or 2 copies of the minor allele of each SNP. The analyzed models were first adjusted for age (continuous), fasting status (≥8 hours or fewer), year of blood return (1-year interval), trial randomization date (12-month difference), and random treatment assignment (aspirin vs. placebo, vitamin E vs. placebo). The multivariate models were additionally adjusted for risk factors for colorectal cancer assessed at baseline, including menopausal status (premenopause, postmenopause, uncertain), hormone therapy use (never, past, current), body mass index (BMI, kg/m2, continuous), physical activity (energy expenditure in kcal/week, continuous), family history of colorectal cancer in a first-degree relative (yes, no), history of colorectal polyps (yes, no), smoking status (never, ever), alcohol consumption (g/day, continuous), and total energy intake (kcal/day, continuous). Because the crude models with adjustment for age, time and status of blood return, and treatment assignment yielded very similar results to the multivariate models (data not shown), we reported here only the results from the multivariate models.
Haplotype frequency and subject-specific expected haplotype indicators were calculated based on the unphased genotype data using the expectation-maximization (EM) algorithm as implemented in SAS procedure PROC HAPLOTYPE (45). We used unconditional logistic regression with additive scoring (ie, number of the copies of specific haplotypes) and the most common haplotype as the referent to estimate haplotype-specific ORs and 95% CIs (46, 47). Haplotypes with estimated frequencies of <5% in all subjects were pooled together into a single category. We performed a likelihood ratio test to examine the association between common haplotypes and colorectal cancer risk with a comparison of the log likelihood of the two models with and without additive effects of each haplotype.
To explore the possible joint effects of variation in the 5 genes on colorectal cancer risk, we evaluated multi-locus interaction by performing multifactor dimensionality reduction (MDR) analysis using the MDR software vesion1.0.Orc1. (48-50). The MDR methods perform exhaustive searches of all possible combinations of n loci with cross-validation procedures and reduce n-dimensional data to a single dimensional variable with 2 levels (high vs. low risk). In the present analysis, we allowed the MDR to choose up to 4 loci simultaneously and repeat 10-fold cross validation 10 times. P values were calculated by permutation testing with 1000 permutations.
We also assessed effect modifiers including age (<70, ≥70 years), BMI (<25, ≥25 kgm2), and HT use (current or never) on genetic association with colorectal cancer risk. We performed unconditional logistic regression analysis according to these factors with adjustment for the covariates described above. We performed a global likelihood ratio test with a comparison of the log likelihood of the two models with and without the interaction terms.
To account for the multiple comparison testing made in the study, we calculated the false discovery rate (FDR) for the analysis on each gene using SAS procedure PROC MULTTEST (51). All analyses except for the MDR were performed using SAS statistical package (version 9.1 for window; SAS Institute). All tests were two sided.
RESULTS
We first compared the baseline characteristics of colorectal cancer patients and control subjects (Table 1). Cancer patients were less likely than control subjects to be current users of hormone therapy. However, there was no difference in distribution between cases and controls with respect to several risk factors for colorectal cancer including smoking, history of colon polyps, family history of colorectal cancer, and screening exams. Both cases and controls were also similar in BMI, physical activity, and in total intakes of alcohol, total calories, and fiber.
Table 1.
Baseline characteristics (mean ± standard deviation or %) among colorectal cancer cases and control subjects in the Women's Health Study.
| Characteristics | Cases (N=158) | Controls (N=563) | Pvalue |
|---|---|---|---|
| Age, year (SD) | 59.2 (8.5) | 58.5 (8.4) | 0.34 |
| Body mass index, kg/m2 (SD) | 26.2 (5.4) | 25.8 (5.0) | 0.38 |
| History of colorectal polyps, % | 4.4 | 2.1 | 0.11 |
| Family history of colorectal cancer, % | 11.4 | 11.4 | >0.9 |
| Colonoscopy/sigmoidoscopy exam, %* | 3.2 | 4.6 | 0.42 |
| Current smoking, % | 7.0 | 11.7 | 0.09 |
| Postmenopause, % | 73.4 | 67.1 | 0.52 |
| Current use of hormone therapy, % | 36.1 | 48.0 | <0.01 |
| Current use of multivitamins, % | 31.4 | 30.8 | >0.9 |
| Physical activity, kcal/week (SD) | 970 (1145) | 1065 (1649) | 0.41 |
| Alcohol intake, g/day (SD) | 4.4 (9.2) | 4.4 (8.3) | >0.9 |
| Total energy intake, kcal/day (SD)† | 1,724 (523) | 1,732 (524) | 0.87 |
| Total fiber intake, g/day (SD)† | 19 (6) | 20 (6) | 0.43 |
| Tumor location | |||
| Proximal colon (right-sided), N | 63 | N/A | |
| Distal colon + rectum (left-sided), N | 92 | N/A | |
| Dukes’ stage | |||
| Dukes’ A, N | 52 | N/A | |
| Dukes’ B, N | 41 | N/A | |
| Dukes’ C, N | 60 | N/A |
Information was obtained at the 12-month follow-up questionnaire.
Nutrient intakes were energy-adjusted.
There were Two (rs2982712 in ESR1 and rs723012 in HSD17B2) among the total 63 SNPs were not in agreement with Hardy-Weinberg equilibrium (HWE) among controls before the FDR correction (p<0.05). After the correction for multiple comparisions, however, these two SNPs did not significantly deviate from HWE (FDR p values ≥0.25) and were retained for further analysis (Supplementary Table 1).
For the 63 SNPs evaluated, three SNPs, rs10046 in CYP19A1 and rs2911422 and rs2042429 in HSD17B2 genes, were marginally associated with colorectal cancer risk (p <0.05) (Table 2). Women who carried two copies of the minor allele for each of the two SNPs were at a greater risk for developing colorectal cancer as compared with non-carriers of the minor allele (Table 2). However, none of these 3 SNPs remained significant after the FDR correction for multiple testing (FDR p values ≥0.25). For the SNPs residing in the ESR1, ESR2, and PGR genes, there was no evidence of an association with colorectal cancer risk (Table 2).
Table 2.
Variation in sex-steroid receptors and hormone-synthesizing enzymes with colorectal cancer risk in the Women's Health Study.
| Gene/Variant | 0 copies |
1 copy |
2 copies |
Ptrend | FDR P | |||
|---|---|---|---|---|---|---|---|---|
| case/control | OR | case/control | OR | case/control | OR | |||
| ESR1 | ||||||||
| rs2077647 | 37/155 | 1.00 | 82/255 | 1.36 (0.87-2.13) | 29/121 | 1.02 (0.58-1.80) | 0.81 | 0.95 |
| rs6902771 | 40/164 | 1.00 | 79/270 | 1.22 (0.79-1.90) | 33/106 | 1.25 (0.73-2.15) | 0.39 | 0.90 |
| rs2234693 | 37/167 | 1.00 | 83/269 | 1.39 (0.89-2.17) | 31/104 | 1.34 (0.79-2.34) | 0.25 | 0.88 |
| rs9340799 | 52/229 | 1.00 | 78/239 | 1.41 (0.93-2.13) | 18/58 | 1.45 (0.76-2.75) | 0.11 | 0.88 |
| rs712221 | 50/189 | 1.00 | 73/273 | 1.02 (0.67-1.56) | 28/78 | 1.36 (0.78-2.38) | 0.35 | 0.90 |
| rs1709183 | 77/283 | 1.00 | 58/225 | 0.97 (0.65-1.45) | 17/37 | 2.00 (1.04-3.86) | 0.18 | 0.88 |
| rs2347923 | 73/258 | 1.00 | 60/209 | 1.04 (0.69-1.56) | 12/56 | 0.74 (0.37-1.50) | 0.60 | 0.93 |
| rs12199722 | 60/249 | 1.00 | 71/217 | 1.41 (0.94-2.12) | 16/49 | 1.39 (0.72-2.70) | 0.12 | 0.88 |
| rs6912184 | 96/334 | 1.00 | 50/182 | 0.98 (0.65-1.47) | 5/26 | 0.70 (0.25-1.91) | 0.62 | 0.93 |
| rs4583998 | 73/254 | 1.00 | 61/215 | 0.96 (0.64-1.44) | 13/56 | 0.80 (0.41-1.57) | 0.57 | 0.93 |
| rs1801132 | 98/337 | 1.00 | 52/192 | 0.95 (0.64-1.41) | 5/24 | 0.73 (0.27-2.03) | 0.59 | 0.93 |
| rs3020314 | 77/242 | 1.00 | 63/251 | 0.77 (0.52-1.15) | 12/53 | 0.71 (0.35-1.42) | 0.16 | 0.88 |
| rs3020401 | 78/251 | 1.00 | 59/245 | 0.77 (0.51-1.15) | 13/48 | 0.92 (0.46-1.81) | 0.37 | 0.90 |
| rs2982685 | 66/211 | 1.00 | 60/232 | 0.78 (0.51-1.18) | 18/72 | 0.81 (0.44-1.48) | 0.30 | 0.88 |
| rs2982683 | 61/268 | 1.00 | 78/227 | 1.57 (1.06-2.33) | 11/43 | 1.29 (0.61-2.71) | 0.08 | 0.88 |
| rs3020407 | 73/260 | 1.00 | 67/223 | 1.00 (0.67-1.48) | 12/60 | 0.70 (0.35-1.40) | 0.45 | 0.90 |
| rs9340954 | 87/294 | 1.00 | 58/215 | 0.91 (0.62-1.35) | 8/37 | 0.66 (0.29-1.51) | 0.35 | 0.90 |
| rs2982712 | 47/190 | 1.00 | 69/213 | 1.36 (0.88-2.10) | 25/102 | 0.99 (0.56-1.75) | 0.76 | 0.95 |
| rs3020434 | 107/371 | 1.00 | 42/159 | 0.94 (0.62-1.42) | 3/13 | 0.70 (0.19-2.62) | 0.60 | 0.93 |
| rs3020363 | 70/251 | 1.00 | 65/225 | 1.04 (0.70-1.55) | 17/69 | 0.85 (0.46-1.57) | 0.75 | 0.95 |
| rs3020370 | 94/326 | 1.00 | 52/188 | 0.98 (0.65-1.45) | 6/26 | 0.77 (0.30-1.98) | 0.68 | 0.95 |
| rs3798577 | 44/142 | 1.00 | 77/284 | 0.93 (0.60-1.45) | 31/120 | 0.85 (0.49-1.47) | 0.56 | 0.93 |
| ESR2 | ||||||||
| rs1255953 | 102/399 | 1.00 | 46/129 | 1.46 (0.96-2.21) | 4/13 | 0.99 (0.31-3.18) | 0.17 | 0.77 |
| rs928554 | 48/176 | 1.00 | 68/254 | 1.01 (0.65-1.55) | 35/109 | 1.12 (0.67-1.87) | 0.70 | 0.77 |
| rs1256031 | 42/163 | 1.00 | 70/256 | 1.12 (0.72-1.76) | 40/120 | 1.28 (0.77-2.15) | 0.34 | 0.77 |
| rs3020450 | 72/258 | 1.00 | 60/230 | 1.05 (0.70-1.58) | 20/57 | 1.29 (0.71-2.33) | 0.46 | 0.77 |
| PGR | ||||||||
| rs473409 | 82/320 | 1.00 | 66/193 | 1.32 (0.89-1.95) | 7/28 | 0.94 (0.39-2.26) | 0.39 | 0.91 |
| rs484389 | 90/329 | 1.00 | 58/190 | 1.09 (0.73-1.610 | 7/24 | 1.00 (0.41-2.44) | 0.77 | 0.91 |
| rs500760 | 90/329 | 1.00 | 57/188 | 1.09 (0.74-1.62) | 7/24 | 1.00 (0.41-2.45) | 0.76 | 0.91 |
| rs1042839 | 103/403 | 1.00 | 50/129 | 1.52 (1.00-2.29) | 2/9 | 0.87 (0.18-4.24) | 0.10 | 0.91 |
| rs545835 | 38/145 | 1.00 | 84/265 | 1.18 (0.75-1.86) | 31/132 | 0.89 (0.52-1.53) | 0.70 | 0.91 |
| rs1042838 | 104/402 | 1.00 | 49/129 | 1.47 (0.98-2.23) | 2/9 | 0.86 (0.18-4.20) | 0.13 | 0.91 |
| rs1824128 | 98/393 | 1.00 | 51/133 | 1.54 (1.02-2.32) | 4/15 | 1.15 (0.36-3.68) | 0.08 | 0.91 |
| rs477151 | 78/275 | 1.00 | 62/214 | 1.02 (0.69-1.50) | 8/48 | 0.62 (0.28-1.40) | 0.46 | 0.91 |
| rs559700 | 103/401 | 1.00 | 48/132 | 1.42 (0.94-2.15) | 2/10 | 0.76 (0.16-3.67) | 0.21 | 0.91 |
| rs516693 | 120/419 | 1.00 | 31/117 | 0.87 (0.55-1.38) | 4/7 | 2.13 (0.50-9.02) | 0.98 | 0.99 |
| rs508533 | 58/211 | 1.00 | 77/252 | 1.11 (0.74-1.65) | 19/79 | 0.94 (0.51-1.71) | 0.98 | 0.99 |
| rs572483 | 69/239 | 1.00 | 70/242 | 0.98 (0.66-1.45) | 14/61 | 0.82 (0.42-1.61) | 0.64 | 0.91 |
| rs543215 | 102/376 | 1.00 | 44/152 | 1.04 (0.68-1.58) | 7/14 | 1.95 (0.74-5.15) | 0.38 | 0.91 |
| rs613120 | 38/139 | 1.00 | 75/280 | 0.93 (0.58-1.47) | 40/120 | 1.22 (0.72-2.06) | 0.46 | 0.91 |
| rs529359 | 55/211 | 1.00 | 73/246 | 1.15 (0.76-1.72) | 20/78 | 0.99 (0.55-1.80) | 0.82 | 0.91 |
| rs481775 | 67/227 | 1.00 | 75/243 | 1.05 (0.7-1.55) | 40/66 | 0.61 (0.30-1.22) | 0.35 | 0.91 |
| rs1379130 | 71/231 | 1.00 | 71/251 | 0.93 (0.63-1.37) | 13/68 | 0.64 (0.33-1.26) | 0.26 | 0.91 |
| rs3740753 | 98/386 | 1.00 | 46/129 | 1.37 (0.90-2.09) | 2/8 | 1.02 (0.21-5.02) | 0.20 | 0.91 |
| CYP19A1 | ||||||||
| rs4646 | 84/283 | 1.00 | 57/220 | 0.82 (0.55-1.22) | 10/37 | 0.82 (0.38-1.77) | 0.34 | 0.93 |
| rs10046 | 31/153 | 1.00 | 77/270 | 1.26 (0.78-2.03) | 42/120 | 1.72 (1.01-2.95) | 0.05 | 0.64 |
| rs2414096 | 31/148 | 1.00 | 76/250 | 1.32 (0.82-2.13) | 37/119 | 1.39 (0.80-2.42) | 0.24 | 0.90 |
| rs727479 | 61/226 | 1.00 | 68/252 | 1.03 (0.69-1.54) | 24/63 | 1.59 (0.90-2.80) | 0.20 | 0.88 |
| rs1008805 | 43/179 | 1.00 | 77/262 | 1.32 (0.85-2.03) | 32/99 | 1.29 (0.75-2.21) | 0.29 | 0.91 |
| rs749292 | 56/175 | 1.00 | 66/264 | 0.79 (0.52-1.28) | 28/104 | 0.85 (0.50-1.44) | 0.44 | 0.93 |
| rs936306 | 109/378 | 1.00 | 34/136 | 0.82 (0.46-1.46) | 4/12 | 1.33 (0.41-4.37) | 0.64 | 0.95 |
| rs3751591 | 103/380 | 1.00 | 42/150 | 1.05 (0.69-1.60) | 6/15 | 1.39 (0.56-1.90) | 0.59 | 0.93 |
| rs1004984 | 62/224 | 1.00 | 70/258 | 0.96 (0.64-1.43) | 19/60 | 1.03 (0.56-1.90) | 0.99 | 0.99 |
| rs2445762 | 84/296 | 1.00 | 59/214 | 0.91 (0.61-1.34) | 9/34 | 0.84 (038-1.88) | 0.55 | 0.93 |
| rs2470144 | 44/156 | 1.00 | 81/274 | 1.02 (0.66-1.58) | 27/115 | 0.79 (0.46-1.37) | 0.46 | 0.93 |
| rs2445765 | 106/381 | 1.00 | 44/137 | 1.10 (0.72-1.66) | 1/19 | 0.17 (0.02-1.28) | 0.48 | 0.93 |
| rs2446405 | 108/383 | 1.00 | 44/141 | 1.02 (0.68-1.55) | 0/20 | -- | 0.19 | 0.88 |
| HSD17B2 | ||||||||
| rs4445895 | 60/226 | 1.00 | 70/248 | 1.01 (0.68-1.52) | 22/66 | 1.30 (0.73-2.32) | 0.47 | 0.77 |
| rs6564961 | 50/202 | 1.00 | 69/243 | 1.13 (0.74-1.74) | 23/71 | 1.43 (0.79-2.56) | 0.25 | 0.59 |
| rs2911422 | 38/173 | 1.00 | 76/278 | 1.28 (0.82-2.00) | 33/88 | 1.76 (1.01-3.06) | 0.05 | 0.25 |
| rs2042429 | 39/162 | 1.00 | 68/269 | 1.08 (0.69-1.71) | 39/94 | 1.83 (1.07-3.10) | 0.04 | 0.25 |
| rs723012 | 63/203 | 1.00 | 74/272 | 0.91 (0.61-1.36) | 13/62 | 0.70 (0.35-1.41) | 0.34 | 0.62 |
| rs9939740 | 62/178 | 1.00 | 68/273 | 0.71 (0.47-1.07) | 19/84 | 0.69 (0.38-1.24) | 0.11 | 0.39 |
*. OR=odds ratio; 95% CI= 95% confidence interval.
†. Multivariate models were adjusted for age (continuous), fasting status (≥8 hours or fewer), year of blood return (1-year interval), trial randomization date (12-month difference), random treatment assignment (aspirin vs. placebo, vitamin E vs. placebo), and additionally for risk factors for colorectal cancer assessed at baseline, including menopausal status (premenopause, postmenopause, uncertain), hormone therapy use (never, past, current), body mass index (kg/m2, continuous), physical activity (energy expenditure in kcal/week, continuous), family history of colorectal cancer in a first-degree relative (yes, no), history of colorectal polyps (yes, no), smoking status (never, ever), alcohol consumption (g/day, continuous), and total energy intake (kcal/day, continuous).
The MDR analysis also revealed no evidence of interaction among the SNPs for colorectal cancer risk. Four best models identified by the MDR procedure were one 1-factor (rs2982683 in ESR1), one 2-factor (rs12199722 in ESR1 and rs2042429 in HSD17B2), one 3-factor (rs2234693, and rs12199722 in ESR1, and rs2042429 in HSD17B2), and one 4-factor (rs1709183, rs2982712, and rs3798577 in ESR1, and rs2042429 in HSD17B2) models. However, these 4 models had only moderate prediction accuracy (~50%). The cross-validation consistency was low in most models (20 to 70%) and none of the models was significantly associated with the disease status (p values ≥0.51).
The tagging SNPs included in the present analysis represented a total of 16 regions of strong linkage disequilibrium (blocks) covering the 5 genes. Specifically, ESR2 and HSD17B2 each contained 1 haplotype block, and CYP19A1, ESR1, and PGR had 4, 6, and 4 blocks, respectively. There was no significant association between the 16 representative blocks and colorectal cancer risk (global p values ≥0.13). The risk estimates for specific haplotypes relative to the most common haplotypes are presented in Table 3.
Table 3.
Haplotype-based association of sex-steroid receptors (ESR1, ESR2, and PGR) and hormone synthesizing enzymes (CYP19A1 and HSD17B2) with colorectal cancer risk in the Women's Health Study*.
| Gene/Block # | N case (%) | N control (%) | OR (95% CI) | Ptrend | FDR P |
|---|---|---|---|---|---|
| ESR1 | |||||
| Block 1 | |||||
| C-T-A | 143 (50) | 558 (55) | 1.00 | 0.37 | 0.90 |
| T-C-G | 111 (39) | 343 (34) | 1.29 (0.96-1.75) | ||
| T-C-A | 26 (9) | 106 (10) | 0.97 (0.60-1.57) | ||
| Freq <5% | 4 (1) | 13 (1) | 0.86 (0.26-2.91) | ||
| Block 2 | |||||
| A-A | 99 (36) | 385 (40) | 1.00 | 0.24 | 0.88 |
| A-G | 96 (35) | 293 (30) | 1.34 (0.95-1.88) | ||
| C-A | 79 (29) | 294 (30) | 1.00 (0.78-1.55) | ||
| Block 3 | |||||
| A-G-C | 199 (71) | 706 (69) | 1.00 | 0.96 | 0.97 |
| G-A-G | 55 (19) | 214 (21) | 0.92 (0.64-1.30) | ||
| A-A-C | 27 (10) | 101 (10) | 0.94 (0.59-1.50) | ||
| Freq <5% | 1 (0.4) | 3 (0.3) | 1.21 (0.08-18.7) | ||
| Block 4 | |||||
| T-A | 212 (71) | 729 (67) | 1.00 | 0.50 | 0.93 |
| C-G | 82 (27) | 336 (31) | 0.84 (0.62-1.13) | ||
| Freq<5% | 6 (2) | 21 (2) | 0.92 (0.35-2.39) | ||
| Block 5 | |||||
| G-C | 95 (34) | 363 (36) | 1.00 | 0.21 | 0.88 |
| A-C | 88 (32) | 344 (34) | 0.96 (0.68-1.34) | ||
| A-T | 93 (34) | 291 (29) | 1.29 (0.92-1.82) | ||
| Block 6 | |||||
| T-C-T | 155 (58) | 574 (58) | 1.00 | 0.22 | 0.88 |
| C-T-C | 43 (16) | 167 (17) | 0.91 (0.61-1.36) | ||
| C-C-C | 41 (15) | 161 (16) | 0.91 (0.61-1.35) | ||
| C-C-T | 29 (11) | 74 (8) | 1.62 (1.00-2.63) | ||
| Freq <5% | 0 | 2 (0.2) | -- | ||
| ESR2 | |||||
| Block 1 | |||||
| C-C-G | 127 (43) | 452 (43) | 1.00 | 0.13 | 0.77 |
| T-T-A | 92 (31) | 331 (31) | 1.03 (0.76-1.40) | ||
| T-T-G | 56 (19) | 231 (22) | 0.85 (0.59-1.22) | ||
| Freq <5% | 19 (7) | 42 (4) | 1.92 (1.03-3.59) | ||
| PGR | |||||
| Block 1 | |||||
| A-C-G-C | 99 (33) | 387 (36) | 1.00 | 0.15 | 0.91 |
| A-T-G-C | 73 (24) | 285 (27) | 0.99 (0.09-1.42) | ||
| A-C-G-A | 53 (17) | 153 (14) | 1.33 (0.88-2.01) | ||
| C-T-T-C | 52 (17) | 146 (14) | 1.42 (0.95-2.14) | ||
| C-T-G-C | 21 (7) | 90 (8) | 0.78 (0.43-1.41) | ||
| Freq<5% | 6 (2) | 11 (1) | 2.42 (0.79-7.38) | ||
| Block 2 | |||||
| T-A-G-A | 94 (33) | 377 (36) | 1.00 | 0.32 | 0.91 |
| C-A-G-C | 76 (27) | 298 (28) | 1.03 (0.73-1.47) | ||
| T-G-G-A | 50 (17) | 145 (14) | 1.41 (0.93-2.14) | ||
| T-A-A-A | 34 (12) | 128 (12) | 1.02 (0.64-1.63) | ||
| T-A-G-C | 32 (11) | 99 (9) | 1.29 (0.80-2.08) | ||
| Freq <5% | 0 | 5 (0.5) | -- | ||
| Block 3 | |||||
| A-T-T | 94 (31) | 377 (35) | 1.00 | 0.16 | 0.91 |
| G-T-C | 96 (31) | 357 (33) | 1.07 (0.76-1.50) | ||
| A-C-T | 56 (18) | 178 (17) | 1.26 (0.85-1.87) | ||
| A-T-C | 58 (19) | 162 (15) | 1.47 (0.98-2.20) | ||
| Freq<5% | 2 (1) | 2 (0.2) | 6.99 (0.89-54.8) | ||
| Block 4 | |||||
| G-C-C | 103 (38) | 373 (38) | 1.00 | 0.51 | 0.91 |
| A-T-C | 85 (31) | 349 (35) | 0.87 (0.62-1.22) | ||
| A-C-G | 48 (18) | 134 (14) | 1.26 (0.82-1.94) | ||
| A-C-C | 33 (12) | 130 (13) | 0.88 (0.56-1.39) | ||
| Freq<5% | 1 (0.4) | 2 (0.2) | 1.67 (0.12-22.5) | ||
| CYP19A1 | |||||
| Block 1 | |||||
| G-T-A-T | 125 (47) | 497 (51) | 1.00 | 0.23 | 0.90 |
| G-C-G-G | 58 (22) | 162 (17) | 1.49 (1.01-2.19) | ||
| T-C-G-G | 37 (14) | 154 (16) | 0.95 (0.62-1.46) | ||
| T-C-G-T | 31 (12) | 97 (10) | 1.14 (0.67-1.95) | ||
| Freq < 5% | 15 (6) | 68 (6) | 0.82 (0.43-1.57) | ||
| Block 2 | |||||
| C-C | 135 (46) | 452 (43) | 1.00 | 0.74 | 0.98 |
| T-T | 118 (41) | 461 (43) | 0.87 (0.66-1.16) | ||
| T-C | 39 (13) | 148 (14) | 0.89 (0.59-1.35) | ||
| Freq <5% | 0 | 1 (0.1) | -- | ||
| Block 3 | |||||
| G-T | 217 (75) | 801 (77) | 1.00 | 0.48 | 0.93 |
| G-C | 30 (10) | 79 (8) | 1.39 (0.88-2.22) | ||
| A-C | 24 (8) | 93 (9) | 0.93 (0.55-1.58) | ||
| A-T | 15 (5) | 63 (6) | 0.88 (0.48-1.59) | ||
| Block 4 | |||||
| C-A-A-C-A | 159 (54) | 562 (54) | 1.00 | 0.69 | 0.95 |
| T-G-G-C-A | 63 (21) | 208 (20) | 1.03 (0.72-1.46) | ||
| T-A-G-G-T | 31 (10) | 111 (11) | 0.92 (0.59-1.45) | ||
| C-A-G-C-A | 15 (5) | 54 (5) | 1.10 (0.59-2.06) | ||
| Freq <5% | 24 (10) | 113 (10) | 0.71 (0.43-1.17) | ||
| HSD17B2 | |||||
| Block 1 | |||||
| C-A-G-T-T | 79 (32) | 326 (35) | 1.00 | 0.55 | 0.79 |
| T-G-C-C-C | 88 (35) | 289 (31) | 1.25 (0.87-1.78) | ||
| C-A-G-T-C | 42 (17) | 170 (18) | 1.01 (0.66-1.56) | ||
| C-A-G-T-C | 14 (6) | 0 | -- | ||
| Freq <5% | 27 (10) | 223 (20) | 1.04 (0.65-1.65) |
Multivariate models were adjusted for age (continuous), fasting status (≥8 hours or fewer), year of blood return (1-year interval), trial randomization date (12-month difference), random treatment assignment (aspirin vs. placebo, vitamin E vs. placebo), and additionally for risk factors for colorectal cancer assessed at baseline, including menopausal status (premenopause, postmenopause, uncertain), hormone therapy use (never, past, current), body mass index (kg/m2, continuous), physical activity (energy expenditure in kcal/week, continuous), family history of colorectal cancer in a first-degree relative (yes, no), history of colorectal polyps (yes, no), smoking status (never, ever), alcohol consumption (g/day, continuous), and total energy intake (kcal/day, continuous).
When evaluating the association between the SNPs and colorectal cancer risk according to modifying factors, we observed effect modification by BMI for the associations between several SNPs and colorectal cancer risk (p values for interaction <0.05) (Table 4). Notably, the risk associated with the two HSD17B2 SNPs, rs2911242 and rs2042429, was altered by BMI; the reported elevated risk for colorectal cancer was more pronounced among minor-allele carriers with normal BMI. However, none of the interactions remained statistically significant after correction for multiple testing (FDR p values ≥0.25). Age and HT use also did not appear to modify the association (data not shown).
Table 4.
Associated SNPs with colorectal cancer risk according to risk modifiers for colorectal cancer in the Women's Health Study*.
| Gene/Variant |
# copies |
N cases (%) |
N controls (%) |
OR (95% CI)†‡ |
N cases (%) |
N controls (%) |
OR (95% CI)†‡ |
Pinteraction |
FDR P |
|---|---|---|---|---|---|---|---|---|---|
| |
BMI <25 kg/m2 |
|
|
BMI ≥25 kg/m2 |
|||||
| HSD17B2 | |||||||||
| rs6564961 | 0 | 19 (27) | 115 (42) | 1.00 | 30 (42) | 83 (35) | 1.00 | 0.05 | 0.25 |
| 1 | 41 (59) | 128 (47) | 2.19 (1.14-4.22) | 28 (39) | 110 (47) | 0.65 (0.35-1.22) | |||
| 2 | 10 (14) | 29 (11) | 3.33 (1.27-8.74) | 13 (18) | 41 (18) | 0.88 (0.40-1.93) | |||
| rs2911422 | 0 | 12 (16) | 100 (35) | 1.00 | 25 (34) | 71 (29) | 1.00 | 0.02 | 0.25 |
| 1 | 45 (62) | 147 (51) | 3.02 (1.46-6.26) | 31 (23) | 125 (52) | 0.62 (0.32-1.18) | |||
| 2 | 16 (22) | 41 (14) | 3.86 (1.58-9.43) | 17 (23) | 45 (19) | 0.98 (0.45-2.10) | |||
| rs2042429 | 0 | 12 (16) | 91 (33) | 1.00 | 26 (37) | 71 (30) | 1.00 | 0.01 | 0.25 |
| 1 | 41 (55) | 143 (51) | 2.44 (1.15-5.13) | 27 (38) | 119 (50) | 0.59 (0.31-1.12) | |||
| 2 | 21 (28) | 46 (16) | 3.88 (1.65-9.12) | 18 (25) | 46 (19) | 1.05 (0.50-2.21) | |||
| ESR1 | |||||||||
| rs2077647 | 0 | 24 (32) | 81 (29) | 1.00 | 13 (18) | 71 (30) | 1.00 | 0.04 | 0.88 |
| 1 | 36 (47) | 136 (48) | 0.84 (0.45-1.56) | 46 (65) | 115 (48) | 2.34 (1.15-4.75) | |||
| 2 | 16 (21) | 66 (23) | 0.86 (0.40-1.85) | 12 (17) | 52 (22) | 1.35 (0.55-3.32) | |||
| rs2982685 | 0 | 25 (34) | 115 (42) | 1.00 | 41 (59) | 92 (40) | 1.00 | 0.04 | 0.88 |
| 1 | 37 (50) | 120 (44) | 1.38 (0.75-2.53) | 22 (32) | 109 (47) | 0.40 (0.21-0.76) | |||
| 2 | 12 (16) | 38 (14) | 1.19 (0.51-2.76) | 6 (9) | 31 (13) | 0.38 (0.14-1.03) | |||
| PGR | |||||||||
| rs545835 | 0 | 27 (35) | 77 (27) | 1.00 | 11 (15) | 65 (26) | 1.00 | 0.03 | 0.91 |
| 1 | 37 (48) | 143 (50) | 0.67 (0.36-1.25) | 46 (61) | 117 (48) | 2.20 (1.03-4.70) | |||
| 2 | 13 (17) | 67 (23) | 0.46 (0.21-1.02) | 18 (24) | 63 (26) | 1.78 (0.75-4.24) | |||
| rs477151 | 0 | 46 (62) | 142 (50) | 1.00 | 32 (43) | 128 (53) | 1.00 | 0.05 | 0.91 |
| 1 | 24 (32) | 118 (42) | 0.57 (0.32-1.03) | 38 (51) | 92 (38) | 1.79 (1.00-3.21) | |||
| 2 | 4 (5) | 24 (8) | 0.42 (0.13-1.39) | 4 (5) | 23 (9) | 0.82 (0.25-2.69) | |||
| rs508533 | 0 | 35 (45) | 105 (37) | 1.00 | 23 (31) | 102 (42) | 1.00 | 0.02 | 0.91 |
| 1 | 32 (41) | 146 (51) | 0.60 (0.34-1.07) | 44 (59) | 101 (41) | 2.10 (1.12-3.91) | |||
| 2 | 11 (14) | 36 (12) | 0.79 (0.34-1.85) | 8 (11) | 42 (17) | 1.05 (0.41-2.66) | |||
Included analysis with cell frequency ≥5%.
OR=odds ratio; 95% CI= 95% confidence interval.
Multivariate models were adjusted for age (continuous), fasting status (≥8 hours or fewer), year of blood return (1-year interval), trial randomization date (12-month difference), random treatment assignment (aspirin vs. placebo, vitamin E vs. placebo), and additionally for risk factors for colorectal cancer assessed at baseline, including menopausal status (premenopause, postmenopause, uncertain), hormone therapy use (never, past, current), body mass index (kg/m2, continuous), physical activity (energy expenditure in kcal/week, continuous), family history of colorectal cancer in a first-degree relative (yes, no), history of colorectal polyps (yes, no), smoking status (never, ever), alcohol consumption (g/day, continuous), and total energy intake (kcal/day, continuous).
DISCUSSION
In this study of common and coding variation in 3 sex hormone receptors (ESR1, ESR2, PGR) and 2 hormone-synthesizing enzymes (CYP19A1, HSD17B2) in relation to colorectal cancer risk, we observed little evidence for an association of variation in these genes with colorectal cancer risk after correcting for multiple comparisons. There was also no evidence of joint effects among multiple loci residing in these sex-hormone genes on risk for developing colorectal cancer.
Very few candidate-gene studies have evaluated variation in sex-hormone genes in relation to colorectal cancer risk and findings have been mixed. Two studies reported a potential link of two variants, rs9340799 (or XbaI-351) in ESR1 and rs125593 in ESR2, to colorectal cancer development (31, 32). However, we did not observe such an association. Variation in CYP19A1 was also not associated with colorectal cancer risk in both our study and the other case-control study of middle-aged men and women (33). To date, six phase-design GWAS of colorectal cancer have undertaken a comprehensive scan which identified several novel susceptibility loci mapping to 8q24, 8q23.3, 10p14, 15q13, and 18q21 (34-39). However, none of these detected regions harbor the sex-hormone genes evaluated in our study.
Higher circulating levels of sex hormones such as estrogens have been linked to an increased risk for developing breast cancer in postmenopausal women (52-54). However, variation in sex hormone synthesizers such as CYP19A1 which is predictive of estradiol levels has not been associated with breast cancer development (42, 55, 56). The observations raise the possibility that hormone synthesizers are related to hormonal levels instead of breast cancer susceptibility, although it remains possible that hormone synthesizers could be related to both and more studies are needed.
Evidence from the GWAS of breast cancer showing low odds ratios (<1.5) for the detected loci (57-62) suggests that the effects attributable to common genetic variants on disease susceptibility are likely small. It is, thus, plausible that variation in hormone synthesizers and other sex-hormone genes confers a small contribution to risk for breast cancer development and studies with larger sample sizes are needed for the detection of such small effects (55). A recent breast cancer consortium study of comprehensive analysis on ESR1 variants among >55000 women of European decent reported a weak association (OR=1.05) between an ESR1 variant tagging a conserved region of intron 4 and risk for developing estrogen receptor-positive breast cancer (63); the association with ESR1 variation was later replicated in a genome-wide scan of Chinese women (62). Current studies of other sex hormone-associated cancers such as colorectal cancer, however, are not well powered to detect the potential, possibly small, effects of variation in sex-hormone genes (64).
If genes respond to sex hormones such as estrogens in a similar fashion in colorectal tissue, combination of variants in these genes may have a more measureable effect on colorectal carcinogenesis. In addition to binding to their ligand receptors (ie, ESR1 and ESR2) to exert their biological function in target tissues, estrogens mediate the activity levels of other genes such as vitamin D receptor (VDR) that contain a response element of activation protein-1 (AP-1) (65, 66). A recent pathology report has shown that the joint variation in both ESR1 (eg, XbaI variant) and VDR (eg, bsmI variant) may enhance the association with rectal cancer risk, possibly through the inhibitory effects of these 2 genes on the expression of a downstream growth factor, erbB2, in rectal tumors (31). Replication of this finding as well as evaluation of sex hormone genes and other estrogen-responsive genes in relation to colorectal cancer will be addressed in our future studies.
Limitations of this study include the fact that the genotyped SNPs may not sufficiently cover the entire gene regions. Although we have chosen a commonly used selection threshold (r2 ≥70%) for tagging SNP selection and have also included several functional SNPs, we cannot rule out the possibility that other untyped variants may contribute to risk of developing colorectal cancer. In addition, our single-gene analyses have very limited statistical power after correction for multiple testing. Power is also limited in this study to allow for subgroup analysis according to menopausal status and tumor characteristics such as tumor location and tumor stage.
In summary, our findings offer little support for the association of variation in ESR1, ESR2, and PGR, CYP19A1 and HSD17B2 with colorectal cancer susceptibility in a female Caucasian population. The risks associated with these sex hormone genes are likely small and studies with larger sample sizes are required for such detection.
Supplementary Material
Supplementary Table 1. Characteristics of variation in sex-steroid receptors (ESR1, ESR2, and PGR) and hormone-synthesizing genes (CYP19A1 and HSD17B2) in the Women's Health Study.
Acknowledgments
The work was supported by grants CA112529 and CA47988 and from the National Cancer Institute, and grant HL43851 and HL080467 from the National Heart, Lung, and Blood Institute. We would like to thank the entire staff of the WHS under the leadership of David Gordon, as well as Mary Breen, Susan Burt, Marilyn Chown, Georgina Friedenberg, Inge Judge, Jean Mac-Fadyean, Geneva McNair, David Potter, Claire Ridge, and Harriet Samuelson. We also acknowledge the Endpoints Committee of the WHS (Dr. Wendy Y Chen, Jim Taylor, and Til Stürmer), and Anna Klevak for technical assistance with the manuscript.
REFERENCES
- 1.Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, et al. Cancer statistics, 2008. CA Cancer J Clin. 2008 Mar-Apr;58(2):71–96. doi: 10.3322/CA.2007.0010. [DOI] [PubMed] [Google Scholar]
- 2.Froggatt NJ, Green J, Brassett C, Evans DG, Bishop DT, Kolodner R, et al. A common MSH2 mutation in English and North American HNPCC families: origin, phenotypic expression, and sex specific differences in colorectal cancer. J Med Genet. 1999 Feb;36(2):97–102. [PMC free article] [PubMed] [Google Scholar]
- 3.Ochiai M, Watanabe M, Kushida H, Wakabayashi K, Sugimura T, Nagao M. DNA adduct formation, cell proliferation and aberrant crypt focus formation induced by PhIP in male and female rat colon with relevance to carcinogenesis. Carcinogenesis. 1996 Jan;17(1):95–8. doi: 10.1093/carcin/17.1.95. [DOI] [PubMed] [Google Scholar]
- 4.Odagiri E, Jibiki K, Kato Y, Nakamura S, Oda S, Demura R, et al. Steroid receptors in dimethylhydrazine-induced colon carcinogenesis. Cancer. 1985 Dec 1;56(11):2627–34. doi: 10.1002/1097-0142(19851201)56:11<2627::aid-cncr2820561116>3.0.co;2-j. [DOI] [PubMed] [Google Scholar]
- 5.Gershbein LL. Action of estrogen and adrenocorticoids on adenocarcinoma induction by 1,2-dimethylhydrazine in male rats. Res Commun Chem Pathol Pharmacol. 1993 Jul;81(1):117–20. [PubMed] [Google Scholar]
- 6.Calle EE, Miracle-McMahill HL, Thun MJ, Heath CW., Jr Estrogen replacement therapy and risk of fatal colon cancer in a prospective cohort of postmenopausal women. J Natl Cancer Inst. 1995 Apr 5;87(7):517–23. doi: 10.1093/jnci/87.7.517. [DOI] [PubMed] [Google Scholar]
- 7.Potter JD, McMichael AJ. Large bowel cancer in women in relation to reproductive and hormonal factors: a case-control study. J Natl Cancer Inst. 1983 Oct;71(4):703–9. [PubMed] [Google Scholar]
- 8.Kune GA, Kune S, Watson LF. Children, age at first birth, and colorectal cancer risk. Data from the Melbourne Colorectal Cancer Study. Am J Epidemiol. 1989 Mar;129(3):533–42. doi: 10.1093/oxfordjournals.aje.a115165. [DOI] [PubMed] [Google Scholar]
- 9.Singh S, Langman MJ. Oestrogen and colonic epithelial cell growth. Gut. 1995 Dec;37(6):737–9. doi: 10.1136/gut.37.6.737. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Narisawa T, Magadia NE, Weisburger JH, Wynder EL. Promoting effect of bile acids on colon carcinogenesis after intrarectal instillation of N-methyl-N'-nitro-N-nitrosoguanidine in rats. J Natl Cancer Inst. 1974 Oct;53(4):1093–7. doi: 10.1093/jnci/53.4.1093. [DOI] [PubMed] [Google Scholar]
- 11.McMichael AJ, Potter JD. Host factors in carcinogenesis: certain bile-acid metabolic profiles that selectively increase the risk of proximal colon cancer. J Natl Cancer Inst. 1985 Aug;75(2):185–91. [PubMed] [Google Scholar]
- 12.McMichael AJ, Potter JD. Reproduction, endogenous and exogenous sex hormones, and colon cancer: a review and hypothesis. J Natl Cancer Inst. 1980 Dec;65(6):1201–7. [PubMed] [Google Scholar]
- 13.Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial. Jama. 2002 Jul 17;288(3):321–33. doi: 10.1001/jama.288.3.321. [DOI] [PubMed] [Google Scholar]
- 14.Anderson GL, Limacher M, Assaf AR, Bassford T, Beresford SA, Black H, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. Jama. 2004 Apr 14;291(14):1701–12. doi: 10.1001/jama.291.14.1701. [DOI] [PubMed] [Google Scholar]
- 15.Gunter MJ, Hoover DR, Yu H, Wassertheil-Smoller S, Rohan TE, Manson JE, et al. Insulin, insulin-like growth factor-I, endogenous estradiol, and risk of colorectal cancer in postmenopausal women. Cancer Res. 2008 Jan 1;68(1):329–37. doi: 10.1158/0008-5472.CAN-07-2946. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Clendenen TV, Koenig KL, Shore RE, Levitz M, Arslan AA, Zeleniuch-Jacquotte A. Postmenopausal levels of endogenous sex hormones and risk of colorectal cancer. Cancer Epidemiol Biomarkers Prev. 2009 Jan;18(1):275–81. doi: 10.1158/1055-9965.EPI-08-0777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Giovannucci E. Obesity, gender, and colon cancer. Gut. 2002 Aug;51(2):147. doi: 10.1136/gut.51.2.147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Enmark E, Gustafsson JA. Oestrogen receptors - an overview. J Intern Med. 1999 Aug;246(2):133–8. doi: 10.1046/j.1365-2796.1999.00545.x. [DOI] [PubMed] [Google Scholar]
- 19.Campbell-Thompson M, Lynch IJ, Bhardwaj B. Expression of estrogen receptor (ER) subtypes and ERbeta isoforms in colon cancer. Cancer Res. 2001 Jan 15;61(2):632–40. [PubMed] [Google Scholar]
- 20.Kono S, Honjo S, Todoroki I, Nishiwaki M, Hamada H, Nishikawa H, et al. Glucose intolerance and adenomas of the sigmoid colon in Japanese men (Japan). Cancer Causes Control. 1998 Aug;9(4):441–6. doi: 10.1023/a:1008879920140. [DOI] [PubMed] [Google Scholar]
- 21.Jassam N, Bell SM, Speirs V, Quirke P. Loss of expression of oestrogen receptor beta in colon cancer and its association with Dukes’ staging. Oncol Rep. 2005 Jul;14(1):17–21. [PubMed] [Google Scholar]
- 22.Issa JP, Ottaviano YL, Celano P, Hamilton SR, Davidson NE, Baylin SB. Methylation of the oestrogen receptor CpG island links ageing and neoplasia in human colon. Nat Genet. 1994 Aug;7(4):536–40. doi: 10.1038/ng0894-536. [DOI] [PubMed] [Google Scholar]
- 23.Di Leo A, Linsalata M, Cavallini A, Messa C, Russo F. Sex steroid hormone receptors, epidermal growth factor receptor, and polyamines in human colorectal cancer. Dis Colon Rectum. 1992 Apr;35(4):305–9. doi: 10.1007/BF02048105. [DOI] [PubMed] [Google Scholar]
- 24.Singh S, Sheppard MC, Langman MJ. Sex differences in the incidence of colorectal cancer: an exploration of oestrogen and progesterone receptors. Gut. 1993 May;34(5):611–5. doi: 10.1136/gut.34.5.611. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Oshima CT, Wonraht DR, Catarino RM, Mattos D, Forones NM. Estrogen and progesterone receptors in gastric and colorectal cancer. Hepatogastroenterology. 1999 Nov-Dec;46(30):3155–8. [PubMed] [Google Scholar]
- 26.Berta L, Fronticelli Baldelli C, Fazzari A, Radice E, Bargoni A, Frairia R, et al. Sex steroid receptors, secondary bile acids and colorectal cancer. A possible mechanism of interaction. Panminerva Med. 2003 Dec;45(4):261–6. [PubMed] [Google Scholar]
- 27.Kaklamanos IG, Bathe OF, Franceschi D, Lazaris AC, Davaris P, Glinatsis M, et al. Expression of receptors for estrogen and progesterone in malignant colonic mucosa as a prognostic factor for patient survival. J Surg Oncol. 1999 Dec;72(4):225–9. doi: 10.1002/(sici)1096-9098(199912)72:4<225::aid-jso8>3.0.co;2-s. [DOI] [PubMed] [Google Scholar]
- 28.Fiorelli G, Picariello L, Martineti V, Tonelli F, Brandi ML. Estrogen synthesis in human colon cancer epithelial cells. J Steroid Biochem Mol Biol. 1999 Dec 31;71(5-6):223–30. doi: 10.1016/s0960-0760(99)00144-2. [DOI] [PubMed] [Google Scholar]
- 29.Means GD, Mahendroo MS, Corbin CJ, Mathis JM, Powell FE, Mendelson CR, et al. Structural analysis of the gene encoding human aromatase cytochrome P-450, the enzyme responsible for estrogen biosynthesis. J Biol Chem. 1989 Nov 15;264(32):19385–91. [PubMed] [Google Scholar]
- 30.Clemons M, Goss P. Estrogen and the risk of breast cancer. N Engl J Med. 2001 Jan 25;344(4):276–85. doi: 10.1056/NEJM200101253440407. [DOI] [PubMed] [Google Scholar]
- 31.Speer G, Cseh K, Winkler G, Takacs I, Barna I, Nagy Z, et al. Oestrogen and vitamin D receptor (VDR) genotypes and the expression of ErbB-2 and EGF receptor in human rectal cancers. Eur J Cancer. 2001 Aug;37(12):1463–8. doi: 10.1016/s0959-8049(01)00139-3. [DOI] [PubMed] [Google Scholar]
- 32.Slattery ML, Sweeney C, Murtaugh M, Ma KN, Wolff RK, Potter JD, et al. Associations between ERalpha, ERbeta, and AR genotypes and colon and rectal cancer. Cancer Epidemiol Biomarkers Prev. 2005 Dec;14(12):2936–42. doi: 10.1158/1055-9965.EPI-05-0514. [DOI] [PubMed] [Google Scholar]
- 33.Bethke L, Webb E, Sellick G, Rudd M, Penegar S, Withey L, et al. Polymorphisms in the cytochrome P450 genes CYP1A2, CYP1B1, CYP3A4, CYP3A5, CYP11A1, CYP17A1, CYP19A1 and colorectal cancer risk. BMC Cancer. 2007;7:123. doi: 10.1186/1471-2407-7-123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Tomlinson I, Webb E, Carvajal-Carmona L, Broderick P, Kemp Z, Spain S, et al. A genome-wide association scan of tag SNPs identifies a susceptibility variant for colorectal cancer at 8q24.21. Nat Genet. 2007 Aug;39(8):984–8. doi: 10.1038/ng2085. [DOI] [PubMed] [Google Scholar]
- 35.Zanke BW, Greenwood CM, Rangrej J, Kustra R, Tenesa A, Farrington SM, et al. Genome-wide association scan identifies a colorectal cancer susceptibility locus on chromosome 8q24. Nat Genet. 2007 Aug;39(8):989–94. doi: 10.1038/ng2089. [DOI] [PubMed] [Google Scholar]
- 36.Tomlinson IP, Webb E, Carvajal-Carmona L, Broderick P, Howarth K, Pittman AM, et al. A genome-wide association study identifies colorectal cancer susceptibility loci on chromosomes 10p14 and 8q23.3. Nat Genet. 2008 May;40(5):623–30. doi: 10.1038/ng.111. [DOI] [PubMed] [Google Scholar]
- 37.Tenesa A, Farrington SM, Prendergast JG, Porteous ME, Walker M, Haq N, et al. Genome-wide association scan identifies a colorectal cancer susceptibility locus on 11q23 and replicates risk loci at 8q24 and 18q21. Nat Genet. 2008 May;40(5):631–7. doi: 10.1038/ng.133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Jaeger E, Webb E, Howarth K, Carvajal-Carmona L, Rowan A, Broderick P, et al. Common genetic variants at the CRAC1 (HMPS) locus on chromosome 15q13.3 influence colorectal cancer risk. Nat Genet. 2008 Jan;40(1):26–8. doi: 10.1038/ng.2007.41. [DOI] [PubMed] [Google Scholar]
- 39.Broderick P, Carvajal-Carmona L, Pittman AM, Webb E, Howarth K, Rowan A, et al. A genome-wide association study shows that common alleles of SMAD7 influence colorectal cancer risk. Nat Genet. 2007 Nov;39(11):1315–7. doi: 10.1038/ng.2007.18. [DOI] [PubMed] [Google Scholar]
- 40.Zhang SM, Buring JE, Lee IM, Cook NR, Ridker PM. C-reactive protein levels are not associated with increased risk for colorectal cancer in women. Ann Intern Med. 2005 Mar 15;142(6):425–32. doi: 10.7326/0003-4819-142-6-200503150-00008. [DOI] [PubMed] [Google Scholar]
- 41.Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics. 2005 Jan 15;21(2):263–5. doi: 10.1093/bioinformatics/bth457. [DOI] [PubMed] [Google Scholar]
- 42.Haiman CA, Stram DO, Pike MC, Kolonel LN, Burtt NP, Altshuler D, et al. A comprehensive haplotype analysis of CYP19 and breast cancer risk: the Multiethnic Cohort. Hum Mol Genet. 2003 Oct 15;12(20):2679–92. doi: 10.1093/hmg/ddg294. [DOI] [PubMed] [Google Scholar]
- 43.Pearce CL, Hirschhorn JN, Wu AH, Burtt NP, Stram DO, Young S, et al. Clarifying the PROGINS allele association in ovarian and breast cancer risk: a haplotype-based analysis. J Natl Cancer Inst. 2005 Jan 5;97(1):51–9. doi: 10.1093/jnci/dji007. [DOI] [PubMed] [Google Scholar]
- 44.Cox DG, Bretsky P, Kraft P, Pharoah P, Albanes D, Altshuler D, et al. Haplotypes of the estrogen receptor beta gene and breast cancer risk. Int J Cancer. 2008 Jan 15;122(2):387–92. doi: 10.1002/ijc.23127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Fallin D, Schork NJ. Accuracy of haplotype frequency estimation for biallelic loci, via the expectation-maximization algorithm for unphased diploid genotype data. Am J Hum Genet. 2000 Oct;67(4):947–59. doi: 10.1086/303069. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Zaykin DV, Westfall PH, Young SS, Karnoub MA, Wagner MJ, Ehm MG. Testing association of statistically inferred haplotypes with discrete and continuous traits in samples of unrelated individuals. Hum Hered. 2002;53(2):79–91. doi: 10.1159/000057986. [DOI] [PubMed] [Google Scholar]
- 47.Kraft P, Cox DG, Paynter RA, Hunter D, De Vivo I. Accounting for haplotype uncertainty in matched association studies: a comparison of simple and flexible techniques. Genet Epidemiol. 2005 Apr;28(3):261–72. doi: 10.1002/gepi.20061. [DOI] [PubMed] [Google Scholar]
- 48.Ritchie MD, Hahn LW, Moore JH. Power of multifactor dimensionality reduction for detecting gene-gene interactions in the presence of genotyping error, missing data, phenocopy, and genetic heterogeneity. Genet Epidemiol. 2003 Feb;24(2):150–7. doi: 10.1002/gepi.10218. [DOI] [PubMed] [Google Scholar]
- 49.Hahn LW, Ritchie MD, Moore JH. Multifactor dimensionality reduction software for detecting gene-gene and gene-environment interactions. Bioinformatics. 2003 Feb 12;19(3):376–82. doi: 10.1093/bioinformatics/btf869. [DOI] [PubMed] [Google Scholar]
- 50.Moore JH. Computational analysis of gene-gene interactions using multifactor dimensionality reduction. Expert Rev Mol Diagn. 2004 Nov;4(6):795–803. doi: 10.1586/14737159.4.6.795. [DOI] [PubMed] [Google Scholar]
- 51.Benjamini Y, Yekutieli D. Quantitative trait Loci analysis using the false discovery rate. Genetics. 2005 Oct;171(2):783–90. doi: 10.1534/genetics.104.036699. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Hankinson SE, Willett WC, Manson JE, Colditz GA, Hunter DJ, Spiegelman D, et al. Plasma sex steroid hormone levels and risk of breast cancer in postmenopausal women. J Natl Cancer Inst. 1998 Sep 2;90(17):1292–9. doi: 10.1093/jnci/90.17.1292. [DOI] [PubMed] [Google Scholar]
- 53.Zeleniuch-Jacquotte A, Bruning PF, Bonfrer JM, Koenig KL, Shore RE, Kim MY, et al. Relation of serum levels of testosterone and dehydroepiandrosterone sulfate to risk of breast cancer in postmenopausal women. Am J Epidemiol. 1997 Jun 1;145(11):1030–8. doi: 10.1093/oxfordjournals.aje.a009059. [DOI] [PubMed] [Google Scholar]
- 54.Kaaks R, Rinaldi S, Key TJ, Berrino F, Peeters PH, Biessy C, et al. Postmenopausal serum androgens, oestrogens and breast cancer risk: the European prospective investigation into cancer and nutrition. Endocr Relat Cancer. 2005 Dec;12(4):1071–82. doi: 10.1677/erc.1.01038. [DOI] [PubMed] [Google Scholar]
- 55.Haiman CA, Dossus L, Setiawan VW, Stram DO, Dunning AM, Thomas G, et al. Genetic variation at the CYP19A1 locus predicts circulating estrogen levels but not breast cancer risk in postmenopausal women. Cancer Res. 2007 Mar 1;67(5):1893–7. doi: 10.1158/0008-5472.CAN-06-4123. [DOI] [PubMed] [Google Scholar]
- 56.Dunning AM, Dowsett M, Healey CS, Tee L, Luben RN, Folkerd E, et al. Polymorphisms associated with circulating sex hormone levels in postmenopausal women. J Natl Cancer Inst. 2004 Jun 16;96(12):936–45. doi: 10.1093/jnci/djh167. [DOI] [PubMed] [Google Scholar]
- 57.Easton DF, Pooley KA, Dunning AM, Pharoah PD, Thompson D, Ballinger DG, et al. Genome-wide association study identifies novel breast cancer susceptibility loci. Nature. 2007 Jun 28;447(7148):1087–93. doi: 10.1038/nature05887. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Hunter DJ, Kraft P, Jacobs KB, Cox DG, Yeager M, Hankinson SE, et al. A genome-wide association study identifies alleles in FGFR2 associated with risk of sporadic postmenopausal breast cancer. Nat Genet. 2007 Jul;39(7):870–4. doi: 10.1038/ng2075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Schumacher FR, Feigelson HS, Cox DG, Haiman CA, Albanes D, Buring J, et al. A common 8q24 variant in prostate and breast cancer from a large nested case-control study. Cancer Res. 2007 Apr 1;67(7):2951–6. doi: 10.1158/0008-5472.CAN-06-3591. [DOI] [PubMed] [Google Scholar]
- 60.Stacey SN, Manolescu A, Sulem P, Rafnar T, Gudmundsson J, Gudjonsson SA, et al. Common variants on chromosomes 2q35 and 16q12 confer susceptibility to estrogen receptor-positive breast cancer. Nat Genet. 2007 Jul;39(7):865–9. doi: 10.1038/ng2064. [DOI] [PubMed] [Google Scholar]
- 61.Thomas G, Jacobs KB, Kraft P, Yeager M, Wacholder S, Cox DG, et al. A multistage genome-wide association study in breast cancer identifies two new risk alleles at 1p11.2 and 14q24.1 (RAD51L1). Nat Genet. 2009 May;41(5):579–84. doi: 10.1038/ng.353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Zheng W, Long J, Gao YT, Li C, Zheng Y, Xiang YB, et al. Genome-wide association study identifies a new breast cancer susceptibility locus at 6q25.1. Nat Genet. 2009 Mar;41(3):324–8. doi: 10.1038/ng.318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63.Dunning AM, Healey CS, Baynes C, Maia AT, Scollen S, Vega A, et al. Association of ESR1 gene tagging SNPs with breast cancer risk. Hum Mol Genet. 2009 Mar 15;18(6):1131–9. doi: 10.1093/hmg/ddn429. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 64.Chung CC, Magalhaes W, Gonzalez-Bosquet J, Chanock SJ. Genome-wide Association Studies in Cancer - Current and Future Directions. Carcinogenesis. 2009 Nov 11; doi: 10.1093/carcin/bgp273. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 65.Watters JJ, Campbell JS, Cunningham MJ, Krebs EG, Dorsa DM. Rapid membrane effects of steroids in neuroblastoma cells: effects of estrogen on mitogen activated protein kinase signalling cascade and c-fos immediate early gene transcription. Endocrinology. 1997 Sep;138(9):4030–3. doi: 10.1210/endo.138.9.5489. [DOI] [PubMed] [Google Scholar]
- 66.Gilad LA, Bresler T, Gnainsky J, Smirnoff P, Schwartz B. Regulation of vitamin D receptor expression via estrogen-induced activation of the ERK 1/2 signaling pathway in colon and breast cancer cells. J Endocrinol. 2005 Jun;185(3):577–92. doi: 10.1677/joe.1.05770. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Table 1. Characteristics of variation in sex-steroid receptors (ESR1, ESR2, and PGR) and hormone-synthesizing genes (CYP19A1 and HSD17B2) in the Women's Health Study.