Estrogen Metabolites and the Risk of Breast Cancer in Older ... : Epidemiology (original) (raw)
A meta-analysis of 9 prospective studies reported a strong relation between serum estradiol concentrations and breast cancer.1 Estradiol is oxidized to yield estrone, which is then hydroxylated primarily via 1 of 2 irreversible pathways, forming either 16α hydroxyestrone (16α-OHE1) or 2-hydroxyestrone (2-OHE1); an increase in 1 pathway will reduce the amount in the competing pathway.2 It has been hypothesized that the extent of 2-OHE1 versus 16α-OHE1 hydroxylation could influence a woman’s risk of breast cancer.3 In animal models, 16α-OHE1 is correlated with the incidence of mammary tumors.4 In humans, the relation between estrogen metabolism and the risk of breast cancer is uncertain.
Several case-control studies have reported a higher 16-α-OHE1 concentration or a lower ratio of 2-OHE1 to 16-α-OHE1 among breast cancer cases compared with controls5-8; others have not.9,10 Only 2 prospective studies of this association have been published. A longitudinal report of the Guernsey III population-based study showed that postmenopausal women who developed breast cancer over a follow up of 19 years had a 15% lower ratio of 2- to 16α-OHE1 than matched controls but the association was modest.11 In contrast, after an average follow up of 5.5 years in the Hormones and Diet in Etiology of Breast Cancer (ORDET) study, there was no association between the 2- to 16α-OHE1 ratio and breast cancer in postmenopausal women.12 Higher hydroxylation of 16α has also been reported in women at high risk for breast cancer, including women with a family history of breast cancer and in women with atypical hyperplasia.10
We addressed this question using cases and controls from a prospective cohort study comparing serum metabolites in 272 women who developed incident breast cancer and 291 randomly selected controls.
METHODS
All women were participants in the Study of Osteoporotic Fractures, a prospective study of 9704 white, community-dwelling women who were at least 65 years of age at the beginning of the study.13 Women were recruited from 1986 to 1988 in Portland, Oregon, Minneapolis, Minnesota, Baltimore, Maryland, and the Monongahela Valley, Pennsylvania (near Pittsburgh) through mailings to population-based lists of age-eligible women (voter registration, driver’s license lists, and health maintenance organization member lists). Black women were excluded because of their lower incidence of fractures; white women who had undergone bilateral hip replacement were also excluded. Details of the study cohort have been published previously.13 For the current analysis, we excluded women who reported hormone replacement therapy or prevalent breast cancer at baseline. Incident cases of breast cancer were identified by self-report on annual follow-up questionnaires and by review of death records obtained from state health departments. Follow up of the cohort is more than 95% complete. During an average of 8.7 years of follow up,14,15 we confirmed 272 cases of breast cancer through physician review of medical records and pathology reports. These 272 confirmed cases included 39 cases of carcinoma in situ, 206 cases of stage 1 or stage 2 cancer, 12 cases of stage 3 or 4 cancer, 14 cases in which the stage was unknown, and 1 case in which information was missing. Information on estrogen receptor status was available for 214 cases; of these, 87% were estrogen receptor-positive. Using a case-cohort approach, 291 women were randomly chosen from the cohort to serve as controls. The study was approved by the Institutional Review Board at each institution. All participants gave written informed consent.
Serum was obtained at a baseline examination in 1986 to 1988 and stored at −120°C until assays were performed. To minimize lipemia, all participants were instructed to adhere to a fat-free diet overnight and on the morning of the examination.
The 2-hydroxyestrone and 16a-hydroxyestrone estrogen metabolites were measured in blinded serum samples by Immuna Care Corporation (Bethlehem, PA) with the ESTRAMET 2/16 enzyme immunoassay kits (ELISAs).16 The assays for urinary estrogen metabolites have been validated against gas-chromatography-mass spectroscopy (GC-MS) methods.16,17 However, GC-MS methods are not sufficiently sensitive to measure the lower levels of estrogen detected in serum. Therefore, serum assays were validated directly against the urine assays by adding known amounts of urinary 2-OHE1 and 16α-OHE1 collected from pregnant women to pooled human serum samples and then performing the serum ELISAs. Agreement between the urine and serum methods was within the interassay variability.
The sensitivity of the 2-OHE1 and 16a-OHE1 assays is approximately 20 pg/mL and 10 pg/mL, respectively. In this study, variability of within-assay duplicates for positive control sera were less than 5% and the between-assay variability was less than 15% for both estrogen metabolite assays. The reproducibility of serum levels of 2-OHE1 and 16a-OHE1 from 25 postmenopausal women measured in blind duplicate in different batches demonstrated correlations of r = .98.
Weight and height were measured at baseline to calculate the body mass index (BMI; in kilograms per square meter). Information on reproductive and medical history, family history of breast cancer, and lifestyle was obtained at baseline by a self-administered questionnaire that was reviewed by a trained interviewer during a clinic visit.
The estrogen metabolites were not normally distributed. A natural log transformation was used to normalize the metabolites for the test of mean differences between cases and controls, after adjusting for age and for age plus BMI, using analysis of covariance. We report the geometric mean and 95% confidence intervals (CI). The relative hazard for breast cancer was calculated and tested for trend across quartiles of estrogen metabolites using a modification of the Cox proportional hazards model that accounts for the case-cohort sampling design.18 Cut points for quartiles were based on the distribution within the random subset of the cohort. We adjusted for age and BMI.
Estrogen metabolites and their ratio were compared across a number of descriptive variables using the nonparametric Wilcoxon 2-sample test. Continuous variables (age, BMI) were dichotomized at the median.
RESULTS
Women who developed breast cancer were slightly younger and heavier (both in weight and BMI) than women who did not develop cancer (Table 1). Cases also reported slightly more alcohol consumption than controls and were more likely to report a history of fibrocystic breast disease. Cases and controls were similar with respect to other risk factors for breast cancer. Within the control group, estrogen metabolite concentrations were similar in women by age (<70, ≥70 years), family history of breast cancer, past estrogen use, and education (data not shown). Comparing women with BMI of more than 26.8 with those with a lower BMI, no differences were found in 2-OHE1. However, 16α-OHE1 concentrations were higher in women with a higher BMI (237 pg/mL versus 224 pg/mL).
Descriptive Characteristics of the Women Who Developed Breast Cancer and the Random Sample of Controls From the Cohort
The mean level of 2-OHE1 was 4% higher among women who developed breast cancer compared with the controls and the mean level of 16α-OHE1 was 3% higher (Table 2). There was, however, no difference in the ratio of 2- to 16α-OHE1 between cases and controls.
Serum Estrogen Metabolite Levels by Breast Cancer Status
The relative risk of breast cancer was greatest among women with the highest 2-OHE1 metabolite concentration, whereas for the 16α-OHE1 metabolite, the highest breast cancer risk was found for the third quartile (Table 3). Among quartiles of the metabolite ratio, the relative risk was slightly elevated in the highest quartile. However, all CIs were wide and include 1.0.
Relative Hazard for Breast Cancer by Concentration of Serum Estrogen Metabolites
DISCUSSION
Bradlow and Fishman and their colleagues3,4,19,20 have suggested that increased formation of 16α-hydroxyestrone could be associated with an increased risk of breast cancer. The results of our study do not support their hypothesis. We found no associations between the extent of 16α-hydroxylation of estrone relative to the 2-hydroxylation pathway and breast cancer risk.
Most previous studies of estrogen metabolism and breast cancer were case-control studies, results of which have been inconsistent.5-10 Most of these studies were based on fewer than 50 cases with noncomparable controls. The largest case-control study was nested within an established population-based study.9 Urinary estrogen metabolites were measured in 66 breast cancer cases identified by a registry and compared with 76 random controls. The urine samples were obtained 3 to 7 years after the clinical diagnosis of breast cancer. This earlier study found no difference in estrogen metabolites between the cases and controls. Our results, obtained prospectively, are consistent with these findings.
In 2 prospective studies among premenopausal women, a higher ratio of 2-OHE1 to 16α-OHE1 at baseline was associated with a decreased risk of breast cancer, but CIs overlapped the null value11,12 However, discrepant results have been reported for postmenopausal women. The Guernsey III study found that a higher 2:16 ratio was associated with a lower risk of breast cancer in postmenopausal women, similar to their findings in premenopausal women.11 In contrast, there was no association in the ORDET study between estrogen metabolism and breast cancer in postmenopausal women.12 The women in our study were on average 20 years past menopause and our results are consistent with the ORDET study findings. Other risk factors for breast cancer such as insulin-like growth factor l have been shown to differ in pre- and postmenopausal women.21
Previous studies of 2- and 16α-hydroxyestrone relied on measurements of urinary metabolites. The assays for these conjugates in urine were validated against gas chromatography-mass spectroscopy methods.16,17 Urinary measurements, however, are sensitive to changes in excretory patterns and function. Concentrations of unconjugated estrogens such as estrone are very low in the systemic circulation, which probably reflects rapid conjugative metabolism (glucuronidation, sulfonation, O-methylation) followed by urinary excretion.22 In contrast, the serum assays that we used measure conjugated forms of the 2- and 16α-OHE1 metabolites, which are found at higher concentrations. We validated the accuracy of the serum assays against urinary metabolites and found excellent agreement. This is the first published study that used the serum assay for the estrogen metabolites. Our results need to be confirmed in other studies using the serum assay.
An alternative hypothesis could be that the 4-hydroxylation pathway is important. Although estradiol is primarily metabolized to 2-OHE1 or 16α-OHE1 in the liver, estradiol is also metabolized via the 4-hydroxylation pathway. The 4-hydroxylation pathway could be a dominant pathway in extrahepatic tissues, including the uterus and breast.2,22 Liehr and Ricci23 have hypothesized that the 4-hydroxylation pathway promotes breast cancer and that the ratio of 2-hydroxyestradiol to 4-hydroxyestradiol could be critical. Prospective studies are needed to test this hypothesis.
Although our prospective design avoids some of the pitfalls of case-control studies, our study has several limitations. Our cohort consists primarily of healthy community-dwelling older white women who could have more positive health behaviors such as routine mammograms. The overall rate of breast cancer was similar to that observed in white women aged 65 years of age or older in the United States.24
In conclusion, our prospective study found no evidence that women who metabolize a greater proportion of their estrogen via the 16α-hydroxylation pathway, compared with the 2-hydroxylation pathway, are at an increased risk of breast cancer.
REFERENCES
1. The Endogenous Hormones, Breast Cancer Collaborative Group. Endogenous sex hormones and breast cancer in postmenopausal women: reanalysis of nine prospective studies. J Natl Cancer Inst. 2002;94:606–616.
2. Zhu B, Conney A. Functional role of estrogen metabolism in target cells: review and perspectives. Carcinogenesis. 1998;19:1–27.
3. Bradlow H, Hershcopf R, Martucci C, et al. Estradiol 16α-hydroxylation in the mouse correlates with mammary tumor incidence and presence of murine mammary tumor virus: a possible model for the hormonal etiology of breast cancer in humans. Proc Natl Acad Sci U S A. 1985;82:6295–6299.
4. Bradlow H, Herschcopf R, Martucci C, et al. 16α-hydroxylation of estradiol: a possible risk marker for breast cancer. Ann NY Acad Sci. 1986;464:138–151.
5. Schneider J, Kinne D, Fracchia A, et al. Abnormal oxidative metabolism of estradiol in women with breast cancer. Proc Natl Acad Sci U S A. 1982;79:3047–3051.
6. Ho G, Luo X, Ji C, et al. Urinary 2/16α-hydroxyestrone ratio: correlation with serum insulin-like growth factor binding protein-3 and a potential biomarker of breast cancer risk. Ann Acad Med Singapore. 1998;27:294–299.
7. Kabat G, Chang C, Sparano J, et al. Urinary estrogen metabolites and breast cancer: a case-control study. Cancer Epidemiol Biomarkers Prev. 1997;6:1–6.
8. Zheng W, Dunning L, Jin F, et al. Urinary estrogen metabolites and breast cancer: a case control study. Cancer Epidemiol Biomarkers Prev. 1997;6:500–504.
9. Ursin G, London S, Stanczyk F, et al. Urinary 2-hydroxyestrone/16α-hydroxyestrone ratio and risk of breast cancer in postmenopausal women. J Natl Cancer Inst. 1999;91:1067–1072.
10. Osborne M, Bradlow L, Wong G, et al. Upregulation of estradiol C16a-hydroxylation in human breast tissue: a potential biomarker of breast cancer risk. J Natl Cancer Inst. 1993;85:1917–1920.
11. Meilahn E, Stavola B, Allen D, et al. Do urinary estrogen metabolites predict breast cancer? Br J Cancer. 1998;78:1250–1255.
12. Muti P, Bradlow HL, Micheli A, et al. Estrogen metabolism and risk of breast cancer: a prospective study of the 2:16α-hydroxyestrone ratio in premenopausal and postmenopausal women. Epidemiology. 2000;11:635–640.
13. Cummings SR, Nevitt MC, Browner WS, et al. Risk factors for hip fracture in white women. Study of Osteoporotic Fractures Research Group. N Engl J Med. 1995;332:767–773.
14. Cauley JA, Lucas FL, Kuller LH, et al. Elevated serum estradiol and testosterone concentrations are associated with a high risk for breast cancer. Study of Osteoporotic Fractures Research Group. Ann Intern Med. 1999;130:270–277.
15. Zmuda J, Cauley J, Ljung BM, et al. Bone mass and breast cancer risk in older women: differences by stage of diagnosis. J Natl Cancer Inst. 2001;93:930–936.
16. Klug T, Bradlow H, Sepkovic D. Monoclonal antibody-based enzyme immunoassay for simultaneous quantitation of 2- and 16α-hydroxyestrone in urine. Steroids. 1994;59:648–655.
17. Ziegler R, Rossi S, Fears T, et al. Quantifying estrogen metabolism: an evaluation of the reproducibility and validity of enzyme immunoassays for 2-hydroxyestrone and 16α-hydroxyestrone in urine. Environ Health Perspect. 1997;105(suppl 3):607–614.
18. Prentice R. A case-control design for epidemiologic cohort studies and disease prevention trials. Biometricka. 1986;73:1–11.
19. Bradlow H, Telang N, Sepkovic D, et al. 2-hydroxyestrone: the ‘good’ estrogen. J Endocrinol. 1996;150:S259–S265.
20. Fishman J, Martucci C. Biological properties of 16alpha-hydroxyestrone: implications in estrogen physiology and pathophysiology. J Clin Endocrinol Metab. 1980;51:611–615.
21. Hankinson S, Willett W, Colditz GA, et al. Circulating concentrations of insulin-like growth factor I and risk of breast cancer. Lancet. 1998;351:1393–1396.
22. Zhu B, Bui Q, Weisz J, et al. Conversion of estrone to 2- and 4-hydroxyestrone by hamster kidney and liver microsomes: implications for the mechanism of estrogen-induced carcinogenesis. Endocrinology. 1994;135:1772–1779.
23. Liehr J, Ricci M. 4-hydroxylation of estrogens as marker of human mammary tumors. Proc Natl Acad Sci U S A. 1996;93:3294–3296.
24.Ries LAG, Eisner MP, Kosary CL, et al. SEER Cancer Statistics Review, 1973–1999. Bethesda, MD: National Cancer Institute; 2002. http://seer.cancer.gov/csr/1973_1999/. Accessed July 8, 2003.
Keywords:
epidemiologic methods; genetic predisposition to disease; neoplasm; questionnaires
© 2003 Lippincott Williams & Wilkins, Inc.