Common variants among genes coding for enzymes in sex steroid biosynthetic pathways may influence the risk of endometrial cancer. We examined the association between endometrial cancer risk and estrogen replacement therapy (ERT) by CYP17 genotype using 51 incident cases and 391 randomly selected controls from a multiethnic cohort in Hawaii and Los Angeles, California. The relative risk of endometrial cancer was calculated for ever use versusnever use of ERT by CYP17 genotype (TT,TC, and CC). We found that women who reported ever taking ERT were more than twice as likely to develop endometrial cancer as women who never took ERT [odds ratio (OR), 2.24; 95%confidence interval (CI), 1.19–4.23]. Among these women, the risk of endometrial cancer was higher for women homozygous for the CYP17T allele (OR, 4.10; 95% CI,1.64–10.3), but not for women with the C allele (OR,1.31; 95% CI, 0.53–3.21). These preliminary findings suggest that CYP17 or other variants in estrogen biosynthesis or metabolism pathways may be potential markers of endometrial cancer susceptibility due to ERT.

Endometrial cancer risk is influenced by factors that alter serum estrogen and progesterone levels. Specifically, risk increases after exposure of endometrial tissue to estrogen that is unopposed by progesterone (1, 2). Estrogen causes proliferation of epithelial cells, and this effect can occur due to endogenous or exogenous sources of exposure. ERT,3in which estrogen is given to postmenopausal women without progestins,was first shown to increase endometrial cancer risk by epidemiological data in the United States in the 1970s. After 1976, progestins were increasingly added to ERT to reduce or possibly eliminate this risk(3). Other risk factors associated with excess endogenous production of estrogens and/or low progesterone (obesity, anovulatory cycles, and polycystic ovaries) have been repeatedly associated with increased risk of endometrial cancer (4). Consistent with these data, endometrial cancer cases have significantly higher levels of serum estrogens and lower levels of sex hormone-binding globulins than controls (5). Given the strong relationships between estrogen, progesterone, and endometrial cancer risk, we hypothesized that common variants among genes coding for enzymes in sex steroid biosynthetic pathways may increase risk of endometrial cancer.

One potential marker of endometrial cancer susceptibility is a common variant in cytochrome P450 (CYP17), a gene that codes for a key enzyme (cytochrome P450c17α) in a rate-limiting step of estrogen biosynthesis (6). A single-bp polymorphism in the 5′ untranslated region of CYP17 (27 bp downstream from the transcription start site) has been used to identify two alleles, T (formerly designated as A1) and C (formerly designated as A2). Of relevance to endometrial cancer risk, at least two studies have found that the C allele is associated with elevated levels of circulating estrogens in pre- and postmenopausal women (7, 8). Furthermore, we found that CYP17is closely associated with patterns of postmenopausal hormone replacement therapy, such that women who carry two copies of the C allele are about half as likely as women with the T allele to be current users of hormone replacement therapy(9).

In the current study, we examined the association between endometrial cancer risk and ERT (without progestins) by CYP17 genotype using cases and controls from a MEC in Hawaii and Los Angeles, California.

The participants included in these analyses were selected from a large, ongoing MEC study in Hawaii and Los Angeles initiated with emphasis on diet and lifestyle characteristics in the etiology of cancer. Details of the study have been published previously(10). Briefly, the cohort was developed beginning in 1993 from driver’s license files in Hawaii and Los Angeles. The cohort totals 215,251 men and women, ages 45–75 years at the time of enrollment, and includes primarily African-Americans, Japanese,Hawaiians, Latinos, and non-Latino whites. Baseline data were collected on all cohort members via a mailed questionnaire that contained five sections: (a) background, including medical history and family cancer history; (b) diet history; (c)medication use; (d) physical activity; and (e)female reproductive history, including the use of hormones.

This case cohort study included 51 incident cases of endometrial cancer and 391 randomly selected MEC controls who reported never taking ERT or exclusively taking estrogens without progestins(ERT). Both cases and controls were postmenopausal women ages 60 years and over at the time of blood draw from the four primary cohort racial/ethnic groups (African American, Japanese, Latina, and non-Latina white). Cases and potential controls were contacted by letter and phone call, followed by a home visit to collect a blood specimen. Blood draw was completed in the morning, typically at the person’s home, after informed consent was obtained. Participation rates for providing a blood sample on request were 74% for cancer cases and 66% for cohort controls. Case ascertainment was completed through the Surveillance, Epidemiology and End Results program in Los Angeles and Hawaii. Controls had no history of endometrial, breast, or ovarian cancer and reported no prior hysterectomy.

Genotyping.

DNA was purified from buffy coats of peripheral blood samples. The CYP17 assay has been described previously(11). A PCR fragment containing the bp change was generated using the following primers: CYP-1, 5-CATTCGCACTCTGGAGTC-3;and CYP-2, 5-AGGCTCTTGGGGTACTTG-3. PCR reactions were carried out in 25-μl aliquots containing about 50 ng of genomic DNA, 50 pmol of each primer, 1× reaction buffer, 100 μmdeoxynucleotide triphosphates, and 1 unit of Taq polymerase(Pharmacia). The amplification was performed for 30 cycles of denaturation at 94°C for 1 min, annealing at 57°C for 1 min, and extension at 72°C for 1 min. An initial denaturation step of 5 min at 94°C and a final extension at 72°C for 5 min were used. The PCR products were digested for 3 h at 37°C using MspAI,separated by agarose gel electrophoresis, and stained with ethidium bromide to identify the bp change.

Statistical Analysis.

The relative risk of endometrial cancer was calculated for ever use versus never use of ERT by CYP17 genotype(TT, TC, and CC). Logistic regression,conditional on CYP17 genotype, was used to calculate summary and stratum-specific ORs and 95% CIs controlling for age (quartiles)and racial/ethnic group as categorical variables in the model. Controlling for weight as a categorical variable (quartiles) did not change the effect estimates. Two sided Ps were calculated to test associations between ERT and endometrial cancer risk for summary and genotype-specific estimates. A one-sided test for homogeneity across strata of CYP17 was calculated to test whether women with the TT genotype were significantly more likely to develop endometrial cancer while taking ERT than women with the TC or CC genotypes.

The study sample included 51 incident cases of endometrial cancer and 391 controls. The distribution of cases and controls was as follows: African American, 41% and 33%; Japanese, 10% and 20%;Latina, 31% and 33%; non-Latina white, 18% and 14%, respectively. The difference between the mean age of cases (69.3 years) and controls(68.8 years) was not significantly different. Cases were statistically significantly heavier than controls after adjusting for age and racial/ethnic group (169 pounds versus 147 pounds; P = 0.0001).

Women who reported ever taking ERT in this study were more than twice as likely to develop endometrial cancer than women who never took ERT(Table 1). Among these women, the risk of endometrial cancer was significantly elevated for women with the TT genotype (OR, 4.10) but was only slightly elevated for women with the TC or CC genotype (OR, 1.31). Using a one-sided test for homogeneity across strata, women with the TT genotype were at significantly higher risk of developing endometrial cancer with ERT use than women with the TC or CC genotypes. The effect of the T allele in never ERT users was small (data not shown) and did not reach statistical significance after adjustment for age and racial/ethnic group (OR for the TT versus CC genotype, 1.30; 95% CI, 0.39–4.26).

The overall positive association between postmenopausal use of ERT and endometrial cancer risk found in these data has been well described in previous publications (12). In the current analysis,the risk of endometrial cancer was much higher for women homozygous for the CYP17T allele, but not for women with the C allele. We found only a small nonsignificant increase in risk for the T allele among never users of ERT. We would expect no change in endometrial cancer risk as long as any increase in estrogen is adequately opposed by progesterone. The small changes in premenopausal E1, estradiol, and possibly progesterone levels due to CYP17 may not significantly alter endometrial cancer risk as long as progesterone levels are high enough to induce epithelial shedding in each menstrual cycle.

One potential explanation of the difference in risk for endometrial cancer by CYP17 genotype for postmenopausal ERT users may be due to individual differences in estrogen binding and metabolism, such that the T allele of the T27C marker may distinguish women whose overall estrogen biosynthetic and metabolism systems are down-regulated (or up-regulated for the C allele). Based on normal, circulating estrogen levels by CYP17 genotype(7, 8), women with the T allele may bind and eliminate circulating estrogens less efficiently than women with the C allele. If the efficiency of a woman’s metabolic system to process E1 differs according to normal endogenous E1 production levels, the addition of a similar dose of postmenopausal ERT may have a greater effect on women with the T allele (who would normally transport and metabolize relatively low endogenous levels of E1) compared with women with the Cvariant. However, to accurately interpret these findings, more testing on circulating levels of endogenous estrogens is necessary to see whether estrogen levels are correlated with CYP17 genotype among women taking a standard dose of ERT.

The relationship between risk of endometrial cancer and use of estrogen and progestin replacement therapy is extremely important, given findings that estrogen and progestin replacement therapy is protective for endometrial cancer risk but may increase a woman’s risk of breast cancer (13). The risks and benefits of prescribing unopposed estrogen or estrogens in combination with progestins for women in relation to their risk for endometrial and breast cancer would be better clarified if markers of disease and individual responsiveness to exogenous hormones were available. The current findings suggest that CYP17 or other variants in the estrogen biosynthesis or metabolism pathway may be potential markers of endometrial cancer susceptibility due to ERT. The current findings are based on small numbers and need to be replicated in larger epidemiological studies. Although the functional relevance of the CYP17 gene is still unproven, these preliminary data suggest that CYP17 could be a marker of up-regulated estrogen biosynthesis and metabolism that may be of utility in studies of endometrial cancer risk.

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.

1

Supported by National Cancer Institute Grants CA63464 and CA54281.

3

The abbreviations used are: ERT, estrogen replacement therapy; OR, odds ratio; CI, confidence interval; MEC,multiethnic cohort; E1, estrone.

Table 1

Odds of endometrial cancer with use of ERT by CYP17 genotype

CYP17 genotypeERT (%)No ERT (%)ORa,b95% CIP
All women      
Cases (%) 20 (39) 31 (61)    
Controls (%) 88 (23) 303 (78) 2.24 1.19–4.23 0.013 
TT      
Cases (%) 12 (50) 12 (50)    
Controls (%) 26 (19) 113 (81) 4.10 1.64–10.27 0.003 
TC or CC      
Cases (%) 8 (30) 19 (70)    
Controls (%) 62 (25) 190 (75) 1.31 0.53–3.21 0.56 
CYP17 genotypeERT (%)No ERT (%)ORa,b95% CIP
All women      
Cases (%) 20 (39) 31 (61)    
Controls (%) 88 (23) 303 (78) 2.24 1.19–4.23 0.013 
TT      
Cases (%) 12 (50) 12 (50)    
Controls (%) 26 (19) 113 (81) 4.10 1.64–10.27 0.003 
TC or CC      
Cases (%) 8 (30) 19 (70)    
Controls (%) 62 (25) 190 (75) 1.31 0.53–3.21 0.56 
a

Adjusted for age and racial/ethnic group.

b

One-sided test for homogeneity across strata of CYP17 genotype adjusted for age and race; P = .037.

1
Henderson B. E., Ross R. K., Bernstein L. Estrogens as a cause of human cancer: the Richard and Hinda Rosenthal Foundation Award Lecture..
Cancer Res.
,
48
:
246
-253,  
1988
.
2
Henderson B. E., Ross R. K., Pike M. C., Casagrande J. T. Endogenous hormones as a major factor in human cancer..
Cancer Res.
,
42
:
3232
-3239,  
1982
.
3
McKean-Cowdin R., Feigelson H. S., Pike M. C., Ross R. K., Henderson B. E. Declining cancer rates in the 1990s..
J. Clin. Oncol.
,
18
:
2258
-2268,  
2000
.
4
Thomas D. Do hormones cause breast cancer?.
Cancer (Phila.)
,
53
:
595
-604,  
1984
.
5
Potischman N., Hoover R., Brinton L., Pentti S., Dorgan J., Swanson C., Berman M., Mortel R., Twiggs L., Barrett R., Wilbanks G., Persky V., Lurain J. Case-control study of endogenous steroid hormones and endometrial cancer..
J. Natl. Cancer Inst. (Bethesda)
,
88
:
1127
-1135,  
1996
.
6
Brentano S. T., Picado-Leonard J., Mellon S. H., Moore C. C., Miller W. L. Tissue-specific, cyclic adenosine 3′,5′-monophosphate-induced, and phorbol ester-repressed transcription from the human P450c17 promoter in mouse cells..
Mol. Endocrinol.
,
4
:
1972
-1979,  
1990
.
7
Feigelson H. S., Shames L. S., Pike M. C., Coetzee G. A., Stanczyk F. Z., Henderson B. E. Cytochrome P450c17a gene (CYP17) polymorphism is associated with serum estrogen and progesterone concentrations..
Cancer Res.
,
58
:
585
-587,  
1998
.
8
Haiman C. A., Hankinson S. E., Spiegelman D., Colditz G. A., Willett W. C., Speizer F. E., Kelsey K. T., Hunter D. J. The relationship between a polymorphism in CYP17 with plasma hormone levels and breast cancer..
Cancer Res.
,
59
:
1015
-1020,  
1999
.
9
Feigelson H. S., McKean-Cowdin R., Pike M. C., Coetzee G. A., Kolonel L. N., Nomura A. M., Le Marchand L., Henderson B. E. Cytochrome P450c17a gene (CYP17) polymorphism predicts use of hormone replacement therapy..
Cancer Res.
,
59
:
3908
-3910,  
1999
.
10
Kolonel L. N., Henderson B. E., Hankin J. H., Nomura A. M., Wilkens L. R., Pike M. C., Stram D. O., Monroe K. R., Earle M. E., Nagamine F. S. A multiethnic cohort in Hawaii and Los Angeles: baseline characteristics..
Am. J. Epidemiol.
,
151
:
346
-357,  
2000
.
11
Carey A. H., Waterworth D., Patel K., White D., Little J., Novelli P., Franks S., Williamson R. Polycystic ovaries and premature male pattern baldness are associated with one allele of the steroid metabolism gene CYP17..
Hum. Mol. Genet.
,
3
:
1873
-1876,  
1994
.
12
Pike M. C., Peters R. K., Cozen W., Probst-Hensch N. M., Felix J. C., Wan P. C., Mack T. M. Estrogen-progestin replacement therapy and endometrial cancer..
J. Natl. Cancer Inst. (Bethesda)
,
89
:
1110
-1116,  
1997
.
13
Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormone replacement therapy: collaborative reanalysis of data from 51 epidemiological studies of 52,705 women with breast cancer and 108,411 women without breast cancer. Lancet, 350: 1047–1059, 1997.