Get this document via a secure connection
The NCI will no longer support CancerMail after November 2002. If you have comments about the NCI's cancer information delivery services, contact us by e-mail at cancer.govstaff@mail.nih.gov or call 301-496-9096.
###########################################################################
Levels of Evidence for preceding statement:
1aii: Evidence obtained from at least one well-designed and conducted randomized controlled trial that has a cancer incidence endpoint
2aii: Evidence obtained from well-designed and conducted nonrandomized controlled trials that have a cancer incidence endpoint
3aii: Evidence obtained from well-designed and conducted cohort or case-control analytic studies, preferably from more than one center or research group, that have a cancer incidence endpoint
5: Opinions of respected authorities based on clinical experience or reports of expert committees
Levels of Evidence for preceding statement:
3aii: Evidence obtained from well-designed and conducted cohort or case-control analytic studies, preferably from more than one center or research group, that have a cancer incidence endpoint
5: Opinions of respected authorities based on clinical experience or reports of expert committees
Levels of Evidence for preceding statement:
1aii: Evidence obtained from at least one well-designed and conducted randomized controlled trial that has a cancer incidence endpoint
3aii: Evidence obtained from well-designed and conducted cohort or case-control analytic studies, preferably from more than one center or research group, that have a cancer incidence endpoint
5: Opinions of respected authorities based on clinical experience or reports of expert committees
There are several other potentially modifiable factors which have been associated with increased risk of endometrial cancer including obesity, diabetes, hypertension, and a diet high in fat. Breast feeding and physical activity are associated with decreased risk of endometrial cancer.
Level of evidence for preceding statement:
3aii: Evidence obtained from well-designed and conducted cohort or case-control analytic studies, preferably from more than one center or research group, that have a cancer incidence endpoint
In the mid 1970s, the diagnosis of approximately 15,000 cases of postmenopausal endometrial cancers in excess of those expected on the basis of the underlying secular trend, has been related to the use of exogenous estrogen therapy.[3] In addition to the risk of developing endometrial cancer in association with the use of estrogen replacement therapy unaccompanied by progesterone, a number of additional risk factors have been identified and often appear to be related to estrogenic effects. Among these factors are obesity, a high-fat diet, reproductive factors like nulliparity, early menarche and late menopause, and tamoxifen use.
Women with hereditary nonpolyposis colorectal cancer (HNPCC) syndrome have a markedly increased risk of endometrial cancer compared to women in the general population. Among women who are HNPCC mutation carriers, the estimated cumulative incidence of endometrial cancer ranges from 20% to 60%.[4-6]
Factors that have been associated with a decreased incidence of endometrial cancer include parity, lactation, use of combined oral contraceptives, a diet low in fat and high in plant foods, and physical activity. Accumulating information about factors that may contribute to an increase in endometrial cancer risk indicates the target populations where preventive interventions are most needed. In some cases the etiologic evidence has suggested strategies that may be pursued to reduce endometrial cancer risk.
There are a limited number of randomized controlled trials. Some randomized controlled data were provided by the Postmenopausal Estrogen Progestin Interventions (PEPI) Trial.[16] There were nearly 600 participants in PEPI who took part in a 3-year, multicenter, randomized trial that was double-masked and placebo-controlled. In addition to the placebo arm of the trial, 1 group received only conjugated equine estrogen (CEE) daily, while 3 other groups received CEE with progestin (medroxyprogesterone acetate (MPA) or micronized progesterone (MP) for the first 12 days of every 28-day cycle, or MP continuously). Endometrial biopsies were obtained at baseline and annually, or at unscheduled visits when clinically indicated. For the women who received progestin with CEE, the occurrence of uterine hyperplasia was similar to that seen with placebo; however, women given estrogen alone had more endometrial tissue abnormalities than women taking placebo (simple hyperplasia: 27.7% versus 0.8%, adenomatous hyperplasia: 22.7% versus 0.8%, and atypical hyperplasia: 11.8% versus 0%, respectively). These data showed that CEE at a dose of 0.625 mg daily was associated with the development of endometrial hyperplasia, which could be prevented by adding MPA or MP to the treatment regimen.
The acceptance of progestin as an intervention to prevent CEE-associated endometrial cancer depends on the role of the endometrial hyperplasia as a precursor of endometrial cancer. A retrospective analysis of 170 women with endometrial hyperplasia who were followed for a mean of 13.4 years demonstrated that the risk of progression to endometrial carcinoma is dependent of the type of hyperplasia. Only 2% of women with endometrial hyperplasia without atypia developed carcinoma. By contrast, 23% of those with atypical hyperplasia developed carcinoma.[17] The vast majority of these tumors were low grade and minimally invasive. Therefore, endometrial hyperplasia with cytologic atypia is considered a precursor of certain types of well-differentiated endometrial carcinomas.[18]
A large population-based case-control study conducted in Los Angeles County concluded that administration of progestin in sequential estrogen-progestin replacement therapy should continue for at least 10 days per month in order to effectively nullify the increased risk of endometrial cancer associated with estrogen replacement therapy. For women who received estrogen replacement therapy, the adjusted odds ratio (OR) was 2.17 (95% confidence interval (CI)= 1.91-2.47) per 5 years of ERT use. Women who received sequential estrogen-progestin replacement therapy with progestin given less than 10 days per month had only a slightly lower risk, OR= 1.87 (95% CI= 1.32-2.65) per 5 years of use. In contrast, when progestin was given for more than 10 days per month, there was no associated increase in risk, OR= 1.07 (95% CI= 0.82-1.41) per 5 years of use.[19] Another population-based case-control study confirmed the decreased risk of endometrial cancer for women who received combination estrogen-progestin replacement therapy with progestin given for more than 10 days per month. However, the results suggest that this protective relationship is not sustained for long-term users. For women who received fewer than 10 days of progestin per month for 5 years or more, the risk was 3.7 fold higher (95% CI= 1.7-8.2). For women using progestin for more than 10 days per month for a period of 5 years or more, the risk was 2.5 fold higher (95% CI= 1.1-5.5).[20]
A prospective cohort study was conducted among Swedish women who had received prescriptions for replacement hormone therapies. For women who reported using medium-potency unopposed estrogens for 6 years or more, the relative risk (RR) of invasive endometrial cancer was 4.2 (95% CI= 2.5-8.4) while the risk for women using a progestin-combined treatment for the same length of time was not significantly elevated, RR= 1.4 (95% CI= 0.6-3.3).[21]
A meta-analysis examining the risk of endometrial cancer in relation to combined oral contraceptive use, included 10 case-control studies and 1 prospective study. Among the studies reporting duration of use (10 of 11), 7 case-control studies demonstrated a decrease in risk estimates with increasing duration of COC use. Overall, 4 years of COC use was associated with a reduced risk of endometrial cancer of approximately 56%, 8 years with a reduced risk of 67% and 12 years of use of approximately 72%. Of note, the prospective study did not show a dose-response, however, the relative risk was consistently reduced by 80% after 9 years of follow-up.[25]
A population-based case-control study in Sweden reported a 30% decreased risk (OR= 0.7, 95% CI= 0.5-0.9) of endometrial cancer among women who had used any type of oral contraceptive; progestin-only pills were associated with a further reduction in risk by approximately 60% (OR= 0.4, 95% CI= 0.2-1.4). For women who reported 3 or more years of use of COCs, the OR was 0.5 (95% CI= 0.3-0.7) and continued to decline to OR= 0.2 (95% CI= 0.1-0.4) for women reporting 10 or more years of use. The risk of endometrial cancer decreased by 10% per year of COC utilization and was observed for atypical hyperplasias as well as for all grades of invasive tumors.[26]
A report was published in 1989 concerning second primary cancers observed in patients treated as part of a randomized Swedish study of adjuvant breast cancer treatment with 40 mg of tamoxifen daily for 2 or 5 years versus no endocrine treatment.[2] With a patient population of 1,846 and a median follow-up of 4.5 years, there was an increase in uterine cancer cases among the 931 tamoxifen-treated women, consistent with a relative risk (RR) of 6.4 (p< .01). In a Danish trial, women identified at greater risk of breast cancer recurrence on the basis of tumor-bearing axillary nodes or tumors greater than 5 cm were randomized to be treated adjuvantly with either chest wall radiation or radiation and 48 weeks of tamoxifen at a dose of 30 mg daily after undergoing modified radical mastectomy.[4] The women who received tamoxifen were compared to those who did not and also to a lower risk group who were not considered for tamoxifen therapy. A statistically significant difference was not demonstrated, but the cumulative incidence of endometrial cancer was 1% in the tamoxifen treated group compared to 0.3% among patients who did not receive tamoxifen. A meta-analysis of Scandinavian trials including the Swedish and Danish studies confirmed a 6-fold increase in endometrial cancer with a relative risk of 4.1 at a median follow-up between 8 and 9 years for 4,914 patients.[3] In the NSABP B-14 trial comparing tamoxifen to placebo as adjuvant therapy for women with node negative estrogen receptor positive breast cancer, in the group of approximately 1,400 patients randomized to receive tamoxifen, the hazard of developing endometrial cancer was 1.6/1,000/year.[7] The relative risk within the trial was 7.5, but the investigators were suspicious of an endometrial cancer detection bias since the rate of 0.2/1,000 in the placebo group was much lower than expected on the basis of population-based data from SEER. The endometrial tumors observed in the B-14 study did not have a different distribution of histologies or prognoses compared to endometrial tumors in nontamoxifen treated patients. An overview of the frequency of occurrence of endometrial cancer in tamoxifen-treated women was presented at the Food and Drug Administration's Oncologic Drug Advisory Committee meeting in June, 1994.[8] The NSABP Breast Cancer Prevention Trial P-1 Study confirmed an increased incidence of endometrial cancer in women at high risk for invasive breast cancer, who received tamoxifen, as compared to women who received placebo. The average annual rate of endometrial cancer was 2.3 per 1,000 among women who received tamoxifen and 0.91 per 1,000 in the placebo group. These data represent a 2.53 greater risk for women taking chemopreventive tamoxifen (95% confidence interval (CI)= 1.35-4.97). The increase in risk differed according to menopausal status; for women aged 49 years or younger RR was 1.21 (95% CI= 0.41-3.60) compared to 4.01 (95% CI= 1.70-10.90) for women aged 50 years of age and older. All of the invasive endometrial cancer cases which occurred among women taking tamoxifen were International Federation of Gynecology and Obstetrics (FIGO) stage I. Similarly, 14 out of 15 (93%) of the invasive endometrial cancer cases diagnosed among women taking placebo were FIGO stage I.[9]
Average Annual Rates of Invasive and In Situ Endometrial Cancer
------------------------------------------------------------------------
| Placebo | Tamoxifen | | |
| | | | |
-----------|------|----------|-------|----------|------|---------------|
Type of |No. of| Rate per | No. of| Rate per | Risk | 95% Confidence|
Event |Events| 1,000 | Events| 1,000 | Ratio| Interval |
| | Women | | Women | | |
-----------|------|----------|-------|----------|------|---------------|
Invasive | 15 | 0.91 | 36 | 2.30 | 2.53 | 1.35-4.97 |
Cancer | | | | | | |
-----------|------|----------|-------|----------|------|---------------|
Age | | | | | | |
<= 49 | 8 | 1.09 | 9 | 1.32 | 1.21 | 0.41-3.60 |
-----------|------|----------|-------|----------|------|---------------|
>= 50 | 7 | 0.76 | 27 | 3.05 | 4.01 | 1.70-10.90 |
-----------|------|----------|-------|----------|------|---------------|
In Situ | 3 | 0.18 | 1 | 0.06 | 0.35 | 0.01-4.38 |
Cancer | | | | | | |
------------------------------------------------------------------------
Table adapted from Fisher et al. Tamoxifen for Prevention of Breast Cancer:
Report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study.
Journal of the National Cancer Institute 90(18): 1371-1388, 1998.
A prospective study was conducted in France to assess the risk of relapse and mortality in women with breast cancer as a function of tamoxifen treatment duration. Patients were randomized to short-term (n= 1,882) treatment which entailed 3 years of tamoxifen or long-term (n= 1,911) treatment which was defined as 10 years of tamoxifen administration past randomization. The results indicate that there was virtually no difference in the incidence of endometrial cancer between short- and long-term tamoxifen users.[10]
There have been several case-control studies looking at the level of risk for developing endometrial cancer after treatment with tamoxifen for breast cancer. One study looked at data for 98 patients who developed endometrial cancer subsequent to a diagnosis of breast cancer and reported that women who had used tamoxifen for more than 2 years had a relative risk of 2.3 (95% CI= 0.9-5.9).[5] In this study, there were statistically significant trends according to duration of use and cumulative tamoxifen dose. A population-based study of endometrial cancer subsequent to a diagnosis of breast cancer based on 42 such cases confirmed that there was relatively low risk for durations of tamoxifen use less than 2 years, with an odds ratio of 0.6 (95% CI= 0.2-1.9).[6]
The question of whether tamoxifen is associated with high-grade lesions of the endometrium was raised by investigators at Yale University on the basis of a hospital-based case series of 53 breast cancer patients who subsequently developed endometrial cancer.[11] Of these 53 patients, 15 had received tamoxifen (40 mg per day). Of the women receiving tamoxifen, 67% had poorly differentiated endometrioid carcinomas or carcinomas associated with a poor outcome (papillary serous, clear cell, or mixed mullerian). Results from this series of patients may not be generalizable, since they were seen at a referral hospital and had also received a higher dose of tamoxifen than usually prescribed in the United States.
A Dutch study of 309 breast cancer patients who developed endometrial cancer compared with 860 age- and time-matched controls showed increased endometrial cancer risk associated with tamoxifen use, and an association with duration of use. Most of these women were treated with 30 or 40 mg daily. Tamoxifen was also associated with an increased risk for higher stage cancers, poor-prognosis histology, and endometrial-cancer-specific mortality.[12]
A case series from M.D. Anderson Cancer Center involved 72 patients with endometrial cancer following a diagnosis of breast cancer.[13] Of these patients, 15 had received tamoxifen and 5 of them had papillary serous histology. These results contrast with the results from randomized clinical trials and other observations. As with the NSABP B-14 trial, a clinicopathologic description of 17 endometrial cancer cases in the Swedish trial indicated that most were not high grade, and the 10-year actuarial survival after post-tamoxifen endometrial cancer was 73%.[14] Of 16 tumors that were evaluable according to the WHO grading system, only 1 patient had advanced disease (stage IV).
Additional experience with the histologic evaluation of endometrial cancer after tamoxifen use has been provided by patients from the Memorial Sloan-Kettering Cancer Center.[15] Risk from tamoxifen-associated uterine cancer in 23 patients at this institution (66% stage I, 74% endometrioid), was similar to that for 50 breast cancer patients who developed endometrial cancer, but had not used tamoxifen. Preliminary results from a laboratory evaluation of tumor samples suggests that there may be a higher incidence of abnormal p53 expression in the tamoxifen-treated patients.[16] A review of the combined experience from 6 studies with 118 patients with tamoxifen-associated endometrial cancer reported that 13% were stage IV or unstaged, 33% had high-risk histology, and 15% died from uterine cancer.[15]
Having comprehensively examined the issue and observed a modest increase in risk of developing endometrial cancer after treatment with tamoxifen, cancer specialists have concluded that the benefits to breast cancer patients, in suppression of breast cancer recurrence, substantially outweigh the risks. Remaining are questions that involve the level of endometrial surveillance needed for breast cancer patients who are using or have used tamoxifen. Also in question is the potential for preventive interventions to obviate endometrial cancer risk. These issues are now assuming great importance, since tamoxifen is being used in breast cancer prevention trials. From 1 of the breast cancer prevention trials using tamoxifen, a randomized cohort of 111 postmenopausal women was recruited for endometrial sampling. In this cohort, endometrial hyperplasia and polyps occurred more frequently in the tamoxifen-treated participants (61) than in the placebo group (50). Sixteen percent of the tamoxifen users (10 patients) had atypical hyperplasia and 8% (5 patients) had a polyp. This compared to no atypical hyperplasia in the control group and a single patient with a polyp. As more basic scientific data become available, the chance of eliminating or controlling tamoxifen-associated endometrial cancer may be improved. Clinical trials are being used to investigate the possibility of using progestin to prevent the development of uterine tissue abnormalities in women taking tamoxifen. Results from a randomized controlled trial in the UK indicate that the levonorgestrel- releasing intrauterine system was successful in preventing the development of endometrial polyps in women who had received at least one year of tamoxifen treatment, as compared to control participants without the device (p= 0.05).[17]
Tamoxifen is a member of a group of drugs known as SERMs (selective estrogen receptor modulators) and recognized for divergent effects in different target organs. Evidence of divergent effects by tamoxifen has been provided with experiments involving implants of estrogen-dependent cells in nude mice. In contradistinction to the tamoxifen-stimulated growth observed for endometrial tumor cell implants, simultaneous growth of estrogen receptor positive (ER+) breast cancer cell implants was inhibited.[18] Laboratory evidence of an estrogen-like effect of tamoxifen on the endometrium is also demonstrated by the induction of c-myc and c-fos expression by tamoxifen in endometrial tissue. Progesterone receptors in the endometrium are upregulated by tamoxifen, with decidual changes occurring after administration of progesterone.
The understanding of estrogen-like effects in the uterus has been advanced by the laboratory investigation of estrogen receptor (ER) ligands, e.g., 17B-estradiol and SERMs, that participate in cell regulation.[19] In the nucleus, the ER/estradiol complex binds to a response element as a prelude to gene transcription. In the nuclei of endometrial cells, ER complexed with either 17B-estradiol or tamoxifen reacts with AP1 sites, mediators of estrogen response. With tamoxifen, an agonistic effect through binding at AP1 occurs in uterine, but not in breast cell lines.[20] Affinity of the ER/tamoxifen complex in the nuclei of uterine cells for AP1 may explain why tamoxifen has an estrogenic effect in this particular tissue. Through its interaction with AP1, the tamoxifen/ER complex is thought to increase the transcriptional efficiency of jun and fos. Other SERMs may differ from tamoxifen in their effect on the uterus. It is reported that raloxifene, a second generation SERM approved for prophylaxis against postmenopausal osteoporosis, does not have an estrogenic effect on the uterus and preliminary evidence indicates that the ER/raloxifene complex does not bind to AP1.[21] In 1999, results from the Multiple Outcomes of Raloxifene (MORE) Randomized Trial were published, indicating that raloxifene reduced the risk of estrogen receptor-positive breast cancer. Unlike tamoxifen, raloxifene did not appear to increase the risk of developing endometrial cancer (RR= 0.8, 95% CI= 0.2-2.7) after 40 months of follow-up.[22] The ideal SERM has yet to be developed, but it appears that uterine cancer may be preventable by selecting a SERM that is not uterotrophic or by following the PEPI model of demonstrating a uterine protective role for progestins in conjunction with SERMs that stimulate the uterus.
Presumably, body weight is a modifiable risk factor, which accounts for a substantial proportion of endometrial cases worldwide. A study conducted among European countries estimated that between 26% and 47% of endometrial cancer cases can be attributed to overweight and obesity. The same group conducted a meta-analysis of studies which examined the relationship between obesity and endometrial cancer. Eleven out of the 12 (5 cohort and 7 case-control) studies reviewed concluded that there is a positive association between endometrial cancer and excess weight.[5]
Relative risks associated with obesity range from 2 to 10.[1] Some studies have concluded that upper-body and central weight confer a higher risk than peripheral body weight, even after consideration of BMI.[1,6-8] However, other studies have failed to confirm such an association. Several studies have observed a stronger association between endometrial cancer and obesity close to the time of diagnosis compared to obesity earlier in life.[9-12] Although the role that obesity plays in the underlying etiology of endometrial cancer is not fully understood, several mechanisms have been proposed. According to the estrogen theory, postmenopausal women accumulate most of their plasma estrogen from the aromatization of androstenedione to estrone and estradiol, which occurs in adipose tissue.[2] This may explain the strength of the relationship observed between weight-gain late in life and endometrial cancer. Alternatively, the association between central adiposity and endometrial cancer may be due to underlying hyperinsulinemia; this would negate obesity as a causal factor in endometrial cancer.[13-15]
The first prospective investigation of endogenous estrogens and the risk of endometrial cancer has been conducted. This case-control study was nested within the New York University Women's Health Study.[16] Results suggest an increased risk of endometrial cancer associated with postmenopausal levels of endogenous hormones including estradiol, percent free estradiol, and estrone. Conversely, risk was decreased with higher levels of percent sex-hormone-binding globulin (SHBG)-bound estradiol and SHBG. Analyses conducted prior to adjustment for hormone levels indicated a positive association with BMI. After adjustment for estrone level, the positive association of BMI with risk of endometrial cancer was attenuated, suggesting that hormone levels may be an intermediate effect of body weight.
|
Browse and Submit Oncology Conferences |
|
Sponsors:
|
small video clip about our work (200 Kbit/s)
|
| Back to the Cancernet
contents overview Questions? Mail them to us! |
We subscribe to the HONcode
principlesof the Health On the Net Foundation
|