Well-known risk factors of endometrial cancer are nulliparity, obesity, endogenous hyperestrogenism, hereditary nonpolyposis colon cancer syndrome, history of breast cancer, tamoxifen use, unopposed estrogen use, and inadequate progestin addition to estrogen. A large international variation in incidence rates of endometrial cancer indicates that many of those risk factors may be modifiable.
It is known that approximately 80% of the endometrial cancers are preceded with endometrial hyperplasia. These type I tumors are mostly low-grade adenocarcinomas with a fairly good prognosis. In contrast, type II tumors, such as serous papillary carcinomas, clear cell adenocarcinomas or squamous carcinomas, originate from polyps or other nonhyperplastic endometrial lesions. They are more malignant and usually diagnosed at a more advanced stage than type I tumors. Except for advanced age, no other confirmed risk factors exist for type II tumors, although estrogen may also be involved.
Previous data uniformly show that unopposed estrogen therapies or an inadequate progestin complement to estrogen is accompanied by an elevation in the risk for endometrial cancer. In these studies, progestins as part of hormone therapy (HT) have mainly been given orally, and the role of long-term use of transdermal progestin in endometrial protection is open.
Moreover, various progestins used in commercial combined HT preparations show large differences in their capacity to bind progesterone and other receptors and therefore differences in endometrial safety between various progestins as part of HT are possible. There are no long-term follow-up data on this issue so far.
Due to the large national differences in the use and content of HT its effect on endometrial cancer burden should ideally be studied in each country separately. The objective of our study was to estimate the risk of endometrial cancer in all Finnish postmenopausal women using various forms of estradiol-progestin therapy.
Taken as a whole, the use of estradiol-progestin therapy for at least 6 months was associated with a 54% elevation in the risk of endometrial cancer in the Finnish postmenopausal female population. However, only monthly and long-cycle sequential estradiol-progestin regimen use for more than 5 years were associated with significant elevation (69% and 276%, respectively) in the risk of endometrial cancer, whereas the use of continuous estradiol-progestin therapy showed a protective efficacy (approximately 70%) against endometrial cancer. Transdermal sequential estradiol-progestin therapy did not differ from oral therapy, and norethistoterone acetate, medroxyprogesterone acetate, and dydrogesterone did not differ from each other in endometrial safety.
We compared the incidence of endometrial cancer in estradiol-progestin therapy users with that in the entire same-aged female population, also including those using any type of postmenopausal hormone therapy. About 11% of postmenopausal women use estradiol-progestin therapy for more than 5 years. Such a small proportion of women with a partly elevated and partly decreased risk for endometrial cancer in the reference background population should not markedly alter the risk estimates and certainly should not invalidate the conclusions. It is acknowledged that our present study cannot control important confounding factors, such as parity, weight, or the use of oral contraceptives, but lack of major differences between HT users and nonusers in these factors in Finland is reassuring.
The purchases of estradiol-progestin therapy could be accurately traced from the national medical Reimbursement Registry from 1994 onward, but one can argue if the users actually used estradiol-progestin regimens. Because only a minor portion of the estradiol-progestin therapy price is reimbursed, it is apparent that women, spending their own money for estradiol-progestin regimens really used them. The whole history of modes, route of administrations, and exposure times of estradiol-progestin therapy were known only for women entering the registry as fresh starters at 1995, but even this group was big (n=121,036) for meaningful subgroup analyses.
It was vital that we could discontinue the follow-up of estradiol-progestin therapy users at hysterectomies. The coverage of the Finnish Cancer Registry is almost 100% but it was conspicuous that only 3% of endometrial cancers were classified as type II cancer. The lower proportion of this cancer in our study than in some other studies may also derive, at least in part, from national differences in the histological criteria of type II endometrial cancer.
Previous studies on the effect of sequential estradiol-progestin therapy have reported both elevations and no effect in the risk of endometrial cancer. In these studies, various progestins for 10-14 days have been added to conjugated equine estrogens, or estradiol, primarily each month. Our data show that monthly estradiol-progestin therapy was associated with significant risk elevations after 5 years (69%) and especially after 10 years (156%) of exposure. The opinions on the optimal duration of the monthly progestin phase in estradiol-progestin therapy are not uniform but based on our data, the progestin phases of 10-14 days each month, or for 14 days at 3 month-intervals, fail in endometrial protection. The highest risk for endometrial cancer in long-cycle estradiol-progestin therapy users in our study is in line with previous data.
Previous studies rather uniformly indicate that the use of continuous estradiol-progestin therapy is protective against endometrial cancer. The protection was 29% in a British study after 3.4 years exposure and 20% in the Women’s Initiative Study after 5.6 years exposure. In our study, the use of continuous estradiol-progestin therapy was associated with a marked risk reduction (approximately 76%) of endometrial cancer. The total accumulated doses of progestins in sequential and continuous estradiol-progestin regimens do not necessarily show drastic differences. Yet the sequential estradiol-progestin therapy increases, and continuous estradiol-progestin therapy reduces the risk of endometrial cancer. Thus, some estradiol-induced endometrial proliferations may fail to be eliminated with a subsequent progestin phase, and they may proceed to cancer. This may imply that, for truly effective protection, progestin must be present in the endometrium daily. The use of continuous combined estradiol-progestin therapy may eliminate some premalignant endometrial changes. This may further be substantiated by our findings that the risk of endometrial cancer already started to reduce during the first 3 years of continuous estradiol-progestin therapy use.
Oral and transdermal estradiol and progestins undergo different enterohepatic metabolism, and the hormonal milieu is thus different between oral and transdermal users of estradiol-progestin therapy. Therefore, the endometrial effects of oral and transdermal regimens may not be uniform. We could compare norethistoterone acetate containing sequential regimens, given orally or transdermally; they showed a similar endometrial cancer risk.
Progestins bind differently to their own and other steroid receptors, and significant variations exist in the potency of progestins to control endometrial bleedings. We present here evidence that the most commonly used progestins-norethistoterone acetate, medroxyprogesterone acetate and dydrogesterone-show no difference in endometrial safety. Endometrium and breast tissue may differ in this regard; norethistoterone acetate as part of estradiol-progestin therapy was associated with a higher risk of breast cancer than some other progestins.
Endometrial cancer in monthly sequential estradiol-progestin therapy users tended to be detected in a localized phase slightly more often than in the reference population, and this is in line with previous data. Such a difference was not seen for long-cycle or continuous estradiol-progestin therapy users. This phenomenon may be related to the different age pattern of estradiol-progestin therapy users; women using sequential monthly estradiol-progestin regimen are usually younger than those wishing to avoid bleedings or to have them seldom. It is possible that endometrial pathology in estradiol-progestin therapy users expresses itself with earlier bleeding in the younger sequential users with a still reactive endometrium. The extra bleedings lead to diagnostic tests, such as endometrial ultrasonography and biopsy. In Finland, endometrial biopsies are taken only for medical indication and not routinely in symptomless women (eg, before the initiation of estradiol-progestin therapy).
Type I cancer deriving from the reactive endometrium dominated in our series, but also the type II cancer showed a clear increasing trend in estradiol-progestin therapy users (23% rise). This is in line with data in recent studies on obese women but in contrast with the prevailing opinion.
To sum up, continuous estradiol-progestin therapy is protective against endometrial cancer, whereas the use of sequential estradiol-progestin therapy is accompanied with significant risk elevation for endometrial cancer. Various progestins or routes of administration do not differ in this regard. In absolute terms, the use of estradiol-progestin therapy by 1,000 women for 10 years would mean eight extra cases of endometrial cancer if the estradiol-progestin regimen was monthly sequential, but three to four fewer cases if the estradiol-progestin regimen was continuous. All this information should be balanced with knowledge of the progestin-associated subjective side effects and late effects on the risk of other diseases (eg, breast cancer and cardiovascular disease) related to the use of estradiol-progestin therapy.
Jaakkola, Susanna MD; Lyytinen, Heli MD; Pukkala, Eero PhD; Ylikorkala, Olavi MD
1. Amant F, Moerman P, Neven P, Timmerman D, Van Limbergen E, Vergote I. Endometrial cancer. Lancet 2005;366:491-505.
2. Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin 2005;55:74-108.
3. Kaaks R, Lukanova A, Kurzer MS. Obesity, endogenous hormones, and endometrial cancer risk: A synthetic review. Cancer Epidemiol Biomarkers Prev 2002;11:1531-43.
4. Prat J. Prognostic parameters of endometrial carcinoma. Hum Pathol 2004;35:649-62.
5. Brown L. Pathology of uterine malignancies. Clin Oncol (R Coll Radiol) 2008;20:433-47.
6. Bjorge T, Engeland A, Tretli S, Weiderpass E. Body size in relation to cancer of the uterine corpus in 1 million Norwegian women. Int J Cancer 2007;120:378-83.
7. Weiss JM, Saltzman BS, Doherty JA, Voigt LF, Chen C, Beresford SA, et al. Risk factors for the incidence of endometrial cancer according to the aggressiveness of disease. Am J Epidemiol 2006;164:56-62.
8. McCullough ML, Patel AV, Patel R, Rodriguez C, Feigelson HS, Bandera EV, et al. Body mass and endometrial cancer risk by hormone replacement therapy and cancer subtype. Cancer Epidemiol Biomarkers Prev 2008;17:73-9.
9. Lethaby A, Suckling J, Barlow D, Farquhar CM, Jepson RG, Roberts H. Hormone replacement therapy in postmenopausal women: Endometrial hyperplasia and irregular bleeding. The Cochrane Database of Systematic Reviews, Issue 3. Art No.: CD000402.
10. Stanczyk FZ. Percutaneus administration of progesterone: blood levels and endometrial protection. Menopause 2005;2:232-7.