Postmenopausal Hormone Therapy and Breast Cancer: French Cohort Study
Postmenopausal Hormone Therapy and Breast Cancer: French Cohort Study
Abstract & Commentary
By Leon Speroff, MD, Editor, Professor of Obstetrics and Gynecology, Oregon Health and Science University, Portland, is Editor for OB/GYN Clinical Alert.
Synopsis: When combined with synthetic progestins, even short-term use of estrogens may increase breast cancer risk. Micronized progesterone may be preferred to synthetic progestins in short-term HRT. This finding needs further investigation.
Source: Fournier A, et al. Breast cancer risk in relation to different types of hormone replacement therapy in the E3N-EPIC cohort. Int J Cancer. 2005;114:448-454.
E3N is a French prospective study of cancer risk factors—Etude Epidémiologique de femmes de la Mutuelle Générale de l’Eduction National. The study population, after exclusions, consists of 54,548 women born between 1925 and 1950, and all belong to a health insurance program that primarily covers teachers. The data are derived from self-administered questionnaires. Hormonal treatments were identified by a booklet of photographs, but doses were not requested. The average age at the beginning of the study was 52.8, ranging from 40.0 to 66.1 years. Follow-up averaged 5.8 years, but the range was wide: 0.1 to 10.6 years. During this follow-up, 948 cases of primary invasive breast cancer were identified. The overall relative risk of breast cancer comparing users to nonusers was elevated, 1.2 (CI = 1.1-1.4).
| Cases | Adjusted Relative Risk | |
| Estrogen alone |
30 | 1.1 (0.8-1.6) |
| Estrogens and micronized progesterone |
55 | 0.9 (0.7-1.2) |
| Estrogens and “synthetic progestins” |
268 | 1.4 (1.2-1.7) |
| Nonoral |
187 | 1.4 (1.2-1.7) |
| Oral |
80 | 1.5 (1.1-1.9) |
Conclusions: Giving estrogen by oral or nonoral routes made no difference when combined with a synthetic progestin. There was no increased risk associated with the use of estrogen alone or the use of progesterone.
Fournier and colleagues concluded that nonoral estrogen increased the risk of breast cancer only when combined with synthetic progestins.
| Use less than 2 years |
2-4 years | 4 or more years |
| Any use | ||
| 185 cases 1.2 (1,0-1,5) |
115 cases 1.2 (1.0-1.5) | 72 cases 1.2 (0.9-1.6) |
| Oral E and synthetic progestins | ||
| 36 cases 1.2 (0.9-1.8) |
27 cases 1.6 (1.1-2.3) | 17 cases 1.9 (1.2-3.2) |
| Non-oral E and progesterone | ||
| 26 cases 0.9 (0.6-1.4) |
13 cases 0.7 (0.4-1.2) | 16 cases 1.2 (0.7-2.0) |
| Non-oral E and Synthetic progestins | ||
| 95 cases 1.6 (1.3-2.0) |
57 cases 1.4 (1.0-1.8) | 35 cases 1.2 (0.8-1.7) |
Commentary
In contrast to almost every other country in the world, oral hormone therapy is not popular in France—55% of the women in the study used percutaneous gels and 45% used transdermal patches. The progestational agents used were mainly micronized progesterone or progesterone derivatives (medroxyprogesterone acetate, cyproterone acetate, retroprogesterone, nomegestrol acetate, or promegestone). Hardly anyone used conjugated equine estrogens, and only 7.6% used progestins from the 19-nortestosterone family (norethindrone acetate or lynestrenol—labeled as "synthetic progestins)."
Fournier et al’s main conclusion is that the association between hormone therapy and the risk of breast cancer varies according to the type of progestin used and not with the route of administration of the estrogen. More specifically, that no increase was observed with micronized progesterone.
The hormone users (29,420 women) differed considerably compared with the non-users (25,128): earlier menarche, earlier menopause, less nulliparity, more children born before age 30, more benign breast disease, less obesity, more education, and a greater previous use of oral contraceptives and progestins. Fournier et al stated that interactions with the characteristics that differed in the 2 groups of women did not change their findings—unfortunately, these manipulations are not presented, because it is surprising that a greater prevalence of recognized risk factors for breast cancer in the hormone user group would not influence the results, and it leaves a clinician uncertain that multiple mathematical adjustments can compensate for these differences.
A major problem with this study is the rapidity at which cases of breast cancer were identified. Fournier et al even calculated a relative risk associated with less than one year of exposure of nonoral estrogens combined with synthetic progestins and amazingly found an increase that was statistically significant: 1.7 (1.3-2.3). I don’t think anyone, epidemiologists or clinicians, can believe that hormone therapy causes new cancers to grow in less than one year’s time.
Fournier et al make the harmful statement that the "carcinogenic effect of the CEE plus MPA association in continuous administration was proved by the WHI trial and recent observational studies in the USA." The use of the word carcinogenic indicates a causal effect of hormone therapy and this is still not certain—an impact on preexisting tumors remains a very real possibility.
Do these data truly indicate a difference among various progestational agents? Perhaps, but there is an important unanswered question. To differentiate among the various agents one would have to be certain that the doses administered represented bioequivalent doses in terms of target tissue impact. Comparing micronized progesterone to progesterone and 19-nortestosterone derivatives must account for individual variability in absorption and metabolism. This would be a formidable task.
There is another possible interpretation of this report. The very rapid appearance of breast cancer cases is the most noteworthy feature of this study. Is it possible that the combination of estrogen with specific progestins leads to earlier detection of preexisting tumors, whereas lesser potent doses of progestins or qualitative differences among progestins are either lacking in this impact or it takes longer to emerge. It was surprising to me that Fournier et al didn’t consider the issue of early detection in their discussion.
Fournier et al divided their hormone users into two groups: incident users who began treatment after and prevalent users who began treatment before the year preceding the beginning of the study. The risk of breast cancer was not increased in the prevalent users, and Fournier et al labeled this group as a less susceptible group of women. Is it possible that early use in the prevalent group led to detection and exclusion from the study?
A cohort study from the Kaiser health program in Southern California has reported an adjusted 31% reduction in all cause and a 48% reduction in breast cancer mortality in users of combined estrogen-progestin therapy who developed Stage I disease.1 In women with Stage II breast cancer, the all-cause reduction was 47% and the breast cancer mortality reduction of 31%. No reduction was observed in users of estrogen only. Nonhormone users presented with larger tumors that were more likely poorly differentiated. Is it possible that this beneficial effect requires exposure to estrogen and progestin, and is it possible that the response is affected by the specific progestin used?
It seems to me that this study raises many questions, and cannot be used to support a differential impact of different progestational agents. But it does support the possibility that hormone therapy affects preexisting tumors and results in earlier detection.
Reference
- Chen W, et al. Mortality following development of breast cancer while using oestrogen or oestrogen plus progestin: a computer record-linkage study. Br J Cancer. 2005;93:392-398.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.