Contemporary Hormonal Contraception and the Risk of Breast Cancer

Last week a story broke in the News media about an article that appeared in the New England Journal of Medicine entitled, “Contemporary Hormonal Contraception and the Risk of Breast Cancer.” I received a number of queries about this news story and told patients that I needed to see the original article itself and spend a good deal of time dissecting the “fine print.” It is a very long and complicated article but let me make the following observations:

• It was a study done on a registry of Danish women between 15 and 49 years of age. Denmark has a population of 5.3 million people, is relatively homogeneous, and has very good national health registries and statistics so that this kind of database mining is relatively easy. However, the population is not very heterogeneous nor very diverse. Many of these families have lived in the same village for many generations. One could facetiously say that if both your parents are Danish that the data derived from a Danish population would be more likely to apply to you and your genetic background.
• Overall, the study claimed a 20% increase in the risk of breast cancer in patients who had used any form of hormonal contraception including the newer levonorgestrel releasing IUDs (like Mirena). Levonorgestrel is a particular type of synthetic progesterone also known as a progestin.
• The New York Times article goes into great detail to try to make readers understand the difference between a relative risk and an absolute risk. To many people, a 20% increase in risk sounds formidable, although, if a risk to begin with is relatively small, an increase of 20% is still a very small risk. Overall, the data presented here would seem to indicate 1 extra breast cancer for every 8,000 women.
• There are some interesting points that they do not make in the article itself but are present in the Tables of the data. The percent of smokers for various groupings ranged from a low of 20% to a high in some groups of 60%!! This is certainly not typical of my patient population, whatsoever. In addition, there was no description of the amount of alcohol consumption, also a known risk factor for breast cancer, which is greater in the Danish population than most of my patients.
• One of the things that was most interesting and yet something I have known for quite some time was that the use of the levonorgestrel (synthetic progestin) releasing IUD resulted in a 21% statistically significant increase in breast cancer. The makers of that devise would have one believe that this only acts locally in the uterus but many of us have long know that there, in fact, is systemic absorption of the progestin.
• Along those lines, it is interesting that the risk of a 50-microgram estrogen pill was similar to the risks of those pills that contain 20-40 micrograms of estrogen, if they employed the same progestin that they were paired with. In other words, this is very strong data pointing to it being the type of progesterone, not the estrogen dose that is most important for breast cancer risk. I have told many of you that the problem in the Women’s Health Initiative was not the estrogen employed (Premarin) but the progesterone (Provera) that it was matched with. This information from this article is very important in that many patients assume a lower dose of hormone would, in fact, be safer. This study would indicate that the choice of the progestin may be much more important that the dose of the estrogen. I’ve also told many of you that the lowest dose pills are too low for women under 30-35 years of age because they do not result in enough bone growth at a time when bone is growing geometrically. Thus, this data is reassuring that low dose pills need not be resorted to for women under 30-35 years of age, but rather, perhaps we should be thinking about which progesterones are being employed.
• As I have explained to many of you, virtually all pills contain the same type of estrogen. It is called ethinylestradiol. What differs in various pills is the synthetic progesterone (progestin) that the estrogen is paired with. More than half the women in this Danish study were taking pills whose progestin was one called gestodene. This is not marketed in any United States birth control pills. I find this reassuring in that, as mentioned, it accounted for more than half of the women in this entire study.
• Norethindrone as in the Lo Estrin family and its generics had no statistically significant increase in breast cancer. Drospirenone as in Yaz and Yasmin and its generics had no significant increase in breast cancer. Norgestimate as in the Ortho-Cyclen family and its generics did show a small increase as did levonorgestrel present in Seasonale, Nordette and its generics as well as the Mirena IUD did show a small increase in breast cancer. This amounted to approximately 1 case in every 5,600 women. It is important to note, however, that progestins like Norgestimate have other secondary benefits such as excellent control of acne.
• There was no information on Demulen and its generics (Kelnor, Zovia) for those formulations are not available in Europe.
• Furthermore, there are well known reductions in cancers of the ovary, uterus and colon in women who use birth control pills later in life. Thus, any small increase in breast cancer may well be totally offset.
• For those of my patients having good, annual radiologic surveillance of their breasts, I have seen no decrease in breast cancer over the last two decades but I have seen virtually no deaths in my practice. Many of you have heard me tell that I never want to “feel breast cancer again.” I expect the breast imagers to pick up early stage disease four to five years before anyone could possibly palpate it. Thus, currently, although, I see lots of breast cancer in hormone users as well as non-hormone users, virtually all that I see in my practice is Stage 1 disease where a simple lumpectomy and a single sentinel lymph node are all that are required and virtually all these patients survive with virtually no death rate.

In summary, I still strongly believe that the short and long term benefits of hormonal cycle control far outweigh any potential risks. In some cases, this data may cause me to choose some formulations which appear to have less risk that other formulations. This appears to be related to the progestin in the pill rather than the dose of estrogen in the pill. I encourage everyone not to smoke, moderate your alcohol intake and continue your exercise program. Clearly, there are more deaths from the long- and short-term sequelae of hip fracture in patients as they age than breast cancers.

As usual the devil is in the details and there was more to this paper than simply what the media reports. If you have any questions or concerns, do not hesitate to contact the office.

Dr Steven R. Goldstein is a Perimenopause Specialist in NYC