A new study has been published suggesting a link between breast cancer and the Mirena IUS (intrauterine system). The Mirena IUS is used commonly in many European countries and is increasing in use in the United States (with good reason, it is highly effective and has a very high satisfaction), so I think a review of the study and the data on any connection between the Mirena and breast cancer is deserved before any connection becomes urban myth. This is especially important given all the great disservice done by the Supreme Court in the Hobby Lobby case as the Mirena IUS does not cause abortions or affect a fertilized egg.
The short answer: The worst case scenario is there may be a 19% increased risk of breast cancer for women who had a Mirena IUS inserted between the ages of 30 and 49. In reality this study only suggests a correlation and provides no proof whatsoever of causation. There are many reason women with a Mirena IUS could have a higher risk of breast cancer when compared with all other women and be completely unrelated to the hormone in the IUD. The most obvious reason is that
women who have an IUD may have other significant risk factors for breast cancer not accounted for in the study such as: fewer pregnancies, later first pregnancy, or they may
have had a strong family history of breast cancer and chose the Mirena because they believed it to be safe than taking hormones by mouth or a shot.
I had a Mirena inserted at the age of 41 and had it for thee years and I’m not losing any sleep over the study. I am somewhat peeved at the design (more on that below. This study actually contributes very little to what we know on the subject.
Real life verdict: studies designed not to answer the question (i.e. does the Mirena IUS cause breast cancer) can’t answer the question. We need the right study because we all know how unsubstantiated medical beliefs about IUDs play out in America.
For those of you who want more than a summary keep on reading, but it’s cool if you don’t want to geek out on a more in-depth analysis!
Depo provera, an injected progestin (same family as levonorgestrel) birth control, is associated with an increased risk of breast cancer for women who use it for more than 1 year. Once the medication is stopped the risk returns over time to baseline. None of the studies that have identified this risk are randomized trials, but given there are at least five that support a connection between one progestin and breast cancer it is valid to ask, about other methods so the idea behind the new study is valid.
It is important to keep in mind that the amount of levonogestrel released in the blood stream is so minute that a new ultra sensitive assay had to be developed to detect it, so we certainly cann’t extrapolate data from depo provera to the Mirena. Different drug and way different dose.
This latest study looked at women in a national registry in Finland. It is an observational study, so they identified all women who had received a Mirena between the ages of 30 and 49 (> 93,000 women) and then looked to find who had a diagnosis of cancer (they apparently have registries for about everything in Finland, they must have more epidemiologists per capita than anywhere else except Framingham). There were no specific controls, but a percentage of the women had given some personal data as part of another study and from that they derived basic demographic data that they extrapolated to the entire cohort. The cancer incidence that was used as to compare the IUD users was derived from the corresponding cancer incidence of the population, not from a specific control group. What the investigators found was a 19% increase in the risk of breast cancer among women who had used the Mirena IUS between the ages of 30 and 49 and a decrease in endometrial (lining of the uterus), ovarian, pancreatic and lung cancers. The pancreatic and lung cancer reduction was completely unexpected.
There is a major issue with this study, the lack of control group. Using an IUD inserts all kinds of confounders, some very important ones include age at first delivery, breastfeeding data, and number of pregnancies. Women using a highly effective method of contraception are likely to have fewer children and may have them later and these are two major risk factors for breast cancer. We have no idea if Mirena users are more likely to breast feed or not. The issue of family history is also huge. What if women with a family history self selected for Mirena because they were worried oral contraceptive pills might raise their cancer risk? Maybe women with more medical issues chose the Mirena over other methods? Maybe…maybe…maybe. Arggh, too many maybes are a concern!
Fortunately, someone did that study. They compared Mirena IUS users with an appropriate control, copper IUD users. Interestingly, many of these patients were also from a Finnish registry. This study also had detailed demographic data on each patient such as body mass index (BMI), age when periods started, reproductive history, breastfeeding, hormonal exposure [contraception, hormone replacement therapy], period of IUD use, family history of breast cancer, and data on health behaviors that are associated with breast cancer (such as alcohol consumption, physical exercise, and smoking). They had data on > 5,000 women who used the Mirena and > 20,000 who used the copper IUD. What did they find? 22% of 5113 women who had breast cancer cases had used the Mirena sometime before their cancer diagnosis and 1130 (22%) had used a copper IUD. The exact same percentage. They also found no difference when the first IUD insertion was 13–60 months prior to breast cancer diagnosis (so recent use) versus the use of IUDs preceding the breast cancer diagnosis by more than 60 months (remote use). The only issue with this study is that it was powered to find a 50% increase and the newer Finnish study only had a 19% increase. So, this other study can’t disprove the newer one, but it is a model of how the study should be done and can help women feel assured that if (and that’ s a big if) there is an increased risk it is indeed low.
I’m not sold on the new study. It is clearly biologically plausible (given what we know about progestins) that even minute doses could increase the risk of cancer. Whether the Mirena raises the breast cancer risk slightly, this study can’t tell us. It is really a shame that they did all that work and didn’t identify copper IUD users as controls. Even without demographic data comparing IUD users to IUD users would have been so much better. I wrote a letter to the journal that published the Finnish study asking why a copper IUD control group wasn’t used. You simply can’t compare a Mirena IUD user to both the non-contracepting and contracepting general population without far more demographic and health data.