A study was released last week that claims to have found a possible genetic link between breast cancer and larger bra sizes. While it hasn’t gotten much coverage, most of the headlines I’ve seen about it make it sound like a done deal: those of us with large breasts are more likely to develop breast cancer. But is this true, or simply pandering for pageviews by misrepresenting the findings of a study that wasn’t terribly rigorous in the first place?
First, let’s look at the study. Researchers at the personal genetics company 23andMe studied women of European descent whose genomes had previously been mapped by the company. They had them fill out an online questionnaire; according to the study as published online in BMC Medical Genetics:
Out of the 16,175 participants, all but 3 reported age, over 15,000 reported bra band size, about 12,000 reported breast surgery status (augmentation, reduction, mastectomy, or none), about 6,000 reported if they had ever been pregnant, and 4,000 reported if they were currently pregnant or breastfeeding.
Presumably all reported their bra cup size, as that was the focus of the research. The researchers then looked at the genes that influence breast size to figure out which alleles corresponded to larger breasts. They then compared these findings to the genes that are known to increase the risk of breast cancer and found two that overlapped, and a third gene known to increase cancer risk that might be linked to larger breasts, but not conclusively. (The UK’s NHS Choices website explains the genetics in more detail; I freely admit this part is a bit over my head.) In conclusion, they announced that:
These results provide insight into the genetic factors underlying normal breast development and show that some of these factors are shared with breast cancer. While these results do not directly support any possible epidemiological relationships between breast size and cancer, this study may contribute to a better understanding of the subtle interactions between breast morphology and breast cancer risk.
Fairly measured summary, right? They found some genes that might be related to both breast size and cancer, but without further study it’s impossible to say if there’s a causal relationship or if it’s just a coincidence. Of course, that wasn’t how the study was covered in the media, or even in 23andMe’s press release announcing the publication of the study. Lead author Nicholas Eriksson, Ph.D., explains the findings thusly:
The findings in this study show that some of the same biological pathways underlie both normal breast growth and breast cancer. Some studies have found that larger breast size as a young woman is associated with a slightly higher risk for breast cancer. The genetic factors we found support this concept that breast size and breast cancer are related.
To me, Eriksson’s statement seems to imply a stronger connection than the study warrants. And of course, headline writers ran wild with the information, even on articles that did present a fairly nuanced and accurate view of the study. Several went with “Breast cancer ‘linked to bra size'” and the articles that didn’t use that exact wording mostly used something very similar. The most amusingly wrong one was actually titled “Women with Big Breasts More Likely to Get Cancer.” No, that’s not even close to accurate based on this study! (But it did at least helpfully go on to say that if you have larger breasts due to implants, that wouldn’t affect your genetic risk of cancer. I’ll be banging my head against a desk now.) Journalists should have a responsibility to accurately reflect the contents of an article in its headline so as to not mislead the public. Anyone just glancing at these headlines would think they had reason to worry about this purported link, when actually it’s far too early to say for sure what, if any, genetic risk of breast cancer can be attributed to bra size.
There are also numerous problems with the study’s methodology. Relying on self-reported data is frequently problematic, especially since not everyone answered every question. For participants who skipped questions, the authors assigned them the average value of the people who did answer. Given that only three people skipped the age question, that’s not a big deal and should have no influence on the data. However, since cup size varies greatly by band size, assigning the average band size of 36.2″ to non-respondents (or people whose numbers were discarded due to the researchers assumption that they had answered in centimeters if they gave band size numbers above 70) could throw off the data somewhat. Assigning the average means women with a 32C bra were assumed to have the same breast volume as women with a 40C, when that isn’t at all the case. Since the average was used for perhaps 1,000 women, that could throw off the data since they looked at the genes of individual women to figure out which alleles corresponded to increased bra size and thus some genes’ activities could have been over- or under-estimated. Never mind that even in two women wearing identical bras, they could still have different breast densities or shapes and thus different volumes.
Further complicating this, of course, is the fact that many women aren’t even wearing the correct bra size. I tried 3 different bra size calculators online with the same random numbers and got three vastly different results. For a 38″ ribcage and 42″ breasts, one site recommended a 40D, another said 38D, and according to both Playtex’s regular and plus-size calculators those measurements would require a 46 band but the breasts were too small to assign any cup size(?!). The study does admit repeatedly that band size isn’t a perfect indicator of breast volume per cup size, but what if both are reported incorrectly?
Some of the test questions were poorly designed as well. While the participants were asked whether they’d undergone any breast surgeries, the study didn’t instruct them on whether to report their current bra size or the size they wore before their augmentation, reduction, or mastectomy. Some women noted that they had in fact reported their pre-surgery size, but a well-designed study would not have left any room for confusion. The women also weren’t asked at what stage in pregnancy/breastfeeding they were; women in early pregnancy aren’t as likely to have changed bra size as women in the late third trimester, and breast size varies throughout nursing as well. Again, some women noted that they gave their pre-pregnancy bra size, but not all.
It’s also highly problematic that only women of European descent were eligible for participation. I don’t know why on earth they would try to announce any sort of general trend about bra size and breast cancer while only looking at data from white women. Perhaps they didn’t have enough WOC’s DNA on file with their company to have a good sample size, but it very much rubs me the wrong way that they would limit the pool by ancestry in this way. While white women do have the highest incidence of breast cancer in the U.S, rates among black women aren’t much lower and they die in much higher numbers. Only studying one race without any explanation as to why is dangerous and implies that the researchers don’t care why non-whites develop the disease.
Perhaps the most glaring anomaly was that the study didn’t ask participants if they had ever been diagnosed with breast cancer or had a family history of the disease. Obviously many of the women may go on to develop it later in life, so their current status as cancer-free wouldn’t negate any genetic links. And of course, since there are many different kinds of breast cancer it might have muddied the data a bit. But it seems like it would be helpful to see if the possible links they found were reflected in any cancer survivors in the pool. Presumably many if not all of the 1.7% of respondents who had undergone mastectomies had either been diagnosed with breast cancer in the past or were found to be carrying other known genetic risk factors and had gotten the procedure to lower future chance of its development. It just still seems very odd that a cancer study wouldn’t ask about cancer.
So how much do we need to worry about the results of this study? Right now, I’d say not at all. There are so many factors that influence breast size and even more that influence breast cancer. While further studies may determine that those specific genes can in fact cause both breast growth and cancer, the data is far from conclusive at the moment. Phew!