Genetics linked to breast size and breast cancer risk

With Nick Eriksson

Social norms and preferences aside, it turns out that breast size matters — but not quite in the way you think. 23andMe’s latest research paper, published in the journal BMC Medical Genetics, shows that genetic factors influence whether women have double As or double Ds. This might sound a bit frivolous at first but our research uncovered surprising connections between the genetics of breast size and the genetics of breast cancer. Analyzing data from more than 16,000 women of European ancestry, we compared their answers to survey questions with their genetic data at millions of SNPs. We found a total of seven genetic factors significantly associated with breast size, three of which are strongly correlated with SNPs already linked to breast cancer. How do you study breast size anyway? In our case, no one’s personal space was violated — all of our data was self-reported by female 23andMe customers of European descent who have opted into research and filled out an online survey. We specifically asked about bra cup size as an approximation for breast size using a 10-point scale ranging from “Smaller than AAA” to “Larger than DDD”. We also took into account age, breast-related surgeries (augmentation, reduction, or mastectomy), pregnancy and breastfeeding status, and bra band size (which is correlated with body mass index (BMI)). Most of the genetic factors we identified for breast size lie in regions of great importance for breast cancer. One of the SNPs, rs12173570, is known to regulate the expression of the estrogen receptor gene, which plays a vital role in breast growth and in the majority of breast cancer cases (known as “ER+” or “hormone positive” types). Another one of these SNPs, rs7816345, is located in a region of the genome (8p21) that often shows abnormalities in people with a certain subtype of breast cancer. Scientists also recently reported genetic factors influencing breast density — a risk factor for breast cancer — and our research suggests that one of those factors may also be associated with breast size. These findings show that some of the same biological pathways underlie both normal breast growth and breast cancer. This isn’t a huge surprise if you think of cancer as unrestrained growth. But the relationship between breast size and breast cancer is complicated. Some studies have found that larger breast size as a young woman is associated with a slight increase in breast cancer risk, but only in women who were lean at a young age. The genetic factors we found aren’t enough to explain this association, but support the idea that breast size and breast cancer are related. While the topic of breast size raised a few eyebrows when we first put out the survey (some thought it was too personal, or couldn’t be serious science), this paper demonstrates that important scientific insights can come from the most unlikely of places. Although the connections between these genetic factors, breast size, and breast cancer aren’t fully understood, our findings give clues to the function of some of these genes and regions that might be useful in combating breast cancer.
Help Us Learn More About Metastatic Breast Cancer: The InVite Study

23andMe and Genentech have launched the InVite Study to understand how genes influence response to treatment for metastatic breast cancer. Anyone who has had metastatic breast cancer and received a specific treatment may be eligible. Learn more at or email
Editor’s note: Pending an FDA decision, 23andMe no longer offers new customers access to health reports referred to in this post. Customers who purchased prior to November 22, 2013 will still be able to see their health reports, but those who purchased after that time will not. Those customers will have access to ancestry information as well as access to their uninterpreted raw data.
  • jpf

    Sooooo, what are you guys saying the relationship is? And how is it not what we think (according to your headline)?

  • Barbara

    I agree with jpf. Your thesis is missing. I think there’s a hidden message in the “some papers maybe might suggest that maybe largerish breasts may might maybe have slightly higher risk of breast cancer maybe.” This is not news, then, and you wasted everyone’s time.

  • rachel

    This is vague. All i can gather out of this summary is that there is likely an association, but it’s not clear what the relationship is.

    • SuperY

      Indeed. Based on what little info we have here, perhaps it’s something along these lines: cancer being abnormal growth, and large breasts also involving somewhat unusual (not abnormal) growth as well as hormones, there are certain cancers that involve a syndrome related to hormone-positivity.

  • Raj Michael

    I just read your paper. Very interesting.

  • kk

    Good job. It is good to identify SNPs that contribute to breast size. Some may think it is too personal, but none-the-less it is good science as it is something can be done with DNA data and questionnaires.

    It is important that we gather as much information between SNPs and traits. If it some how contributes to the understanding of breast cancer, it is a big bonus.

    Also with clientele of 23andMe being mostly European, it leave a question whether any of the analyse can be extend to non-Europeans.

  • Francois

    jpf and Barbara – why don’t you read the paper? No time? Well… as I understand it the “message” is that there is a small set of genes that both influence breast size and breast cancer risk. Simutaneously. These relationships are not as simple as “big breasts = cancer risk” and need to be analysed further. I think this is an interesting message, don’t you?
    PS: I got to this interpretation without reading the paper either.

    • SuperY

      Chill on the condescension, please. One can be smart without being insecure about it.

  • Jacques Beaugrand

    could it be simply that the more tissue there is in a breast the more (higher probability) of developping a cancer?

    Thus rs7816345 and rs4849884 regulate the expression of the estrogen receptor gene

    and larger breasts would have greater chances of developping cancer when auxilliary conditions are present (epigenetic conditions).

  • lamed vav

    I look normal and do not need to wear bras.
    If i do wear a bra I need
    E, F, G or the biggest cup I can get and
    I am free of the breast cancer genes and no one
    in my family ever had BrCa So I would say large
    is healthy and good.

  • John Echeverria

    larger breast size as a young woman is associated with a slight increase in breast cancer risk, but only in women who were lean at a young age. Interesting.

  • Gail

    I have hypotheses that certain ethnic backgrounds have larger breast sizes — do you have any stats on this?

  • anonymous for reasons related to the topic

    so, it says I should have average size breasts, as I am CC. However, I would say my Ta Ta’s are merely TT’s. If you catch my drift….

  • KK

    @ Gail and others with similar question.

    The paper said that the result is based on european women. My first question is “Do these SNPs perform the same function in non-European?” I would guess that the answer is yes. This can be confirmed by supplementing the research to include non-European women. This would imply these SNPs have different prevalence in different ethnic groups. This one can be easily done by 23andme by comparing the existence of SNPs with the percentages of Asian, African makeup.

  • xyz

    Well, I have to wonder about self reported data — having two kids my breast size went from C -> D -> A -> C -> A -> B over period of 5 years — and it seems that most of my girlfriends experience such variation (but for some net effect is bigger, for some like me net effect is lower).

  • SteppyLou

    In both results, I am CC – one means average-sized compared with smaller, and the other means… average-sized when compared with larger. Wondering if I belong in a Russ Meyer film as I am a 34-36 E+. I’m pretty sure that’s not average, because I can never find bras that fit in stores. And I’m not augmented, nor have I had kids. Hmm.

  • Liz

    These snp’s show myself, and my female relatives as having average sized breasts. Well, at the age of 16, I was 115 lbs, and I wore a C cup. As an adult, I have varied in size from 32/34D to 36-C. My daughter is also a 32-D, and is very thin. I have a few of the risk alleles for breast cancer FGFR2, 16q12, and stxbp4. They put my risk at 17.3 vs 13.5. I’m certainly not average in breast size, and neither is my daughter. My mother is somewhat average, however she received the same results as my daughter and I received.

  • Maureen

    I’m not sure that I understand this. I have CC and CT for these two tests, indicating that I have smaller breasts than average, or average breasts. Are you saying that I am at a lower risk for breast cancer? But, I already have breast cancer!

  • Gypped in the Nip

    Maureen — what the hell, right? I have a CC at both the SNPs listed, but am actually an A cup (was AA until I started getting older) and my other genetic data suggests I’m at higher risk for developing breast cancer.

    It sounds to me that the combination of genetic BLUE PRINT, actual genetic EXPRESSION (what size you ended up), environmental factors in your life (pollution, chemicals, medication, distress), and then any other bad luck factors is too complicated to make any kind of accurate prediction.

    Best thing for any of us ladies to do is to get those mammograms and do our own checking as well. The research presented here is too maybe/kinda/sorta to be helpful.

  • thequeenbee

    sorry ladies..CC and DD and EE cups (for that matter) are not really large breasts. I ma a 32 K or 32 JJ or a 32 III cup. Renenber the number measures around the chest wall, the letters measures the DEPTH to hold the girls.

    I have large breasts. I was also born in Europe so let’s see..

    1. Born in Europe? Check
    2. skinny kid? check
    3. skinny adult check
    4. breast size genetic? NO–on both sides the women normally are 32Bs on down the line–I hit the boobie jackpot (all natural)
    5. Breast cancer genes in me? NO
    6. breast cancer as of yet (I am over 50) NO
    7. Breast cancer in my family? NO

    Okay.. in other news, scientist also found that breast cancer was more likely to occur in females, and that most breasts had nipples, and of course that all fat was bad, aspirins prevented heart attacks and that eggs were the complete food.

    Junk science is when one goes only so far as observations and correlations without any real challenges or actual factual data. The only bit of truth in all of this as far as I can see, is that there is a cancer gene or gene precursor–all the rest is fishing and needless drivel–because the fact is, fat and skinny girls get cancer–women with big tatas and little bitty ones get cancer–and women with no genetic markers or family history of cancer get cancer–regardless of how big their boobs are.

    Junk Science–there ought to be a law that fines any touted study or research that bases their extrapolations on sheer guesses and associations.

    • SuperY

      Same thing with humblebragging, wouldn’t you agree?

  • It is possible that you are on to something with the SNPs or whatever you criteria is, but it seems that you have a long way to go before you publish as you are possibly scaring the bejeepers out of a lot of women and girls about something they have no control over. Their mothers would have control over what they feed them when they are young. I think you make a big mistake by leaving diet out of the equation. Diet and ethnicity is a study unto itself. What mothers feed their children to keep them quiet and how it affects their development is another thing to take in to consideration. Narrow studies are for flour and fauna, not human beings.

  • Afiq Zamri

    This is a very interesting article but your thesis did not really gives evidence about genetic factors causing breast cancer. I cannot really see the relationship.

  • Afiq Zamri

    This is a very interesting article but your thesis did not really gives evidence about genetic factors causing breast cancer. I cannot really see the relationship. Maybe more evidence can be shown to prove that breast size that is controlled by genetic factors can lead to higher chance of having breast cancer.

  • catherine

    I first began to develop at 10, & got periods at 12.I was at least a year ahead of most of my classmates, it was the mid 60`s, & I lived in a poor area where girls tended to develop late. I wasn`t particularly well built or tall. & my mother didn`t reach menarch until her mid-teens. I was bullied at school for developing earlier than my classmates, & this resulted in major issues with my body for all my young adult life. I wasn`t emotionally mature enough to deal with puberty then, & I really believe that we should find a way to safely delay sexual maturity until kids are ready to cope with it. Doing this would save so many social problems, like teen pregnancy, & pscycological difficulties in teenagers.