Monday, September 20, 2010

Ladybusiness anthropology: Physical activity and prevention of reproductive cancers

ResearchBlogging.orgI use PubMed to make RSS feeds out of searches on topics relevant to my research, because I cannot be trusted to regularly look this up on my own (I also have feeds from many of the main journals in my field, natch). One of my searches, “inflammation endometrium,” turned up an interesting article today from the European Journal of Cancer. Friedenreich et al review the literature on physical activity and cancer prevention and find a consistent relationship between physical activity and reduced risk of several cancers. Obviously, I want to focus on breast, endometrial and ovarian cancers, since those are the stuff of ladybusiness.

Breast cancer

The authors found seventy three separate studies on breast cancer (they say separate, I assume, to differentiate multiple publications on the same data set). Three quarters of the studies found physical activity to have a positive effect on breast cancer risk; within those, the risk reduction tended to be around twenty five percent when comparing most to least active participants. Interestingly, postmenopausal physical activity seemed to have the strongest effect, which was a bit of a surprise to me. What this says, to me, is that adjustments to lifestyle made late in life can still have significant effects on health. If you were a couch potato when young but are committed to physical activity now, all is not lost.

Endometrial cancer

For endometrial cancer, the authors concluded that “physical activity probably protects against endometrial cancer” (p. 2594, anyone else find that wording odd?). There were twenty studies of endometrial cancer, so far fewer than of breast cancer, but the reduction in risk was very similar, of twenty to thirty percent. Sedentary behavior appears to increase risk (like jobs where you sit all day), and more intense or longer bouts of activity tend to have more positive effects.

Ovarian cancer

Here, the authors found conflicting evidence, where some studies found risk reduction with physical activity, some found no effect, and one found a risk increase with physical activity. There were about twenty studies that looked at ovarian cancer, but the authors claim that the sample size was often small.

What surprised me about this section was what little attention they paid to mechanism. Ovarian cancer is different from breast or endometrial cancer: a big part of what researchers think causes it is the “incessant ovulation” of industrial societies. That is, women in industrial populations, due to a positive energy balance, low rates of childbearing and low rates of breastfeeding, have as many as 400 menstrual cycles in their lives, compared to only 50 in a forager population, which would have a later age at menarche and women who are often pregnant or breastfeeding. Continual insults to the ovarian tissue, from the rupture of the ovarian wall during ovulation, and its subsequent repair, increase the risk of cell mutations. Small differences in sedentary or active behavior don’t tend to be enough to keep a woman from ovulating.

Why parse out physical activity and sedentary behavior?

You may have noticed that I mention different kinds of physical activity, but also sedentary behavior. In the paper the authors go into far more detail about different types of behavior, from domestic activities to occupational activities to recreational ones, and whether they are physically active or sedentary. It is important to consider both of these factors because they contribute to an overall lifestyle that can tip the balance towards a more positive (more calories in than out) energy balance, or a more negative (more calories out than in) energy balance.

Let’s take my Polish field site for instance, the Mogielica Human Ecology Study Site, directed by Grazyna Jasienska. These women burn on average more calories than urban US women (Clancy et al 2009). Women there work alongside the men to do all the agricultural work needed to run their farms. You could just stop there and say that is very different from many populations within the US. But these women also do all their own housework and cooking, they have gardens, if they work they walk or take the bus. They are often sweeping and mopping instead of vacuuming. They spend a lot less time in front of screens, like televisions or computers (they have them for sure, but there are fewer desk jobs in front of a computer there). At least from my observations, they are less likely to order things online, and instead go to the store.

So the lifestyle differences, comparing an urban, sedentary person with a desk job in the US to a farmer in Poland, become far more striking, as do the number of calories they likely burn each day in their daily lives.

Is there population variation in cancer rates?

I thought you’d never ask! The main reason I decided to write about this article was so that I could highlight one of the most elegant, simple little graphs I have ever read, from a 2001 article by Jasienska and Thune. They compare population averages in progesterone (a female reproductive hormone) with breast cancer rates. Here’s the graph:

Jasienska and Thune 2001

Guess what is one of the biggest predictors of progesterone concentrations? Physical activity.

Finally, why someone else should always read your manuscripts

There were a few tough sentences in here that a copyeditor could have really improved. The authors wrote:
“There is strong and consistent evidence that physical activity reduces the risk of several of the major cancer sites, and that between 9% and 19% of cancer cases could be attributed to lack of sufficient physical activity in Europe” (p. 2593, abstract)
Which I tweeted, and to which Alex Wild of Myrmecos responded:
“@KateClancy So if Europeans were more active we'd all have less cancer?#wordingfail”
And then as I was preparing this post, I found:
“Physical activity reduces breast cancer risk when performed at any age throughout life, but activity done after the age of 50 years does have a stronger effect on risk than activity done earlier in life and sustained lifetime activity is of benefit” (p. 2594).
So, sustained lifetime activity isn’t as good as activity after 50, or this is a point independent of which time of activity matters? I thought perhaps they were trying to make points about both? Copyeditor, please!

Anyway, this article was a fun way to get to share a broader perspective on reproductive cancer rates in women across the world. Share your thoughts or questions in the comments!


Clancy, K., Ellison, P., Jasienska, G., & Bribiescas, R. (2009). Endometrial thickness is not independent of luteal phase day in a rural Polish population Anthropological Science, 117 (3), 157-163 DOI: 10.1537/ase.090130

Friedenreich CM, Neilson HK, & Lynch BM (2010). State of the epidemiological evidence on physical activity and cancer prevention. European journal of cancer (Oxford, England : 1990), 46 (14), 2593-604 PMID: 20843488

Jasienska, G., & Thune, I. (2001). Research pointers: Lifestyle, hormones, and risk of breast cancer BMJ, 322 (7286), 586-587 DOI: 10.1136/bmj.322.7286.586


  1. Kate, you asked for comments, so here is one. It's great to discover your blog, and I'm excited to see what more you do with it.

    I was wondering why physical activity leads to lower progesterone.

    And I was also thinking about the role of physical activity and the stuff I've read, though not deeply, about how it can reduce chronic inflammation, which also seems to be a cancer predictor. But I could be way off there, just thoughts bumping around in my head right now.

  2. Thanks, Daniel, for coming by, commenting, and featuring me on your wonderful blog Neuroanthropology.

    Physical activity tends to produce energetic stress, which leads to ovarian suppression. The mechanism for this appears to have something to do with disruption of the hypothalamic-pituitary-ovarian axis due to stressors. On a broader scale, female reproduction is costly, and needs to "think" long-term, so suppression of ovarian function will occur if there is a sense that there will not be sufficient energy to not only ovulate, but maintain the endometrium, implant, make a placenta, gestate, give birth, and lactate. So it's an adaptive, sensitive process.

    What you're saying about physical activity and inflammation is also important. I'm not sure if the relationship is because adipose tissue and inflammation are correlated (and so more physical activity presumably decreases fat tissue), or if there is an independent relationship. But increasingly we are finding that inflammation plays a role in just about everything. My current work focuses on the relationship between inflammation and endometrial functioning (hence the PubMed RSS feed search), because I think that's one of the ways we will be able to examine a strong relationship between inflammatory processes and reproductive functioning in women.

    Again, thanks for your comment and I hope to see you around here more!

  3. Interesting, then, how physical activity becomes an important life history parameter, which isn't exactly how it's thought of in the literature.

    And thanks for the comment over at Neuroanthropology!

  4. Hmm, maybe it depends on where in the literature you're looking. In human reproductive ecology, I would say it's considered the most important thing, what explains the greatest amount of variation in women's reproductive functioning. But in medicine, public health, maybe other subfields of anthro, perhaps it's not looked at in the same way.

  5. Do you know how progesterone relates to breast cancer in post-menopausal women? Is there a need to reduce or banish progesterone supplements?

  6. Silver Fox, that's a great question. I am not an expert on cancer by any means, but my understanding of the relationship between progesterone and breast cancer is that it is correlative, not causative. Lower progesterone indicates a higher degree of ovarian suppression, OR a population that has a lower relative set point due to a different (generally more active) lifestyle. I wonder if populations with lower average progesterone concentrations have more anovulatory cycles, which might mean less estrogen exposure and then lower risk of breast cancer.

    One of the problems in trying to figure this all out is that, even in anthropology where we love ourselves some variation, we tend to only allow women into our studies who have at least somewhat regular cycles, and often exclude those cycles that are later found to be anovulatory. This is significant if there is variation between populations in the number of women who may be anovulatory at any given time (which I suspect is the case). It also means that we likely REDUCE the amount of hormonal variation we see between populations, since we're only looking at the cycles where women have ovulated.

    So, not a perfect answer, but, I think, the only one I can really give! I'm not crazy about exogenous hormone supplements, but I have been trying to examine that discomfort in order to be critical of the times it's problematic, and more forgiving when it's not. I think I would worry more about post-menopausal estrogen than progesterone. But again, this is definitely not my specialty.