Cancer: What’s Luck Got To Do With It?Grandfather smoked like a chimney, ate bacon everyday and lived to be 90. Yet, your best friend, a lifetime vegan who exercised regularly, succumbed to breast cancer in her thirties.
We’ve all used these anecdotes to try to make sense of the deadly scourge that is cancer. So do this week’s headlines in the popular press, Biological bad luck blamed in two thirds of cancer cases
give us license to bring on the bacon and booze? Not so fast, as the study just published in Science
by Johns Hopkins researchers Cristian Tomasetti and Bert Vogelstein was much more nuanced than the headlines suggest.
♦ The premise of the study was the puzzling observation
that some tissues give rise to cancers a million times more frequently than others: for example, the lifetime risk of being diagnosed with cancer is 6.9% for lung, 1.08% for thyroid, 0.6% for brain, 0.003% for pelvic bone and 0.00072% for laryngeal cartilage. Even within the digestive tract, cancers of the colon (4.8%) are much more common than stomach (0.86%) despite both tissues being exposed to the same carcinogens and dietary insults. To make sense of this, the researchers turned to cancer stem cell theory
: that malignancy is caused by mutations in a small number of stem cells that retain a lifetime ability to divide. They then searched the literature to estimate the number of stem cells in each tissue. What they found was that tissues with higher populations of stem cells were more prone to cancer
. This linear correlation (R=0.8; R^2=0.65) was pretty good, extended over 5 orders of magnitude (see graph), and makes sense since we already know that every time DNA replicates there is a finite chance of making errors and that the more mutations in DNA the greater the chance of some of them triggering cancer. In simple terms, a large part (estimated two-thirds; see R^2) of the variation in cancer risk between tissues is due to the difference in their stem cell population. This does not translate into “two-thirds of an individual’s risk of cancer is due to dumb luck”!
risk factors pile on top of the basic risk of random mutations from stem cell divisions. To identify cancer types (red circles
) in which the contribution of environmental and inherited factors was especially high relative to the random DNA replication-driven component, researchers used an unbiased clustering algorithm that used the product of the log values of the x- and y-axes in the graph below. What they found was consistent with what we already know about some cancers. Smoking
greatly increases risk of lung cancer by ~18-fold for both sexes (23-times in men, 13-times in women), as seen by the higher risk incidence of lung cancer for smokers compared to non-smokers in the chart. People with familial mutation
in the APC gene have a 100% rate of colorectal cancer unless the colon is removed. Infection with Hepatitis C
increases risk of liver cancer by 10-fold.
♦ This new analysis explains some puzzling facts: the same APC mutation has a much higher chance of giving rise to colon cancer instead of duodenal cancer because there are 150 times more stem cell divisions in the former compared to the latter. Another example is that basal epidermal cells and pigment cells of the skin (melanocytes) are exposed to the same carcinogen (UV radiation) at the identical dose. Yet, basal cell carcinomas are much more common than melanomas. The authors argue that this is explained by the underlying difference in rates of stem cell division.
♦ Why should we care about this analysis?
Understanding the underlying risks contributing to each cancer type should determine the best public health strategy to tackle it. Early detection
should be the main focus for prevention of cancer types largely driven by random errors in DNA replication, whereas vaccines
against infectious agents or altered lifestyle
will be key to reducing incidence in cancers with high environmental risk.
Inset image: http://www.sciencephoto.com/media/254142/view #ScienceSunday #cancer