Frontline Cancer: Examining the luck factor


Some of you may have read news stories recently that suggested getting cancer is simply a matter of bad luck, and that nothing can be done to significantly reduce one’s cancer risk.

This conclusion is based upon a study conducted by researchers at Johns Hopkins University and published in the estimable journal Science in January. The study garnered lots of headlines, from the front pages of U-T San Diego to The New York Times. It generated discussion and more than a little confusion.

The reason for the confusion is simple: The research was extremely complicated, but the media, by and large, distilled the results into headlines that were attention grabbing, but also misleading.

In the service of science and, well, the facts, I would like to take this space to discuss the results of the Johns Hopkins study, why the findings should be interpreted with caution and why it is not in your best interest to abandon a healthy lifestyle and resign yourself to “luck” or “fate” in the effort to prevent cancer.

A little background about the causes of cancer will be helpful.

There is a great deal of scientific evidence that cancers are genetic diseases — caused by an accumulation of mutated or other alterations in DNA.

It has been known for many decades that there are three primary causes of cancer. First, environmental factors, such as overexposure to sun or smoking. Second, hereditary genetics like the BRCA1 factor in breast cancer. And third, random errors that occur during routine cell division.

The Science paper from Cristian Tomasetti and Bert Vogelstein at Johns Hopkins was a purely mathematical model attempting to quantify the importance of these factors. Their model concluded that 65 percent of cancer is caused by random error occurring during normal cell division. According to their research, environmental effects or hereditary factors cause only one-third of cancers and the other two-thirds are just the result of bad luck. And thus, there is not too much to be done about it other than to rely on early detection and screening.

However, there are several important points to keep in mind with this study. The mathematical model is based on estimated stem cell numbers and division rates of these cells in each organ. It is a number that is extremely difficult to accurately determine. Moreover, the model assumes that these rates are fixed and do not change over a lifetime, which is inconsistent with our understanding of cancer development.

The model is primarily based on stem cell numbers in humans, but the researchers also extrapolated from numbers in mice. In some cases, the origin of stem cell numbers is not clear. A significant number of leading scientists who work with stem cells in cancer research do not agree with Tomasetti’s and Vogelstein’s figures. They are understandably cautious about accepting the results — for a number of reasons.

For one, the model used by Tomasetti and Vogelstein did not look at certain cancer subtypes that come from the same stem cell. In cancers like leukemias, which arise from hematopoietic stem cells, you would expect equal numbers of different types of leukemias based on their stem cell mathematical model, but that isn’t what happens. The same is true for brain cancers, which arise from neural stem cells. You would expect similar numbers of the different types of brain tumors to arise from neural stem cells, but brain tumor types vary widely in the rate of incidence.

Additionally, it’s not clear that all cancers arise from actual stem cells. For example, this is an ongoing debate regarding pancreatic cancer.

Problematically, the study excluded breast and prostate cancer, which are among the most common cancers, but included several less studied, rare tumor types. Random stem cell replication errors may indeed be the initiating event in certain relatively rare cancers, but the fact is we just don’t know how they arise. The two-thirds of cancers considered to be “bad luck” cancers included these rare tumors.

The study model did not account for the known effects of infection, inflammation or obesity on stem cell division and cancer development. There were inconsistencies with other cancers that fell into the “bad luck” category, such as head and neck and esophageal cancers that were put into this unfortunate group, but research also shows that smoking and alcohol play big roles in the development of these cancers.

These conflicting results underscore the concern about the accuracy of the stem cell numbers and assumptions used in the mathematical model — and if the resulting calculations were inaccurately magnified.

It is admittedly very difficult to decipher the findings of this work, and even more so to effectively translate them for both public and healthcare professionals. Nonetheless, it seems that the magnitude of the findings was overstated in some tumors.

It is not news that cancers can arise spontaneously from random genetic changes. What is new in this study is the magnitude of the claim that the majority of cancers are just bad luck. The study raises many questions and the public health message should not be that there’s nothing you can do about getting cancer. Even the authors agree that some cancers are preventable and every effort must be made to decrease the risk of cancer.

With advances in technology and our understanding of stem cell biology, it will be interesting to see if the conclusions of this study remain intact. In the meantime, it seems prudent to make your own luck by living a healthy lifestyle and decreasing exposure to the toxins known to provoke DNA replication errors.

— Scott M. Lippman, M.D., is director of UC San Diego Moores Cancer Center. His “Frontline Cancer” column on medical advances from the front lines of cancer research and care appears monthly in La Jolla Light. You can reach Dr. Lippman by e-mail: