While the paper by Tomasetti and Vogelstein is certainly intriguing and will help guide future research on targeting prevention and early detection efforts, it is still a preliminary finding and does not reverse the long-term conclusion that is supported by decades of well-designed research in people: that cancer is largely preventable.
It’s been hard to avoid the headlines this past week touting the conclusions of a new study in the journal Science that cancer is a disease largely caused by “bad luck.” Some of these headlines have been sensationalized; some have been more measured. Together, they’ve seemed nearly ubiquitous. And in some ways that’s been positive, demonstrating that cancer – and its causes and prevention – remain an important topic to both the public and the media.
Unfortunately, the conclusions of the paper – and the subsequent coverage they’ve received – seem to reach well beyond what the actually findings of the study support, and in doing so, have incorrectly called into question what decades of overwhelming evidence shows: that over half of all cancers – and up to three quarters of some cancers – could be prevented by regular screening and healthy lifestyle choices.
The new study by Johns Hopkins researchers Cristian Tomasetti and Bert Vogelstein gathered data on the number of stem cell divisions that occur in various tissues in the body and compared these data to annual incidence rates of cancers in those tissues. Stem cells divide to replace cells in tissue that die or are lost. This maintains the tissue. Some tissues need a lot of cell divisions for maintenance. Others need less. When cells divide, the DNA in cells also replicate and divide, opening up the chance for random mutations that could eventually lead to cancer.
The researchers found a strong correlation between tissues’ number of stem cell divisions and cancer in those tissues, attributing 65% of cancers to these divisions, and the random DNA mutations that can occur with them.
While the study is well-designed and executed, it also has some key limitations that don’t fully support its overarching conclusions.
First, the study was not done directly in humans. It correlated data on cancer rates with tissue cell division data generated from cell analyses done in labs. It is essentially a laboratory-based study, and lab studies in general are fantastic ways to generate hypotheses on cancer causes but bad ways to generate conclusive evidence on actual causes of cancer in humans. The research world is littered with promising findings in petri dishes and mice that were never proved in actual people.
Second, many of the cancers included in the study are those already known to have few, if any, known lifestyle risk factors. And while the study included two common preventable cancers, lung and colorectal, it also did not include two other major cancers with key lifestyle factors – breast and prostate cancer. Together, lung, colorectal, breast, and prostate cancers account for approximately half of newly diagnosed cancers in the United States, a number much higher than the other serious cancers included in the study. The study itself even showed that colorectal cancers and lung cancers (in smokers) were less due to “bad luck” mutations than most others in the study. Other analyses based on long-term studies in humans have found that up to 75 percent of lung and colon cancers could be prevented, as well as at least half of breast cancers.
Finally, simply because there is a correlation between stem cell division and rates of cancer does not mean that random DNA mutations are the precipitating cause of cancer. Multiple cell divisions could accelerate the cancer-causing effects of lifestyle factors. And many lifestyle factors – from obesity to smoking – are thought to increase the risk of cancer by fostering cell division and DNA changes that begin the cascade to cancer. So what is described as random “bad luck” may actually have its genesis in lifestyle choices.
While the paper by Tomasetti and Vogelstein is certainly intriguing and will help guide future research on targeting prevention and early detection efforts, it is still a preliminary finding and does not reverse the long-term conclusion that is supported by decades of well-designed research in people: that cancer is largely preventable. In our own 2012 paper that also appeared in a Science publication – Science Translational Medicine – we conclude that over half of the 572,000 deaths from cancer in the United States in 2011 were caused by key modifiable risk factors, such as smoking, obesity, and lack of physical activity.
Of course, “bad luck” plays a role in any cancer – any disease, for that matter. We know people who smoke and drink and weigh more than they should who live until their 90s, and we know the healthiest people in the world who are struck down by disease early in life. Healthy choices, though, can greatly increase our odds of living longer and healthier lives. It’s no guarantee, but it helps to stack the deck in our favor.
For more on the preventability of cancer, see our page: Preventing Cancer – Today