By Jenny Song
Diet and cancer may be linked, but it’s difficult for researchers to tell us how
By Jenny Song
Genetic risk for cancer may be unchangeable, yet we know that lifestyle choices, such as smoking, almost certainly increase one’s risk. But is the same true of diet? Can what you eat—or not eat—cause the disease or prevent you from developing it?
Cancer rates vary around the world and over time, even when genetics remain the same. That suggests diet may play a role in the disease, says Meir Stampfer, a professor of epidemiology and nutrition at the Harvard School of Public Health in Boston. “When migrants go from one country to another, their cancer rates change. So something in the environment is driving those differences, and diet is a very likely suspect.”
But proving a cause-and-effect relationship between food and cancer prevention is difficult, says epidemiologist Regina Ziegler, a senior investigator at the National Cancer Institute.
Why? For one thing, food consumption can fluctuate. “If you think about how much your diet varies from day to day, from week to week and from year to year,” she says, “it’s very hard to characterize adult diets because they change so much and have so many parameters.”
Another problem is that diet and lifestyle are often intertwined. People who have a healthy diet tend to exercise more and are more likely to seek medical care. How can researchers know whether it is diet—or one of the other factors, or a general healthier lifestyle—that contributes to a reduced risk of cancer? “If you say, ‘Let’s look at the people who eat more soy,’ they could also have lower body weight, be more active, drink more Chinese tea,” says Ziegler. “And you can try to control for that, but it’s hard to separate the one factor.”
Through observational studies, researchers may ask people what they eat, keep track of diseases the participants later develop, and then compare the diets of people who develop cancer to those who don’t. These studies can suggest a link between food and cancer, but because of the complex factors involved in diet and lifestyle, an observational study doesn’t provide conclusive evidence. A more rigorous way to demonstrate a connection would be through a randomized controlled clinical trial, in which patients are randomly divided into groups. Each group is then assigned to consume a different strictly controlled diet (or regimen of dietary supplements) and monitored. But these studies are expensive, and there’s a question of the feasibility of conducting a good randomized trial to examine links between diet and disease.
“I mean how are we going to randomize children to ‘high soy intake’ and ‘low soy intake’ and wait a number of years for them to develop breast cancer?” Ziegler says. Instead, she says, investigators must use a combination of other well-designed studies, including test tube studies, animal studies and human studies, and evaluate whether their results suggest a risk or a benefit to people who eat a particular type of food.
So what are we to think about the latest findings headlining the daily news that tout the benefits of green tea, tomato products and other “cancer preventive” foods? How seriously should the public consume these announcements?