Why do we assume that diet plays such a major role in cancer etiology? And by “we,” I don’t refer to just the lay public, but to scientists as well. At least the initial idea stems from global variations in cancer incidence. Certain cancers occur at high rates in the US and Europe—breast cancer, colorectal cancer, prostate cancer—while these same cancers, at least decades ago when these hypotheses were formulated, occurred at much lower rates in countries in east Asia, notably Japan. Other cancers, like gastric and esophageal cancer, had a reverse pattern.
A geographic difference of this type could either result from genetic or environmental influences. However, it was known that when Japanese or Chinese people migrated to the West, they rapidly acquired the cancer rates of their new home. Genetic changes could not possibly occur that rapidly, so it was apparent that the differences in cancer rates were due to environmental or lifestyle factors.
The most obvious lifestyle factor that differed between the East and the West was diet, inasmuch as the two diets were dramatically different. Of course, there are many differences that one could explore, but it was also apparent that the two regions differed in a similar way in the occurrence of coronary heart disease. And it was understood that fat and other lipids played a major role in that disease, so it was thought that fat may be responsible in part for the geographic differences in these cancers. A number of studies have been conducted to examine that question.
One of the earliest and best was a study from Harvard published in 1987 from the Nurses’ Health Study, possibly the most important single epidemiologic study ever done and still ongoing today. It was led by Walter Willett and Meir Stampfer, who is the current head of the study and the son of a prominent rabbi from Portland, Oregon. The Nurses’ Health Study is a cohort study—a cohort study collects information on its subjects at the start of the study, including information on the exposure being investigated, and then follows the subjects forward in time to see who gets the disease of interest. If there is a connection between the exposure (in our case, high dietary fat intake) and breast cancer, then we will expect that the subjects who have high fat intake at the start of the study will go on to have more cases of breast cancer than the subjects who had low intake of fat at the start of the study.
As its name implies, the NHS recruited over 120,000 nurses at its inception. For the purposes of this study, 89,538 nurses, who had filled out food frequency questionnaires, were included. This method ascertained the major components of diet. These women were then followed for four years, during which 601 cases of breast cancer occurred. The cohort was subdivided into five groups by the amount of daily intake of fat and then compared for breast cancer risk. Just for context, around the time this study was conducted, the average daily intake of fat for an American was about 40% of daily calories (it has since decreased substantially). The five groups were, on average, 32%, 36%, 39%, 41%, and 44% total fat as a percentage of total calories. Assigning a risk of breast cancer of 1 to the lowest group (the 32% group), each higher group had a relative risk for breast cancer of 0.80, 0.88, 0.80, and 0.82. None of these risks were statistically different from 1. In essence, the study showed no association between intake of dietary fat and risk of breast cancer. Similar studies by others have also found absence of an association. So is it true that there is no association between dietary fat intake and breast cancer risk?
In my last article, I noted that one potential problem in epidemiologic studies of diet and cancer is that a lack of variability may inhibit a proper understanding of reality. That may be an issue here. The range of fat intake in this study went from an average of 32% to 44% of calories. This is a relatively narrow range of intakes—thus, while there was no difference across this range, our conclusions must be constrained. Very responsibly, these investigators concluded their study as follows, ”These data… do not exclude a possible influence of fat intake… at levels lower than 30 percent of calories.” In other words, if a woman’s intake of fat were lower than 30%, perhaps she would have a lower risk of breast cancer; this study was unable to explore that possibility.
Let us return to the reason for our fat hypothesis—the difference in fat intake between Japan and the US. In the 1970s, the average fat intake in the US was about 40% of calories. What was it in Japan? Let us return to this discussion in our next episode of Thoughts on Cancer.
Alfred I. Neugut, MD, PhD, is a medical oncologist and cancer epidemiologist at Columbia University Irving Medical Center/New York Presbyterian and Mailman School of Public Health in New York.
This article is for educational purposes only and is not intended to be a substitute for professional medical advice, diagnosis, or treatment, and does not constitute medical or other professional advice. Always seek the advice of your qualified health provider with any questions you may have regarding a medical condition or treatment.
