Cancer, a word that for decades was whispered as taboo, has become front and center in the medical community. Cancer is the number one killer of Americans, at least those <85 years old, even ahead of cardiovascular disease (1). We have thought that diet may be an important component in preventing cancer. Is diet a plausible approach?
An article published in the New York Times, entitled “An Apple a Day and Other Myths,” questions the validity of diet in the prevention of cancer. This article covers cancer in general, which is a huge and daunting topic.
The article’s author references a comment by Walter Willet, M.D., a professor and chair of the Harvard School of Public Health’s Epidemiology and Nutrition Department as indicating that the research is inconsistent when it comes to fruits and vegetables. The article goes on to state that even fiber and fats may not play significant roles in cancer.
I don’t necessarily disagree with their assessment. However, I would like to emphasize that Dr. Willet also commented that there are no large, well-controlled diet studies. This leaves the door open for the possibility that diet does have an impact on cancer prevention. I would like to respond.
As Dr. Willet hinted, the problem with answering this question may lie with the studies themselves. The problem with diet studies in cancer, in particular, is that they rely mainly on either retrospective (backward-looking) or prospective (forward-looking) observational studies.
Observational studies have many weaknesses. Among them is recall bias, or the ability of subjects to remember what they did. Durability is also a problem; the studies are not long enough, especially with cancer, which may take decades to develop. Confounding factors and patient adherence are other challenges, as are the designs and end points of the studies (2). Plus, randomized controlled trials are very difficult and expensive to do since it’s difficult and much less effective to reduce the thousands of compounds in food into a focus on one nutrient. Let’s look at the evidence.
The EPIC trial
Considered the largest of the nutrition studies is the European Prospective Investigation into Cancer and Nutrition (EPIC). It is part of what the author is using to demonstrate his point that fruits and vegetables may not be effective, at least in breast cancer. This portion of the study involved almost 300,000 women from eight different European nations (3). Results showed that there was no significant difference in breast cancer occurrence between the highest quintile of fruit and vegetable consumption group compared to the lowest. The median duration was 5.4 years.
Does this study place doubt in the diet approach to cancer? Possibly, but read on. The most significant strength was its size. However, there were also many weaknesses. The researchers were trying to minimize confounding factors, but there were eight countries involved, with many different cultures, making it almost impossible to control. It is not clear if participants were asked what they were eating more often than at the study’s start. Risk stratification was also not clear; which women, for example, might have had a family history of the disease.
Beneficial studies with fruits and vegetables
Also using the same EPIC study, results showed that fruit may have a statistically significant impact on lung cancer (4). Results showed that there was a 40 percent decrease in the risk of developing lung cancer in those that were in the highest quintile of fruit consumption, compared to those in the lowest quintile. However, vegetables did not have an impact. The results were most pronounced in the northern European region. I did say the answer was complex.
Ironically, it seems that some other studies, mostly smaller studies, show potentially beneficial effects from fruits and vegetables. This may be because it is very difficult to run an intensive, well-controlled, large study.
Dean Ornish, M.D., a professor of medicine at UC San Francisco Medical School, has done several well-designed pilot studies with prostate cancer. His research has a focus on how lifestyle affects genes. In one of the studies, results of lifestyle modifications showed a significant increase in telomere length over a five-year period (5).
Telomeres are found on the end of our chromosomes; they help prevent the cell from aging, becoming unstable and dying. Shorter telomeres may have an association with diseases, such as cancer, aging and morbidity (sickness). Interestingly, the better patients adhered to the lifestyle modifications, the more telomere growth they experienced. However, in the control group, telomeres decreased in size over time. There were 10 patients in the lifestyle (treatment) group and 25 patients in the control group — those who followed an active surveillance-only approach.
In an earlier study with 30 patients, there were over 500 changes in gene expression in the treatment group. Of these, 453 genes were down-regulated, or turned off, and 48 genes were up-regulated, or turned on (6). The most interesting part is that these changes in gene transcription occurred over just a three-month period with lifestyle modifications.
In both studies, the patients had prostate cancer that was deemed at low risk of progressing into advanced or malignant prostate cancer. These patients had refused immediate conventional therapy including hormones, radiation and surgery. In both studies, the results were determined by prostate biopsy. These studies involved intensive lifestyle modifications that included a low-fat, plant-based, vegetable-rich diet. But as the researchers pointed out, there is a need for larger randomized controlled trials to confirm these results.
A meta-analysis involving a group of 24 case-control studies and 11 observational studies, both types of observational trials, showed a significant reduction in colorectal cancer (7). This meta-analysis looked at the effects of cruciferous vegetables, also sometimes referred to as dark-green, leafy vegetables.
In another study that involved a case-control observational design, cruciferous vegetables were shown to significantly decrease the risk of developing multiple cancers, including esophageal, oral cavity/pharynx, breast, kidney and colorectal cancer (8). There was also a trend that did not reach statistical significance for preventing endometrial, prostate, liver, ovarian and pancreatic cancers. The most interesting part is that the comparison was modest, contrasting consumption of at least one cruciferous vegetable a week with none or less than one a month. However, we need large, randomized trials using cruciferous vegetables to confirm these results.
In conclusion, it would appear that the data are mixed in terms of the effectiveness of fruit and vegetables in preventing cancer or its progression. The large studies have flaws, and pilot studies require larger studies to validate them. However, imperfect as they are, there are results that indicate that diet modification may be effective in preventing cancer. I don’t think we should throw out the baby with the bath water.
There is no reason not to consume significant amounts of fruits and vegetables in the hopes that it will have positive effects on preventing cancer and its progression. There is no downside, especially if the small studies are correct.
(1) CA Cancer J Clin. 2011;61(4):212. (2) Nat Rev Cancer. 2008;8(9):694. (3) JAMA. 2005;293(2):183-193. (4) Int J Cancer. 2004 Jan 10;108(2):269-276. (5) Lancet Oncol. 2013 Oct;14(11):1112-1120. (6) Proc Natl Acad Sci U S A. 2008 Jun 17;105(24):8369-8374. (7) Ann Oncol. 2013 Apr;24(4):1079-1087. (8) Ann Oncol. 2012 Aug;23(8):2198-2203.
Dr. Dunaief is a speaker, author and local lifestyle medicine physician focusing on the integration of medicine, nutrition, fitness and stress management. For further information, go to the website www.medicalcompassmd.com or consult your personal physician.