Data Review Shows Fruits and Vegetables Can Block Major Cancers

September 1, 1996

When organizers asked Tim Byers, MD, MPH, to speak at the American Cancer Society's National Conference on Cancer Prevention and Early Detection, he liked their proposed title for his talk--"Nutrition: The Data Are There."

When organizers asked Tim Byers, MD, MPH, to speak at the AmericanCancer Society's National Conference on Cancer Prevention andEarly Detection, he liked their proposed title for his talk--"Nutrition:The Data Are There."

"I thought that was pretty fair," said Dr. Byers, ofthe University of Colorado School of Medicine, Denver. He thenproceeded to drive the point home as he reviewed some 150 studiesof nutrition's role in upper gastrointestinal (including oral,esophageal, and gastric), colorectal, breast, prostate, and lungcancers. The diseases included in his review of large case-controland prospective studies account for almost two-thirds of the cancermortality in the United States.

British and US studies, he noted, indicate that nutrition andtobacco contribute equally to cancer's toll, with about one-thirddue to nutrition and one-third resulting from tobacco. "Amongstthe three quarters of Americans that don't smoke, nutrition becomesthe most important preventable cause of cancer," Dr. Byerssaid.

"I haven't tried to be selective in terms of the outcome,only selective in terms of the study size and design," hetold the conference, sponsored, in part, by the Centers for DiseaseControl and Prevention.

Upper Gastrointestinal: Thirty-four studies--done at differenttimes, by different investigators, in different countries, indifferent populations, using different methods of dietary intake--"consistentlyshow a 30% to 40% reduction in risk for those who eat five servingsof fruits and vegetables a day vs those who eat less than twoa day," Dr. Byers said.

These results are independent of tobacco and alcohol use, he noted."So there is something healthful about fruits and vegetables."

Colorectal: His analysis of 27 "of the better studies"suggested a risk reduction from a high fruit and vegetable dietalso on the order of 30% to 40% on average. "In fact, therisk reduction for colon cancer may be substantially more than50%, but conservatively, I think, there's going to be a halvingof risk," Dr. Byers said.

Conversely, "the strongest studies" suggest a doublingof risk from a diet high in fat. "Epidemiologic studies,despite their uncertainties, paint for colorectal cancer a fairlycompelling picture of a high-fat diet that generates increasedrisk and a high fruit and vegetable diet associated with decreasedrisk," he said. "Fruits and vegetables do reduce cancerrisk at all the GI sites, from the mouth to the anus."

Breast: Clearly, the issue of whether a high-fat diet increasesbreast cancer risk stands today as one of the most controversialnutritional questions. A summary of 22 studies reveals a mixedpicture, with some showing an increased risk on the order of 50%and some showing a reduced risk, he reported. "The best evidenceto date is that modest reduction in fat intake, of the types werecommend for colon cancer, heart disease, and weight control,will not appreciably affect the breast cancer risk," he said.

However, this does not mean dropping fat intake to 20% of calories,as in the National Institutes of Health's Women's Health Initiativetrial, will have no effect, he stressed. "There are treatmenttrials of very low fat diets that I think are very reasonableexperiments to do."

As for fruits and vegetables, the data suggest a slight protectiveeffect for breast cancer, but much weaker than for colon and lungcancer.

Alcohol intake presents quite a different story. An updated meta-analysisshowed a 40% increase in breast cancer risk for women who consumealcohol, Dr. Byers said, although a number of studies suggestthat moderate drinking helps protect against coronary heart disease.

"Despite the fact that one of my studies is one of the negatives[in showing an alcohol-cancer risk], I am convinced that alcoholis a risk factor for breast cancer, but I am equally convincedit's a protective factor for heart disease," he said. "Ithink women who are at high-risk of breast cancer could reasonablymake an informed decision not to drink alcohol. For women whoare at average risk for breast cancer and heart disease, it'sa trade-off."

Prostate: In his review of 12 prostate studies, Dr. Byersfound them inconsistent and indicative that fruits and vegetablesdo little or nothing to ward off prostate cancer.

With regard to dietary fat, however, "the strongest studiessuggest about a 30% to 40% elevation in risk for prostate cancerfor being in the upper quintile compared to the lowest quintileof dietary fat intake. Dietary fat is a risk factor for prostatecancer and is probably mediated through fat's effects on testosteronemetabolism."

Lung: His analysis of 31 studies concluded a link existsbetween diet and lung cancer. All studies that controlled fortobacco exposure revealed "a fairly consistent lower riskfor those in the upper quintile compared to those in the lowerquintile of fruit and vegetable intake in the diet," he said."And, indeed, there is a relationship as well with fat, butI don't think it's nearly as strong."

A number of substances in food have been suggested as cancer preventers.But Dr. Byers noted that studies to date have failed to identifyany single one as a "magic bullet." He cited six majorstudies of beta-carotene. Only one, carried out in China in apopulation with high gastric cancer rates and micronutrient deficiencies,showed any benefit--and it used a combination that included seleniumand vitamin E as well as beta-carotene. Two studies actually suggestedthat beta-carotene may have increased lung cancer risk.

Dr. Byers believes that these apparent ill effects of beta caroteneconfirm "that there is something beneficial in the naturalgoodness of whole foods and when you disturb that balance, youactually may trigger cancer to occur."