Cancer Expert Faults Reductionist Thinking in Diet-Cancer Research

American Institute for Cancer Research
Friday, 9 June 2000

Urges New Directions in Study of Diet-Cancer Connection

In a "State-of-the-Research" presentation at the May 29 National Nutrition Summit, cancer researcher Cheryl Rock, Ph.D., R.D, of the University of California at San Diego's Cancer Prevention and Control Program and the Department of Family and Preventive Medicine, outlined the progress of investigations into the diet-cancer connection and described some inherent limitations of dietary analysis.

Addressing a roomful of summit attendees, Dr. Rock specifically decried a "reductionist" research approach that, for example, fails to address how compounds within foods like vegetables, fruits, whole grains and beans work together in complex ways to fight cancer. "Any study of the diet is difficult," she said, "because of the complex ways that foods - and the natural substances they contain - interact. Scientific studies have a hard time addressing themselves to this kind of interaction, because their focus is often too narrow."

"The Focus should be on Food, not Nutrients"

Dr. Rock pointed out that the scientific method itself is based upon the concept of analysis, a word that derives from the Greek "to break apart." "Science tells us to test one factor of the diet so we can isolate and quantify its effects, and then move on to another. And when we try this, when we give human subjects isolated nutrients like vitamin E, beta-carotene or fiber, we tend not to see cancer-protective effects."

According to Dr. Rock, the failure of these so-called "single-agent" studies is not surprising. "The ultimate goal is to understand the influence of the whole diet upon human cancer. So even if these trials with isolated nutrients and phytochemicals started showing consistent cancer-protective effects, we still couldn't say much about how they work in the human diet. After all," she said, "the focus should be on food, not nutrients."

Every Kind of Scientific Study Has Limitations

Rock said that much of the evidence linking diet to cancer risk is based on the results of so-called epidemiological studies, which compare the diets and cancer rates of groups of individuals or even whole populations. Results of these studies have consistently associated diets high in fruits, vegetables, whole grains and beans with lower risks for many cancers. But epidemiological data are only a first step, in that they can suggest associations, not pinpoint specific causal relationships.

Laboratory tests - which involve cell or tissue cultures and animal models - have so far uncovered several biological processes ("mechanisms") by which diet may reduce the risk or possibly slow the progression of cancer. Although these results may or may not correspond to human biology, such findings do strongly support the mounting evidence for a diet-cancer link.

To date, most of the clinical human trials that have tested the influence of diet upon cancer risk have involved single nutrients. Dr. Rock believes the negative results of these tests indicate that single-agent studies may be the wrong approach.

Even studies that investigate the effects of a single food prove problematic, she said. "Whether you're isolating nutrients, or isolating foods, you come up against the same basic issue. Eating more of one type of food is always associated with eating less of another type of food."

Another consideration is that these studies often rest upon self-reported dietary data; test subjects record food intake themselves, without independent verification. In self-reported dietary studies, subjects have been documented underreporting the amount and type of food they eat by as much as 30%.

"There is a shift underway in the study of diet," said Dr. Rock, "one that applies new methods and measures to study the effects of overall dietary patterns."

Strategies for the Future

Researchers are currently working to identify biomarkers - measurable biological factors that can be used to precisely track and measure dietary change without having to rely on self-reported food intakes. One such biomarker is the recent discovery that a person's fruit and vegetable intake can be reliably determined by examining the level of carotenoids in his or her blood.

Such biomarkers are already offering new insights. In one recent study published in the New England Journal of Medicine, researchers attempted to gauge the effect of prescribing a low-fat, high-fiber diet upon the recurrence of colon polyps. After four years, subjects in the so-called "Polyp Prevention Trial" who were assigned the prescribed diet experienced the same degree of recurrence as subjects who followed a "normal" diet.

These findings were widely interpreted as "proof" that diet has no effect upon colon cancer. In her talk at the National Nutrition Summit, however, Dr. Rock pointed out that the biomarkers utilized in the study - blood carotene levels - indicate that, despite any differences in reported intakes of fat and fiber, most of the study subjects do not appear to have increased their vegetable and fruit intake by any significant degree.

The ability to verify dietary data is only one advantage of using biomarkers as measures in clinical studies. They also help scientists design new research approaches that reflect the real-world influences of diet upon the human body.

Dr. Rock went on to say that more research needs to be devoted to studying the interplay of diet and genetics. Different people experience different levels of cancer risk because of interactions between their unique genetic makeup and environmental factors such as diet. As scientists continue to map the human genome and identify more genes that carry a predisposition to certain cancers, the study of dietary influences will gain new insights and advances.

"When it comes to cancer, genetics and diet are inextricably linked," said Dr. Rock. "Our efforts to understand and combat the disease should reflect that fact."

Urgent Need for Accurate, Complete Science Reporting

Finally, Dr. Rock addressed public perceptions of diet-cancer research. "It is crucial that people understand how this work is progressing," she said, "but it is equally important that this information come within a clear, understandable context. It's not enough to merely report what the latest study says. We have to help people understand how to evaluate all of this incoming information, how to weigh it against what has gone before."

Brief reports about the often conflicting results of single-agent studies promote public confusion, frustration, and, ultimately, a feeling of helplessness. "And that's a shame, because right now we know something very important about cancer: people can take small, positive steps to lower their risk. They can be physically active, maintain a healthy weight, avoid smoking, and consume a diet high in vegetables and fruit, whole grains and beans. That message is too important to get muddled by incomplete or inaccurate perceptions. "When you look at all the research, the best advice to lower colon cancer risk remains the same: eat plenty of vegetables, and get regular physical activity."

For more information, or to contact American Institute for Cancer Research, see their website at: www.aicr.org