Colon Cancer and Diet: Devil in the Details
by Marty Root, Ph.D.

The April 20 New England Journal of Medicine published two reports concluding that high fiber supplements and a low-fat high-fiber diet did not protect against recurring adenomas polyps, a precursor of colon cancer. These results are disappointing in light of the high rate of colon cancer incidence and death in this country. These reports share similar results with earlier smaller studies of dietary interventions for polyps. However, much excellent research over the years has strongly suggested that dietary patterns practiced in other parts of the world are indeed protective and that countries that consume a Western-style diet high in meat and low in vegetables have high rates of colon cancer. We will examine these two new reports more closely and then ask where the field of dietary prevention of colon cancer stands at this point.

Two Studies: A Brief Summary

In the high-fiber wheat-bran study a number of people who had recently had an intestinal polyp removed were asked to consume either a high fiber or a low fiber breakfast cereal or snack for 4 years. Their colons were again examined for polyps and they found no difference between the high-fiber group and the low-fiber group for recurring polyps.

In the more comprehensive second study, people with polyps recently removed were asked to be in a study of diet and polyp recurrence. Half were given extensive instruction and encouragement in changing their eating habits to a low-fat high-fiber diet that might reduce recurring polyps; the others were not. After 4 years they found no difference between the two groups.

Both were expensive and difficult studies that started with high hopes that certain changes in diet could reduce the high colon cancer incidence in the Western countries.

What Went Wrong?

There is no reason to doubt the conclusions of the studies, at least for what they were designed to test. Both were well devised, executed and reported. The basic problem with these studies appears that the dietary changes were "too little too late" and that more far-reaching changes in diet--shown in other research to have measurable effects on colon cancer -- were not undertaken and studied here.

The wheat-bran study asked subjects to change their eating habits to include a breakfast cereal or snack bar that would give an extra 2 grams (control group) or 13.5 grams (treatment group) of fiber per day. But there were a few problems. First, only about one-third of the subjects completed the study. Then, while researchers aimed for an 11.5 gram difference in fiber intake, a dietary analysis of the subjects near the end of the study showed that the difference was only 9.4 grams -- only about a 50% increase over their starting level of 19 grams per day. Dietary fiber intake in developing countries where colon cancer is low is often 30-50 grams per day. Next, the dietary changes required them to eat a breakfast cereal, presumably with milk, and to eat snack food. In general these would not really be considered an entirely healthy habits. This was not so much a dietary supplement as a dietary addition or alteration.

One problem with this kind of study is that diets are not like medicines. In a study of a new medicine, the control group would take no medicine. However, in dietary studies can we ask the control group to eat no diet? People will always eat something and giving them something new to eat means that other compensating changes get made in the diet. This makes interpretation of dietary experiments difficult.

This was also an older group with an average age of 66. This means that many subjects in the study probably started eating their current diet when Herbert Hoover was in the White House. Will just four years of a sugary fibrous cereal really make an impact on this lifetime dietary pattern? Compare this to a typical man in China who has been eating rice and vegetables for as many years and does not spend a minute of his waking time worrying about colon cancer. This wheat-bran dietary treatment was probably "too little too late."

Cultural Bias and the Western “Low-Fat, High-Fiber” Diet

The second study was much more difficult and potentially more rewarding. But the results were the same: no dietary effect on recurring polyps. Getting people to change their diets is difficult. To their credit these researcher got their subjects to dramatically change their diet to a healthier low-fat high-fiber diet. They saw a 10% drop in fat intake to an astonishingly low level -- for Americans -- of 24% of calories. Fiber intake went up about 70% to 35 grams per day. Fruit and vegetable intake increased about 2 servings per day and the subjects lost 2 1/2 pounds of weight over 4 years compared to the control group.

The problems they had with their study are the typical problems of doing dietary research in the US. The diet-colon cancer connection was first suggested in studies observing African populations, where colon cancer rates are very low. It was also based on numerous other observational studies showing that when individuals in these low-risk populations adopt a Western diet, they assume the same far higher risk of colon and other cancers found in Western countries. Thus it would seem to make sense to test the dietary habits of these low-risk populations against the common high-risk dietary pattern in the US to determine how the non-Western diets are protective. However, this study did not, and probably could not, attempt this.

While the researchers, familiar with the international studies, may have wanted to get their subjects to adopt a dietary pattern that looked like rural Africa or China, they knew that this was impossible given the subjects they had to work with. They chose only to test what actually amounted to modest changes compared to the non-Western low-risk diets, since if they put these older American subjects on the vegetarian or near-vegan diets found to be protective in Asia and Africa, the subjects might drop out of the study.

The relatively modest change in diet habits of the study participants explains why their cholesterol levels had almost no change. They were still eating their meats. On average the treatment group subjects were eating one less mouthful of meat (15 grams) per day compared to controls and dairy intakes were not even reported. Their serum carotenoid levels were not increased much, indicating that they were not eating many additional orange, yellow, or green vegetables and fruits.

More Questions than Answers

The authors themselves gave several reasons why they did not get the results they were hoping for. The study examined the recurrence of intestinal adenomas. While everyone agrees that some of these growths may eventually turn cancerous, this is only one part of a long process of getting and dying from cancer. Maybe fiber and low fat have their primary effects on some other, probably earlier, stage of carcinogenesis. For example, higher fiber and lower fat have been shown to decrease the transport of mutagens to the colon cells, a very early step in the cancer process. Other biochemical studies support this view.

Furthermore, the treatment group knew that they were supposed to be eating a low-fat high-fiber diet. Therefore when the researchers came around to ask them what they were actually eating, they probably tended to err in their reporting on the healthy side. Therefore the published fiber intakes may have been overestimates of actual intake.

Were the changes the researchers were asking for from their subjects enough to see an effect? Probably not. Could they have asked their subjects to make even bigger changes? Probably not.

Did the researchers choose the right nutrients and foods to modify? Numerous medical studies indicate that low meat consumption reduces risk of colon cancer. If the researchers had focused on reducing meat, dairy, and eggs rather than increasing fiber, they might have seen a reduction in polyps.

It is possible that preventing colon cancer with fiber and low fat should be done in the teenage years -- when these study participants (average age of 61) were busy dancing in bobby socks. Many of these folks were also overweight, with an average BMI of 27. This is another problem of doing research in a country where the average adult is often more than 20 pounds overweight. Being overweight in itself is a known risk factor for colon cancer.

Thinking Outside the Box

There is a possible analogy between these studies and a study completed a few years ago in Finland. Smokers were randomized to receive, or not receive, a dietary supplement of beta-carotene that was widely believed to prevent lung cancer. Not only did the beta-carotene not have an effect -- but it appeared to actually increase the rate of lung cancer in these smokers. While the researchers suggested several reasons why the beta-carotene did not work, one explanation that was widely accepted is that smokers are so "predestined" to have lung cancer that a simple pill given very late in the process is just overwhelmed by the cancer. What we are seeing with fiber and colon cancer may be the same situation -- the cancer process is too far advanced to be stopped by modest dietary changes.

Do we wish that the results of these studies were different? Yes. Much previous research supports the hypotheses they were testing and we, as a nation, need to really get serious about this disease. Are we surprised at these negative results? Not really, after a careful reading of the reports. The diets, which appear protective in Asia and Africa and in studies of low-risk groups in the US -- are shown to be vegetarian or near-vegan diets. These are the diets that should be studied further, now that it seems clearer that a modest modification in the Western diet is not protective. Merely adding some fiber and lowering some fat -- while continuing to consume large amounts of meat and dairy -- does not appear protective against colon cancer, according to these studies.

Researchers must begin to think outside the box, and get beyond their own (and our own) cultural and traditional dietary biases. After all, individuals at high risk of colon cancer might happily adopt a no-meat and no-dairy diet if large, well-funded studies like these confirmed the protective quality of a vegan diet. But where and how to do these studies remains a problem.

I invoke the mantra of all researchers; "More research is needed."

Comments? Please post them on the New Century Nutrition Discussion board.

Marty Root has a PhD in Nutrition from Cornell University, and works as a Senior Research Scientist at BioSignia, Inc. in Chapel Hill, NC. His work involves making statistical models that predict the onset of chronic diseases such as heart disease and the cancers. He also writes for New Century Nutrition. Dr Root is married and has three children.