Russell

Colorectal cancer continues to be a major public health issue throughout the developed world. In the U.S. alone, approximately 120,000 new cases are diagnosed and 55,000 Americans die from this cancer each year. It continues to be the second most common cause of cancer death in the U.S., as well as in many other countries. Only very modest declines in the death rates from colon cancer have been realized, despite decades of supposed advances in surgery, radiotherapy and chemotherapy. This raises the real question of whether we ought to be expending our efforts to prevent the cancer, as opposed to treating it.

Since colorectal cancer, among all the major cancers, is thought to be the one that is most intimately tied to dietary habits, there is considerable interest and research going into ways in which the cancer might be prevented through either alterations in dietary habits or by dietary supplementation.

Joel, you are an expert and researcher in the area of colon cancer prevention. Despite two recent, well publicized studies which we'll discuss, that seemed to cast doubt on the role of dietary fiber, I think you'd agree that just about everyone accepts the concept that certain dietary patterns appear to be protective against colon cancer. This beneficial dietary pattern, in general, would be a diet high in fruits and vegetables, high in fiber, low in calories and low in animal fat. Today, we will be talking about specific components of the diet that may underlie some of the protective effects seen in the above, general "healthy" dietary pattern.

Among these components are calcium, vitamin D, folate, vitamin E, fiber, selenium and garlic. Before getting into the specific components, I would like to ask you a few questions about the link of animal fat and red meat intakes to colon cancer. I think there is some fairly strong epidemiologic evidence that a diet high in saturated animal fat and red meat (beef, pork and lamb) is detrimental.

But the question is, how much animal fat are we talking about and how many servings of red meat are acceptable? What are the guidelines that the doctor should be giving to patients about this aspect of their diets?

Mason

There are considerable observational data, as well as animal studies, to support the premise that red meat and saturated animal fat are independent risk factors for the development of colorectal cancer. And it is worth emphasizing that the red meat component of a diet seems to operate as an independent risk factor apart from the mere fat content contained in that meat. Roughly speaking, a diet containing less than 40 grams of animal fat per day conveys one-half the risk of colon cancer of a diet containing >65 grams per day. Similarly, a diet containing <60 grams of red meat per day has half the risk of a diet containing >130 grams per day.

It is also worth mentioning that it is rather clear that heavily cooked meat, which produces a brown or black surface on the meat, contain a lot of heterocyclic amines, which in an in vitro setting are clearly mutagenic. Again, there are some observational epidemiologic data that meat which is heavily cooked might further increase the risk of cancer, although interventional human studies haven't been done in this regard.

Fiber: Has Its Glow Faded?

Russell

There has been quite a bit of interest in dietary fiber as being a protective component of the overall healthy diet that I spoke about above. However, a couple of recent intervention studies have appeared to throw a monkey wrench into the gears and caused quite a bit of shock and discomfort in people who have been studying this relationship.(1),(2)

Joel, could you expand on these recent studies as to the role of fiber or the lack of role of fiber in protecting against adenoma polyp recurrence?

Mason

Two, back-to-back papers presented in The New England Journal of Medicine (Spring 2000) each described a well-designed clinical intervention trial with subjects who had had a prior colonic adenoma. In the Arizona Polyp Trial,(1) approximately 1300 subjects were followed prospectively over a period of approximately three years. The subjects were divided into two groups -- one group received a low wheat bran fiber supplement on a daily basis, providing a total of 18 g of dietary fiber from the supplement and from their diets each day, while the other group received a high fiber supplement which provided a total of about 28 g/fiber/day when added to the dietary intake (the wheat bran supplements amounted to 2.0 and 13.5 g per day, respectively). After three years, there was no significant difference in the recurrence rate of adenomas in these two groups of individuals.

Similarly, in the National Polyp Trial,(2) approximately 1900 individuals were prospectively examined over three years. In the intervention group, they were extensively counseled on the adoption of a high fiber, high fresh fruit and vegetable, and low fat diet; whereas, no such counseling occurred in the control group. The appropriate changes in diet in the intervention group in the latter trial were well documented and shown to be considerably different from their baseline diet. Again, there was no significant difference in adenoma recurrence in the intervention group compared to the control group.

Many individuals have assumed that these two trials have definitively proven that fiber is not effective in the chemoprevention of colorectal cancer. I would disagree.

First of all, the trials were only conducted over a three- to four-year follow-up period. The natural history of colorectal cancer is much longer. The evolution from a diminutive polyp to a frank colorectal cancer is thought about in terms of decades, not a period of just three to four years.

More importantly, the "end point" in both of these studies was recurrence of a colorectal adenoma, which is considered to be a predictor of subsequent occurrence of colorectal cancer. However, the concordance between the two is actually only modest. Proving that fiber does or does not prevent colorectal adenomas does not necessarily predict its ability to provide protection against colorectal cancer.

Given the huge body of intervention data, as well as some smaller observational trials that preceded these two studies that suggested a potential role for fiber in the prevention of colorectal cancer, I am not yet willing to close the door on the possibility that fiber might have a role to play. I continue to feel that if a health professional is going to counsel a patient regarding the dietary chemoprevention of colon cancer, it is reasonable, at this present date, to suggest that the individual increase their dietary consumption of fiber to between 25-30 g of fiber per day, emphasizing wheat bran. For most individuals, this should be done in a gradual manner, since the precipitous increase in fiber intake often leads to a variety of gastrointestinal symptoms.

Role of Calcium and Vitamin D

Russell

Calcium and vitamin D have also been of great interest with regard to their possible protective effects against colon cancer. But most of the intervention studies in patients with adenomatous polyps have concerned calcium. Nevertheless, there are a number of analytic or observational epidemiologic studies that have investigated the association between calcium and colon cancer.

Joel, would you expand on what we have learned from these various studies? Could you also address the issue of how important vitamin D is in prevention of colon cancer, in that vitamin D is intimately and obviously related to calcium metabolism?

Mason

There has been growing literature, over the past decade, indicating that increased amounts of dietary calcium are protective against either colorectal cancer or the adenomatous polyp. There is also a fair bit of experimental data in animal models which concurs.

Some small intervention trials have been performed in humans using surrogate endpoint biomarkers of colorectal cancer that have been suggestive of a protective effect of calcium, although there are also some negative studies in this regard. The endpoints in these studies were, largely, indicators of the rate of proliferation in the colonic mucosa and other related biochemical or histologic changes that are said to precede the neoplastic process itself.

The most compelling study in this field, at the present time, is one that was published in The New England Journal of Medicine by the epidemiology group at Dartmouth Medical School.(3) They administered, in a prospective, controlled, randomized fashion, 1200 mg of elemental calcium over a several year period, during which intermittent colonoscopies were performed. After approximately four years of follow-up, there was about a 15% decline in adenomas (highly statistically significant) in the intervention group, as compared to the placebo group.

We don't have a good handle on how this will translate into reduction of colorectal risk because, as I mentioned, the reduction in recurrence of colonic adenomas does not perfectly correlate with a reduction in colorectal cancer risk. Nevertheless, these observations do provide fairly convincing evidence that we can reduce the risk of a neoplastic process in the colon, at least to a modest degree, with calcium supplementation.

The issue of vitamin D has not been well studied. There are both observational studies in human populations, as well as supportive work in animal models of colorectal cancer, which suggest that supplemental levels of vitamin D might also convey a beneficial effect. However, randomized trials in humans have not been performed to validate vitamin D's beneficial influence. Nevertheless, the intimate interplay between vitamin D and calcium metabolism certainly suggests that vitamin D could play a similarly important role in chemoprevention.

Is Folate the Magic Bullet?

Russell

Joel, now I want to return to your true love, that is folate and colon cancer prevention.

There is certainly just as much interest in folate vis a vis colon cancer as there has been about calcium in the prevention of colon cancer. Can you tell us a bit about the theory that underlies the prevention of cancer by folate and also could you review what evidence there is in humans that folate exerts a protective effect?

Mason

Rob, I will plead guilty to being somewhat folate-centric but I have not gone so far as to modify Ptolemy's model to include folate at the center of the universe!

Interestingly, folate is a critical cofactor in the synthesis of both DNA and RNA, as well as playing other critical roles in metabolism. Therefore, it is not too surprising that folate inadequacies in the diet lead to certain abnormalities in these critical nucleotides. Since there is a considerable consensus that cancer arises because of insufficiently repaired defects in DNA, it is not hard to understand, at least on a superficial level, that inadequate amounts of folate might lead to DNA abnormalities and a subsequent enhanced risk of cancer.

The first human observations along these lines were made in the late 1980s in people with chronic colitis, who, as you know, are prone to folate deficiency, as well as to a several fold increase of colorectal cancer. Since then, there have been many observational epidemiologic studies that suggested a 50-60% decline in the relative risk of colorectal cancer or colorectal adenomas with higher amounts of folate in the diet, as opposed to those individuals who have very low amounts of folate in their diet. There are now a total of 18 such epidemiologic studies, and the consistency of the results among these different studies is impressive [reviewed in reference(4)].

Work in animal models of colorectal cancer has also been surprisingly consistent in showing a protective effect of dietary folate in the laboratory rodent. Intervention studies in humans are at a relatively early stage in this field, though there have been some small preliminary intervention studies that have been promising. Given the small size of these studies and the use of endpoints that are not yet validated, I think we certainly have less than definitive evidence that folate can prevent colorectal cancer.

To my knowledge there are now four prospective randomized trials that are underway, which are using either polyp recurrence or molecular markers in the colon as endpoints. Over the next few years, we ought to have results from those studies and, hopefully, this will give us a definitive answer about folate.

Antioxidants: Conflicting Data

Russell

The data on antioxidants in the prevention to colon cancer has, oftentimes, been conflicting. There have been a number of research reports implying reduced risk because of micronutrients, such as vitamin C or carotenoids, but other studies have failed to confirm these positive results.

In general, I think most researchers feel that the relationship between lower cancer risk and the consumption of these antioxidants is explained mostly by the consumption of diets high in plant foods As I said at the beginning, such diets are protective against colon cancer even though we don't know the specific components that are active in that protection.

For vitamin E, also, the data are mixed. This is an example where observational studies have led to interesting hypotheses but, when tested with intervention studies, these hypotheses turned out to be unfounded. I certainly wouldn't recommend any kind of high dose antioxidant supplements to individuals with the intent of lowering colon cancer risk.

However, I would like to consider selenium, an antioxidant for which there is some evidence of possible protection.

Joel, what do you think of the role of selenium at this point?

Selenium Shows Promise

Mason

Rob, you are right -- among all the antioxidant micronutrients, selenium, at the present time, seems to be the only one that shows promise in the prevention of colorectal cancer.

Aside from some provocative and supportive studies in animal models of colorectal cancer, there exists one large prospective randomized trial in which individuals were provided with either a 200 ug selenium supplement or a placebo over a period of several years. Although the primary endpoint of this trial was not colorectal cancer, there was a 25% decline observed in the occurrence of colorectal cancers during the course of this study.(5) This study, at present, stands alone but it is very compelling and certainly needs to be followed up with other studies.

If there is a strong intent to counsel a patient regarding a nutritional-chemoprevention of colon cancer, I think it would be reasonable to counsel the patient to take 200 ug of selenium per day. This is below the recommended safe upper limit of 400 ug/day. I should also mention that the Linxian study in China, which was a cooperative study conducted by China and the National Cancer Institute, showed a significant reduction in adenocarcinoma of the stomach with a combination of antioxidant micronutrients (vitamin E, beta-carotene and selenium). The applicability of findings from this rather unique population -- both in regard to its very high incidence of stomach cancer and high prevalence of micronutrient deficiencies -- to the American population is, however, questionable.

Russell

Are there any other food components that have been implicated in protection against colon cancer, realizing that there are not nearly sufficient data to make definitive statements?

Is Garlic Good for You?

Mason

Rob, some items that are currently under investigation include garlic, which in a large observational study (The Iowa Women's Health Study]) was shown to be a significant independent predictor of colorectal cancer. There are, in addition, several animal studies using the reported active component of garlic, allium isothiscyanate, that support a protective effect. Nevertheless, I think the data fall far short of being convincing. Thus, I would not make a recommendation to use garlic as a chemoprotective factor.

Lastly, it is worth remembering that aspirin, as well as a lot of other non-steroidal inflammatory agents, have now been shown to be chemopreventive against colorectal cancer. In this regard, it is useful to note that various different forms of salicylates, even acetylsalicylic acid, are contained in certain food groups, although foods high in salicylates, such as raisins, have not been examined carefully for their protective effects.

What About Exercise?

Russell

Joel, another promising protective factor that is not strictly part of the diet, but is related to a healthy diet, is that of exercise. To my mind, this relationship has never really been explained, although it seems rather robust. Do you have any theories about how exercise might influence the rate of colon cancer?

Mason

That is a wonderful question but there's still no definitive answer. On the one hand, physical activity seems to be protective because a physically active lifestyle frequently leads to less obesity. People who are obese have higher circulating levels of insulin. One of the theories that is prominent for the role of obesity, as well as for the role of physical activity in modulating the risk of colorectal cancer, is that an excessively high and chronic level of circulating insulin serves as a trophic factor for the colonic epithelium. In some studies, physical activity is an independent risk factor even when the phenomena of obesity or body weight is taken into account.

Theories as to how physical activity might modulate colorectal cancer risk in the absence of body weight also pertain to alterations in other hormones. It is known, for instance, that exercise can modify some of the circulating levels of sex hormones and since colorectal carcinogenesis is thought, in part, to be suppressed by estrogen, scientists surmise that, perhaps, physical activity modulates cancer risk in this regard.

Alcohol and Caffeine

Russell

Our discussion wouldn't be complete unless we talked about some of the other components of the diet that have been implicated as causing an increased risk in colorectal cancer. We have mentioned heterocyclic amines from meat preparation but also I am wondering about alcohol and caffeine. There is some evidence that these substances are associated with increased risk of colorectal cancer. With regard to alcohol, I was wondering if you feel that relationship may be tied into folate metabolism or an interruption in folate metabolism?

Mason

There are sufficient data to strongly suggest that alcohol is an independent risk factor for colorectal cancer, particularly in regard to cancers of the distal sigmoid colon or rectum. As you mentioned, alcohol is well known to interfere with folate metabolism in a number of ways and there are some data in animals, as well as in humans, suggesting that excessive and chronic alcohol use might specifically inhibit folate metabolism in the distal colon.

Regardless of how alcohol exerts its influence, it is worth counseling patients that adherence to the U.S. Dietary Guidelines -- no more than one alcoholic drink for women and two drinks a day for men -- would be a wise not only for general health purposes but for the prevention of colorectal cancer.

In regard to the use of caffeine or coffee or tea, I think the data that largely exist at this point are observational and not consistent enough to make any suggestions in regard to whether modifications in coffee ought to be made.

Russell

Joel, this has been an interesting conversation but I am wondering what advice one should give to a patient who has, for example, a family history of colorectal cancer? What can this person do for himself or herself to try to lower the risk that he or she would get colorectal cancer?

Mason

First, in regard to dietary counseling to the patient, I would first suggest a high fruit and vegetable diet (5+ servings/day, emphasizing fresh vs. cooked items), one that is high in whole grain fiber and low in red animal meat and saturated fat. It would also be worthwhile for this individual to take a multivitamin each day that provides an additional 400 ug of folate and sufficient vitamin D. Finally, it would be reasonable to suggest that they take a 1200 mg supplement of elemental calcium, as well as a 200 ug supplement of selenium.

It is important, moreover, that patients aim for a desirable body weight, both through the use of regular physical exercise and through eating a modest diet that is appropriate in calories. I'd also recommend adherence to the U.S. Dietary Guidelines for alcohol consumption. Lastly, although it is not directly applicable to dietary interventions, such individuals could potentially benefit from taking a single 325 mg tablet of aspirin, three or more times a week. The other on-steroidal that has been extensively tested is sulindac which has also been quite promising in clinical intervention trials.

I would like to emphasize that all of these interventions probably require adherence over a period of many years, most likely at least a decade, to be truly effective. Patients ought to be made aware that protection requires adoption of a long term change in lifestyle rather than one that just lasts a matter of weeks or months.