Fiber for the Prevention of Colon Cancer: SOL

“The hypothesis that fiber intake is important in reducing colon cancer risk was interesting, but the reality is that the data have not provided much support for that”

–Walter Willett, M.D., Dr.P.H. (professor of epidemiology and nutrition at the Harvard School of Public Health, Boston).

In this Article:

  • A History of the Hypothesis that Fiber Protects Against Colon Cancer
  • The First Major Trials
  • The Lyon Diet Heart Study: Supportive?
  • The NEJM Publishes Two Strikes Against Fiber, the Lancet Published One
  • Japanese Researchers Provide Additional Evidence Against Fiber
  • Conclusion

A History of the Hypothesis that Fiber Protects Against Colon Cancer

Let’s start off with a condensed history of where the idea that fiber might protect against colon cancer is thought to have come from before we look at clinical trials testing such an idea.

In the early to middle 1900’s, a Surgeon named Denis Burkitt studied the eating and disease habits of England and some African countries and came to suspect fiber was a protective element against colon cancer, given the relative scarcity of the disease in the high fiber eaters of Africa compared to the U.K. and other countries where lower fiber diets were common. Burkitt’s work is often credited with popularizing the idea that fiber could prevent colon cancer.

It’s a simple conclusion, easy to understand, not difficult to believe. Additional, better observational evidence, e.g. prospective cohort studies, would also strengthen this theory by finding colon cancer more likely to develop in people following low fiber diets. Additionally, animal research would demonstrate some types of fiber would protect rats and mice from colon cancer induced by cancer causing chemicals (i.e. azoxymethane).

So far it seemed quite reasonable. But these forms of evidence are simply not enough. Randomized, controlled trials on human beings were needed.

Thankfully, such studies were conducted.

The First Major Trials

In 1989 what was arguably the first study published examining the role of fiber in colon cancer prevention observed that subjects with a genetic predisposition to colon cancer saw beneficial effects on rectal polyp development from a wheat bran fiber supplement (10). However, due to the small number of subjects, poor adherence, and other oddities this study would seem to be weak evidence of fiber’s protective effect on colon cancer risk. Still, this study would be a promising start for a hypothesis whose run at clinical confirmation would prove incredibly awkward and ultimately unsuccessful.

In 1994 the Toronto Polyp Prevention trial would be published (1). The TPP study would report a very slight, non-significant increase in the return of neoplastic polyps with a low fat, high fiber diet. When broken down by gender the women had non-significantly less of these polyps when eating the low fat, high fiber diet and the men saw non-significantly more of these polyps following the low fat, high fiber diet. Due in part to the small number of subjects either or both of these results could have been due to chance. Additionally, this study did not study fiber’s effect alone, but with a low fat diet, which means any effect could be due to either factor.

In 1995 the Australian Polyp Prevention Study was published that finally attempted to evaluate fiber’s ability to prevent glandular tumors of the colon, aka adenomas (2). A low fat diet, B-carotene supplements, and dietary wheat bran (for fiber), were all tested alone and in combination to examine their effect on colon adenoma development. Unfortunately, the number of subjects was a bit small and perhaps partly as a result no significant effects on colon tumor incidence were found. A trend towards a negative effect of fiber on adenoma occurrence was seen, except when combined with a low fat diet, which showed a trend towards a positive effect. Again, these results were still not statistically significant. Clearly more studies were needed.

The Lyon Diet Heart Study: Supportive?

In 1999 the Lyon Diet Heart study was published (3,4). It observed a lower number of digestive tract cancers in a Mediterranean diet compared to a control diet lower in fiber (cancers: 1 vs. 4). Of course, a few things make this study a dubious one to support the alleged benefit of fiber:

A) Many dietary factors were changed beyond fiber intake, not the least of which was a decrease of linoleic acid intake in the Mediterranean diet group, which some evidence suggests may protect against cancer, including of the digestive tract (5). Other factors, like a greater intake of oleic acid or omega-3 fatty acids for example, may also have played a role.

B) This effect on digestive tract cancer may have due to statistical chance.

C) Those in the control group diagnosed with cancer actually had a fiber intake similar to the Mediterranean diet group (≈20 grams/day) whereas those in the control group who didn’t end up with cancer ate roughly 5 grams of fiber per day less than both the members of the control who did get cancer and the Mediterranean diet group.

Due to this, additional studies, more controlled and sufficiently powered, were needed.

The NEJM Publishes Two Strikes Against Fiber, the Lancet Publishes One

In the April of 2000 two studies were published in highly respected New England Journal of Medicine which would cast the protective image of fiber into doubt.

The Polyp Prevention Trial would examine the effect of a low fat, high fiber diet on colorectal adenomas in nearly 2000 people who had already had one (6). However, this study was changing more than just fiber intake. Subjects in the experimental group would end up eating more fiber (from whole grains, legumes, fruits and vegetables) and also less fat from, but not limited to, red meat. This study reported that such a diet had no effect on the return of colorectal adenomas and, additionally, colon cancer was non-significantly increased by roughly 250 percent (colon cancer rates were 10 in the low fat, high fiber group and 4 in the control group). These results would prove strikingly similar to the results of the Women’s Health Initiative several years later, which also found no reduction in the risk of colon tumors or colon cancer in women following a lower fat, higher fiber diet (7). The Women’s Health Initiative also reported a non-significant 26% increase in the rates of death from colorectal cancer.

However, as with the Lyon Diet Heart Study, both the Polyp Prevention Trial and the Women’s Health Initiative modified both fat and fiber simultaneously and thus should be interpreted cautiously when discussing only the effect of fiber. However, they do provide us with some interesting evidence regarding fiber in context.

Still, if we want a truly controlled study testing fiber the other NEJM paper, the Phoenix Colon Cancer Prevention Trial, fits that bill (8). The Phoenix Colon Cancer Prevention study recruited well over 1000 subjects who had previously had a colorectal adenoma and used wheat bran to increase the fiber intake of more than half the subjects. This study found fiber exerted no significant effect on the prevention of colorectal adenoma recurrence and, in line with the Polyp Prevention Trail, colon cancer was non-significantly higher in the high fiber group at a rate of 7 cases versus only 2 in the low fiber group.

A few months after the NEJM would publish these studies on fiber a similar study was published in the equally respected journal the Lancet. If any astute readers were wondering if wheat bran was perhaps a bad fiber source to test and wanted to see if a different type of fiber would show different effects in a controlled trial this study helps answer that question. Known as the European Cancer Prevention Orginisation Study, it used a supplement of ispaghula husk, or psyllium fiber, to see if it could prevent the return of colorectal adenomas in nearly 200 subjects compared to a control group receiving a placebo (9). If you were hoping for a happy ending you don’t know nutrition science: the fiber supplement increased the return rate of adenomas by a 67%.

Japanese Researchers Provide Additional Evidence Against Fiber

In 2005 Japanese researchers published a study examining the ability of fiber from wheat bran to prevent colorectal tumors (11). If you’ve been paying attention it should come as no big shock that by the end of the trial those in the wheat bran group had a not quite significant 31% increase in colon tumors.

Although the idea that fiber can prevent colon cancer was clearly questionable before, this study certainly did nothing to support such a crappy theory.

Cochrane Review: A Nail in the Coffin?

When a Chochrane Review analyzed the clinical studies looking at fiber’s effect on colon cancer (12) there no clear evidence fiber prevented tumor incidence. Furthermore, it was reported that a statistically significant increase in colon cancer rates occurred in higher fiber diets when two of the trials were assessed together.


There are a few main things I hope people take away from this post:

1. None of these studies looked at the effects of very low fiber diets (<10 grams/day) on colon cancer risk. It may be at this level fiber becomes protective against colon cancer. It might or it might not, but we simply can’t say based on this evidence.

2. I’m more trying to cast doubt on a theory than suggest an alternative one. If you take from this that “fiber” is bad for colon cancer risk you do so of your own accord. I’m primarily trying to get people to forget about “fiber” and focus on other factors when they make food choices to prevent colon cancer. Right now more evidence is needed before we can speak of “fiber” as being good or bad for cancer risk. There are many types of fiber which may have different effects, so we should be careful not to throw the baby out with the bath water. Insoluble fiber sources like wheat bran and psyllium husk? Not helpful. Other types of fiber like inulin and resistant starch? Hard to say right now.

*The intermittent replacement of the terms colon and colorectal are used in this post for simplicity, despite their technically being different terms.




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s