Colorectal Cancer in Primary Care

The incidence of colorectal cancer is highest in Western populations. It is thought that the reason for this is largely modifiable lifestyle factors. Primary care practitioners are uniquely positioned to influence the habits of their patients. Screening and early diagnosis have profound effects on the prognosis of individuals with colorectal cancer, and recognizing those at high risk may save lives through early detection. In addition, many evidence-based nutritional interventions exist that may reduce the risk of developing colorectal cancer in those at high risk. 

 

Colorectal cancer (CRC) is the third most common cancer¹ and the second leading cause of cancer deaths in the United States.² Overall risk of colon cancer is approximately 5% in the general population without family history, with 92% of colorectal cancers diagnosed after age 50.³ Unfortunately, 20% of those diagnosed with colon cancer are diagnosed in late, less treatment-responsive stages.³ It is estimated that routine screening of everyone over the age of 50 could save 18,800 lives annually.⁴ Primary care practitioners are uniquely positioned to influence adherence to screening procedures, recognize those at highest risk of disease, and address modifiable risk factors in their patients.

 

Those at greatest risk for colon cancer are carriers of 1 of 2 familial genetic disorders, familial adenomatous polyposis (FAP) or hereditary non-polyposis colon cancer (HNPCC), also known as Lynch syndrome. These genetic disorders account for 1–2% and 3–5% of all colorectal cancers, respectively. While the incidence of colorectal cancer cases directly attributable to these genetic syndromes is small, they are highly penetrant, with a lifetime risk of 100% and 80% of developing colorectal cancers in those with FAP and Lynch syndrome respectively.⁵

Inflammatory bowel diseases (IBD), including Crohn’s disease and ulcerative colitis, also raise the risk of colorectal cancers considerably. While IBD-related colorectal cancers account for only 1–2% of all CRC cases, CRC accounts for 10–15% of deaths in patients with IBD.⁶ Risk reduction for these patients should focus on controlling the underlying disease, thus reducing the chronic inflammation that is an integral part of the carcinogenic process. 

In addition to genetic syndromes and IBD, established risk factors include age, personal history of adenomas, family history of colorectal cancer, diets high in fat and low in fruit/vegetables, obesity, sedentary lifestyle, smoking, and excessive alcohol intake.³ A recent meta-analysis has suggested that diabetes mellitus is also an independent risk factor.⁷ Up to 70% of colorectal cancers may be due to preventable habits involving diet and lifestyle.⁸ The National Cancer Institute (NCI) has created an online risk assessment tool to calculate the risk of colon cancer in patients over 50 years old.⁹ Identifying the risk factors relevant for individual patients provides an opportunity for clinicians to educate patients on how to modify their risk. 

Common symptoms of colorectal cancer include alterations in bowel habit, rectal bleeding, weight loss, abdominal pain, obstructive symptoms, and changes in the caliber/shape of stool (See Table I). The general nature and prevalence of benign abdominal complaints make recognizing which patients may have a malignancy very challenging. Interestingly, a recent review of 248 published papers found that the only symptoms with positive predictive value (PPV) for colorectal cancers were bleeding and weight loss.¹⁰ Another study found that 96% of patients that present with rectal bleeding to a primary care clinic do not have a malignancy.¹¹ However, there is greater PPV for colorectal cancer when rectal bleeding is accompanied by changes in bowel habits, or when unaccompanied by perianal symptoms.¹¹ The presence of severe anemia in those patients presenting with rectal bleeding also increases the PPV for colorectal cancer in a primary care setting.¹² One study from the United Kingdom showed that concomitant signs and symptoms with rectal bleeding doubles the likelihood of a colorectal cancer diagnosis.¹³

Symptoms of colorectal cancer may differ depending on the location of the tumor, from proximal to distal colon. Those presenting with proximal colon tumors may present without changes in bowel habits, as the highly liquid stool may pass easily around even large masses. Tumors of the sigmoid and/or rectum may result in blood that is bright red and easily mistaken by the patient for hemorrhoid-derived bleeding. Difficulty on defecation or constipation can result from the mass effect of tumors in the distal colon where the stool is more bulky and unable to pass. All patients with bright red blood in the stool should undergo a digital rectal exam to assess whether hemorrhoids are present. Any blood in the stool in the absence of hemorrhoids should prompt an immediate referral to gastroenterology for colonoscopy. 

Many of the risk factors for colorectal cancer can be ascertained quite easily. Obesity,¹⁴ sedentary lifestyle,¹⁵ diabetes,¹⁶ age, personal history of adenomas, smoking, and excessive alcohol intake should all be recognized and addressed as risk factors for colorectal cancers as well as several other cancers. In patients with existing IBD, there should always be a higher level of suspicion, and colonoscopies should be a routine part of care in concert with the gastroenterologist involved. Since more 90% of colorectal cancers occur in those over age 50, aging is considered the major risk factor in those without IBD. For patients with a family history of colorectal cancer in a first-degree relative, especially with a relative diagnosed before age 55, the risk of colorectal cancer is nearly double that of patients without a family history.¹⁷

Race and ethnicity of patients should also be considered when assessing risk. The Multi Ethnic Cohort study looked at 165,711 African Americans, Japanese Americans, Latinos, Native Hawaiians, and whites in Hawaii and California.¹⁸ The relative risk of each ethnic group was compared to whites. Japanese American men had a 27% increased relative risk, Japanese American women a 49% increased relative risk, and African American women a 48% increased relative risk. Further, African American and Japanese American women were at higher risk of advanced disease on diagnosis. 

The ethnic variation in risk is likely due to epigenetic changes induced in the luminal microenvironment of the colon. In one study of African Americans, the authors suggested the increased risk may be due to greater consumption of red meat and saturated fat, combined with higher prevalence of pathogenic bacteria in the gastrointestinal tract.¹⁹ This data is in keeping with the theory that the microorganisms inhabiting the colon (microbiota) have a distinct role in the benefit (or lack of benefit) derived from the diet.²⁰ A follow-up study found that Native Africans have higher levels of protective short chain fatty acids (SCFAs; eg, butyrate) produced from fiber by beneficial bacteria in the gut.²¹

 

What about the small number of patients at very high risk, those with genetic syndromes such as FAP and Lynch syndrome? While FAP and Lynch account for approximately 5% of all colorectal cancers, they disproportionately affect those under 50 years old, so clinicians must be aware of how to recognize these patients early on. Individuals affected by FAP can form hundreds of polyps by age 30, resulting in an average age of colorectal cancer diagnosis of 39.³ Those affected by Lynch syndrome are an average of 44 years old at the time of diagnosis. Therefore, a family history of colorectal cancer diagnoses in a patient’s relatives under the age of 50 should raise the level of suspicion and prompt further investigation. 

FAP can be found through colonoscopy examination, as the polypoid colon can be visualized. Lynch syndrome, however, is not marked by excess number of polyps or adenomas, but a genetic mutation in mismatch repair genes that allow mutations in DNA to accumulate and increase the likelihood of polyps transitioning to carcinoma. Therefore, Lynch syndrome is more difficult to ascertain, so a distinct criteria, called the Amsterdam criteria, has been developed to determine whether genetic testing is appropriate. A patient meets the Amsterdam criteria for genetic testing if he/she has all of the following: 3 or more relatives with confirmed diagnoses of colorectal cancer (1 of whom is a first degree relative of the other 2), 2 successive generations have been affected, at least 1 of the relatives was diagnosed under 50 years old, and FAP has been excluded. 

Lynch syndrome is associated with increased risk of several other cancers as well, including endometrial, ovarian, stomach, small intestine, hepatobiliary, urinary tract, brain, and skin cancers. Thus, Amsterdam II criteria was created. Amsterdam II criteria is identical to the original Amsterdam criteria but substitutes the presence of any combination of the associated cancers in 3 or more relatives for the first criterion. While the Amsterdam criteria are a useful clinical guide for raising the level of suspicion for Lynch syndrome, it should be noted that many affected by Lynch syndrome do not meet the criteria and that the genetic defects can also arise de novo. Regardless, all patients who meet the criteria should undergo genetic testing for Lynch syndrome.

The Centers for Disease Control and Prevention (CDC) have stated that screening for colorectal cancer in those over 50 years old is “the key to preventing colorectal cancer.” CDC recommendations for those over 50 include annual high-sensitive fecal occult blood test (FOBt), flexible sigmoidoscopy every 5 years, and/or colonoscopy every 10 years.²² These screenings are recommended for asymptomatic individuals, and all primary care practitioners should monitor and ensure compliance. The US Preventative Services Task Force has determined that the risks of colonoscopy do not warrant its use as a screening tool for those over 75 years of age.²³ Of course, patients of any age with signs and/or symptoms suggestive of colorectal cancer should be referred to gastroenterology for further work-up. 

Neither computed tomographic colonography (“virtual colonoscopy”) nor double contrast barium enemas are recommended for screening purposes. Virtual colonoscopies require a full bowel prep, cause radiation exposure through use of computed topography, and are not sensitive for smaller or more sessile lesions.²⁴ Further, if polyps or adenomas are found, endoscopic colonoscopy must then be performed to remove them, requiring another procedure. Double contrast barium enema is still used as a diagnostic tool when indicated, but its effectiveness as a screening tool has not been evaluated. 

Testing of the stool for unique genetic changes found only in colorectal cancers is now available commercially. DNA mutation testing of stool is not recommended by any official body (eg, CDC) as a screening tool. However, patients over 75 who are at high risk of colorectal cancer due to family or personal history may be one subpopulation of patients for whom the test should be considered clinically.²⁵ Age, after all, is one risk factor we cannot influence in our patients, and data suggest the risk of CRC approximately doubles every decade of life from 40–80 years old.²⁶

Pill cams (ie, colon capsule endoscopes), capsule-sized cameras that are swallowed by the patient, can collect pictures of the mucosa throughout the colon. The pictures are collected and analyzed remotely, making this a convenient test for patients. This procedure still requires a bowel prep to clear the colon contents. It is not recommended as a screening tool at this time. While early generations of the pill cam had inferior sensitivity compared to colonoscopy in detecting polyps, advances in technology have resulted in improvements in all aspects of pill cam colon visualization.²⁷ For now, the pill cam appears to be less sensitive in detecting polyps, adenoma, and colorectal cancer, but future improvement in the technology may render it a useful, if not a commonplace, tool someday.²⁸

 

Removal of polyps and adenomas has been shown to reduce the risk of developing colorectal cancer.²⁹ This interventional component of colonoscopies allows the practitioner to explain to patients that the procedure is not just “looking for cancer” but in fact may lessen the risk of developing it. Many patients are not aware of this persuasive fact.

There is compelling data regarding an inverse relationship of physical exercise and colorectal cancer risk. Some studies have shown regular exercise cuts the relative risk of colorectal cancer approximately in half.^{30, 31} A meta-analysis of 52 studies published in 2009 found an inverse association between physical activity and colorectal cancer with an overall relative risk (RR) of 0.76 (95% confidence interval (CI): 0.72, 0.81).³² The data regarding the reduction in colorectal cancer risk with exercise is consistent and convincing. It is estimated that 16% of colon cancers diagnosed globally are due to physical inactivity.³³ The most motivating means of getting patients to exercise regularly is to speak with them about what activities they enjoy or have enjoyed. Deriving pleasure from the activity becomes the primary reward, making the fact that it is “exercise” a secondary consideration.

A vast amount of information exists regarding the role of diets and dietary components in relation to development of colorectal cancers. Unfortunately, there is little consensus and much of the data is contradictory. Looking at trends in the data, diets high in trans and saturated fat, sugar, and refined starches and low in fruits and vegetables are associated with higher risk of colorectal cancer.^{34, 35} However, there has been more inconsistency than one may expect in the data. Regarding fat, one Japanese study restricting fat energy to 18–22% of the diet found that colorectal cancer recurrence was significantly higher with this low-fat diet.³⁶ Fiber, often assumed to lower the risk of developing colon cancer, has surprisingly conflicting data. A review of 5 studies involving 4,394 subjects found no reduction in risk of adenomatous polyps with higher fiber intake.³⁷ The most likely reason that dietary data is contradictory is that consumption of dietary components, especially fat and fiber, interface with the microbiota of the gut to form unique compounds dependent of the taxonomy of the microbiota. For example, short chain fatty acids (SCFAs), which are essential for colonic epithelium homeostasis, are metabolic products of specific microbiota derived from fiber in the lumen. Without the presence of particular strains of microbiota, the fiber may not be metabolized into SCFAs, thus disturbing the homeostasis. Such microbiota-food interfacing has only recently been explored, and future studies will likely attempt to account for this confounding variable. Meanwhile, recommending diets high in fruits and vegetables provides chemopreventative phytonutrients in addition to fiber, so a plant-based diet should be recommended despite the debatable role of fiber itself. 

There is evidence that higher calcium intake lowers the risk of developing colon cancer.^38,39 Mechanistically, calcium may bind to carcinogenic bile acids and other nefarious compounds, rendering them inactive,⁴⁰ or it may be effecting cellular differentiation within colonic crypts. In a pooled analysis of 2 randomized, placebo controlled trials (N=1,346 subjects), supplementation with 1,200–2,000mg/day of calcium for 3–4 years resulted in a 26% reduction of recurrent colorectal adenomas (OR: 0.74, CI: 0.58, 0.95).⁴¹ In another study, the protective effects from adenoma formation was found as much as 5 years after stopping supplementation with calcium.⁴² Not all studies, however, have found a risk reduction with supplementation,⁴³ and the National Cancer Institute states “evidence is inadequate to determine whether calcium supplementation reduces the risk of CRC.”⁴⁴

Despite the inconsistency in the data, there is compelling reason to recommend calcium use for prevention given its relative safety when combined with vitamin D. For example, The Polyp Prevention Trial followed 1,905 participants with a history of polyps over 4 years. Calcium and vitamin D intake was ascertained through questionnaire.⁴⁵ There was a consistent, though not statistically significant, trend for reduction of risk of recurrent adenomas with higher intakes of calcium and vitamin D. This trend was more pronounced in those taking supplemental calcium and vitamin D. Also, in a meta-analysis of 3 randomized, placebo controlled trials of subjects (N=1,279) with previous adenomas, calcium supplementation was associated with a significant reduction in adenomas (RR: 0.80, CI: 0.68, 0.93; P=0.004).⁴⁶

Of interest, when looking at normal epithelial turnover in the colonic mucosa, normalization of apoptosis is required for orderly renewal of cells. A study comparing the effects of supplemental vitamin D and calcium in 92 healthy men found that apoptosis was enhanced in the presence of each agent alone and in combination.⁴² The synergism of calcium and vitamin D on systemic absorption is well known. Perhaps lesser known, calcium and vitamin D may also optimize uptake by colonic mucosa cells via increased receptor expression. In a clinical trial using 2 g/day of elemental calcium and/or 800 IU/d of vitamin D3 versus placebo, the calcium-only group had 27% increased expression of calcium receptors (CaRs), the vitamin D3 supplemented group had a 39% increase in calcium receptors, and in patients receiving both agents, the vitamin D receptor (VDR) increased 19% and CaR expression increased 24%.47

Milk product consumption has also been associated with protection from colorectal cancers. A study assessing the milk product consumption in school and the risk of developing colorectal cancer in New Zealand found that those children who consumed the most milk products were 38% less likely to have colorectal cancers later in life than those consuming the least.⁴⁸ Other studies have corroborated that higher milk intake reduces risk of colon cancer development significantly.^49,50

The idea that sun exposure reduces mortality from cancer was first postulated by Frank Apperly in a 1941 paper titled, “The relation of solar radiation to cancer mortality in North America.”⁵¹ The hypothesis, however, was honed and extended to vitamin D status in 1980, when Garland and Garland hypothesized that the lower rates of mortality from colon cancer in the sunny southern regions of the United States may be a result of higher vitamin D levels.⁵²

Unlike with calcium, there is little debate regarding the protective role of vitamin D in colorectal cancer prevention. A plethora of epidemiological and clinical trial data has consistently demonstrated higher intake and higher circulating levels of vitamin D are inversely associated with CRC risk.^53–56 The World Cancer Research Fund published a meta-analysis of prospective randomized trials of vitamin D intake/status and risk of colorectal cancer through June 2010.⁵⁷ They found an inverse association with dietary vitamin D intake, supplemental vitamin D, and vitamin D status. The Institute of Medicine has also recently stated that observational data support the inverse association between serum vitamin D levels and colorectal cancer incidence.⁵⁸

The evidence for vitamin D intake/status clearly favors an inverse association with CRC, but there is no consensus on what the ideal intake or status amount should be. While the Institute of Medicine has stated that vitamin D confers protective benefit from CRC, their recommendation for vitamin D intake is still based on the level needed to avoid osteomalacia in 97.5% of individuals (20 ng/mL). However, 20 ng/mL does not meet the minimal circulating amount needed for prevention of cancers, colon cancer included.^59,60

Optimal serum levels of 25-hydroxyvitamin D [25(OH)D] have been proposed to be 30–60 ng/mL, which is approximately 1,000–4,000 IU of vitamin D/day received through sun exposure and supplement use.⁶¹ The originators of the vitamin D-cancer connection, Garland and Garland, agree that 25(OH)D levels should be well over 20 ng/mL for cancer prevention. In 2009, they published a thorough review of the topic and estimated ideal circulating levels to be 40–60 ng/mL.⁶² They claim that if serum levels of 25(OH)D were maintained above 34 ng/mL (85 nmol/L) in all US residents, 50% of colon cancers could be prevented. 

In observational studies, low folate has been linked to increased risk of colon cancer. However, some data suggest that high intake of folate may have a paradoxical effect, raising the risk of developing colorectal cancer in some individuals. Since dietary folate is molecularly different than supplemental folate, it is possible that this is confounding the data. Another hypothesis is that folate compounds, while preventing the initiation of the carcinogenic process, may stimulate later stages of its development.⁶³ Lastly, it has also been postulated that folate may stimulate the growth of colorectal tumors once present,⁶⁴ as it is common for tumors to increase their folate receptors in an effort to keep up with the increased need for DNA replication processes.⁶⁵

One pooled analysis of 13 prospective studies, totaling 725,134 participants and 5,720 incident colon cancers, found that higher folate intake was associated with reduced colon cancer risk.⁶⁶ The relative risk (RR) for the highest consumers versus the lowest consumers of dietary folate was 0.92 (95% CI: 0.84–1.00; P value, test for between-studies heterogeneity=0.85) and the RR for total folate intake was 0.85 (95% CI: 0.77–0.95; P-value, test for between-studies heterogeneity=0.42). The authors found that there was a 2% risk reduction (95% CI: 0–3%) for every 100 mcg increase in total folate intake.

A large prospective cohort study published in 2011 was designed to assess risk for dietary, supplemental, and total folate intakes pre- and post-grain fortification in the United States.⁶⁷ Using data from the NIH-AARP Diet and Health Study involving 525,488 individuals aged 50–71 beginning in 1995, the amount of dietary and supplemental folate was calculated pre- and post-fortification (before and after July 1, 1997) from food questionnaires. For the follow-up through December 31, 2006 (mean follow up 9.1 years), 7,212 incident colorectal cancers were identified. In the post-fortification analysis, the hazard ratio for those consuming >900 mcg/d versus <200 mcg/d of total folate was 0.70 (95% CI: 0.58–0.84). A reduction of risk was found for both supplements and dietary folate when looked at separately. The authors concluded that increased folate was associated with decreased risk of colorectal cancers; however, they concede additional follow-up is needed as the adenoma-carcinoma sequence may take more than 10 years to develop.

There is evidence that folate may increase adenoma formation in those with a history of adenomas. The Aspirin-Folate Polyp Prevention Study used 1.0 mg folate/day in patients with a history of adenomas and found after approximately 5 years the number and severity of the adenomas in the folate group was higher than the placebo group.⁶⁸ Interestingly, data from the same trial showed that those in the placebo group with the highest folate intakes and folate status were less likely to develop adenomas in the future.⁶⁹ This is in keeping with the hypothesis that there are paradoxical effects dependent on whether the folate is acting on normal mucosa or already-formed adenomas.

A 2011 analysis of data from the Nurses’ Health Study and Health Professionals Follow-Up Study prospectively examined associations between folate intake (assessed every 2–4 years by questionnaire) and risk of colorectal cancers or adenomas from 1980 to 2004.⁷⁰ Comparing intakes >800 mcg/d to <250 mcg/d, they found an association between total folate intake 12–16 years before diagnosis and lower risk of colorectal cancer (RR=0.69, 95% CI: 0.51–0.94). However, they found no association with recent folate intake and colorectal cancer risk. Regarding adenomas, both long- and short-term intake of total folate was associated with reduced risk of formation, with a strong association 4–8 years prior to diagnosis (OR: 0.68; 95% CI: 0.60–0.78; >800 mcg/d compared with < 250 mcg/d). Lastly, they found that current use of multivitamins for >15 years was associated with lower risk of colorectal cancer and that shorter use was associated with less adenoma formation. This is in keeping with previous studies suggesting long-term multivitamin use, but not short-term use, reduces the risk of colorectal cancer.⁷¹ Since this analysis began with data from 1980, it included a large amount of time before food fortification with folate. The authors conclude, “We did not observe an adverse effect of total folate or synthetic folic acid on the risk of colorectal cancer or adenoma even during the folic acid fortification era.”

It should be noted that there is little controversy regarding the protective role of folate repletion in those consuming excessive alcohol. Low folate status in heavy drinkers is associated with greater risk of colorectal cancer.⁷² This may be exacerbated by polymorphisms in methylenetetrahydrofolate reductase enzyme and/or alcohol dehydrogenase enzyme.⁷³ Therefore, repletion in patients that continue to drink heavily may be indicated if dietary sources are suspected of being inadequate.

While the latest analyses lean toward the safety of higher folate intake, it is still prudent to encourage patients to derive the majority of their requisite folate from foods and avoid high doses of folic acid supplementation. Newer forms of folic acid, such as 5-methylenetetrahydrofolate are not tested or proven to be equivalent to food sources, and further research into their use is needed before definitive conclusions can be drawn on their role in affecting colorectal cancer risk.

In several studies, selenium status was lower in those with adenomas and colon cancer versus controls.^74–76 One study of selenium-deficient patients with a history of adenomas showed that repletion of selenium corrected both selenium status and activity of glutathione peroxidase in the colonic mucosa.⁷⁷ Higher selenium status appears to be associated with less risk of colon cancer and adenoma formation.⁷⁸ Smokers, and those who have quit within the last 10 years, appear to derive greater protective effect from higher selenium status than non-smokers.⁷⁹

It is quite consistent in the data that more is not necessarily better in the case of selenium repletion. Therefore, avoiding selenium deficiency is the goal of treatment and doses, if supplemented, should be kept below 200 mcg/day. A small study comparing 200 mcg/d to 400 mcg/d showed that while circulating levels of selenium were higher in those taking 400 mcg/d, a 25% reduction in cancers was found only in those taking the lesser amount.⁸⁰

Curcumin, the collective term for the 3 curcuminoids that give Turmeric (Curcuma longa) its distinct yellow color, may be the most well-characterized chemopreventative agent from any spice. Spices are collectively composed of a myriad of chemopreventative phytochemicals, including phenolics, terpenoids, and flavonoids.⁸¹ The intake of spices, including turmeric, has been postulated to be one of many reasons the rate of colon cancer is much lower in India and other high-spice consumption regions versus Western countries.⁸² While there are broad physiological effects of compounds from spices, much of the chemopreventative effect is thought to be achieved from blocking activation of nuclear factor kappa B (NFkB).^83,84 NFkB affects the expression of hundreds of genes involved in cancer development from carcinogenesis to metastasis.⁸¹ As simple as it seems, encouraging patients to use high-quality spices in their kitchens may be a risk-free means of reducing their risk of developing cancer.

Looking at curcumin specifically, the preponderance of in vitro and animal evidence shows curcumin is a potent chemopreventative and therapeutic agent against various cancers, including CRC;^85–88 however, data from prospective trials is still needed. What we can glean from the data to date is that doses higher than 3.6 g/day are needed for the colonic mucosa to achieve sufficient levels for possible therapeutic benefit.⁸⁹ This is considered a high dose but appears safe, since phase I studies suggest that single daily doses of up to 12 g/day do not induce any toxic symptoms.^90–92

In a small pilot study of 10 subjects (5 with ulcerative proctitis and 5 with Crohn’s disease), the anti-inflammatory effect of curcumin showed significant improvement in symptoms and lower sedimentation rate values.⁹³ In another pilot study of 5 patients with FAP, a combination of curcumin (480 mg) and quercetin (20 mg) 3 times daily resulted in a 60.4% (P<0.5) decrease in the size and a 50.9% (P<0.5) decrease in number of polyps compared to baseline.⁹⁴ A recent phase II trial showed that curcumin is able to increase the expression of the tumor suppressor p53 in patients with existing colon cancer.⁹⁵ While the evidence is far from conclusive, the relative safety of curcumin warrants its consideration in high-risk patients.

Omega 3 fatty acids eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) have potent anti-inflammatory effects. There is also evidence for direct protective effects regarding proliferation, apoptosis, angiogenesis, and metastasis in colon cancer cells.⁹⁶

One may expect that high doses of EPA and/or DHA would confer benefit by virtue of anti-inflammatory effects. While epidemiological data supports the protective role of high fish diets,^97–100 there is a paucity of randomized trial information to support or refute this. One publication from 1998 shared 3 cases of colon cancer in patients with FAP that the authors felt may have been partly influenced by oral supplementation with DHA.¹⁰¹ In a mouse model of colitis, in which an infection with Helicobacter hepaticus can produce colitis and adenocarcinoma of the colon, dietary fish oil enriched with DHA led to more severe colitis and a greater number of adenocarcinomas forming, likely due to changes in immune mediators.¹⁰²

In contrast, an interventional trial using enterically coated EPA in 55 FAP subjects found a 22.4% (95% CI: 5.1–39.6%) reduction in the number and a 29.8% (95% CI: 3.6–56.1%) decrease in the sum of the size of polyps.¹⁰³ In another trial of 30 patients with a history of adenomas, purified EPA (2 g/day) significantly reduced crypt cell proliferation and increased apoptosis compared to placebo.¹⁰⁴ There is some evidence that DHA may have synergistic effects with butyric acid, the short-chain fatty acid derived from fermentation in the colon.¹⁰⁵ Once again, the interface of dietary constituents with colonic flora at the site of the colonic crypts is thought to create the hostile or favorable environment needed for carcinogenesis or healthy cells, respectively. Supplementation with EPA/DHA without the support of a healthy flora may not confer the benefit that is intended. Moderate consumption of fish, particularly cold-water fish, appears to be a reasonable recommendation that may reduce risk. Supplementation with fish oil should be kept at levels comparable to dietary intake, namely less than 2.5–3 grams daily, as supra-physiological doses have no evidence for greater benefit.

The anticarcinogenic effects of garlic are thought to be derived primarily from organosulfur compounds, including the odiferous diallyl sulfide that gives garlic its distinct smell.¹⁰⁶ Garlic and its constituents have been well proven to lessen CRC carcinogenesis, reduce proliferation, block angiogenesis, induce differentiation and apoptosis, inhibit cyclooxygenase-2 and squelch free radicals.^107–110 In one study of subjects with a history of adenomas, the use of aged garlic extract (2.4 ml/d) significantly reduced the incidence and size of future adenomas versus controls, who consumed 0.16 mL/d.¹¹¹ The Iowa Women’s Health Study found that those with the highest consumption of garlic had a 32% reduction in risk of colorectal cancer compared to those with the lowest consumption.¹¹² A systematic review of the literature published in 2007 concluded, “On balance, there is consistent scientific evidence … reporting protective effects of garlic on CRC despite great heterogeneity of measures of intakes among human epidemiological studies.”¹¹³ It is clear that encouraging our patients to consume garlic in their diets has little downside, and it should be included in a healthy diet. 

As mentioned above, the interplay between dietary components and the microbiota of the colon is thought to have a integral role in carcinogenic processes of the colonic epithelium,¹¹⁴ as well as overall health and disease.¹¹⁵ The short-chain fatty acid butyrate is a product of microbiota metabolism of fiber from the diet, and butyrate is involved in colonic homeostasis of colonic crypts.¹¹⁶ Many other compounds formed from microflora may influence the inflammatory environment of the colon, systemic immune function,¹¹⁷ and the presence of bioactive compounds locally that may promote or inhibit carcinogenic processes.¹¹⁸ By definition, probiotics “confer a health benefit on the host,”¹¹⁹ so the ingestion of probiotics has little to no risk. Further, specific studies have shown high doses of probiotics are capable of reducing local inflammation in Crohn’s disease,¹²⁰ so their use as a supplement is perhaps most indicated for those with IBD. All patients should be encouraged to consume cultured food products containing live probiotic organisms. For patients unwilling or unable to consume the whole foods, supplements are available and should be recommended at the practitioner’s discretion.

Green tea components such as polyphenols (eg, epigallocatechin-3-gallate, EGCG) can affect colorectal carcinogenic processes.¹²¹ Animal models demonstrate the antitumorogenic properties of GTE on mouse models of various cancers, including colon cancer.¹²² One pilot trial of 136 patients demonstrated that those who consumed the GTE equivalent to >10 cups of green tea daily had less development of adenomas compared to those consuming a lesser amount of green tea.¹²³ The use of green tea extract may be considered in high-risk patients, and all patients can be encouraged to consume green tea as part of a diet high in antioxidants.

Cyclooxygenase-2 (cox-2) is overexpressed in the majority of colorectal tumors, and cox-2 inhibition may reduce the production of adenomas or cancer.¹²⁴ The role of aspirin in preventing colon cancer is fairly well established.^125–128 In addition to greater rates of prevention, those consuming aspirin also have lower rates of mortality once diagnosed with colon cancer.^129–131 NSAIDs and cox-2 inhibitors, such as celecoxib, have also demonstrated protective effects from the development of adenomatous polyps.¹³² However, unlike many of the natural agents, these drugs do not come without possible harm, including cardiovascular risk (cox-2 inhibitors),¹³³ gastrointestinal bleeding, and renal impairment. The US Preventative Services Task Force currently recommends against their use for chemoprevention of colorectal cancer in persons with average risk based on risk/benefit analysis.¹³⁴ A recent publication of the use of 600 mg aspirin per day for an average of 25 months versus placebo in individuals with Lynch syndrome showed a substantial reduction in colorectal cancer after a mean follow up of 55.7 months (HR: 0.41; P=0.02).¹³⁵ Further studies are needed to determine which patients derive enough benefit to warrant the risk. Until the data emerges, the clinician must weigh the risks and benefits on an individual patient basis.

Primary care practitioners are uniquely positioned to affect their patients’ risk of colorectal cancer. All patients should be screened at age 50, as symptoms of colorectal cancer may not be evident until late in the disease process. Earlier screening should be considered in patients at higher risk. Careful attention should be paid to every patient’s family history as this can often be the most reliable means of recognizing there may be a genetic predisposition to development of colorectal tumors at an early age. Recognizing risk factors and counseling patients on how to affect risk factors such as obesity and sedentary lifestyle are paramount to affecting outcomes. Many of the recommendations regarding diet, exercise, and normalization of weight will have ramifications far beyond prevention of colorectal cancers and must be implemented as part of an overall wellness goal with patients.