The purpose of this study is to help develop diets for colon cancer prevention. This study will compare the Mediterranean diet to the Healthy People 2010 diet in 120 subjects with increased risk for colorectal cancer.

There is substantial epidemiological evidence that dietary patterns influence colorectal cancer risk. The associations of any particular nutrient with increased or decreased risks, however, may not be due to that nutrient per se but to the whole foods that are rich in that nutrient. Simultaneously, reducing intakes of foods associated with increased risk while increasing foods identified in preventive diets may be the best approach for prevention. The Cretan-Mediterranean diet in particular appears to hold great promise for cancer prevention. The major components of the traditional Cretan diet have been associated with decreased colon cancer. Relative to the American diet, this diet has lower n-6/n-3 and n-6/n-9 fatty acid ratios, lower polyunsaturated fatty acid intake, lower red meat intake, and higher intakes of plant-based foods and monounsaturated fatty acids.

The hypothesis of this study is that adherence to a Mediterranean type of diet will result in a decrease in n-6 fatty acids and increased n-3 and n-9 fatty acids in human colorectal mucosa. This together with aspects of the diet such as increased intakes of fruits and vegetables, is expected to modulate eicosanoid metabolism and epithelial proliferation in normal mucosa. 120 persons, with an increased risk for colorectal cancer, will be randomized to a modified Mediterranean diet or a Healthy People 2010 diet for six months.

Inclusion Criteria:

• Prior adenomatous polyp.
• Prior resected early (Dukes A, B, or C) colon cancer. With the exception of curative surgery for small lesions, such as endoscopically removed cancers, eligible subject will be at least two years post treatment for colorectal cancer.
• A history of colon cancer in a primary relative or in two secondary relatives.
• Good general health and not expecting major lifestyle changes in the next 6 months.
• Age 21 or older.
• Not expecting a change in hormonal therapies over the next 6 months.
• Taking less than 81 mg/day or 325 mg aspirin every other day for prevention of cardiovascular disease.
• Dietary intake that is within the usual range for a typical American diet.
• Read and understand English.
• Sign the consent and willing to comply with all study procedures.
• Have a telephone.
• At least 5 years post any type of treatment for any other cancer except cancers that were removed completely by surgery and no other treatment was undergone.
• No more than occasional use (< 25% of the time) of pain medications and willing to take only regular strength acetaminophen while on study except for 81 mg/day or 325 mg every other day of aspirin for prevention of cardiovascular disease.

Exclusion Criteria:

• On medically prescribed diets or following a diet that would require extensive counseling to correct nutritional deficiencies.
• Taking supplements or medications that might obscure our ability to detect an effect of diet (eg. lipid-lowering medications, insulin, fish oils, mega-vitamins).
• Are pregnant or lactating or planning to get pregnant.
• Previous diagnosis of HIV or hepatitis C.
• Have cancer at the present time.
• Being treated with or taking therapies or supplements that could obscure our ability to detect diet effects, such as fish oils.
• Previous advanced cancer (Duke's D) or hereditary and familial polyposis (HNPCC/FAP) because the latter are rare conditions with unique etiology.
• Due to the effects of inflammation on biomarker levels in mucosa, persons with Crohn's disease or inflammatory bowel disease will be excluded.
• Persons with BMI < 18.5 or > 35 kg/m2 since low BMI could indicate eating disorders and high BMI values, above the midpoint of the obesity range, could indicate more prevalent health problems and these persons can be more difficult to counsel.
• Persons taking very high levels of aspirin or non-steroidal anti-inflammatory agents (NSAIDS) for conditions such as arthritis, a chronic inflammatory condition, will be excluded since it will preclude our ability to detect a further decrease in PGE2.

5 years