This purpose of this study is to investigate whether the number and size of rectal polyps can be reduced in patients with Familial Adenomatous Polyposis (FAP) by using a highly-purified form of a naturally occurring substance, the omega-3 fatty acid, eicosapentaenoic acid (EPA).

It has been found that people who consume a large amount of oily fish tend to have a lower risk of developing colon cancer. This is thought to be due to the omega-3 fatty acids present in oily fish, one of which is EPA. The effect of taking a 99% pure form of EPA (2g per day) compared with placebo capsules on the number and size of polyps in the rectum over a six month period will be investigated.

FAP is an inherited susceptibility to diffuse colorectal adenomas and colorectal carcinoma, occurring in close to 100% of unresected colons. It is caused by a germline mutation in the Adenomatous Polyposis Coli (APC) gene located in the long arm of chromosome 5. To prevent cancer development it is recommended that patients with FAP undergo colectomy with ileo-anal or ileo-rectal anastomosis (or colectomy and end-ileostomy) at a socially convenient time before polyp progression to malignancy and before the age of 25. Patients with the attenuated FAP phenotype, often associated with mutations at the 5' terminus (exon 4 and proximally), have fewer polyps and may often delay colectomy. Patients with an ileo-rectal anastomosis are still susceptible to polyp formation in the remaining rectal stump and require 6 monthly check-ups with a flexible sigmoidoscope with removal of any polyps that develop. Therefore, an effective chemopreventative agent with a favourable side-effect profile would be of benefit to FAP patients with ileo-rectal anastomosis (IRA) and recurrent rectal polyps in addition to young adults who prefer to delay colectomy. If such an agent were to be effective in FAP patients in the prevention of colonic polyps, it may also be of benefit to the larger population of patients with sporadic colorectal adenomatous polyps who are also at risk of colorectal cancer.

Colorectal polyps are thought, at least in part, to arise from an imbalance in the levels of cell proliferation and apoptosis (natural cell death) in the colonic mucosa. It has been suggested that omega-3 polyunsaturated fatty acids (PUFAs) in fish oil can manipulate the high levels of colonic-mucosal cell proliferation rates associated with colonic adenomas.

The rationale for this trial is based on the increasing evidence linking inflammatory processes and the development of a number of cancers, including bowel cancer. This has focused attention on the role of inflammatory mediators in the development of cancer. In particular, the family of eicosanoids (including 2-series prostaglandins, 4-series leukotrienes and thromboxanes) produced through conversion of the omega-6 PUFA, arachidonic acid, via cyclo-oxygenase-2 (COX-2) is believed to contribute to the physiological processes of inflammation and the development of tumours. Prostaglandin E2, a product of the conversion of arachidonic acid via the COX-2 pathway, has been implicated in tumourigenesis through:

1. promotion of angiogenesis
2. anti-apoptotic properties
3. increasing expression of matrix metalloproteinases and hence the ability of a tumour cell to undergo metastasis
4. altering the cytokine expression profile of cells.

The class of eicosanoid synthesised will depend on the PUFA substrate. Whilst arachidonic acid is converted to 2-series prostaglandins and 4-series leukotrienes, EPA is converted to 3-series prostaglandins and 5-series leukotrienes. Overall, the latter eicosanoids are less potent as inflammatory mediators than those derived from arachidonic acid.

Increasing daily intake of EPA, the omega-3 PUFA analogue of arachidonic acid, alters the balance between the cell content of these fatty acids. This results in reduced production of the more active inflammatory/tumourigenic products of arachidonic acid metabolism. This is supported by the results of recent work at St George's Hospital Medical School, London. In patients with a history of colonic adenomas, daily dosing with a highly purified, free-fatty acid form of the EPA produced a significant reduction in cell proliferation and increase in apoptosis in the colonic mucosa. This preparation of EPA has the proprietary name "Alfa" and is referred to here as EPA.

This proposed study will be based upon the randomised, placebo-controlled National Cancer Institute sponsored study in which three groups of FAP patients were assigned to one of two doses of celecoxib (a COX-2 inhibitor) or placebo. The results showed a reduction in the polyp burden of the group taking the higher (400mg twice daily (bd)) dose. However, there is evidence to suggest that COX-2 inhibitors carry significant potential for side-effects. Adopting a similar design, EPA will be substituted for celecoxib in this randomised, placebo-controlled trial, comparing 2g EPA to placebo, with reduction in polyp burden as the primary objective.

Inclusion Criteria:

• Subjects must have a known diagnosis of Familial Adenomatous Polyposis (FAP) and have had a previous colectomy with ileo-rectal anastomosis.
• Males or females aged 18 and over
• If the participant is female and of child bearing potential, she agrees to participate in this study by providing written informed consent, has been using adequate contraception (e.g. abstinence, condom, Intra-uterine device (IUD), birth control pill, diaphragm and spermicidal gel combination) since her last menses and will use adequate contraception during the study, is not lactating, and agrees to undergo a serum pregnancy test at baseline and month 6. Sexually active males must agree to use an accepted method of contraception.
• Rectal polyp status: the subject has an endoscopically assessable rectal segment.
• Subjects must show a willingness to abstain from regular use of non-steroidal anti-inflammatory medication for the duration of the study. A cardioprotective dose of aspirin (75mg) will be permitted.
• Subjects must have provided written informed consent to participate.
• Subjects must have assessable rectal polyps post baseline flexible sigmoidoscopy.
• Subjects must have the following rectal polyp burden at the conclusion of the baseline endoscopy:
• Rectum - 3 or more quantifiable polyps ≥2mm diameter
• In the rectum quantifiable polyps are defined as being within a composite "cloverleaf" photograph that includes a tattoo.

Exclusion Criteria:

• Subjects who are due to undergo an anticipated colectomy within 8 months of randomisation
• History of invasive carcinoma in the past 5 years other than resected Dukes' A/B1 colon cancer or resected non-melanomatous skin cancer
• Partial or complete colectomy within 12 months prior to enrolment.
• History of pelvic radiation
• Subjects who are allergic to fish
• Subjects who have diabetes mellitus
• Subjects who are pregnant or breast-feeding
• Subjects taking aspirin or other non-steroidal anti-inflammatory drugs on a regular basis other than low dose (75 mg) cardioprotective dose.
• Subjects who have aspirin-sensitive asthma
• Subjects suffering from haemorrhagic disorders
• Subjects who are taking warfarin or other anticoagulants
• Subjects who have significant abnormalities on their screening blood tests
• Subjects taking lipid lowering medication
• Subjects with gastrointestinal malabsorptive disease
• Subjects with known or prior coagulopathy
• Subjects with uncontrolled hypercholesterolaemia
• Subjects who are taking other fish-oil supplements (e.g. cod liver oil) who are unwilling to stop them for the duration of the study. Subjects previously taking fish oil must have a washout period of 1 month prior to study enrolment.
• Subjects who are deemed mentally incompetent, or have a history of anorexia nervosa or bulimia
• Subjects with a history of alcohol or drug abuse, including laxative abuse which would render the subject unreliable.
• Subjects considered by their physician unlikely to be able to comply with the protocol.
• Subjects who have taken part in an experimental drug study in the preceding 3 months.
• Subjects who have a positive pregnancy test within 14 days prior to baseline visit.

The baseline characteristics are presented for the full analysis set which constitutes 55 of the 58 randomised subjects. The full analysis set includes all subjects who were randomised into the study, took at least one dose of study medication provided both at baseline and at least one treatment measurement for efficacy.

6 months