Colorectal carcinoma is the third most common cause of death from cancer. Approximately, 30% of colorectal carcinomas involve the rectum. Optimizing local control in the pelvis while reducing treatment toxicity remains one of the principal goals of therapy for patients with locally advanced rectal carcinoma. Treatment strategies that achieve this goal will have a significant impact on our society.C linical trials have shown that this type of cancer is less likely to come back if chemotherapy and radiotherapy are added to surgery. A combination of all three types of therapy is now standard.

Celecoxib (Celebrex®) is a drug that lessens the action of an enzyme called cyclooxygenase-2 (COX-2) also known as a "COX-2 inhibitor". It is an anti-inflammatory capsule (drug that reduces irritation) that is commonly used to treat arthritis. It is not a chemotherapy drug. Laboratory experiments have shown that such COX-2 inhibitors may increase the anti-cancer effect of radiotherapy, without increasing radiation side effects. This has not yet been confirmed in humans.The main purpose of this study is to confirm that celecoxib does not increase the side effects when given with radiotherapy and chemotherapy for rectal cancer. We shall also be looking at how effective the combination of radiotherapy, chemotherapy and celecoxib is in shrinking rectal cancer.

Inclusion Criteria:

• patients with resectable or potentially resectable adenocarcinoma of the rectum
• clinical stage T2 N1-2 or T3-4 N0-2 (patients who require diverting loop colostomy are eligible)
• age greater than 18 years, ECOG performance status < 2 (appendix, section 13.1)
• biopsy proven adenocarcinoma, superior margin of the tumour within 15cm of the anal verge on rigid sigmoidoscopy

Exclusion Criteria:

• Distant metastasis, Prior pelvic irradiation, Inflammatory bowel disease, Medical conditions which preclude radical therapy
• History of malignancy within five years (except nonmelanoma skin cancer, CIN cervix)
• Pregnancy
• Hypersensitivity to celecoxib, NSAID, sulfonamides or 5-FU
• Significant comorbid illness
• History of peptic ulcer disease or NSAID-related gastrointestinal bleeding
• Use of aspirin, other NSAID or coxib in the two weeks prior to study entry
• Neutrophil count <1.5x109/L, platelet count <100x109/L, serum bilirubin >1.25xULN (upper limit of normal), AST/ALT >3xULN, serum creatinine >1.25xULN