This thesis evaluated the role of rIL-2 in patients with advanced colorectal cancer and also, as a perioperative regime, in patients with localized colorectal cancer undergoing surgical resection. Twenty patients with advanced colorectal cancer received up to six cycles of chemoimmunotherapy, each consisting of 5-fluorouracil, levamisole and rIL-2 at 18x10<SUP>6</SUP>IU/m<SUP>2</SUP>/24 for 120 hours. Responding patients were found to have significantly lower pre-treatment serum IL-6 and soluble IL-2 receptor levels, compared with non-responders. Differential patterns of host cytokine release were also identified. Haemodynamic monitoring found that indices of rIL-2-mediated toxicity, e.g. weight gain correlated with alterations in serum cytokine concentrations. In a separate study, eighteen patients, undergoing curative surgery for localized colorectal cancer, were randomized to receive placebo or bolus low-dose subcutaneous rIL-2 for three days preoperatively. rIL-2 was found to significantly enhance host antitumour natural cytotoxicity, monocyte activity and immune cell surface activation marker expression (e.g. CD25). Circulating levels of key host cytokines (e.g. interleukin-6, soluble interleukin-2 receptor) were elevated in the immediate postoperative period in these patients. Mesenteric release of key cytokines was determined in patients undergoing resection for benign and malignant colorectal disease through portal sampling at surgery. Higher patterns of release were found in patients with malignancy suggesting local modulation of immune activity. rIL-2 has been found to beneficially enhance host immune reactivity in patients with localized and advanced colorectal cancer. The nephrotoxic potential of rIL-2 therapy was determined through urinary enzyme release, plasma renin and standard blood biochemistry measurements. RIL-2, when administered carefully, may form the basis of further adjuvant immunotherapy in both the peri-operative period and in patients with advanced colorectal cancer.