Studies in animals and an open-label trial have suggested a role for antibodies to tumor necrosis factor alpha, specifically chimeric monoclonal antibody cA2, in the treatment of Crohn's disease. We conducted a 12-week multicenter, double-blind, placebo-controlled trial of cA2 in 108 patients with moderate-to-severe Crohn's disease that was resistant to treatment. All had scores on the Crohn's Disease Activity Index between 220 and 400 (scores can range from 0 to about 600, with higher scores indicating more severe illness). Patients were randomly assigned to receive a single two-hour intravenous infusion of either placebo or cA2 in a dose of 5 mg per kilogram of body weight, 10 mg per kilogram, or 20 mg per kilogram. Clinical response, the primary end point, was defined as a reduction of 70 or more points in the score on the Crohn's Disease Activity Index at four weeks that was not accompanied by a change in any concomitant medications. At four weeks, 81 percent of the patients given 5 mg of cA2 per kilogram (22 of 27 patients), 50 percent of those given 10 mg of cA2 per kilogram (14 of 28), and 64 percent of those given 20 mg of cA2 per kilogram (18 of 28) had had a clinical response, as compared with 17 percent of patients in the placebo group (4 of 24) (p<0.001 for the comparison of the cA2 group as a whole with placebo). Thirty-three percent of the patients given cA2 went into remission (defined as a score below 150 on the Crohn's Disease Activity Index), as compared with 4 percent of the patients given placebo (P=0.005). At 12 weeks, 41 percent of the cA2-treated patients (34 of 83) had had a clinical response, as compared with 12 percent of the patients in the placebo group (3 of 25) (P=0.008). The rates of adverse effects were similar in the groups. A single infusion of cA2 was an effective short-term treatment in many patients with moderate-to-severe, treatment-resistant Crohn's disease.