Background: Autoantigen-specific immunotherapy by mucosal tolerance induction via the intranasal route is an attractive therapeutic option for the treatment of autoimmune diseases, including rheumatoid arthritis (RA). Human cartilage glycoprotein-39 (HC gp-39) has been identified as a potential key autoantigen in RA. Based on animal studies, intranasal administration of the autoantigen is hypothesised to induce immunological tolerance in patients with RA and to ameliorate disease activity. In a phase I/IIA clinical trial in patients with RA, intranasal application of HC gp-39 was safe and well tolerated. Objective: To investigate the efficacy of intranasally administered fully human, recombinant HC gp-39 (Org 39141) by a large clinical study. Methods: In a 13-week multicentre, double-blind, randomised, placebo-controlled, parallel-group, dose-finding, proof-of-concept trial, patients with RA (disease-modifying antirheumatic drug (DMARD) naive or after washout of DMARD treatment) were randomised to receive either intranasal applications of placebo or HC gp-39 in doses of 30, 150, 300 or 600 mu g, once a week. The primary efficacy variable was the 28 joint count Disease Activity Score (DAS28). Results: During the treatment period the DAS28 decreased similarly for all treatment groups-including placebo-indicating lack of efficacy of intranasal HC gp-39 treatment in the current setting. Safety variables were similar for all study groups. Conclusion: It was concluded that with the treatment protocol used (dose levels and frequency of dosing), intranasal treatment with Org 39141 was safe but did not result in more clinical improvement than in placebo-treated patients.