To determine whether rituximab, a monoclonal antibody against the B-lymphocyte antigen CD20, is effective in the treatment of refractory noninfectious scleritis.

Prospective, dose-ranging, randomized, double-masked phase I/II clinical trial.

Twelve patients with noninfectious scleritis refractory to systemic corticosteroid and ≥1 other systemic immunosuppressive agent were enrolled from January 2007 to March 2010.

Subjects were randomly assigned to 500 (n = 5) or 1000 mg (n = 7) dosing arms of rituximab intravenous infusions (500 or 1000 mg), given at study days 1 and 15. Initial responders with breakthrough inflammation after study week 24 were offered treatment with an additional cycle of 2 open-label rituximab 1000 mg infusions.

Primary outcomes were reduction of inflammation, as measured with a validated scleritis disease grading scale (SGS) and reduction in corticosteroid dose by ≥50%. Patients were characterized as responders to study therapy if ≥1 of these endpoints showed improvement and neither showed evidence of worsening. Secondary outcomes were improvement in visual acuity, reduction in pain, and improvement in patient and physician-reported global health assessment.

Of 12 enrolled patients, 9 met the SGS endpoint at or before week 24, and 4 additionally were able to reduce corticosteroid dose by ≥50%. With regard to secondary outcome measures, 11 and 9 patients showed improvement in patient and physician global health scores, respectively, and 7 patients had reduction in pain. Of 9 initial responders, 7 experienced breakthrough inflammation after 24 weeks and were treated with a second cycle of rituximab infusions. Four patients had significant objective or subjective worsening within 8 weeks of receiving rituximab; this event was averted in subsequent patients by treatment with peri-infusional oral corticosteroid. No other significant adverse events were noted. No differences in efficacy, toxicity, or likelihood of retreatment were noted between the dosing arms.

Rituximab was effective treatment for 9 of 12 enrolled patients with refractory, noninfectious scleritis at 24 weeks, although 7 required reinfusion with rituximab to maintain inflammatory control. The treatment was well-tolerated, and peri-infusional inflammatory exacerbations were managed successfully with oral corticosteroids. Further long-term studies are warranted to determine the safety and efficacy of rituximab in treating noninfectious scleritis and other ocular inflammatory diseases.