The clinical efficacy of 3 mg/day prednisone in patients with rheumatoid arthritis: evidence from a randomized, double-blind, placebo-controlled withdrawal clinical trial.

Author(s): Pincus T

Affiliation(s): Department of Rheumatology, New York University Hospital for Joint Diseases, New York, NY 10003, USA. tedpincus@gmail.com

Publication date & source: 2011-09, Clin Exp Rheumatol., 29(5 Suppl 68):S73-6. Epub 2011 Oct 21.

Publication type: Research Support, Non-U.S. Gov't

A randomised, double-blind, placebo-controlled, withdrawal clinical trial was conducted of prednisone <5 mg/ day versus placebo in 31 patients with rheumatoid arthritis (RA). These patients had been treated with long-term 1-4 mg/day of prednisone, 22 with 3 mg/day, in usual clinical care at a single academic clinical setting. Stable clinical status over 12 weeks prior to screening for the trial was documented quantitatively by patient questionnaire scores. The protocol involved three phases: a) 'equivalence' - 1-4 study prednisone 1-mg tablets taken for 12 weeks, to ascertain their efficacy versus the patient's usual prednisone tablets prior to randomisation; b) 'transfer' - substitution of a 1-mg prednisone or identical placebo tablet at a rate of a single 1-mg tablet every 4 weeks (over 0-12 weeks) to the same number as baseline prednisone; c) 'comparison' - observation over 24 subsequent weeks taking the same number of either placebo or prednisone tablets as at baseline. The primary outcome was withdrawal due to patient-reported lack of efficacy versus continuation in the trial for 24 weeks. Thirty-one patients were randomised, 15 to prednisone and 16 to placebo, with 3 administrative discontinuations. In 'intent-to-treat' analyses, 3/15 prednisone and 11/16 placebo participants withdrew (p=0.03). Among participants eligible for the primary outcome of withdrawal for lack of efficacy, 3/13 prednisone versus 11/15 placebo participants withdrew (p=0.02). No meaningful adverse events were reported, as anticipated. These data document statistically significant differences between the efficacy of 1-4 mg prednisone vs. placebo in only 31 patients, which may suggest a robust treatment effect.