Treatment of rheumatoid arthritis patients with interleukin-1 receptor antagonist: radiologic assessment.

Objectives: The radiologic findings of a placebo-controlled, dose-ranging, multicenter, multinational trial have been reported previously. Radiographs were evaluated using the Larsen scoring method and Erosion Joint Count. After completion of the study, a subset of the films was read again using a modified Sharp score. This article will focus on the methodologies, scoring indices, and outcomes of the Larsen and Erosion Joint Count evaluations. Modified Sharp scores are presented in a separate article.

Methods: A 6-month, phase II, randomized, double-blind, placebo-controlled trial was conducted involving 472 patients with active rheumatoid arthritis. Patients from 41 centers in 11 countries were randomly selected to receive 30 mg/d, 75 mg/d, or 150 mg/d of recombinant human interleukin-1 receptor antagonist (IL-1ra) subcutaneously daily or placebo. Radiographic criteria were circulated to all centers, and the same 2 radiologists used the Larsen score and the Erosion Joint Count to score what was essentially a homogeneous film collection. At the completion of the study, a subset of radiographs also was read using the Genant-modified Sharp score. Patients in any of the treatment arms had the option of continuing in an extension trial for an additional 6 months, and those in the placebo arm had the option of being randomly placed into one of the treatment arms.

Results: The Larsen and Erosion Joint Count data from these patients confirm that at 24 weeks, patients receiving placebo worsened by an average of 6.49 Larsen units, whereas those receiving 30, 75, or 150 mg/d of IL-1ra worsened by 3.53, 4.19, and 3.90 Larsen units, respectively. Overall, patients receiving therapy worsened by an average of 3.86 units, achieving statistical significance versus placebo (P = .034). These data are not significantly different from those of the main trial. Mean values were ANOVA-adjusted for country and treatment-group interactions. Similarly, the Erosion Joint Count in placebo patients worsened by an average of 2.64, whereas those receiving 30, 75, or 150 mg/d of IL-1ra worsened by 1.46, 1.05, and 1.70, respectively. The overall therapy and 75 mg/d arm achieved significance versus placebo (P = .002 and P < or = .001, respectively). Preliminary data from the extension study indicate continuing benefit.

Conclusions: Treatment with IL-1ra reduced the rate of joint deterioration and development of new bone erosions.