POS0094 EFFECTS OF RANKL INHIBITION ON PROMOTING HEALING OF BONE EROSION IN RHEUMATOID ARTHRITIS USING HR-pQCT: A 2-YEAR, RANDOMIZED, DOUBLE-BLIND, PLACEBO-CONTROLLED TRIAL

Abstract

Background:Partial repair of bone erosions in rheumatoid arthritis (RA) is known from high-resolution peripheral quantitative computer tomography (HR-pQCT) studies in patients with moderate to high disease activity using biologics [1]. Whether RANKL inhibition by denosumab is efficacious in healing existing erosions in RA patients with low disease activity or in remission on conventional synthetic DMARDs is uncertain.Objectives:To evaluate the effects of denosumab on erosion healing at 2-4 metacarpophalangeal head as determined by HR-pQCT in patients with RA with stable disease.Methods:This was a randomized, placebo-controlled, double-blind study. RA patients with disease activity score 28 joints (DAS28) ≤5.1 were randomized (1:1) to subcutaneous denosumab 60 mg or placebo once every six months for 24 months. The primary outcome was erosion healing at MCP 2-4 on HR-pQCT at 12 months. The effects of denosumab on erosion and joint space parameters on HR-pQCT and radiographs, disease activity and health assessment questionnaire-disability index (HAQ-DI) were also examined.Results:At 24 months, HR-pQCT images were analyzed in 98 patients. Baseline demographic, clinical characteristics and imaging parameters were comparable between the two treatment groups (table 1). Seventeen patients in each group (placebo group: 17/52, 32.6%; denosumab group: 17/50, 34.0%) achieved sustained low disease activity (DAS28 ≤ 3.2) throughout the 24 months. At 12 months, changes in erosion parameters on HR-pQCT were similar between the two groups. At 24 months, new erosions (19% vs 9%, p=0.009) and erosion progression (34% vs 16%, p<0.001) were more common in the placebo group than the denosumab group. Erosion healing was seen in a significantly higher proportion of patients in the denosumab group (20% vs 6%, p=0.045) at 24 months. The details of the changes in HR-pQCT erosion parameters are shown in figure 1. No significant differences in the changes in joint space parameters on HR-pQCT, van der Heijde-Sharp erosion score, DAS28 and HAQ-DI were observed between the two groups at 12 and 24 months.Table 1.Baseline clinical, demographic, disease activity parameters and medicationsPlacebo (n=55)Denosumab (n=55)Total (n=110)Age56.5 ± 7.157.2 ± 8.556.8 ± 7.8Gender (Female)47 (86)41 (75)88 (80)Disease duration (years)8.5 ± 6.87.3 ± 6.97.9 ± 6.8Rheumatoid factor positive40 (72)38 (69)78 (71)ACPA positive43 (78)44 (80)87 (79)DAS28-CRP2.43 ± 0.832.6 ± 0.922.51 ± 0.88DAS28-CRP>3.28 (15)13 (24)21 (19)HAQ-DI (0-3)0.31 ± 0.380.46 ± 0.470.39 ± 0.43csDMARDs49 (89)52 (95)101 (92)Combination csDMARDs26 (47)33 (60)59 (54)Glucocorticoids5 (10)5 (9)10 (9)vdH- Sharp erosion score10.4 ± 18.48.9 ± 13.89.6 ± 16.2vdH- Sharp JSN score12.4 ± 17.711.5 ± 17.211.9 ± 17.4Lumbar spine aBMD, g/cm20.914 ± 0.1470.930 ± 0.1430.922 ± 0.145Total hip aBMD, g/cm20.837 ± 0.1020.847 ± 0.1460.841 ± 0.125Femoral neck aBMD, g/cm20.681 ± 0.0990.695 ± 0.1280.687 ± 0.114Data are reported as mean ± SD or number (%). ACPA: Anti-cyclic citrullinated peptide antibody; DAS28: disease activity score 28; csDMARDs: conventional synthetic disease modifying anti-rheumatic drug. HAQ-DI: health assessment questionnaire disability index; vdH- Sharp score: Van der Heijde- Sharp score; aBMD: areal bone mineral densityConclusion:Although no differences in erosion parameters were observed at 12 months, denosumab was more efficacious than placebo in erosion repair on HR-pQCT after 24 months.