A multicentre randomised controlled trial of zoledronic acid for osteoarthritis of the knee with bone marrow lesions

Background: No disease-modifying drugs are currently available for the treatment of osteoarthritis (OA). Bone marrow lesions (BMLs) visualised on magnetic resonance imaging (MRI) have been identified as a promising therapeutic target. Our pilot study showed that a single infusion of zoledronic acid (ZA) reduced knee pain and BML size in knee OA patients over 6 months. A longer, larger study was required to assess whether these improvements can be reproduced in a larger multicentre design. Objectives: To compare the effect of once-yearly intravenous infusion of ZA to placebo on knee pain and BML size over 24 months in knee OA patients with significant knee pain and BMLs. Methods: The Zoledronic Acid for Osteoarthritis Knee Pain (ZAP2) study is a multicentre, randomised, double-blinded, placebo-controlled trial over 24 months. Patients≥50 years who had significant knee pain (defined as a visual analogue scale (VAS)≥40 mm) and MRI-detected knee BML were randomised to receive either ZA (5 mg in 100 ml saline) or placebo (100 ml saline) once-yearly. Those with severe knee OA (joint space narrowing (JSN) on X-ray of Grade 3 using the Osteoarthritis Research Society International (OARSI) atlas) were excluded. Outcomes included knee pain and function by the Western Ontario and McMasters Universities Osteoarthritis Index (WOMAC), knee pain by VAS and change in knee total BML size (sum of medial femoral, lateral femoral, medial tibial, lateral tibial and patellar sites) by proton density weighted MRI from baseline to 24 months. Effect modification of the absence or presence of radiographic OA (JSN grade 0 or grade 1–2) was pre-specified. Mixed effect modelling using an intent-to-treat design was performed for data analyses. Adjustment for baseline values were performed for knee pain and function outcomes due to baseline imbalances. Results: 223 patients (mean ±SD age 62.0±8.0 years, 117 females) were enrolled. At baseline, mean ±SD knee WOMAC pain (0–500), WOMAC function (0–1700) and VAS pain scores (0–100) were 200.0±105.0, 656.9±352.9 and 51.0±20.5, and median BML size was 489.6 mm. After 24 months, there was no significant difference in changes in knee pain (WOMAC pain: −37.5 vs −58.0, p=0.205; VAS pain: −11.5 vs −16.8, p=0.17) or function scores (−134.9 vs −159.2, p=0.65), or knee BML size (−33.5 mm vs −11.7 mm, p=0.68) between the ZA group and the placebo group. Pre-specified analyses consistently showed that ZA was more effective than placebo in patients without radiographic OA (JSN Grade 0) on changes in WOMAC pain (−88.3 vs −42.6, p=0.21), VAS pain (−21.8 to −8.3, p=0.11), WOMAC function (−296.9 vs −78.5, p=0.06) and BML size (−67.4 vs 98.2, p=0.14). Adverse events were more frequent in the ZA group, primarily flu-like symptoms and musculoskeletal pain and stiffness. Conclusions: Once-yearly infusion of ZA did not significantly reduce knee pain or BML size overall in knee OA patients over 24 months but may have symptomatic benefit in milder disease. References: Laslett LL, et al. Ann Rheum Dis 2012;71:1322–8. Disclosure of Interest: None declared DOI: 10.1136/annrheumdis-2018-eular.1539 Citation: Ann Rheum Dis, volume 77, supplement Suppl, year 2018, page A57Session: Opening plenary abstract session