FRI0534 Radiographic outcomes were associated with pain and function responses: post-hoc analysis from a phase 2 study of a wnt pathway inhibitor, sm04690, for knee osteoarthritis treatment

Background SM04690, a small molecule intra-articular (IA) Wnt pathway inhibitor is in development as a potential disease modifying knee osteoarthritis drug. A phase 2, 52 week, randomised controlled trial evaluated changes in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain and Function and medial joint space width (mJSW). It was hypothesised that observed mJSW increases led to WOMAC subscore responder improvements. To address this question, a concordance analysis was performed. Objectives To evaluate concordance, or level of agreement, between mJSW and WOMAC Pain and Function ‘responders.’ Methods Subjects with ACR-defined knee OA, Kellgren-Lawrence (KL) grades 2–3, received 2 mL IA SM04690 (0.03, 0.07, or 0.23 mg) or placebo (PBO) in the target (most painful) knee. WOMAC Pain [0–50] and Function [0–170] were assessed at Weeks 0, 4, 13, 26, 39 and 52 and knee radiographs at Weeks 0, 26 and 52. Baseline-adjusted logistic regression group analyses estimated concordance between mJSW change and pain and function changes for responders who achieved both WOMAC Pain and Function improvements of >50% and>20 [scaled to 100] points. Receiver-operator characteristic (ROC) curves were generated with area under curve (AUC) to estimate concordance (AUC >0.7=‘acceptable’ and >0.8=‘excellent’ concordance 1 ). ITT and two subgroups were analysed: 1) unilateral symptomatic knee OA (pre-specified: UNI) and 2) unilateral symptomatic knee OA without widespread pain or comorbid symptoms (Widespread Pain Index≤4 and Symptom Severity≤2, post-hoc: UNI-WP). Results 455 subjects were enrolled (mean age 60.3 [±8.7] years, BMI 29.9 [±4.6] kg/m 2 , 268 [58.9%] female, 292 [64.2%] KL Grade 3, 164 [36.0%] UNI knee OA). In the ITT, approximately 53% were responders across all groups. In UNI, 20 (56%) 0.03 mg; 20 (63%) 0.07 mg; 23 (64%) 0.23 mg and 15 (47%) PBO, and in UNI-WP, 15 (56%) 0.03 mg; 16 (62%) 0.07 mg; 19 (70%) 0.23 mg and 12 (44%) PBO were responders. The 0.03 mg (UNI, NS; UNI-WP, p=0.047) and 0.07 mg (UNI, p=0.009; UNI-WP, p=0.013) doses also demonstrated increased mJSW compared to PBO at Week 52. In ITT, no treatment group achieved AUC >0.7 (figure 1). In UNI, the 0.07 mg dose demonstrated ‘acceptable’ concordance between response and mJSW (AUC=0.783). In UNI-WP, the 0.07 mg dose showed ‘excellent’ concordance (AUC=0.825). Conclusions In this post-hoc analysis, treatment with SM04690 maintained or increased mJSW in the 0.03 and 0.07 mg doses compared to PBO over 52 weeks. In UNI and UNI-WP 0.07 mg cohorts, changes in mJSW were concordant with WOMAC Pain and Function response. Reference [1] Hosmer DW, Lemeshow S. Applied Logistic Regression2000. New York: John-Wiley & Sons, Inc. Disclosure of Interest C. Swearingen Shareholder of: Samumed, LLC, Employee of: Samumed, LLC, J. Tambiah Shareholder of: Samumed, LLC, Employee of: Samumed, LLC, A. Gibofsky Shareholder of: AbbVie, Amgen, Johnson and Johnson, GSK, Regeneron, Consultant for: AbbVie, Pfizer, Horizon, Iroko, Celgene, Novartis/Sandoz, Samumed, LLC, Speakers bureau: AbbVie, Celgene, Pfizer, N. Lane Consultant for: Samumed, LLC, T. McAlindon Grant/research support from: Samumed, LLC, Consultant for: Samumed, LLC, Astellas, Flexion, Pfizer, Regeneron, Seikugaku, M. Hochberg Consultant for: Bioberica, EMD Serono, Novartis Pharma AG, Plexxikon, Pfizer, Proximagen, Regeneron, Samumed, LLC, Theralogix LLC