POS0907 EFFECT OF UPADACITINIB ON REDUCING PAIN IN PATIENTS WITH ACTIVE ANKYLOSING SPONDYLITIS AND INADEQUATE RESPONSE TO NONSTEROIDAL ANTI-INFLAMMATORY DRUGS

Abstract

Background:Pain is a debilitating symptom of ankylosing spondylitis (AS) and negatively impacts patient (pt) lives. Upadacitinib (UPA), a Janus kinase (JAK) inhibitor engineered for increased selectivity for JAK1 over JAK2, JAK3, and tyrosine kinase 2, 1 showed significant efficacy vs placebo (PBO) in the randomized phase 2/3 SELECT-AXIS 1 study in active AS.2Objectives:To evaluate the efficacy of UPA on multiple pain assessments through 64 weeks (wks).Methods:SELECT-AXIS 1 (NCT03178487) enrolled adults with active AS, who had an inadequate response, intolerance or contraindications to ≥2 NSAIDs, were biologic DMARD naive and met the modified New-York Criteria. Pts were randomized 1:1 to UPA 15 mg once daily (QD, n=93) or PBO (n=94) for 14 wks (Period 1), followed by open-label UPA 15 mg QD during 90-wk extension (Period 2); reported here are data through wk 64. Pain endpoints included the proportion of pts achieving ≥30%/≥50%/≥70% reduction in Pt’s Global Assessment (PGA) of pain on a 0–10 numeric rating scale (NRS), minimal clinically important difference (MCID, defined as ≥1 point reduction or ≥15% reduction from baseline [BL]) in PGA of pain, and much better improvement (MBI, defined as ≥2 point reduction and ≥33% reduction from BL) in PGA of pain. In addition, mean change from baseline in PGA of pain, BASDAI questions 2 (neck/back/hip pain) and 3 (peripheral pain/swelling), and pt’s assessment of back pain and nocturnal back pain NRS scores (NRS 0–10) were assessed. Non-responder imputation (binary endpoints) and mixed-effects model for repeated measurements (continuous endpoints) were used for missing data/dropouts in Period 1; as-observed analysis was used for Period 2.Results:A significantly higher proportion of pts receiving UPA vs PBO achieved reductions in all PGA of pain assessments as early as wk 2 that was sustained at all time points in Period 1; the only exception was ≥70% reduction in PGA of pain that was significant at wk 4 and sustained thereafter (Figure 1). For ≥30%/≥50%/≥70% reduction and MBI, the response rate increased over time with UPA; the difference for UPA vs PBO also continued to increase over time for ≥50% and ≥70% reduction endpoints. For MCID, an increase from BL to wk 2 was observed and plateaued thereafter. The mean change from BL in PGA of pain, BASDAI Q2, back pain, and nocturnal back pain NRS scores were significantly greater for UPA vs PBO at all time points in Period 1; BASDAI Q3 was significant at wk 8 and 14 (Figure 1). The effect of UPA on pain reduction was sustained through wk 64. PBO pts who switched to open-label UPA at wk 14 generally reached the same level of pain reduction as those initially randomized to UPA (Figure 1).Conclusion:In pts with active AS and an inadequate response/contraindication to NSAIDs, a greater proportion of patients treated with UPA achieved rapid, significant, and clinically meaningful reductions in pain vs PBO through 14 wks across multiple pain assessments. The reductions in pain were sustained over time, and pts who switched from PBO to UPA reached the same level of improvement as the continuous UPA group.