THU0207 SUSTAINABILITY OF RESPONSE TO UPADACITINIB AS MONOTHERAPY OR IN COMBINATION AMONG PATIENTS WITH RHEUMATOID ARTHRITIS AND PRIOR INADEQUATE RESPONSE TO CONVENTIONAL SYNTHETIC DMARDS

Abstract

Background/purpose: To assess long‐term sustainability of responses to upadacitinib (UPA), with (+) or without (‐) background csDMARD(s), in rheumatoid arthritis (RA). Methods: Data are from two phase 3 randomized, controlled trials of UPA‐15 mg or‐30 mg once‐daily in patients with roughly similar baseline characteristics: SELECT‐NEXT (12‐week), csDMARD‐inadequate response (IR) patients +csDMARD(s); SELECT‐MONOTHERAPY (14‐week), MTX‐IR patients‐csDMARD(s). Patients could enter a long‐term extension of 5 years. This post hoc analysis evaluated clinical remission (REM: CDAI ≤2.8; SDAI ≤3.3), low disease activity (LDA: CDAI ≤10; SDAI ≤11), and DAS28 (CRP) <2.6/≤3.2 at first occurrence before Week 84, followed by 3, 6, and 12‐month evaluations for the total number of UPA‐15 patients. Kaplan‐Meier evaluated sustainability among those achieving REM/LDA and defined as time to earliest date of losing response at two consecutive visits or discontinuation. Predictive ability of time to REM/LDA was assessed using Harrell's concordance (c)‐index. Follow up continued until Week 84. Results: Through Week 84, CDAI REM/LDA was achieved in 43%/79% of UPA‐15+ csDMARD(s) (SELECT‐NEXT) and 37%/76% of UPA‐15‐csDMARD(s) (SELECT‐MONOTHERAPY). 35%/25% of UPA‐15+ csDMARD(s) and 27%/23% of UPA‐15‐csDMARD(s) achieved sustained CDAI REM through 6/12 months after first occurrence (Table). 64%/56% of UPA‐15+ csDMARD(s) and 61%/56% of UPA‐15‐csDMARD(s) achieved sustained CDAI LDA. Time to initial REM/LDA did not appear associated with sustained response (c‐indices [95% CI] CDAI REM: +csDMARD[s], 0.541 [0.47, 0.62] and‐csDMARD[s], 0.568 [0.49, 0.65]; CDAI LDA: +csDMARD[s], 0.521 [0.46, 0.58] and‐csDMARD[s], 0.498 [0.43, 0.56]). Through follow‐up, 55% of UPA‐15+ csDMARD(s) patients and 62% of UPA‐15‐csDMARD(s) remained in REM while 72% and 70% of patients remained in LDA, respectively (Figure). Similar results were observed for SDAI REM/LDA and DAS28 (CRP) <2.6/≤3.2. Conclusions: Greater than 25/50% of csDMARD‐IR patients receiving UPA‐15+/‐csDMARD therapy achieved sustained REM/LDA across disease activity measures. UPA‐15 sustainability appeared comparable +/‐csDMARDs through 84 weeks. CDAI, clinical disease activity index; csDMARD, conventional synthetic disease modifying antirheumatic drug; UPA, upadacitinib. n refers to the total number of patients randomized to UPA 15 mg with (+) or without (‐) csDMARD(s). Data were censored at the cut‐off (22 March 2018 for SELECT‐NEXT and 25 May 2019 for SELECT‐MONOTHERAPY, when all patients had reached Week 84 visit); censored = stopped collecting data. Data are summarized, with no adjustments applied to account for between‐study differences. Non‐responder imputation was used for missing data.