Efficacy and Safety of Upadacitinib in Patients With Active Psoriatic Arthritis and Axial Involvement

Results From Two Phase 3 Studies

Xenofon Baraliakos; Roberto Ranza; Andrew Östör; Francesco Ciccia; Laura C. Coates; Simona Rednic; Jessica A. Walsh; Kevin Douglas; Tianming Gao; Koji Kato; In-Ho Song; Fabiana Ganz; Atul Deodhar


Arthritis Res Ther. 2023;25(56) 

In This Article

Abstract and Introduction


Background: The objective of this post-hoc analysis was to assess the efficacy and safety of upadacitinib in psoriatic arthritis (PsA) patients with axial involvement.

Methods: Post-hoc analysis of SELECT-PsA 1 and SELECT-PsA 2 in patients randomized to upadacitinib 15 mg (UPA15), placebo (switched to UPA15 at week 24), or adalimumab 40 mg (ADA; SELECT-PsA 1 only). Axial involvement was determined by investigator judgement (yes or no; based on the totality of available clinical information, such as duration and characteristics of back pain, age of onset, and previous lab investigations and imaging, if available) alone, or investigator judgement and patient-reported outcome (PRO)-based criteria (Bath Ankylosing Spondylitis Disease Activity Index [BASDAI] ≥ 4 and BASDAI Q2 ≥ 4). Efficacy outcomes that describe axial disease activity, including BASDAI endpoints, such as change from baseline in the overall BASDAI score or proportion of patients achieving BASDAI50 (≥ 50% improvement from baseline), as well as Ankylosing Spondylitis Disease Activity Score (ASDAS) endpoints, such as mean change from baseline in overall ASDAS or proportion of patients achieving ASDAS inactive disease or low disease activity, were evaluated at weeks 12, 24, and 56, with nominal P-values shown. Treatment-emergent adverse events (TEAEs) are summarized through week 56.

Results: 30.9% of patients in SELECT-PsA 1 and 35.7% in SELECT-PsA 2 had axial involvement by investigator judgement alone; 22.6% (SELECT-PsA 1) and 28.6% (SELECT-PsA 2) had axial involvement by investigator judgement and PRO-based criteria. Greater proportions of patients achieved BASDAI50 with UPA15 versus placebo using either criterion, and versus ADA using investigator judgement alone, at week 24 in SELECT-PsA 1 (investigator alone: UPA15, 59.0%, placebo, 26.9%, P < 0.0001, ADA, 44.1%, P = 0.015; investigator and PRO-based: UPA15, 60.4%, placebo, 29.3%, P < 0.0001, ADA, 47.1%, P = 0.074), with comparable findings in SELECT-PsA 2. Similar results were observed with UPA15 for additional BASDAI and ASDAS endpoints at weeks 12 and 24, with improvements maintained at week 56. Rates of TEAEs were generally similar across sub-groups irrespective of axial involvement status.

Conclusions: PsA patients with axial involvement determined by predefined criteria showed greater BASDAI and ASDAS responses with UPA15 versus placebo, and numerically similar/greater responses versus ADA. Safety results were generally comparable between patients with or without axial involvement.

Trial Registration: ClinicalTrials.gov: SELECT-PsA 1, NCT03104400; SELECT-PsA 2, NCT0310437.