Efficacy and safety of upadacitinib for active ankylosing spondylitis refractory to biological therapy: a double-blind, randomised, placebo-controlled phase 3 trial

Abstract

Objectives: To evaluate the efficacy and safety of upadacitinib, a Janus kinase inhibitor, in patients with active ankylosing spondylitis (AS) with an inadequate response (IR) to biological disease-modifying antirheumatic drugs (bDMARDs).

Methods: Adults with active AS who met modified New York criteria and had an IR to one or two bDMARDs (tumour necrosis factor or interleukin-17 inhibitors) were randomised 1:1 to oral upadacitinib 15 mg once daily or placebo. The primary endpoint was Assessment of SpondyloArthritis international Society 40 (ASAS40) response at week 14. Sequentially tested secondary endpoints included Ankylosing Spondylitis Disease Activity score, Spondyloarthritis Research Consortium of Canada MRI spine inflammation score, total back pain, nocturnal back pain, Bath Ankylosing Spondylitis Functional Index, Bath Ankylosing Spondylitis Metrology Index and Maastricht Ankylosing Spondylitis Enthesitis Score. Results are reported from the 14-week double-blind treatment period.

Results: A total of 420 patients with active AS were randomised (upadacitinib 15 mg, n=211; placebo, n=209). Significantly more patients achieved the primary endpoint of ASAS40 at week 14 with upadacitinib vs placebo (45% vs 18%; p<0.0001). Statistically significant improvements were observed with upadacitinib vs placebo for all multiplicity-controlled secondary endpoints (p<0.0001). Adverse events were reported for 41% of upadacitinib-treated and 37% of placebo-treated patients through week 14. No events of malignancy, major adverse cardiovascular events, venous thromboembolism or deaths were reported with upadacitinib.

Conclusion: Upadacitinib 15 mg was significantly more effective than placebo over 14 weeks of treatment in bDMARD-IR patients with active AS. No new safety risks were identified with upadacitinib.

Trial registration number: NCT04169373.

Keywords: Magnetic Resonance Imaging; antirheumatic agents; inflammation; spondylitis, ankylosing.

Figure 1

Study design. Study design of the AS bDMARD-IR study of the SELECT-AXIS 2 master protocol is illustrated. *Patients in remission at week 104 could enter a remission-withdrawal period until flare or week 152. AS, ankylosing spondylitis; ASAS40, Assessment of SpondyloArthritis international Society 40 response; bDMARD, biological disease-modifying antirheumatic drug; IR, inadequate response; QD, once daily; SI, sacroiliac.

Figure 2

Patient disposition. *Patients were screened between 26 November 2019 and 20 May 2021, for the SELECT-AXIS 2 master protocol, which used a common screening platform to assign patients either to the AS bDMARD-IR study or nr-axSpA study. †Patients could have multiple criteria or multiple reasons for screening failure. Details of screen failure due to study eligibility criteria are presented in online supplemental table 1). ‡Other reasons included imaging, site, or system issues. §Patients did not fail screening (master protocol details provided in online supplemental methods). ¶Primary reason for discontinuation provided. AS, ankylosing spondylitis; bDMARD, biological disease-modifying antirheumatic drug; IR, inadequate response; nr-axSpA, non-radiographic axial spondyloarthritis; QD, once daily.

Figure 3

Multiplicity-controlled and key secondary endpoints at week 14. (A) ASAS20, ASAS40, ASAS PR and BASDAI50 responses at week 14 based on NRI-MI analysis. (B) Change from baseline in SPARCC MRI spine and sacroiliac joint scores at week 14 based on ANCOVA analysis. SPARCC MRI was assessed in patients with available baseline MRI data up to 3 days after the first dose of study drug and available week 14 MRI data up to the first dose of study drug in the open-label period. (C) Additional multiplicity-controlled key secondary efficacy endpoints at week 14; ANCOVA analysis for BASMI and MMRM analysis for other endpoints. MASES was assessed in patients with baseline enthesitis. (D) Change from baseline in ASQoL and ASAS Health Index at week 14 based on MMRM analysis. ANCOVA/MMRM analyses are based on as observed data. All endpoints were multiplicity controlled and tested sequentially (online supplemental figure 1), except for SPARCC MRI sacroiliac joint score. Error bars show 95% CI. Significant in multiplicity-controlled analysis: ***p<0.0001. Without adjustment for multiplicity (nominal): ††† p<0.0001. ANCOVA, analysis of covariance; ASAS, Assessment of SpondyloArthritis international Society; ASAS20, Assessment of SpondyloArthritis international Society 20 response; ASAS40, Assessment of SpondyloArthritis international Society 40 response; ASAS PR, Assessment of SpondyloArthritis international Society partial remission; ASDAS, Ankylosing Spondylitis Disease Activity Score; ASQoL, Ankylosing Spondylitis Quality of Life Score; BASDAI50, at least 50% improvement from baseline in Bath Ankylosing Spondylitis Disease Activity Index; BASFI, Bath Ankylosing Spondylitis Functional Index; BASMI, Bath Ankylosing Spondylitis Metrology Index; CRP, C-reactive protein; MASES, Maastricht Ankylosing Spondylitis Enthesitis Score; MMRM, mixed-effect model for repeated measures; NRI-MI, non-responder imputation incorporating multiple-imputation; QD, once daily; SPARCC, Spondyloarthritis Research Consortium of Canada.

Figure 4

ASAS40 response through week 14. NRI-MI analysis was used. Error bars show 95% CI. Significant in multiplicity-controlled analysis: ***p<0.0001. Without adjustment for multiplicity (nominal): †p<0.05; †††p<0.0001. ASAS40, Assessment of SpondyloArthritis international Society 40 response; NRI-MI, non-responder imputation incorporating multiple imputation; QD, once daily.

Figure 5

ASDAS responses at and through week 14. (A) Proportion of patients with ASDAS responses at week 14 was based on NRI-MI analysis. ASDAS low disease activity was defined as ASDAS (CRP) <2.1 and ASDAS inactive disease as ASDAS (CRP) <1.3. (B) Mean change from baseline in ASDAS (CRP) through week 14 was based on MMRM analysis, and the numbers of patients were as observed at each visit. Error bars show 95% CI. Significant in multiplicity-controlled analysis: ***p<0.0001. Without adjustment for multiplicity (nominal): †††p<0.0001. ASDAS, Ankylosing Spondylitis Disease Activity Score; CRP, C-reactive protein; MMRM, mixed-effect model for repeated measures; NRI-MI, non-responder imputation incorporating multiple-imputation; QD, once daily.