Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Feb;12(1):173-202.
doi: 10.1007/s40744-024-00736-4. Epub 2025 Jan 6.

Real-World Persistence and Effectiveness of Upadacitinib versus Other Janus Kinase Inhibitors and Tumor Necrosis Factor Inhibitors in Australian Patients with Rheumatoid Arthritis

Peter Youssef  1   2   3 , Sabina Ciciriello  3   4 , Talib Tahir  5 , Joanna Leadbetter  6 , Belinda Butcher  6   7 , Miriam Calao  8 , Nicole Walsh  8 , Catherine O'Sullivan  3 , Tegan Smith  3 , Geoffrey Littlejohn  9   10   11
Affiliations

Real-World Persistence and Effectiveness of Upadacitinib versus Other Janus Kinase Inhibitors and Tumor Necrosis Factor Inhibitors in Australian Patients with Rheumatoid Arthritis

Peter Youssef et al. Rheumatol Ther. 2025 Feb.

Erratum in

Abstract

Introduction: This study sought to describe treatment patterns, persistence, and effectiveness of upadacitinib (UPA) alone and compared to other Janus kinase inhibitors (JAKis) or tumor necrosis factor inhibitors (TNFis) in patients with rheumatoid arthritis (RA).

Methods: This retrospective, non-interventional study used the OPAL dataset, derived from electronic medical records. Patients initiated UPA (N = 2624), other JAKis (baricitinib and tofacitinib [N = 925]), or TNFis (adalimumab, etanercept, certolizumab, golimumab, infliximab [N = 3540]) between May 2020 and March 2023. Median persistence (Kaplan-Meier) and effectiveness (Disease Activity Score 28-joint C-reactive protein, three variables [DAS28CRP{3}]) were evaluated for UPA-treated patients and in three propensity score-matched cohorts: UPA monotherapy versus combination therapy, UPA versus other JAKis, and UPA versus TNFis.

Results: In patients prescribed UPA, 41.3% were ≥ 65 years old, 33.8% were prescribed as first-line advanced therapy, and 27.2% were prescribed monotherapy. Persistence on UPA was 26.6 months (95% confidence intervals: 24.4, 29.9) and longest in earlier lines of therapy. The DAS28CRP(3) remission rate was 73% at 3 months, with improvements observed across lines of therapy. UPA monotherapy and combination therapy had similar persistence (27.8 [23.5, 33.4] versus 30.4 months [22.1, 35.3], p = 0.84) and effectiveness. UPA showed longer persistence than other JAKis (28.8 [25.6, 32.4] versus 17.2 months [14.9, 19.8], p < 0.001) and TNFis (26.6 [24.9, 30.8] versus 13.3 months [11.5, 14.5], p < 0.001). DAS28CRP(3) remission rates were greater at 3 months for UPA than other JAKis (75.0% versus 61.5%) and TNFis (72.7% versus 59.5%). In unmatched subgroups, compared to cycling between TNFis, switching to UPA from other JAKis or TNFis resulted in longer persistence (JAKi-to-UPA: 25.3 [16.1, not reached]; TNFi-to-UPA: 27.8 [23.2, 35.4]; TNFi-to-TNFi: 9.6 [8.4, 10.7]) and greater DAS28CRP(3) remission rates over 9 months.

Conclusions: Overall, the breadth and depth of data from this large real-world dataset continue to support a favorable clinical profile of UPA for the treatment of RA and may inform treatment choices in everyday clinical practice.

Keywords: Australia; Biologic disease-modifying antirheumatic drug (bDMARD); Janus kinase inhibitor (JAKi); Persistence; Real-world effectiveness; Remission; Rheumatoid arthritis (RA); Targeted synthetic disease-modifying antirheumatic drug (tsDMARD); Tumor necrosis factor inhibitor (TNFi); Upadacitinib (UPA).

Plain language summary

This study of patients treated in clinical practice looked at how a targeted medication, a Janus kinase inhibitor (JAKi) called upadacitinib, was used to treat rheumatoid arthritis, a disease that causes joint pain and damage. The researchers wanted to see how long patients continued treatment with upadacitinib, how effective it was at reducing symptoms, and how these outcomes compared to other JAKis (baricitinib and tofacitinib) and tumor necrosis factor inhibitors, a different type of advanced medication. Patients continued upadacitinib treatment for a median time of over 2 years. Those patients who had never previously used JAKis or tumor necrosis factor inhibitors continued upadacitinib treatment longer than those patients with previous experience. Patients treated with upadacitinib alone or in combination with less advanced medications like methotrexate continued treatment for a similar length of time. Compared to other JAKis and tumor necrosis factor inhibitors, patients prescribed upadacitinib stayed on their treatment the longest. Upadacitinib helped to lessen the symptoms of rheumatoid arthritis, whether patients took it alone or in combination with medications like methotrexate. After 3 months, more patients treated with upadacitinib experienced reduced symptoms than patients treated with other JAKis or tumor necrosis factor inhibitors. Additionally, patients who switched to upadacitinib after using other JAKis or tumor necrosis factor inhibitors reduced their symptoms and continued treatment for longer than those switching between tumor necrosis factor inhibitors. Overall, patients treated with upadacitinib continued treatment for longer and saw greater improvements in the symptoms of rheumatoid arthritis than patients prescribed other advanced medications.

PubMed Disclaimer

Conflict of interest statement

Declarations. Conflict of Interest: Financial arrangements of the authors with companies whose products may be related to the present manuscript are listed, as declared by the authors. Peter Youssef: has received honoraria for lectures and consulting from AbbVie, AstraZeneca, Eli Lilly, Celltrion, Pfizer, Sandoz, Janssen, and Novartis. Sabina Ciciriello and Talib Tahir: Nothing to disclose. Belinda Butcher: Acts as a statistical consultant to many pharmaceutical and device companies, including those who manufacture products for RA. Joanna Leadbetter: Acts as a statistical consultant to many pharmaceutical and device companies, including those who manufacture products for RA. Tegan Smith: Former employee of OPAL Rheumatology Ltd, current employee of Hunter Medical Research Institute, Newcastle, NSW, Australia. Catherine O’Sullivan: Former employee of OPAL Rheumatology Ltd. Miriam Calao: Employee of AbbVie Pty Ltd. Nicole Walsh: Employee of AbbVie Pty Ltd. and may hold stock or options. Geoffrey Littlejohn: has received honoraria for lectures, presentations, consulting, or support/sponsorship for scientific conferences from AbbVie, AstraZeneca, Gilead, and GSK. Ethical Approval: The activities of OPAL Rheumatology Ltd have received overarching ethics approval from the University of New South Wales Human Research Ethics Committee (HC17799), based on opt-out patient consent. This protocol was approved by the University of New South Wales Human Research Ethics Committee (HC230281). Inclusion of data in the OPAL dataset was based on opt-out consent, all data are de-identified.

Figures

Fig. 1
Fig. 1
Sankey diagram of treatment patterns for patients prescribed UPA as monotherapy or in combination with csDMARDs. Data include the treatment switching patterns of all patients prescribed UPA as monotherapy or in combination with csDMARDs through 21 months. Patients who stopped using UPA entirely are labeled as stopped. Patients with no data after a given timepoint, either because they were lost to follow-up or because the next timepoint was beyond the end of the study window, are labeled as no further follow-up. csDMARDs conventional synthetic disease-modifying antirheumatic drugs, UPA upadacitinib
Fig. 2
Fig. 2
Treatment persistence for patients prescribed UPA from the full analysis set and by line of advanced therapy. Kaplan–Meier analysis estimating treatment persistence through 33 months is presented for all patients prescribed UPA in the a full analysis set and b categorized by the line of advanced therapy. The number of patients at risk is the total number of patients included at each timepoint. UPA upadacitinib
Fig. 3
Fig. 3
Proportion of patients prescribed UPA in each DAS28CRP(3) category from the full analysis set overall and by line of advanced therapy. Data include the proportion of patients categorized by DAS28CRP(3) score for patients who initiated UPA in a the full analysis set and categorized by b first-line, c second-line, and d third-line advanced therapy at index and 3-, 9-, 15-, and 21-months post-index. All data are reported as observed. The number of patients with data available for calculation of a DAS28CRP(3) score and the number of patients with a missing score are reported below. The categories for DAS28CRP(3) score were defined as: clinical remission (< 2.6), low disease activity (LDA; ≥ 2.6 to ≤ 3.2), moderate disease activity (MDA; > 3.2 to < 5.1), and high disease activity (HDA; ≥ 5.1). DAS28CRP(3) disease activity score, 28-joint C-reactive protein (three variables); UPA upadacitinib
Fig. 4
Fig. 4
Treatment persistence and proportion of patients in DAS28CRP(3) categories for patients prescribed UPA as monotherapy and UPA in combination with csDMARDs from the propensity score-matched cohort. a Kaplan–Meier analysis estimating treatment persistence through 33 months and b the proportion of patients in each DAS28CRP(3) category at index and 3-, 9-, 15-, and 21-months post-index for patients in the UPA monotherapy and UPA combination therapy treatment group from the propensity score-matched cohort. All data are reported as observed. For the Kaplan–Meier curves, the number of patients at risk is the total number of patients included at each timepoint. P values were calculated from a log-rank test between treatment groups. For DAS28CRP(3) categories, the number of patients with data available for calculation of a DAS28CRP(3) score and the number of patients with a missing score are reported below. The categories for DAS28CRP(3) score were defined as: clinical remission (< 2.6), low disease activity (LDA; ≥ 2.6 to ≤ 3.2), moderate disease activity (MDA; > 3.2 to < 5.1), and high disease activity (HDA; ≥ 5.1). Differences in the proportion of patients in each category at each timepoint for the 2 treatment groups were determined by ANOVA. csDMARDs conventional synthetic antirheumatic disease-modifying drugs; DAS28CRP(3) 28-joint disease activity score based on C-reactive protein (three variables); UPA upadacitinib
Fig. 5
Fig. 5
Treatment persistence and proportion of patients in DAS28CRP(3) categories for patients prescribed UPA and other JAKis from the propensity score-matched cohort. DAS28CRP(3), 28-joint disease activity score based on C-reactive protein (three variables). a Kaplan–Meier analysis estimating treatment persistence and b the proportion of patients in each DAS28CRP(3) category at index and 3-, 9-, 15-, and 21-months post-index for patients in the UPA or other JAKi treatment groups from the propensity score-matched cohort. All data are reported as observed. For the Kaplan–Meier curves, the number of patients at risk is the total number of patients included at each timepoint. Nominal P values were calculated from a log-rank test between UPA and other JAKis treatment groups. For DAS28CRP(3) categories, the number of patients with data available for calculating a DAS28CRP(3) score and the number of patients with a missing score are reported below. The categories for DAS28CRP(3) score were defined as: clinical remission (< 2.6), low disease activity (LDA; ≥ 2.6 to ≤ 3.2), moderate disease activity (MDA; > 3.2 to < 5.1), and high disease activity (HDA; ≥ 5.1). Differences in the proportion of patients in each category at each timepoint for the two treatment groups were determined by ANOVA. JAKi Janus kinase inhibitor; UPA upadacitinib. *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001
Fig. 6
Fig. 6
Treatment persistence and proportion of patients in DAS28CRP(3) categories for patients prescribed UPA and TNFis from the propensity score-matched cohort. DAS28CRP(3), 28-joint disease activity score based on C-reactive protein (three variables). a Kaplan–Meier analysis estimating treatment persistence and b the proportion of patients in each DAS28CRP(3) category at index and 3-, 9-, 15-, and 21-months post-index for patients in the UPA or TNFis treatment groups from the propensity score-matched cohort. All data are reported as observed. For the Kaplan–Meier curves, the number of patients at risk is the total number of patients included at each timepoint. Nominal P values were calculated from log-rank test between UPA and TNFi treatment groups. For DAS28CRP(3) categories, the number of patients with data available for calculating a DAS28CRP(3) score and the number of patients with a missing score are reported below. The categories for DAS28CRP(3) score were defined as: clinical remission (< 2.6), low disease activity (LDA; ≥ 2.6 to ≤ 3.2), moderate disease activity (MDA; > 3.2 to < 5.1), and high disease activity (HDA; ≥ 5.1). Differences in the proportion of patients in each category at each timepoint for the two treatment groups were determined by ANOVA. TNFi tumor necrosis factor inhibitor; UPA upadacitinib. *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001
Fig. 7
Fig. 7
Treatment persistence and proportion of patients in DAS28CRP(3) categories for patients who switched treatment in second line of advanced therapy. a Kaplan–Meier analysis estimating treatment persistence and b the proportion of patients in each DAS28CRP(3) category at index and 3-, 9-, 15-, and 21-months post-index for patients who switched from other JAKis to UPA, from TNFis to UPA, or cycled from TNFi to TNFi. All data are reported as observed. For the Kaplan–Meier curves, the number of patients at risk is the total number of patients included at each timepoint. Nominal P values were calculated from log-rank tests between the subgroup TNFis to UPA versus other JAKis to UPA and versus TNFi to TNFi. For DAS28CRP(3) categories, the number of patients with data available for calculation of a DAS28CRP(3) score and the number of patients with a missing score are reported below. The categories for DAS28CRP(3) score were defined as: clinical remission (< 2.6), low disease activity (LDA; ≥ 2.6 to ≤ 3.2), moderate disease activity (MDA; > 3.2 to < 5.1), and high disease activity (HDA; ≥ 5.1). DAS28CRP(3) 28-joint disease activity score based on C-reactive protein (three variables); JAKi Janus kinase inhibitor; TNFi tumor necrosis factor inhibitor; UPA upadacitinib
See this image and copyright information in PMC

References

    1. Youssef P, Ciciriello S, Tahir T, Smith T, O’Sullivan C, Leadbetter J et al. Persistence, effectiveness and treatment patterns of upadacitinib in over 2600 Australian rheumatoid arthritis patients: a retrospective analysis from the OPAL dataset - ACR meeting abstracts. Arthritis Rheumatol. Available from: https://acrabstracts.org/abstract/persistence-effectiveness-and-treatmen....
    1. Ciciriello S, Youssef P, Tahir T, Smith T, O’Sullivan C, Leadbetter J et al. Upadacitinib vs TNFi and other JAKi treatment outcomes in Australian rheumatoid arthritis patients: descriptive comparison of persistence and effectiveness using the OPAL dataset - ACR meeting abstracts. Arthritis Rheumatol; p.76. Available from: https://acrabstracts.org/abstract/upadacitinibvs-tnfi-and-other-jaki-tre....
    1. Smolen JS, Aletaha D, McInnes IB. Rheumatoid arthritis. Lancet. 2016;388:2023–38. - PubMed
    1. Smolen JS, Aletaha D, Barton A, Burmester GR, Emery P, Firestein GS, et al. Rheumatoid arthritis. Nat Rev Dis Prim. 2018;4:18001. - PubMed
    1. Lin Y-J, Anzaghe M, Schülke S. Update on the pathomechanism, diagnosis, and treatment options for rheumatoid arthritis. Cells. 2020;9:880. - PMC - PubMed

LinkOut - more resources