Sustained corticosteroid-sparing effects of upadacitinib maintenance therapy in patients with moderate-to-severe crohn’s disease: 2-year results from the U-ENDURE long-term extension study

Background: Long-term use of corticosteroids (CS) is a concern for patients (pts) with Crohn’s disease (CD), as prolonged use is associated with increased mortality and adverse outcomes. Upadacitinib (UPA), an oral, selective, reversible JAK inhibitor, approved for moderate-to-severe CD, demonstrated CS-sparing efficacy with a consistent safety profile through maintenance week (wk) 52. [1] [2] [3] [4] We evaluated the long-term efficacy of UPA in achieving CS-free clinical and endoscopic outcomes in pts in the U-ENDURE long-term extension (LTE) study.

Methods: Pts who completed the U-ENDURE 52-wk maintenance study were eligible for the LTE study where they continued their assigned treatment (1x daily UPA 15mg [UPA15] or UPA 30mg [UPA30]). Efficacy was assessed from LTE wk0-48, among pts with 2 year (yr) (100wk) of total maintenance therapy, within the overall pts and in pts with baseline (BL) CS use (induction wk0). CS-free endpoints (without CS use for≥90 days), included clinical remission (per stool frequency/abdominal pain score [SF/APS] and CDAI), endoscopic response, and endoscopic remission, were evaluated using both as-observed and nonresponder imputation (NRI) methods, with NRI analysis presented in text. Safety was not evaluated here but was reported previously [3].

Results: Among pts with 2 yr of total maintenance therapy, 35.7% (87/244) received CS at baseline, with 13.1% (32/244) receiving CS at any time during the LTE regardless of BL CS use. Among pts taking BL CS and those who received CS at any time during the LTE (n=20), the median (range) time of CS use was 127.5 (2-1589) days. CS-free clinical remission at LTE wk0 and wk48 per SF/APS (overall pts: UPA15, 71.7%, 58.9%; UPA30, 81.8%, 57.7%; pts with BL CS use: UPA15, 71.8%, 53.8%; UPA30, 79.2%, 54.2%, NRI, [Fig. 1A, B]) and CDAI criteria (overall pts: UPA15, 74.8%, 60.7%; UPA30, 83.9%, 65.0%; pts with BL CS use: UPA15, 74.4%, 56.4%; UPA30, 81.3%, 58.3%, NRI, [Fig. 1C, D]). At LTE wk48, high rates of UPA-treated pts achieved CS-free endoscopic response (overall pts: UPA15, 48.6%; UPA30, 55.5%; pts with BL CS use: UPA15, 38.5%; UPA30, 60.4%, NRI) and CS-free endoscopic remission (overall pts: UPA15, 33.6%; UPA30, 46.0%; pts with BL CS use: UPA15, 28.2%; UPA30, 54.2%, NRI, [Fig. 2]).

Fig. 1 Corticosteroid-Free Clinical Remission in Patients Treated With Upadacitinib Through Week 48 of the U-ENDURE Long-Term Extension Study. Patients on CS at baseline of induction began a mandatory taper at induction wk 4. CS-free clinical remission was evaluated throughout LTE among patients who abstained from CS use for 90 days before assessment. Patients were blinded until the last patient completed wk 52 of maintenance. For NRI analysis on binary variables, patients were categorized as “nonresponder” after initiation of any protocol rescue medications or missing data. For AO analysis, all available baseline measurements before initiation of open-label UPA 30 mg QD rescue in LTE were used for analysis, and no missing data were imputed. A and B) SF/APS clinical remission was defined as average daily, very soft, or liquid SF≤2.8 and average daily, AP score≤1.0, and both not greater than induction baseline. C and D) Clinical remission per CDAI was defined as CDAI<150. AO, as observed; CDAI, Crohn’s disease activity index; CI, confidence interval; CS, corticosteroid; LTE, long-term extension; NRI, nonresponder imputation; QD, once daily; SF/APS, stool frequency/abdominal pain score; UPA, upadacitinib, wk, week.

Fig. 2 Corticosteroid-Free Endoscopic Outcomes in Patients Treated With Upadacitinib Through Week 48 of the U-ENDURE Long-Term Extension Study. Patients on CS at baseline of induction began a mandatory taper at induction wk 4. CS-free endoscopic outcomes were evaluated at LTE wk 0 and wk 48 among patients who abstained from CS use for 90 days before assessment. Patients were blinded until the last patient completed wk 52 of maintenance. For NRI analysis on binary variables, patients were categorized as “nonresponder” after initiation of any protocol rescue medications or missing data. For AO analysis, all available baseline measurements before initiation of open-label UPA 30 mg QD rescue therapy in LTE were used for analysis, and no missing data were imputed. A and B) Endoscopic response was defined as a decrease in SES-CD>50% from baseline of the induction study (or for patients with an SES-CD of 4 at baseline of the induction study, at least a 2-point reduction from baseline), as scored by a central reviewer. Endoscopies were performed annually. C and D) Endoscopic remission was defined as SES-CD≤4 and at least a 2-point reduction from baseline and no subscore>1 in any individual variable baseline, as scored by a central reviewer and measured up to LTE wk 48. AO, as-observed; CDAI, Crohn’s disease activity index; CI, confidence interval; CS, corticosteroid; LTE, long-term extension; NRI, nonresponder imputation; QD, once daily; SES-CD, simple endoscopic score for patients with Crohn’s disease; UPA, upadacitinib; wk, week.

Conclusion: Pts with moderate-to-severe CD who received long-term UPA maintenance treatment sustained high rates of CS-free clinical and endoscopic outcomes, suggesting that UPA may serve as an effective long-term CS-sparing therapy.

Publication History

Article published online:
04 September 2025

• References

• 1 RINVOQ [package insert], extended-release tablets, FDA Prescribing Information. AbbVie Inc; 2022. Accessed 1 June 2022.
• 2 RINVOQ [package insert], prolonged-release tablets. EMA Summary of Product Characteristics. AbbVie; 2022. Accessed 1 June 2022.
• 3 Loftus EV, Panes J. et al. N Engl J Med 2023; 388: 1966-1980
• 4 Dubinsky MC. et al. Journal of Crohn&apos;s and Colitis 2023; 17: i104-i106