Multicentre, prospective, observational study on the effectiveness and safety of upadacitinib for active perianal fistulising Crohn’s disease in patients with prior anti-tumour necrosis factor failure: a study protocol

Abstract

Background

Perianal fistulising Crohn’s disease (pfCD) is a debilitating phenotype and remains a therapeutic challenge. Current guidelines recommend anti-tumour necrosis factor agents as first-line therapy for pfCD; however, there is still no consensus on the optimal pharmacological strategy once anti-TNF therapy fails. Upadacitinib, an oral selective Janus kinase 1 inhibitor, has demonstrated efficacy in moderate-to-severe Crohn’s disease. Nevertheless, real-world data specifically addressing its effectiveness and safety in the refractory pfCD population remain scarce. This study aims to evaluate the real-world effectiveness and safety of upadacitinib in patients with active pfCD who have failed anti-TNF therapy.

Methods and analysis

This is a multicentre, prospective, observational cohort study. A total of 90 adult patients with Crohn’s disease-associated perianal fistulas with active drainage and refractory to prior therapy with infliximab or adalimumab will be enrolled. Patients with severe luminal disease, undrained abscesses >2 cm, anal stenosis or diverting stomas will be excluded. Participants will receive upadacitinib 45 mg/day during the induction phase, followed by 15 or 30 mg/day during maintenance. All treatment decisions are made according to routine clinical practice and local product labels, not mandated by the study protocol, including upadacitinib dosing, dose adjustments and concomitant therapies. The primary outcome is a composite of clinical and radiological fistula response at week 12. Secondary outcomes, assessed at weeks 12 and 52, include perianal fistula-related, CD-related and safety outcomes.

Ethics and dissemination

Shanghai Jiao Tong University School of Medicine, Renji Hospital Ethics Committee has reviewed and approved this study (Approval No: LY2025-024-A). Findings will be disseminated through publication in peer-reviewed journals, presentations at scientific conferences and summaries provided to the participating institutions and relevant patient communities.

Trial registration number

ChiCTR2600117355.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The prospective cohort design allows for pre-specified, uniform data capture at multiple time points, minimising recall and information bias and enabling comprehensive longitudinal assessment of both effectiveness and safety outcomes over follow-up.

• A composite endpoint consisting of clinical and radiological assessments was used as the primary outcome.

• By using a prospective multicentre design and rigorous multi-modal endpoints (combining clinical and TOpCLASS-based radiological assessments), this study will bridge the gap between clinical trials and real-world practice.

• Due to the limitations inherent in observational study designs, it is possible that some confounding factors and sources of bias remain unaccounted for, both in the design phase and in subsequent statistical analysis.

Introduction

Crohn’s disease (CD) is a chronic, progressive inflammatory bowel disease with an increasing global burden, notably in East Asia.^{1 2} As one of the most common and challenging complications of CD, perianal fistulas occur in 30% to 70% of patients. This condition not only serves as a marker of more severe disease but also leads to symptoms such as pain, purulent drainage and faecal incontinence, which often result in diminished quality of life, increased disability risk and higher rates of surgical intervention.^3–5

Anti-tumour necrosis factor-alpha (TNF-α) agents, such as infliximab and adalimumab, have been the cornerstone of biological therapy for perianal fistulising Crohn’s disease (pfCD).⁶ However, primary non-response and secondary loss of response to these agents are frequent, affecting 30%–40% of patients initially and accumulating over time.⁷ This poses significant therapeutic challenges. Although alternative biologics, such as anti-integrins (eg, vedolizumab) and anti-interleukin (IL)-12/23 agents (eg, ustekinumab), are available, their effectiveness for fistula closure is not universal, leaving many patients with refractory disease.^8–11 Consequently, there is a persistent and critical need for effective and novel therapies for this population.

Upadacitinib is an oral, selective inhibitor of Janus kinase (JAK) 1 that was recently approved for treating moderate-to-severe CD. In a phase 3 clinical trial programme (U-EXCEL, U-EXCEED and U-ENDURE), upadacitinib demonstrated superior efficacy over placebo in achieving clinical remission and endoscopic response.¹² A post hoc analysis of these phase 3 trials further suggested that upadacitinib may be effective in patients with perianal fistulising Crohn’s disease, demonstrating higher rates of fistula closure and resolution of drainage compared with placebo.¹³ As a small molecule, upadacitinib offers the potential advantages of convenient oral administration and a rapid onset of action.

Despite the promising data from randomised controlled trials (RCTs),^{14 15} their stringent eligibility criteria may limit generalisability. Real-world evidence is essential for understanding the effectiveness and safety of upadacitinib in routine clinical practice, which includes a more heterogeneous patient population with complex comorbidities and prior drug exposures. Currently, real-world evidence for upadacitinib in anti-TNF-failed pfCD remains scarce. Existing studies are mostly retrospective or small-scale and lack unified radiological standards.^16–18 Therefore, a multicentre, prospective study in China is of great clinical value.

Objectives

Primary objective

To assess the real-world effectiveness of upadacitinib, as measured by the proportion of patients achieving combined fistula response (clinical and radiological fistula response) at week 12.

Secondary objectives

1. To evaluate the long-term effectiveness of upadacitinib in the treatment of CD-related perianal fistulas at week 52.

2. To characterise changes from baseline in the following measures: disease activity scores (Perianal Disease Activity Index (PDAI) and Crohn’s Disease Activity Index (CDAI)), patient-reported outcomes (Inflammatory Bowel Disease Questionnaire (IBDQ) and Wexner incontinence score (WIS)) and inflammatory biomarker levels (C-reactive protein (CRP) and faecal calprotectin).

3. To document the safety and tolerability profile of upadacitinib over the 52-week study period by monitoring the incidence, type and severity of adverse events (AEs).

Methods and analysis

Study design and setting

This prospective, multicentre, observational cohort study was launched on 13 March 2025, with planned completion on 30 June 2027, across nine tertiary care centres in China. This real-world study does not interfere with routine clinical care. All treatment decisions (including upadacitinib initiation, dosing and discontinuation) are made by treating physicians per standard practice. A unified investigator training programme has been conducted prior to study initiation across all participating centres to standardise upadacitinib prescribing practices, patient data collection procedures and follow-up. Clinical management and follow-up procedures reflect standard care practices at participating centres. Variability in clinical management is expected and recorded as part of routine clinical documentation. The coordinating centre will conduct regular on-site monitoring visits and remote data audits to oversee protocol adherence and data quality across all participating sites. A panel of 3–5 senior inflammatory bowel disease (IBD) specialists will centrally adjudicate treatment outcomes to ensure standardised and objective efficacy assessment and treatment decisions. All patients will receive multidisciplinary management for perianal Crohn's disease according to standardised institutional practice. Management may include abscess drainage, examination under anaesthesia (EUA), seton placement or removal, antibiotic therapy and other clinically indicated surgical procedures.

The study relies on routine clinical examinations, radiological imaging and laboratory tests to evaluate drug safety and efficacy. A study flow diagram is presented in figure 1.

Figure 1. Flow diagram of the study design. CDAI, Crohn’s Disease Activity Index; IBDQ, Inflammatory Bowel Disease Questionnaire; PDAI, Perianal Disease Activity Index; UPA, upadacitinib; WIS, Wexner incontinence score.

Participants

Participant recruitment commenced on 13 March 2025 and will continue until 30 June 2026. The final follow-up assessments for the last enrolled patient are planned to conclude on 30 June 2027. This trial has been registered with the Chinese Clinical Trial Registry (ChiCTR2600117355). Consecutive adult patients with active pfCD and a history of anti-TNF therapy failure will be screened for eligibility if they present to these centres and their treating physician has determined to initiate upadacitinib according to the local label and standard clinical practice.

All patients will receive upadacitinib in accordance with routine clinical medication practice in a real-world setting. According to the product label, the recommended regimen consists of 45 mg once daily for 12 weeks (induction phase), followed by 15 mg or 30 mg once daily for the subsequent 40 weeks (maintenance phase). Dose adjustments or discontinuation of therapy will be at the discretion of the IBD expert panel and will be documented in the medical records.

The study duration is 52 weeks. Study visits and assessments are scheduled at baseline and weeks 2, 4, 8, 12, 16, 24, 36 and 52. The study schedule, including enrolment, exposures and assessments, is presented below in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement. Laboratory tests include, but are not limited to, complete blood count, serum chemistry, CRP, faecal calprotectin, lipid profile and creatine kinase levels.

Eligibility criteria

Inclusion criteria

Patients must meet all of the following criteria:

1. At baseline (week 0), male or female participants aged 18 to 60 years (inclusive) with entirely Chinese descent.

2. Established diagnosis of CD for at least 3 months prior to baseline (week 0), confirmed by clinical, endoscopic, radiological or histological examination.

3. Prior to baseline (week 0), participants must have experienced failure of anti-TNF-α therapy, defined as either primary non-response or secondary loss of response, with documented inadequate response to one or more of the following biologic regimens: at least one 6-week infliximab induction regimen (intravenous infusions ≥5 mg/kg at baseline, week 2 and week 6) or at least one 4-week adalimumab induction regimen (a single 160 mg subcutaneous injection at baseline followed by a single 80 mg subcutaneous injection at week 2).

4. Active perianal fistulas with at least one CD-related perianal fistula actively draining for a minimum of 4 to 12 weeks, confirmed by pelvic MRI within the baseline period.

5. Negative screening for active tuberculosis (TB) or, if latent TB is diagnosed, completion of at least three consecutive weeks of appropriate prophylactic therapy prior to baseline.

6. Willing and able to provide written, informed consent.

Exclusion criteria

Patients will be excluded if they meet any of the following:

1. Current diagnosis of ulcerative colitis or inflammatory bowel disease unclassified.

2. Pregnancy or lactation.

3. Presence of severe, uncontrolled comorbid conditions; active opportunistic infections; immune system deficiencies; history of malignancy; or history of demyelinating disorder.

4. A past or current history of thromboembolism (including deep vein thrombosis, pulmonary embolism and arterial thrombosis).

5. A past or current history of serious adverse cardiovascular events (myocardial infarction or stroke), or moderate-to-severe congestive heart failure (New York Heart Association (NYHA) class III or IV) within the past 6 months.

6. Presence of severe hepatic impairment (Child-Pugh class C) or severe renal impairment and end-stage renal disease.

7. Haematological abnormalities showing absolute lymphocyte count <500 cells/mm³, absolute neutrophil count <1000 cells/mm³ or haemoglobin <80 g/L.

8. Presence of an active severe infection (including localised infections).

9. Recent use of prohibited medications: immunomodulators (azathioprine, 6-mercaptopurine, methotrexate and thalidomide) within 14 days before baseline.

10. Receipt of systemic corticosteroids within 1 month prior to baseline (week 0), with an oral corticosteroid dose >20 mg at baseline.

11. Prior treatment with any JAK inhibitor (eg, tofacitinib and baricitinib).

12. Patients who have undergone an ostomy, ileal pouch-anal anastomosis or short bowel syndrome.

13. Patients who have undergone bowel resection or other major abdominal surgery within 3 months before baseline.

14. Known hypersensitivity to upadacitinib or any of its excipients.

15. Patients with undrained fluid collections or abscesses measuring >2 cm between inside walls in at least two dimensions and patients with solid components.

16. The presence of severe luminal disease (Crohn’s Disease Activity Index >450 or Harvey-Bradshaw Index >16).

17. Patients with impassable anal or rectal stenosis.

18. Patients with anovaginal/rectovaginal fistulas and patients with diverting stoma.

19. Any other condition that, in the investigator’s judgement, would make the patient unsuitable for participation in the study.

Outcomes

As research on perianal Crohn’s disease advances, there has been a clear shift from clinical endpoints towards combined clinical and radiological assessments, reflecting recognition of the limitations of clinical evaluation alone—particularly in deep or complex fistulas.^{19 20} Building on this evolution, our study adopts a composite endpoint approach, with the primary endpoint defined as combined fistula response integrating both clinical and radiological outcomes.

To ensure standardised and reproducible clinical assessment, perianal fistulas will be assessed at baseline and each follow-up visit by a trained investigator using standardised digital perianal examination. A draining fistula is defined as a fistulous tract with an external opening that demonstrates visible discharge of fluid either spontaneously or on gentle finger compression. Each draining fistula will be recorded individually. Assessments will be performed by the same investigator whenever possible to reduce inter-observer variability. Radiological assessment will be assessed using pelvic MRI and refers to the TOpCLASS Global Expert Delphi Consensus (2025).²¹

Primary outcomes

Combined fistula response at week 12, defined as the simultaneous achievement of both clinical and radiological fistula response.

1. Clinical fistula response is defined as a ≥50% reduction in the number of draining fistulas compared with baseline.

2. Radiological fistula response is defined by at least one essential criterion on MRI:

   • An increasingly fibrotic fistula tract (compared with baseline).

   • An unequivocal reduction in fistula tract T2-weighted hyperintensity compared with baseline study.

   • An unequivocal reduction in fistula tract diameter.

   • An unequivocal reduction in fistula tract length.

   • An unequivocal reduction in size or complete absence of previously present abscess.

   • Where contrast is used: an unequivocal reduction in fistula tract hyperintensity on contrast T1-weighted fat-saturated images compared with baseline study.

Secondary outcomes

1. Perianal fistula-related outcomes:

   • PDAI.²²

   • Clinical fistula response (as defined above).

   • Clinical fistula remission: 100% closure of all treated external openings, with no occurrence of new fistulas or abscesses, and absence of any fistula drainage (either spontaneous or on gentle compression).

   • Radiologic fistula response (as defined above).

   • Radiologic healing: absence of T2-weighted hyperintense signal, absence of contrast enhancement on post-contrast T1-weighted images (if contrast used) and completely fibrotic fistula tract.

   • The magnetic resonance novel index for fistula imaging in CD (MAGNIFI-CD).^{20 23}

   • Patient-reported outcomes: Wexner incontinence score.²⁴

• 2 1Crohn’s disease-related outcomes²⁵:

  • Clinical remission (patient-reported 0utcome (PRO)-based or CDAI-based) defined by meeting either criterion:

    1. PRO-based remission: mean daily number of liquid/watery stools ≤2.8 and mean daily abdominal pain score ≤1.0. Both not exceeding baseline values.

    2. CDAI-based remission: CDAI <150.

  • Patient-reported outcomes: IBDQ total score.

  • Endoscopic response and remission:

    1. Endoscopic response: >50% reduction from baseline in the Simple Endoscopic Score for Crohn’s Disease (SES-CD) from baseline at weeks 12 and 52 (as assessed by expert panel), or ≥2-point reduction in subjects with baseline SES-CD=4.

    2. Endoscopic remission: SES-CD ≤4, with an absolute reduction of ≥2 points from baseline and no subscore >1, at weeks 12 and 52.

  • Composite endpoints: clinical remission and endoscopic response/remission.

  • Biomarker response: CRP and faecal calprotectin.

• 3 1Safety outcomes²⁶:

  • Incidence, type and severity of AEs and serious AEs (SAEs).

  • Changes in laboratory parameters, vital signs, ECG and chest CT.

Sample size calculation

Due to the lack of published data on the combined clinical and radiological response to upadacitinib in patients with pfCD at week 12, the sample size estimation is based on data from studies of similar therapeutic agents^{11 14 27 28} and the phase III trials of upadacitinib in pfCD.¹³ Based on these data, we anticipate a combined response rate of 30% at week 12. To estimate this response rate with a 95% CI of width no greater than 20 percentage points, a sample size of 81 evaluable patients is required. After accounting for a potential 10% dropout or loss to follow-up during the study period, the final target enrolment is set at 90 patients.

Data collection and management

Data will be captured electronically via a study-specific, standardised electronic case report form (eCRF),²⁹ with primary source data derived from patients’ original medical records. To ensure data integrity, all personnel involved in data entry will undergo standardised training. Although clinical management will remain at the treating physician’s discretion, detailed information on potential confounding variables—including study site, concomitant surgical and medical interventions (timing, indication and procedure type, such as abscess drainage, EUA, seton manipulation, antibiotic therapy and other perianal procedures), therapeutic drug monitoring, anti-drug antibody status, duration of prior anti-TNF therapy, and the number and classes of previous biologic agents—will be prospectively collected using the standardised eCRF. These data will enable characterisation of inter-site and treatment-related variations and adjustment for potential confounding in the statistical analyses.

The data management framework incorporates a multi-layered validation process. First, automated logic and range checks will be built into the eCRF to identify errors or missing entries at the point of capture. Subsequently, a dedicated data manager will oversee query management, flagging inconsistencies, which are to be formally resolved by investigators at each site. Quality assurance will be further maintained through periodic on-site monitoring visits to verify the accuracy and completeness of the eCRF data against primary source documents. Finally, to ensure security and confidentiality, all electronic data will be hosted on secure, password-protected servers, with participants identified solely through unique study codes.

Statistical analysis

Statistical analyses will be performed using SPSS Statistics. Baseline demographic and clinical characteristics will be summarised using mean (SD) or median (IQR) for continuous variables, and counts (percentages) for categorical variables.

The primary endpoint, proportion of patients achieving a combined fistula response at week 12, will be reported as the number of responders along with the corresponding proportion (percentage). Given the descriptive, single-arm nature of this study, no confirmatory hypothesis testing will be performed. Primary inferences will be based on point estimates and their two-sided 95% CIs; p values will not be calculated for the primary efficacy analyses. Secondary continuous endpoints (eg, changes from baseline in PDAI, CDAI, IBDQ and CRP) will be summarised descriptively, including number of observations, mean, SD, minimum, maximum and median. Any p values reported for these exploratory secondary outcomes will be interpreted descriptively, without adjustment for multiplicity, and will not serve as confirmatory evidence of treatment efficacy. Safety data, including all adverse events and serious adverse events, will be summarised descriptively by number of occurrences, categories and severity grades, presented as frequencies and percentages.

Concomitant medications will be described separately for pre-study and on-treatment periods. All perianal interventions will be recorded in detail and explored as hypothesis-generating subgroup analyses.

Any missing data or incomplete assessments will be carefully documented. The potential impact of missing data on study findings will be evaluated using appropriate sensitivity methods. Detailed handling strategies will be pre-specified in a statistical analysis plan (SAP) to be finalised before database lock.

Ethics and dissemination

The study will be conducted in full conformity with the principles of the Declaration of Helsinki, the International Council for Harmonisation Good Clinical Practice guidelines, and applicable local laws and regulations. Written informed consent will be obtained from all participants before any study-specific procedures. To ensure confidentiality, all participant data will be de-identified and managed using unique study codes.

Patients or the public were not involved in the design, conduct or reporting of this research. However, the study findings will be disseminated to participating institutions and relevant patient communities through accessible summaries once the results are published. The findings of this study will be submitted for publication in peer-reviewed international journals and presented at national or international scientific conferences. Authorship will be determined according to the International Committee of Medical Journal Editors (ICMJE) criteria.

This study was retrospectively registered at Chinese Clinical Trial Registry on 22 January 2026 (ChiCTR2600117355), after the initiation of patient enrolment. The registration was completed after study initiation due to administrative delays in the registration process. We confirm that all protocol definitions, including primary and secondary outcomes, were specified prior to any outcome data analysis. No interim analyses influenced the study design or analytical plan.

Trial status

Current protocol version: 1.0, updated 16 February 2026.

Recruitment began 13 March 2025. We expect recruitment to finish 30 June 2026.