Endoscopic retrograde biliary drainage plus glucocorticoids versus glucocorticoid alone for autoimmune pancreatitis with significant obstructive jaundice: study protocol for a multicenter, open-label, non-inferiority randomized controlled trial

Abstract

Background

Patients with autoimmune pancreatitis (AIP) frequently develop obstructive jaundice. Endoscopic retrograde biliary drainage (ERBD) is typically performed before glucocorticoid initiation to prevent biliary infections and worsening cholestasis. While guidelines support glucocorticoid monotherapy without ERBD for mild jaundice, its efficacy and safety remain unestablished in AIP patients with clinically significant obstructive jaundice. This trial aims to determine whether glucocorticoid monotherapy is noninferior to ERBD plus glucocorticoids as initial therapy for AIP patients with significant obstructive jaundice.

Methods/design

In this multicenter, noninferiority, open-label, randomized controlled trial, 78 patients with significant obstructive jaundice without concurrent cholangitis (serum total bilirubin ≥ 3 × ULN) across 13 centers in China will be randomly assigned to receive either ERBD plus glucocorticoids or glucocorticoids alone (39 patients per group). The primary endpoint is the clinical response rate, which was defined as a ≥ 50% reduction in the total bilirubin level within 2 weeks of the intervention. The secondary outcomes are adverse event rates, including ERCP-related complications and glucocorticoid-related adverse events.

Discussion

This trial is expected to reveal whether ERBD is routinely required before glucocorticoid administration in AIP patients with significant obstructive jaundice, which is an important question that has not been addressed via an RCT design. The results will supplement the evidence for clinical decision-making and may provide a more straightforward basis for treatment selection, potentially simplifying the management of these patients.

Trial registration

Chictr.org.cn ChiCTR2400086160. Registered on June 26, 2024.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13063-026-09595-6.

Introduction

Autoimmune pancreatitis (AIP) is a unique, autoimmune-mediated form of chronic pancreatitis characterized clinically by frequent presentation with obstructive jaundice with or without a pancreatic mass and, therapeutically, by a dramatic response to glucocorticoids [1]. It comprises two distinct pathologic entities: type 1 (lymphoplasmacytic sclerosing pancreatitis) and type 2 (idiopathic duct-centric pancreatitis) [2, 3]. Type 1 AIP represents the pancreatic manifestation of immunoglobulin G4 (IgG4)-related disease and is commonly accompanied by IgG4-related sclerosing cholangitis (IgG4-SC) [4, 5]. Approximately, 75% of type 1 AIP patients and 35% of type 2 AIP patients present with painless obstructive jaundice, which may result from IgG4-SC or secondary biliary strictures due to pancreatic head compression in AIP [4, 6–9]. As glucocorticoids may trigger or worsen cholangitis, obstructive jaundice in AIP patients has historically been managed by endoscopic or percutaneous transhepatic biliary drainage (PTCD) before glucocorticoid administration [10, 11]. A multinational study revealed that the majority (71%) of AIP patients with obstructive jaundice underwent endoscopic retrograde cholangiopancreatography (ERCP) with biliary stenting (i.e., endoscopic retrograde biliary drainage, ERBD) for biliary drainage [6]. However, the role of ERBD remains controversial owing to its inherent procedural invasiveness [12].

Since AIP responds promptly to glucocorticoids, several retrospective studies have shown that some AIP patients with obstructive jaundice can be safely treated with glucocorticoids alone without ERBD [9, 11–15]. Some guidelines recommend endoscopic drainage before glucocorticoids, except for patients with mild jaundice [5, 15, 16]. However, existing evidence is derived from retrospective studies with limited sample sizes and a paucity of high-level evidence. The efficacy and safety of glucocorticoid monotherapy without ERBD in AIP patients with significant obstructive jaundice remain unestablished. We designed this first prospective randomized controlled trial (RCT) to investigate whether glucocorticoid monotherapy is noninferior to ERBD plus glucocorticoids for AIP with significant obstructive jaundice.

Methods/design

Trial design

This study is a pragmatic, multicenter, open-label, parallel-group, randomized controlled trial with a noninferiority design. The protocol was developed in compliance with the SPIRIT guidelines [17]. The completed SPIRIT checklist is provided as an additional file with this submission. Trial conduct and reporting will adhere to the CONSORT statement [18, 19]. Ethical approval was obtained from the Ethics Committee of Chinese PLA General Hospital (Approval No. S2022-428-01), and the study was conducted in accordance with the Declaration of Helsinki. The trial is registered at the Chinese Clinical Trial Registry (chictr.org.cn) under the identifier ChiCTR2400086160.

Participants

Participants will be recruited from 13 participating medical centers across China. The recruitment strategy will utilize online platforms and social media campaigns coordinated by the leading center (the first medical center of the Chinese PLA General Hospital), supplemented by coordinated referrals from the collaborating institutions. Patients (≥ 18 years of age) with AIP presenting with significant obstructive jaundice and suspected of having AIP will be screened for eligibility. Initially, local attending physician(s) will assess these patients based on the predefined inclusion and exclusion criteria (Table 1). To confirm the diagnosis and ensure uniform application of standards, all potential participants will undergo a standardized pre-randomization workup including contrast-enhanced abdominal computed tomography (CT), magnetic resonance cholangiopancreatography (MRCP), and measurement of serum tumor markers (CA19-9, CEA). Endoscopic ultrasound (EUS) with or without tissue acquisition is recommended in cases of diagnostic uncertainty. Subsequently, all cases will be reviewed by a central adjudication committee prior to enrollment confirmation to ensure consistent application of the International Consensus Diagnostic Criteria for Autoimmune Pancreatitis (ICDC) and to exclude suspected malignancy. Eligible patient(s) will be informed of the trial and invited to participate.

Table 1.

Inclusion and exclusion criteria

Inclusion criteria
Diagnosis: Definite AIP meeting ICDC (2011) criteria, either type 1 or 2
Clinical presentation: Significant obstructive jaundice (serum total bilirubin ≥ 3 × upper limit of normal [ULN], i.e., ≥ 63 μmol/L; normal range: 0–21 µmol/L)
Age: ≥ 18 years
Consent: Provision of written informed consent
Exclusion criteria
Critical illness with unstable vital signs
Severe cardiopulmonary disease, altered consciousness, ASA class ≥ 3, or ineligibility for sedation-assisted endoscopy based on anesthesiologist assessment
Active malignancies (except curatively resected nonmetastatic tumors > 5 years prior)
Confounding hepatobiliary-pancreatic disorders (e.g., choledocholithiasis, primary biliary cholangitis) or suspected malignancy
Anatomic barriers to ERCP (altered gastrointestinal anatomy, severe stenosis/varices)
Coagulopathy (platelets < 50 × 10⁹/L or INR > 1.5) or inability to withhold anticoagulants > 5 days
Glucocorticoid contraindications: Drug hypersensitivity, uncontrolled neuropsychiatric or cardiometabolic disorders, active infections, advanced osteoporosis, or peptic ulcers
Pregnancy or lactation

INR international normalized ratio, AIP autoimmune pancreatitis, ICDC International Consensus Diagnostic Criteria, ERCP

Inclusion criteria

1. Diagnosis: Define AIP meeting ICDC 2011 [3], either type 1 or 2.

2. Clinical presentation: Significant obstructive jaundice (serum total bilirubin ≥ 3 × upper limit of normal [ULN], i.e., ≥ 63 μmol/L; normal range: 0–21 μmol/L)

3. Age: ≥ 18 years

4. Consent: Provision of written informed consent

Exclusion criteria

1. Critical illness with unstable vital signs

2. Severe cardiopulmonary disease, altered consciousness, ASA class ≥ 3, or ineligibility for sedation-assisted endoscopy on the basis of anesthesiologist assessment

3. Patients with active malignancies (except for curatively resected nonmetastatic tumors > 5 years prior)

4. Confounding hepatobiliary–pancreatic disorders (e.g., choledocholithiasis, primary biliary cholangitis) or suspected malignancy

5. Anatomic barriers to ERCP (altered gastrointestinal anatomy, severe stenosis/varices)

6. Coagulopathy (platelets < 50 × 10^⁹/L or INR > 1.5) or inability to withhold anticoagulants for > 5 days

7. Glucocorticoid contraindications: Drug hypersensitivity, uncontrolled neuropsychiatric or cardiometabolic disorders, active infections, advanced osteoporosis, or peptic ulcers

8. Pregnancy or lactation

Randomization

Investigators will inform patients about the trial procedure. After providing signed informed consent, a standardized dataset will be uniformly collected from all participants across participating hospitals. Randomization will be performed by the Department of Gastroenterology, Chinese PLA General Hospital (the leading center), using a computer-generated sequence with variable block sizes. Allocation will be concealed via a secure, web-based system. To minimize imbalance across multiple centers and baseline severity, randomization will be stratified by participating study site and baseline serum total bilirubin level (63–210 μmol/L vs. > 210 μmol/L). Enrolled patients will be randomly assigned (1:1 ratio) via an online randomization tool to either the glucocorticoid monotherapy group or the ERBD plus glucocorticoids group.

Blinding

This study is an open-label trial. A sham ERCP procedure was considered but deemed ethically problematic and risky for patients with significant obstructive jaundice, as it would expose them to sedation and procedural risks without therapeutic intent. To minimize bias, (1) the primary endpoint is an objective laboratory measure, (2) a central independent adjudication committee, blinded to treatment allocation, will assess all secondary clinical endpoints using predefined criteria, and (3) standardized protocols for concomitant medications, monitoring, and discharge criteria will be implemented across all centers to reduce performance bias.

Interventions

ERBD procedures

Patients randomized to the ERBD group will receive standard preprocedural management, including intravenous antibiotic prophylaxis (e.g., ciprofloxacin) and prophylactic rectal NSAID (diclofenac 100 mg or indomethacin 100 mg) administered within 60 min prior to the procedure, unless contraindicated. The ERCP procedure is subsequently initiated. A duodenoscope is advanced orally to the papilla. The bile duct will then be cannulated with a sphincterotome via the guidewire-assisted technique. A cholangiogram will then be performed, followed by the insertion of a biliary stent. The use of a 10-Fr diameter stent is recommended; the exact length (typically 5–7 cm) is at the endoscopist’s discretion based on anatomy. The performance of a biliary sphincterotomy prior to stent insertion is optional and should be recorded. Key procedural variables, including cannulation attempts, stent type/size/length, use of PEP prophylaxis, and antibiotic regimen, will be captured in the case report form.

Stent management

A planned endoscopic stent removal is scheduled for 4 weeks (±1 week) post-placement. An abdominal X-ray will be performed prior to removal to assess stent position. Spontaneous stent passage will be recorded if confirmed. Management of retained stents will follow standard endoscopic practice.

Glucocorticoid therapy

Patients randomized to the glucocorticoid monotherapy group will receive oral methylprednisolone (40 mg/day or 0.6–1.0 mg/kg/day) according to the standard glucocorticoid regimen in the ICDC [3]. Patients allocated to the ERBD group will initiate oral methylprednisolone therapy (40 mg/day or 0.6–1.0 mg/kg/day) 24 h post-ERCP after excluding immediate postprocedural adverse events (AEs). Adherence to the glucocorticoid regimen will be encouraged through patient education at enrollment and during follow-up contacts. This 24-h delay is a standard safety precaution to assess for early ERCP-related complications (e.g., severe pancreatitis) before starting immunosuppression. Given the rapid action of glucocorticoids in AIP, this brief delay is not anticipated to materially affect the day 14 bilirubin outcome. A sensitivity analysis of the primary endpoint, excluding patients who experienced major post-ERCP complications requiring intervention, will be conducted to assess the robustness of the findings. For patients who develop severe post-ERCP pancreatitis or severe infection after the procedure, glucocorticoid therapy will be initiated as soon as possible following clinical improvement and discussion by the study expert panel. Throughout the study period, no additional concomitant immunomodulatory medications (e.g., rituximab, hydroxychloroquine, or leflunomide) will be administered. After 4 weeks of therapy, the glucocorticoid dosage will be reduced by one tablet (equivalent to 5 mg) every 1–2 weeks on the basis of serial assessment of imaging findings and serological parameters. This tapering regimen will be maintained until a maintenance dose of 5 mg/day is reached within 2–3 months.

Modification or discontinuation criteria for allocated interventions

Allocated interventions may be modified or discontinued under the following circumstances:

1. Serious adverse events (SAEs): Development of severe glucocorticoid-related adverse events (e.g., new-onset psychosis, uncontrolled hyperglycemia, severe infection) or serious ERCP-related complications (e.g., severe pancreatitis, perforation) may necessitate dose reduction, temporary interruption, or permanent discontinuation, as judged by the treating physician and study expert panel.

2. Treatment failure: Treatment failure in the glucocorticoid monotherapy group is defined as meeting any of the following criteria after randomization: (a) < 20% reduction in serum total bilirubin from baseline by day 7, (b) development of cholangitis (as per defined criteria), and (c) clinical deterioration as judged by the treating physician (e.g., worsening abdominal pain, vomiting, or signs of sepsis) potentially related to unresolved biliary obstruction. In such cases, urgent ERBD will be performed as rescue therapy. Crossovers to rescue ERBD will be recorded and analyzed in both the intention-to-treat (ITT) and per-protocol (PP) populations.

3. Participant request: A participant may voluntarily withdraw from the assigned intervention at any time, at which point standard clinical care will be provided.

Outcome

The primary outcome is the clinical response rate, which is defined as a ≥ 50% reduction in the serum total bilirubin from baseline level within 14 days after initiation of therapy. Secondary outcomes are designed to capture patient-important clinical benefits beyond biochemical response and will be analyzed hierarchically as follows:

1. Incidence of cholangitis within 4 weeks

2. Need for urgent or unplanned biliary drainage

3. Symptom improvement (pruritus, fatigue, jaundice-related symptoms) assessed by patient-reported outcomes

4. Length of hospital stay during index admission

5. Thirty-day hospital readmission rate

6. Adverse event rates, including ERCP-related complications (e.g., pancreatitis, cholangitis, bleeding, perforation) and glucocorticoid-related AEs (e.g., infection, worsening hyperglycemia)

ERCP-related complications are defined according to the European Society of Gastrointestinal Endoscopy (ESGE) guidelines [20]. Post-ERCP pancreatitis (PEP) is defined as new or worsened abdominal pain combined with > 3 times the upper limit of normal amylase or lipase at more than 24 h after ERCP and the need for admission or prolongation of a planned admission.

• Cholangitis: New-onset temperature > 38 °C for more than 24 h, accompanied by evidence of biliary obstruction (e.g., rising bilirubin or imaging findings) and at least one of the following: right upper quadrant pain, leukocytosis, or positive blood cultures. Baseline cholangitis is excluded by the absence of fever and normal inflammatory markers prior to randomization.

• Bleeding: Hematemesis and/or melena or hemoglobin drop > 2 g/dL

• Perforation: Evidence of gas or luminal contents outside of the gastrointestinal tract as determined by imaging

A central independent adjudication committee, blinded to treatment allocation, will assess all secondary clinical endpoints and SAEs using predefined criteria. The committee will consist of at least three members (e.g., gastroenterologists not involved in the trial) who will independently review de-identified medical records, laboratory data, and imaging reports provided by the study sites. Any discrepancies in adjudication will be resolved through discussion or by a lead arbitrator. This centralized, blinded adjudication process ensures symmetric and consistent evaluation of these key outcomes across both treatment arms.

Follow-up protocol

Clinical and biological data will be collected on day 1, day 4 (±1 day), and day 7 (±1 day). Assessments will include the following: (1) Laboratory testing: complete blood count, inflammatory markers, liver function tests (LFTs), and pancreatic enzymes and (2) clinical evaluation: symptom improvement and adverse event monitoring. On day 14 (days 13–15), patients will undergo the following assessments: (1) radiological examination: abdominal CT or ultrasonography (US) to evaluate pancreatic morphology and (2) laboratory testing: measurement of serum IgG4 levels and LFTs. The 4-week (±1 week) follow-up visit will include the planned stent removal. Long-term monitoring will be conducted through scheduled clinic visits, telephone consultations, and WeChat communications to track symptom resolution, therapeutic efficacy, and adverse event occurrence (Figs. 1 and 2).

Fig. 1.

Research flow chart

Fig. 2.

Schedule of enrollment, interventions, and assessments. ¹Baseline characteristics: Screening laboratory test results from within 1 week prior to enrollment are permissible for baseline assessment. ²Complete blood count (CBC): Hemoglobin (HGB), erythrocyte (red blood cell, RBC) count, platelet (PLT) count, and leukocyte (white blood cell, WBC) count including differentials (neutrophil and lymphocyte percentages). ³Liver function tests (LFTs): Alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin, direct bilirubin (DBIL), alkaline phosphatase (ALP), and gamma-glutamyl transferase (γ-GGT). ⁴Pancreatic enzymes: Serum amylase and lipase levels. ⁵Medical imaging: Abdominal computed tomography (CT), ultrasonography (US), or magnetic resonance cholangiopancreatography (MRCP) to assess pancreaticobiliary morphology. ^⁶Primary endpoint: Clinical response rate, defined as a ≥ 50% reduction in serum total bilirubin level from baseline within 14 days after initiation of therapy. ^⁷Secondary endpoints: Incidence rates of all adverse events, including procedure-related and drug-related complications. Incidence of cholangitis within 4 weeks, the need for urgent or unplanned biliary drainage, symptom improvement (pruritus, fatigue, jaundice-related symptoms) assessed by patient-reported outcomes, length of hospital stay during index admission, and 30-day hospital readmission rate

Sample size calculation

The sample size was calculated to test our null hypothesis that glucocorticoid monotherapy is noninferior to ERBD plus glucocorticoids for jaundice relief (defined as achieving a 50% reduction from baseline in the total serum bilirubin level within 2 weeks) in AIP patients with significant obstructive jaundice. The target sample size was set at 78 patients (39 patients per group). Retrospective data from nine medical centers (Chinese PLA General Hospital) revealed clinical response rates of 92% in the control group (ERBD plus glucocorticoids) and 89% in the experimental group (glucocorticoid alone) among AIP patients with obstructive jaundice. The margin of noninferiority is set at 20% (Δ = 0.20). This margin was chosen based on clinical reasoning: if glucocorticoid monotherapy is noninferior within this margin, it would indicate that a clinically acceptable proportion of patients (e.g., if the control group response rate is 92%, the monotherapy rate would be ≥ 72%) still achieve the primary endpoint of significant jaundice relief within 2 weeks, potentially obviating the need for invasive ERBD and its associated risks. Importantly, this threshold is consistent with margins used in other noninferiority trials involving invasive alternatives to interventional procedures in benign diseases, supporting its clinical acceptability [21]. With 80% power at a one-sided α = 0.05 level, 74 patients (two equal groups of 37) were needed. To accommodate a 5% dropout rate during the short-term follow-up, we will recruit 78 (74/0.95) patients in total.

Statistical analysis

Descriptive statistics will be used to describe the adequacy of patient characteristics and assess the balance of the randomization process. The primary analysis will test the noninferiority of glucocorticoid monotherapy compared with ERBD plus glucocorticoids regarding the clinical response rate. The risk difference (RD) and its one-sided 97.5% confidence interval (CI) will be calculated for the ITT population. Noninferiority will be declared if the upper bound of the one-sided 97.5% CI for the RD (glucocorticoid alone minus [ERBD plus glucocorticoids]) is less than the pre-specified noninferiority margin (Δ = 20%). If noninferiority is established, a supportive superiority analysis using a chi-square test (two-tailed, α = 0.05) will be performed. The PP population comprises randomized subjects without protocol deviations (such as eligibility violations, nonadherence to the randomized treatment, or missing primary outcomes). Both the ITT and PP populations will be analyzed for the primary noninferiority endpoint. A definitive conclusion of noninferiority will require concordance between the ITT and PP analyses. A pre-specified integrative interpretation plan will be applied: if noninferiority is met for the primary endpoint, the overall benefit-risk profile will be assessed by comparing the rates of ERCP-related complications in the ERBD group with the rates of rescue drainage and glucocorticoid-related AEs in the monotherapy group. A secondary analysis will explore the potential effect modification by study site and baseline total serum bilirubin on the primary outcome measure. The secondary endpoints will be compared between treatment arms via a chi-square test or Fisher’s exact test and a t-test for independent means, as appropriate. Missing data for the primary endpoint will be handled primarily using multiple imputation under the missing-at-random assumption. Sensitivity analyses will include the following: (1) Worst-case scenario imputation (where missing data in the glucocorticoid monotherapy arm are imputed as non-response and missing data in the ERBD plus glucocorticoids arm are imputed as response) and (2) tipping-point analysis to evaluate the robustness of the conclusion to different missing data mechanisms. We will also identify potential factors affecting AEs through univariate and multivariate logistic regression analyses. All tests will be two-tailed, with P < 0.05 considered statistically significant.

Data management

To ensure the integrity and quality of trial data, a multifaceted management plan will be implemented. First, a central adjudication committee will review all candidate cases to confirm eligibility, with a focus on verifying the AIP diagnosis and excluding malignancy. The principal investigator is responsible for the overall conduct of the trial, including appropriate data generation, recording, and reporting, and will implement a comprehensive quality control and quality assurance system.

All data will be collected using a secure, web-based electronic data capture (EDC) system featuring an audit trail and role-based access control. Data entry will be performed by trained site personnel. A detailed data dictionary defining all variables will be provided to all participating sites. To ensure data accuracy, both remote and on-site monitoring will be conducted, which includes source data verification against the EDC entries. A dedicated data center will perform ongoing checks on data entry status and consistency, providing feedback to each site for continuous quality control. The primary database will be stored on secure servers at the leading center.

Laboratory assessments across centers will be harmonized by requiring the use of standardized assays. Laboratory normal ranges and methodologies will be documented for each site to ensure comparability, and the data center will perform consistency checks on all laboratory values.

Confidentiality

All participant data will be de-identified using a unique study identification number. Any documents containing personal identifiers will be stored separately from the study data in locked cabinets at each study site, with access limited to authorized personnel. Electronic data will be stored on secure, password-protected servers with encryption. During data transmission and analysis, only de-identified data will be used. All research team members will receive training on data confidentiality and privacy protection requirements. The principal investigator will be responsible for ensuring the confidentiality of the data and for preserving all the data related to the study for at least 5 years from the date of the report on the completion of the study.

Clinical monitoring

A comprehensive clinical monitoring plan will be implemented to ensure protocol adherence, regulatory compliance, and good clinical practice. Remote verification of the study and regulatory documents will be performed. In addition, both scheduled and for-cause on-site monitoring visits will be conducted at each center to verify source data and assess procedural compliance. No interim analysis is planned. Furthermore, independent auditors will conduct annual audits to ensure trial integrity.

Safety

To ensure comprehensive and symmetric capture of safety data, an active surveillance plan will be implemented for all participants from randomization until the 4-week follow-up visit. This plan includes scheduled clinic visits, telephone consultations, and structured inquiries during follow-up contacts to systematically solicit and document AEs occurring both in-hospital and out of hospital. An adverse event refers to any unintended medical occurrence in a patient administered an investigational product or participating in a clinical study, irrespective of treatment causality. In this trial, all AEs—regardless of severity, preexisting conditions, or relationship to study drugs—will be documented in designated case report form sections. Investigators will assess causality between adverse reactions and interventions. Any unanticipated AEs potentially related to the study must be reported to the Institutional Review Board of the Ethics Committee at Chinese PLA General Hospital.

Dissemination plans

The results of the trial, both positive and negative, will be reported at international conferences and/or published in an international journal.

Discussion

To date, there are no available data from prospective studies comparing the efficacy and safety of glucocorticoids with those of ERBD in AIP patients with obstructive jaundice. This noninferiority RCT will examine whether ERBD is routinely needed before glucocorticoid administration in AIP patients with significant obstructive jaundice (without concurrent cholangitis) by comparing glucocorticoid monotherapy with ERBD plus glucocorticoids.

Biliary stenting is typically employed to manage obstructive jaundice in AIP because of concerns that glucocorticoids may worsen biliary tract infections and exacerbate cholestasis [22]. Furthermore, advanced sclerosing cholangitis with fibrotic strictures often results in glucocorticoid refractoriness [11]. In 2010, the Japanese consensus guidelines for AIP recommended biliary drainage before glucocorticoid therapy in patients with obstructive jaundice [3, 12].

However, ERCP remains a relatively high-risk invasive procedure with potential complications, including pancreatitis, hemorrhage, infection, or perforation [11]. Recently, some studies reported rapid relief from biliary obstruction with glucocorticoids, along with the resolution of jaundice in AIP- and IgG4-related SC patients, without the need for biliary stenting [9, 11, 12]. Several reports have suggested that biliary drainage may not be essential, even in cases of moderate and severe jaundice [14, 23]. However, these investigations have significant limitations, including substantial confounding factors, small sample sizes, and heterogeneity. Crucially, the severity of jaundice and liver dysfunction at enrollment varied widely. Many studies have included a substantial proportion of patients with only mild jaundice [11–14, 22, 23] Current guidelines support the safety of glucocorticoid monotherapy for mild, uninfected obstructive jaundice patients (level B [1] and level D [5] evidence). For moderate and severe obstructive jaundice, however, biliary drainage remains recommended prior to glucocorticoid initiation. Therefore, this study focused on evaluating the efficacy and safety of glucocorticoid monotherapy in patients with significant obstructive jaundice, implementing the inclusion criterion of a total serum bilirubin concentration ≥ 3 times the ULN.

The selection of an appropriate and relevant primary outcome is important for any RCT. Given that AIP results in a rapid response to glucocorticoids, the primary outcome was a ≥ 50% reduction from baseline in total serum bilirubin within a 2-week period (i.e., 14 days), which is consistent with the guideline-recommended timeframe for glucocorticoid trials [3, 24]. This primary outcome is objectively quantified by serum bilirubin measurement and is unlikely to be significantly affected by the open-label design. In a study by Iwasaki S. et al. [13], biliary stents were successfully removed in all 10 AIP patients with jaundice after a mean duration of 16 days (range: 9–30 days). Miyazawa et al. [22] reported intentional stent removal in 46.4% (13/28) of patients postconfirmation of glucocorticoid efficacy, with a median removal time of 16 days (range: 10–143) after glucocorticoid initiation. Thus, our 2-week timeframe may provide prospective data for determining the optimal timing of biliary stent removal.

The trial has a noninferiority design since we wish to evaluate whether glucocorticoid monotherapy is as effective as, but not necessarily more effective than, routine ERBD before glucocorticoid administration. If glucocorticoid monotherapy is as effective as ERBD, the potential benefits include avoiding invasive procedures and their inherent risks and avoiding anesthesia. Additionally, it substantially reduces healthcare expenditures by eliminating both the cost of stent placement in most cases and the need for follow-up abdominal radiography (to confirm spontaneous stent passage) and follow-up upper gastrointestinal endoscopies for retained stent removal [25]. There is no accepted guidance regarding the magnitude of a noninferiority margin for surgical trials [26]. However, in clinical practice, for nonfatal, low-risk benign diseases (such as chronic inflammatory or functional disorders), the noninferiority margin is typically set between 15 and 25% [27]. Given that AIP is a rare, benign condition with a low mortality risk requiring long-term glucocorticoid administration, a noninferiority margin of 20% was chosen. This margin balances the trade-off between glucocorticoid monotherapy (potentially less effective but less invasive) and combined ERBD plus glucocorticoid therapy. This implies that if the safety of glucocorticoid monotherapy is assured, clinicians and patients would accept a ≤ 20% reduction in its efficacy relative to that of combination therapy.

Conclusion

Current studies evaluating glucocorticoid monotherapy versus ERBD for obstructive jaundice in AIP patients are lacking. As the first multicenter RCT assessing its efficacy and safety, this study is expected to provide valuable evidence to facilitate evidence-based practice for optimizing the clinical management of AIP patients with significant obstructive jaundice.

Trial status

This trial was registered on chictr.org.cn (ChiCTR2400086160) on June 26, 2024. Recruitment started on June 26, 2024. The first patient was randomized on June 30, 2024. It is scheduled to end on December 31, 2027. The protocol number is version 3.0 dated April 1, 2025, https://www.chictr.org.cn/showproj.html?proj=210824.

Abbreviations

AIP

Autoimmune pancreatitis

IgG4-SC

IgG4-related sclerosing cholangitis

ERCP

Endoscopic retrograde cholangiopancreatography

ERBD

Endoscopic retrograde biliary drainage

PEP

Post-ERCP pancreatitis

RCT

Randomized controlled trial

AEs

Adverse events

SAEs

Serious adverse events

PTCD

Percutaneous transhepatic biliary drainage

IgG4

Immunoglobulin G4

ICDC

International Consensus Diagnostic Criteria for Autoimmune Pancreatitis

ESGE

European Society of Gastrointestinal Endoscopy

CT

Computed tomography

LFT

Liver function tests

ITT

Intention to treat

EDC

Electronic data capture

Authors’ contributions

Conception and design, YQZ, SXC, and MYL; data collation, SXC, DXC, YWL, MK, YWW, and BGB; statistical analysis and safety monitoring plan, SXC, YMH, MK, and YWL; article writing, SXC; article revision, MYL, YQZ, and JZ; and final approval of the manuscript, all authors.

Funding

No funding was received.

Data availability

The datasets used and analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The research adhered to the ethical guidelines outlined in the 2013 Declaration of Helsinki and good clinical practice and was approved by the Ethics Committee of the Chinese PLA General Hospital and other participating institutions (Approval No. S2022-428-01). All significant protocol amendments were submitted for approval to the lead ethics committee/IRB prior to implementation.

Consent for publication

Not applicable.

Competing interests

We declare that we have no competing interests.