Pancreatic and Gallbladder Adverse Events Associated with Tirzepatide in Obesity Management: A Scoping Review

Abstract

Background

Obesity is a chronic and relapsing condition that affects all demographics and results in comorbidities, rising healthcare costs, and social challenges. Tirzepatide, a novel dual glucose-dependent insulinotropic polypeptide (GIP)/glucagon-like peptide-1 (GLP-1) receptor agonist, has emerged as a promising therapy offering significant weight reduction. However, uncertainty remains regarding its pancreatic and gallbladder safety profile across different populations and evidence sources.

Aim

To systematically map existing evidence on the safety of tirzepatide in obesity management, with an emphasis on pancreatic- and gallbladder-related adverse events (AEs) across randomized controlled trials (RCTs), observational studies, case reports, and pharmacovigilance evidence.

Methods

Following the methodological framework of Arksey and O’Malley (2005), expanded with Joanna Briggs Institute guidance, this scoping review included RCTs, observational studies, and post-marketing safety reports identified between December 2019 and September 2025. The searches were conducted across PubMed, EMBASE, the Cochrane Library, and the FDA Adverse Event Reporting System (FAERS), with additional evidence identified through gray literature, targeted journal searches, and published pharmacovigilance studies based on the EudraVigilance database. ClinicalTrials.gov was subsequently searched as an additional source to identify relevant clinical trials. Studies were eligible if they examined pancreatic or gallbladder AEs in the context of tirzepatide exposure among individuals with obesity. The review included 11 RCTs, 5 observational studies, 4 case reports, 1 EudraVigilance pharmacovigilance analysis, and 575 FAERS reports. Pancreatic and gallbladder AEs were systematically extracted and coded using MedDRA version 28.1, following the predetermined hierarchy. Findings were mapped descriptively across data sources without formal comparative statistical testing.

Results

Pancreatic and gallbladder AEs were reported infrequently in RCTs. Pancreatitis was reported in 11 participants across five RCTs, with study-specific frequencies ranging from 0.2% (3 of 1,896) to 3.7% (1 of 27), while cholelithiasis was reported in 73 participants across 11 RCTs, with frequencies ranging from 0.3% (1 of 335 and 1 of 364) to 3.7% (1 of 27). Among observational studies, pancreatitis was reported in three studies, with proportions of 0.6% (63/10,002), 0.9% (1/110), and 2.1% (5,882/280,094). One retrospective cohort study reported an incidence rate of 3.84 events per 1,000 person-years among 9,193 participants, while cholelithiasis was reported at 11.89 events per 1,000 person-years in the same cohort. In the EudraVigilance pharmacovigilance analysis, pancreatitis represented 363 reported AEs (6.3% of all AEs and 19.4% of gastrointestinal AEs). In the FAERS database, pancreatitis accounted for 155/288 pancreatic-related reports [53.8%] and cholelithiasis for 143/249 gallbladder-related reports [57.4%]. Four case reports described six distinct AEs: edematous pancreatitis, acute pancreatitis, pancreatitis necrotizing, acalculous cholecystitis, gallstones, and gallbladder wall thickening, each reported once. Female participants, White participants, adults aged 18–64 years, and participants from the United States (US) predominated among reported demographic characteristics; however, demographic characteristics and comorbidity profiles were inconsistently reported and varied across studies.

Conclusion

Pancreatic and gallbladder AEs were reported across clinical trials, observational studies, case reports, and pharmacovigilance databases. RCTs generally reported low frequencies of pancreatobiliary AEs, whereas pharmacovigilance databases captured a broader spectrum of reported events. Variability in study populations, comorbidity profiles, reporting methods, and the absence of exposure denominators limited direct comparison across evidence sources. Vigilant monitoring and standardized AE reporting are needed to further characterize the pancreatobiliary safety profile of tirzepatide.

Video abstract

Point your SmartPhone at the code above. If you have a QR code reader the video abstract will appear. Or use: https://youtu.be/4nPzBg3c90s

Access transcript: https://www.dovepress.com/article/video_transcript/637243

Introduction

Obesity is a chronic, progressive, and complex disease¹ that constitutes a major global health burden and is associated with substantial premature morbidity and mortality.² The prevalence of obesity has increased substantially worldwide, with global adult obesity prevalence more than doubling from 1990 to 2022.¹ Amid the rising global prevalence of obesity, tirzepatide has emerged as an important pharmacological treatment option for obesity management.³ Tirzepatide, a dual glucose-dependent insulinotropic polypeptide/glucagon-like peptide-1 receptor agonist (GIP/GLP-1 RA), is indicated for weight management in adults with obesity or overweight accompanied by weight-related comorbidities and is intended to be used alongside a reduced-calorie diet and increased physical activity.^4,5 A systematic review of clinical guidelines for the management of obesity in adults found that lifestyle modification, including dietary intervention and increased physical activity, together with psychological and behavioral interventions, remains a fundamental component of comprehensive obesity management. Medical nutrition therapy was recommended in 9 of 10 guidelines, physical activity in 7 of 10, and psychological and behavioral interventions in 8 of 10, while pharmacotherapy was recommended in 4 of 10 guidelines as an adjunct to lifestyle modification in appropriately selected individuals.⁶ Consistent with the increasing role of pharmacotherapy in obesity management, tirzepatide is recommended by several clinical guidelines as a pharmacological treatment option for adults with obesity, although eligibility criteria based on body mass index (BMI) thresholds and obesity-related comorbidities vary across guidelines.^4,7–9 In particular, the 2026 American College of Physicians guideline recommends tirzepatide as a first-line pharmacological treatment, alongside lifestyle modifications, for non-pregnant adults with obesity (BMI ≥30 kg/m²), supported by moderate-certainty evidence.¹⁰ These recommendations have been accompanied by increasing uptake of tirzepatide in clinical practice since its approval for obesity management.

Following its US approval as an anti-obesity medication in November 2023, an early real-world analysis found that 2,541 of 1,721,466 eligible patients (0.15%) were newly prescribed tirzepatide.¹¹ Subsequent US prescribing data, evaluated from January 2018 to June 2025, showed a 31.98% increase in tirzepatide prescribing between March and June 2025, reaching an overall prescription rate of 1.32%.¹² Other real-world observational data have demonstrated a 371.7% increase in GLP-1-based therapy utilization between 2022 and 2024, with a marked shift toward tirzepatide as the predominant agent in a population in which 83.6% of patients had obesity.¹³ In addition, a real-world survey of people with obesity without type 2 diabetes (T2D) reported high treatment motivation, adherence, and satisfaction among those prescribed tirzepatide.¹⁴

The expanding use of tirzepatide in obesity management highlights the importance of further characterizing its safety profile in the target population. Safety has been evaluated in systematic reviews with different primary objectives and populations. Reviews specifically assessing AEs examined a broad range of outcomes, including gastrointestinal, pancreatic, biliary, metabolic, hypersensitivity, serious, and multisystem events, but predominantly included patients with T2D.^15,16 Other systematic reviews and meta-analyses evaluated safety in mixed diabetes and obesity populations, with gastrointestinal events consistently being the most frequent,^17,18 while obesity-focused systematic reviews^19,20 and post-marketing pharmacovigilance analyses have likewise reported AEs across multiple organ systems.^21,22

Although tirzepatide has a broad safety profile, pancreatic and gallbladder events warrant specific attention as clinically relevant pancreatobiliary outcomes, with acute pancreatitis and acute gallbladder disease specifically addressed in the US Food and Drug Administration (FDA) and European Medicines Agency (EMA) product information.^23,24 Furthermore, the occurrence of pancreatic and gallbladder AEs with tirzepatide is biologically plausible, given the effects of incretin-based therapies on hepatobiliary, and pancreatic physiology. Several biological mechanisms have been proposed to underlie the effects of GLP-1 RA on gallbladder and biliary function, including alterations in cholecystokinin (CCK) secretion, bile acid signaling, and gallbladder motility and refilling.^25,26 Experimental and clinical studies have also demonstrated effects of GLP-1 receptor signaling on pancreatic enzyme regulation.^27–30 However, much of the available mechanistic evidence derives from selective GLP-1 receptor agonism, with more limited evidence regarding GIP-mediated effects on pancreatobiliary physiology. Thus, the extent to which these mechanisms contribute to clinically relevant pancreatobiliary AEs, particularly in the context of tirzepatide’s dual GIP/GLP-1 receptor agonism, remains uncertain.

Taken together, these considerations, along with the recent approval and expanding use of tirzepatide and regulatory warnings regarding pancreatobiliary AEs, underscore the need for a comprehensive exploring of its safety profile, particularly with respect to pancreatic and gallbladder outcomes. We define pancreatobiliary AEs as pancreatic disorders (including acute pancreatitis) and biliary/gallbladder disorders (including cholelithiasis, cholecystitis, and biliary colic). Existing evidence from systematic reviews and meta-analyses, and real-world data remains heterogeneous regarding pancreatic and gallbladder AEs. Several studies found no significant biliary risk or increased incidence of gallstone-related AEs with tirzepatide.^20,31,32 In contrast, other studies identified an association between tirzepatide and gallbladder or biliary tract disorders,^33,34 including cholelithiasis, particularly in weight-loss contexts.³⁵ Regarding pancreatic outcomes, the evidence is generally more consistent, with most systematic reviews reporting no increased risk of pancreatitis with tirzepatide^{20,32–34,36} or only a low incidence of acute pancreatitis.¹⁵ Nevertheless, some recent analyses have identified a statistically significant increase in risk.³⁷

Real-world safety data offer additional insights but remain limited in scope.³⁸ Pharmacovigilance analyses utilizing data from the FAERS database reported no increased pancreatobiliary risk aside from biliary colic,²² whereas analyses derived from the EudraVigilance pharmacovigilance system identified pancreatitis as common but did not capture gallbladder events such as cholelithiasis or cholecystitis.³⁹ Further pharmacovigilance signals for acute pancreatitis have been observed,⁴⁰ and recent FAERS-based analyses of AEs confirmed cholecystitis and cholelithiasis, particularly among women over 45 years of age.⁴¹

These inconsistencies underscore gaps in the current literature, particularly the lack of consistent findings regarding pancreatic and gallbladder AEs. The majority of pharmacovigilance analyses are restricted to data up to Q1 2025 and include combined diabetes and obesity populations,⁴² reinforcing the necessity of incorporating the most recent evidence specifically for overweight and obese cohorts. Furthermore, many existing studies primarily focus on diabetes populations, with relatively few investigations specifically tailored to individuals with obesity.³⁸

Beyond pancreatic- and gallbladder-related AEs, potential malignancies involving these organs represent an additional but less explored area of clinical interest, particularly in the context of long-term treatment.^36,43 The potential cancer risk associated with tirzepatide has primarily been evaluated in the context of its use forT2D, although more recent evidence has included populations with and without diabetes.⁴³ These findings indicate no increased risk of specific cancer types with tirzepatide, consistent with the findings of Cai et al, who similarly reported no significant differences in neoplasm incidence between tirzepatide and control groups.²⁰ Nevertheless, the available comparative evidence addresses cancer incidence at the study level and may not capture the full spectrum of rare neoplastic events reported across clinical and real-world settings.

Overall, important gaps remain, including fragmented evidence across RCTs, observational studies, case reports, and pharmacovigilance databases; inconsistent findings regarding gallbladder and biliary events; generally reassuring pancreatic outcomes in randomized trials, alongside reports of pancreatic AEs in pharmacovigilance data; a broader spectrum of reported AEs in pharmacovigilance databases than that described in randomized trials; limited evidence specifically in overweight and obesity populations compared with T2D populations; and the lack of a comprehensive synthesis mapping the types, patterns, frequencies, and characteristics of pancreatobiliary AEs across diverse evidence sources.

Given the heterogeneity of evidence types, variation in AE reporting and coding practices, and the objective of mapping the available evidence and identifying knowledge gaps rather than generating pooled effect estimates, a scoping review was considered an appropriate approach. Pancreatic and gallbladder AEs were selected due to their biological plausibility with incretin-based therapies, regulatory importance given recent FDA cautions, and inconsistent findings across obesity-focused studies. The potential relevance of neoplastic outcomes to the long-term safety of tirzepatide and the limited evidence in this area supported the inclusion of neoplastic events involving the pancreas, gallbladder, and biliary tract to comprehensively map the available evidence across clinical trials and real-world data.

Accordingly, this scoping review was conducted to systematically map pancreatic- and gallbladder-related AEs associated with tirzepatide use in obesity management across RCTs, observational studies, case reports, pharmacovigilance databases, and gray literature. Specifically, the review aimed to identify, classify, and summarize the types and frequencies of these AEs. In addition, the review examined patterns in AE reporting across patient subgroups, including age, sex, comorbidities, and geographical region; mapped methodological characteristics of the available evidence, including study designs, study populations, and surveillance systems; and identified areas in which safety evidence remains limited, particularly among underrepresented populations. Although regulatory product information addresses acute pancreatitis and gallbladder-related events as potential safety concerns with tirzepatide, regulatory documents primarily provide safety warnings and precautions rather than a comprehensive mapping of clinical and real-world evidence across different study designs and populations. This review incorporates clinical trial, observational, and pharmacovigilance evidence to map pancreatobiliary events in obesity management, thereby complementing regulatory guidance with a broader cross-source synthesis. By examining the types, frequencies, and characteristics of these events across diverse evidence sources, the review provides a more comprehensive characterization of the available safety evidence, highlights demographic and methodological patterns, and identifies areas requiring further investigation. In particular, integrating post-marketing pharmacovigilance evidence with clinical data enables identification of a broader spectrum of reported pancreatobiliary events, including rare events that may be difficult to characterize in randomized trials, while also highlighting differences in the populations and settings in which these events have been reported. This review may inform hypothesis generation, refinement of safety endpoints, and the design of future studies evaluating the long-term pancreatobiliary safety of tirzepatide in populations with obesity. Importantly, by identifying AEs reported in routine clinical practice that may not be observed in RCTs, post-marketing surveillance reinforces its essential role in establishing a comprehensive safety profile.

Materials and Methods

Methodology

Protocol and Registration

This scoping review, guided by the Arksey and O’Malley methodological framework (2005),⁴⁴ expanded with Joanna Briggs Institute (JBI) guidance,⁴⁵ was registered with the Open Science Framework (OSF) Registration DOI 10.17605/OSF.IO/PDZM6 (https://osf.io/zw4ug) on September 11, 2025. The OSF record contains the successive versions of the review protocol, with the registration DOI corresponding to the final Protocol Version 3. The title was refined from “Safety profile” to “Adverse events” in the final manuscript to more accurately reflect the focus of the review. The research question was also refined for clarity, although these editorial refinements did not result in any substantive changes to the review objectives, eligibility criteria, search strategy, data extraction, or methodology following protocol registration.

Study Design

A scoping review methodology was employed to map the breadth, range, and nature of the available literature and to identify potential evidence gaps. This approach facilitated a systematic exploration of broad evidence landscapes, particularly relevant to underexplored areas such as pancreatic and gallbladder AEs associated with tirzepatide in obesity management. The review followed Arksey and O’Malley’s five-stage framework. Reporting adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) Checklist (Supplementary Table 1).

Identifying the Research Question

The research question was developed using the Population–Concept–Context (PCC) framework:

Population: Individuals with obesity across diverse demographics (age, sex, comorbidities, geographic regions).

Concept: Pancreatobiliary AEs, defined as pancreatic disorders (including acute pancreatitis) and biliary/gallbladder disorders (including cholelithiasis, cholecystitis, and biliary colic).

Context: Exposure to tirzepatide across clinical trials, real-world studies, and pharmacovigilance databases.

Based on this framework, the scoping review was designed to address the following research question (RQ): “What pancreatic- and gallbladder-related AEs have been reported in association with tirzepatide use for obesity management across clinical trials, real-world studies, and pharmacovigilance evidence sources?”

Primary Objective:

1. To identify, classify, and summarize the types and frequency of pancreatic and gallbladder-related AEs associated with tirzepatide in obesity management based on data from clinical trials, real-world studies, and pharmacovigilance databases.

Secondary Objectives:

1. To identify trends in AE reporting across different patient subgroups (age, sex, comorbidities, geographical regions).

2. To map methodological characteristics of studies reporting these AEs including study designs, populations, and surveillance systems.

3. To identify gaps in the safety evidence, particularly in underrepresented populations (eg, adolescents, elderly and those with multiple chronic conditions).

Identifying the Relevant Studies

A comprehensive literature search was conducted across three bibliographic databases: PubMed, EMBASE, and the Cochrane Library, as well as pharmacovigilance sources, including the FAERS database. A preliminary search was conducted on July 30, 2025 to assess the availability of relevant literature and inform development of the search strategy. No data were extracted at this stage. Following protocol registration on September 11, 2025, the final search and data extraction were conducted on September 28, 2025. Published pharmacovigilance analysis of EudraVigilance data was identified through bibliographic database searches and gray-literature screening and were eligible for inclusion.

Preliminary screening revealed that several relevant studies were not retrieved through title and abstract searches. To strengthen coverage, ClinicalTrials.gov was subsequently added to the search strategy to capture unpublished or recently completed trials not yet indexed in the primary databases, following the protocol amendment documented on October 19, 2025. In addition, AEs reported in published articles did not encompass the full spectrum of outcomes relevant to this scoping review, further necessitating the inclusion of supplementary sources to ensure comprehensive coverage. The FAERS database was searched on September 28, 2025, for reports submitted from 2022 through September 28, 2025, using tirzepatide as the active substance rather than a specific brand name. The search period began in 2022 to capture reports from the initial use of tirzepatide, including potential off-label use for weight management before its approval for chronic weight management in November 2023. To identify reports relevant to obesity and weight management, the “Reason for Use” field was filtered for Obesity, Overweight, Weight Control, and Weight Decreased.

To collect real-world evidence, gray literature was accessed via Google Scholar, with the first ten pages of results screened and relevant records retained. To supplement database searches and enhance identification of rare AEs, targeted searches were conducted in journals known for publishing case reports and safety-related evidence concerning metabolic and gastrointestinal drug events. The following journals were included: American Journal of Gastroenterology (AJG), Cureus, Endocrine Practice, BMJ Case Reports, Wiley Clinical Case Reports, and JAMA. These journals routinely publish case reports describing pancreatic, gallbladder, and gastrointestinal AEs associated with GLP-1 RA, making them high-yield sources of tirzepatide safety information.

Search Strategy

Search terms (tirzepatide/obesity/pancreatitis/cholecystitis/cholelithiasis/adverse events) were combined using keywords, Medical Subject Headings (MeSH), and Boolean operators to ensure comprehensive coverage of the literature (Supplementary Table 2).

Timeframe: Searches spanned December 2019 (SURMOUNT-1 initiation in obese patients) to September 2025.

Screening and Study Selection

Search results were imported into Rayyan. A two-stage screening process was conducted, comprising title/abstract screening followed by full-text screening. Two reviewers independently screened titles and abstracts, with disagreements resolved by discussion or a third reviewer. The process was documented with a PRISMAScR flow diagram,⁴⁶ following a 20-record pilot test to confirm the clarity and applicability of the inclusion and exclusion criteria.

Inclusion Criteria

Peer-reviewed, studies published between December 2019 and September 2025, including clinical trials, observational studies, case reports and pharmacovigilance data from the FAERS database.

Types of participants: Patients with obesity regardless of their demographic characteristics and concurrent comorbidities.

Concept: Studies reporting at least one pancreatobiliary AE.

Context: Tirzepatide exposure in populations with obesity, regardless of concurrent comorbidities, or where obesity-related indications could reasonably be inferred from the study context across clinical trials, real-world studies, and pharmacovigilance databases. Because obesity frequently coexists with T2D, studies including participants with both conditions were retained when obesity constituted the primary treatment indication. This approach was adopted to enhance the applicability of findings to real-world obesity populations while avoiding evidence derived exclusively from diabetes-focused treatment cohorts.

Types of evidence sources: Primary studies employing quantitative, qualitative, or mixed-methods designs, as well as secondary analyses of pharmacovigilance databases. FAERS was treated separately as a pharmacovigilance data source because adverse-event reports were extracted directly from the database rather than from a published secondary analysis.

Exclusion Criteria

Studies were excluded if they were nonhuman, published in languages other than English, focused exclusively on patients with diabetes, lacked data on pancreatic or gallbladder AEs, or were reviews, systematic reviews, meta-analyses, protocols, editorials, post-hoc analyses, or other secondary publications, except for eligible pharmacovigilance database analyses.

Data Extraction

The relevant data were entered into the Excel data extraction form by two researchers independently (Supplementary Table 3). For each included study, the following details were extracted and charted:

• Study Identification and Source: Author(s) and Year of Publication, Citation, Study Number/Registration, Study title;

• Population Characteristics: Age Group, Gender, Ethnicity, Race, Comorbidities;

• Frequency/incidence/number of reports of Pancreatic- and Gallbladder-related AEs, Severity, Dropouts, Death;

• Study characteristics (Design, Duration, Reporting period, Sample Size, Aim, Geographic region, Population types, Comparator);

• Surveillance methods and Reporting source.

Collating, Summarizing and Reporting the Results

Pancreatic- and gallbladder-related AEs were systematically extracted from RCTs, observational studies, case reports, EudraVigilance-based analyses and the FAERS database, in accordance with the predetermined MedDRA hierarchy applied across these data sources.

Participant counts and study-specific AE frequencies were retrieved from RCTs, while incidence rates, where available, were extracted from observational studies, alongside the numbers of reports with relevant AEs from the FAERS database and the numbers of reported AEs from the EudraVigilance-based analysis. Data were summarized descriptively using frequency counts,⁴⁷ and trends were explored descriptively across data sources without formal comparative statistical testing, consistent with scoping review methodology. Comparator data were extracted but not included in the descriptive quantitative reporting, which focused on AEs among tirzepatide-treated participants.

Counts are not directly comparable across data sources and reflect different reporting units.

Severity was extracted as reported by original sources and summarized descriptively without reclassification. Demographic characteristics, study characteristics, surveillance methods, and reporting sources were mapped to identify dominant approaches and underrepresented areas. Consistent with the aim of this scoping review to map the broad spectrum of pancreatobiliary AEs, neoplastic events involving the pancreas, gallbladder, and biliary tract were also captured when identified in the eligible evidence sources. Gaps in evidence were highlighted, particularly for populations with limited representation and underrepresented study types.

Results were presented narratively and visually, using tables, charts, and a PRISMA-ScR flow diagram to summarize key trends and highlight differences between trial data and real-world pharmacovigilance findings.

Coding

AEs in RCTs were initially reported using different MedDRA versions (Supplementary Table 4). For consistency, all events across study designs were realigned and coded according to MedDRA Version 28.1 (September 2025) and categorized under Preferred Terms (PTs) and System Organ Classes (SOCs), while preserving the established hierarchy. For observational studies, AEs already coded in MedDRA were extracted directly. When MedDRA coding was unavailable, reported events were harmonized by mapping them to the corresponding Preferred Terms (PTs) or System Organ Classes (SOCs) in MedDRA Version 28.1 to ensure consistent terminology.

Where PTs were unavailable, High-Level Terms (HLTs) were applied. In the FAERS database, AEs were reported in accordance with the current MedDRA version. Since AEs in case reports were provided verbatim, definitive pancreatic and gallbladder diagnoses, excluding separate coding of associated signs and symptoms, were standardized using MedDRA Version 28.1, thereby harmonizing terminology and enabling consistent comparison across study designs (Supplementary Table 5).

Ethical Considerations

This scoping review relied exclusively on previously published studies and publicly available data, and therefore ethical approval was not required. All included studies were cited and reported in accordance with established principles of transparency and research integrity.

Results

Search Results

A total of 971 records were identified across multiple databases and sources: PUBMED (25), EMBASE (161), the Cochrane Library (185), ClinicalTrials.gov (12), Google Scholar (13), and the FAERS database (575). After removing 128 duplicates, 843 records remained for title/abstract screening. During this screening phase, 179 records were excluded, leaving 664 articles for full-text eligibility assessment. Of these, 70 were excluded for various reasons, including:

• ●Studies not yet completed
• ●Treatments other than tirzepatide
• ●Lack of relevant AEs reporting
• ●Results unavailable due to withdrawal or termination
• ●Background articles

This resulted in 594 records meeting the inclusion criteria. Additionally, searches from journals American Journal of Gastroenterology (AJG), Cureus, Endocrine Practice, BMJ Case Reports, Wiley Clinical Case Reports, and JAMA identified three cases for inclusion.

In total, 597 records were incorporated into the review (Figure 1).

Figure 1.

PRISMA flow diagram of the study selection process for the scoping review. Abbreviation: PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

A total of 11 RCTs^48–59 (two articles by Jastreboff et al^48,49 represent the same study but differ in reporting periods; SURMOUNT-4 Lead-in and SURMOUNT-4 Double-blind⁵² were counted as one study, as were SURMOUNT-OSA Trial 1 and SURMOUNT-OSA Trial 2),⁵⁷ 5 observational studies,^60–64 1 Pharmacovigilance analysis of individual case safety reports (ICSRs) using the EudraVigilance database,³⁹ 4 case reports^65–68 and 575 reports from the FAERS database⁶⁹ were included in the summarizisation. The extracted data from the included RCTs, observational studies, and case reports are presented in Supplementary Table 6, while the extracted FAERS data are presented in Supplementary Tables 7A–10B. Counts from pharmacovigilance databases reflect reported AEs rather than unique individuals and should not be interpreted as incidence.

To ensure completeness of safety and demographic data, published reports were supplemented with information from ClinicalTrials.gov,^59,70–78 where AEs and demographic details collected during the trials but not reported in the corresponding articles were retrieved and distinguished in Supplementary Table 6. For the descriptive mapping, data from all tirzepatide treatment arms across doses were pooled and summarized for tirzepatide overall. Comparator-arm data were extracted but were not incorporated into the descriptive summary of tirzepatide-associated AEs. Detailed comparator data are provided in Supplementary Table 6. For clarity, each included trial was counted as a single study regardless of the number of associated publications or ClinicalTrials.gov records. Study titles, NCT numbers, acronyms where available, and corresponding references are provided in Table 1. Journal articles retrieved through the initial database searches served as the primary references, while ClinicalTrials.gov records were additionally referenced when they provided relevant information not available in the corresponding publications.

Table 1.

Characteristics and Reference Mapping of Included Studies

No	Evidence Type	Trial Acronym or Title	Trial ID/NCT Number	Publication/Registry	Reference
1	RCT	SURMOUNT-1	NCT04184622	Jastreboff et al, 2022	[48]
	RCT	SURMOUNT-1	NCT04184622	Jastreboff et al, 2025	[49]
	RCT	SURMOUNT-1	NCT04184622	ClinicalTrials.gov	[71]
2	RCT	SURMOUNT-2	NCT04657003	Garvey et al, 2023	[50]
	RCT	SURMOUNT-2	NCT04657003	ClinicalTrials.gov	[72]
3	RCT	SURMOUNT-3	NCT04657016	Wadden et al, 2023	[51]
	RCT	SURMOUNT-3	NCT04657016	ClinicalTrials.gov	[77]
4	RCT	SURMOUNT-4	NCT04660643	Aronne et al, 2024	[52]
	RCT	SURMOUNT-4	NCT04660643	ClinicalTrials.gov	[78]
5	RCT	SURMOUNT-5	NCT05822830	Aronne et al, 2025	[53]
	RCT	SURMOUNT-5	NCT05822830	ClinicalTrials.gov	[76]
6	RCT	SURMOUNT-J	NCT04844918	Kadowaki et al, 2025	[54]
	RCT	SURMOUNT-J	NCT04844918	ClinicalTrials.gov	[74]
7	RCT	SYNERGY-NASH	NCT04166773	Loomba et al, 2024	[55]
	RCT	SYNERGY-NASH	NCT04166773	ClinicalTrials.gov	[70]
8	RCT	SURMOUNT-CN	NCT05024032	Zhao et al, 2024	[56]
9	RCT	SURMOUNT-OSA	NCT05412004	Malhotra et al, 2024	[57]
	RCT	SURMOUNT-OSA	NCT05412004	ClinicalTrials.gov	[75]
10	RCT	SUMMIT	NCT04847557	Packer et al, 2025	[58]
	RCT	SUMMIT	NCT04847557	ClinicalTrials.gov	[73]
11	RCT	A Study to Measure Energy Expenditure and Food Intake in Participants With Obesity Using Tirzepatide	NCT04081337	ClinicalTrials.gov	[59]
12	Observational: retrospective cohort study	Semaglutide vs Tirzepatide for Weight Loss in Adults With Overweight or Obesity	NA	Rodriguez et al, 2024	[60]
13	Observational, retrospective analysis of public Facebook posts	Harnessing Facebook to Investigate Real-World Mentions of Adverse Events of Glucagon-Like Peptide-1 Receptor Agonist (GLP-1 RA) Medications: Observational Study of Facebook Posts From 2022 to 2024	NA	Alibilli et al, 2025	[61]
14	Retrospective cohort study	Weight Loss and Safety of Glucagon-like Peptide-1 Agonists in Inflammatory Bowel Disease Patients S1086	NA	Anderson et al, 2024	[64]
15	Pharmacovigilance analysis of individual case safety reports (ICSRs) using EudraVigilance database	Safety profile of tirzepatide: A real-world pharmacovigilance analysis of EudraVigilance database	NA	Khurana et al, 2024	[39]
16	A retrospective cohort study was conducted using the TriNetX research network	Comparative Risk of Pancreatic Pseudocyst Among Users of GLP-1 RAs: Findings From a LargeScale EHR-Based Study S138	NA	AlQirem et al, 2025	[62]
17	A retrospective cohort study using TriNetX, a global federated network of de-identified electronic health records	Comparative Gastrointestinal Safety of Tirzepatide vs Semaglutide: A Real-World Multicenter Analysis S786	NA	Nwatamole et al, 2025	[63]
18	Case report	Tirzepatide (Mounjaro)‑induced acute pancreatitis and ketoacidosis in an adult nondiabetic patient	NA	Sallam et al, 2025	[65]
19	Case report	A Case of Fatal Fulminant Necrotizing Pancreatitis in a Patient with Recent Tirzepatide Initiation 1698670	NA	Grennan et al, 2024	[66]
20	Case report	A Probable Case of Tirzepatide-Induced Acute Pancreatitis	NA	Sumaruth et al, 2025	[67]
21	Case report	Weight Loss, Unexpected Cost: GLP-1 Receptor Agonist-Associated Acalculous Cholecystitis S3261	NA	Ghassemi et al, 2025	[68]
22	Case reports	NA	NA	FDA Adverse Event Reporting System database	[69]

Abbreviation: NA, not applicable.

Pancreatic-Related Adverse Events

Pancreatic-related AEs were reported across 9 RCTs, including SURMOUNT-1 (NCT04184622),^48,49,71 SURMOUNT-2 (NCT04657003),^50,72 SURMOUNT-3 (NCT04657016),^51,77 SURMOUNT-4 (NCT04660643),^52,78 SURMOUNT-J (NCT04844918),^54,74 SYNERGY-NASH (NCT04166773),^55,70 SURMOUNT-OSA (NCT05412004),^57,75 SUMMIT (NCT04847557)^58,73 and NCT04081337,⁵⁹ 5 observational studies,^60–64 1 EudraVigilance pharmacovigilance analysis,³⁹ 3 case reports^65–67 and 309 pancreatic-related adverse-event reports identified in the FAERS database.⁶⁹ Among the 309 pancreatic-related FAERS reports, 288 were classified within the “Gastrointestinal Disorders” SOC, 13 within “Neoplasms Benign, Malignant and Unspecified (Including Cysts and Polyps)” SOC, 7 within “Investigations” SOC, and 1 within “Infections and Infestations” SOC.

To distinguish unique studies from multiple publications or ClinicalTrials.gov records referring to the same trial, each included study was assigned a unique study identity based on the trial acronym and/or NCT number. The corresponding publications and registry records are mapped to these unique studies in Table 1. Pharmacovigilance findings represent reported events rather than incidence estimates due to the unavailability of exposure denominators. Therefore, direct comparisons with event frequencies observed in clinical trials should be interpreted cautiously.

Across five RCTs, a total of 11 participants experienced pancreatitis, with study-specific proportions ranging from 0.2% to 3.7%, depending on study duration and design: 3/1,896 [0.2%] and 3/762 [0.4%] in SURMOUNT-1 at 72 and 176 weeks, respectively,^48,49 2/623 [0.3%] in SURMOUNT-2,⁵⁰ 1/287 [0.3%] in SURMOUNT-3,⁵¹ 1/142 [0.7%] in SYNERGY-NASH⁷⁰ and 1/27 [3.7%] in NCT04081337.⁵⁹ Among four observational studies reporting pancreatic outcomes,^60,61,63,64 three reported pancreatitis in a total of 5,946 participants, with proportions ranging from 0.6% to 2.1%: 63/10,002 [0.6%], 1/110 [0.9%] and 5,882/280,094 [2.1%]; one study reported an incidence rate of 3.84 events per 1,000 person-years among 9,193 participants. In the EudraVigilance pharmacovigilance analysis,³⁹ pancreatitis accounted for 363 reported AEs, representing 6.3% of all reported AEs and 19.4% of AEs within the gastrointestinal disorders system organ class. In the FAERS database, pancreatitis accounted for more than half of the pancreatic-related adverse-event reports within the “Gastrointestinal Disorders” SOC, representing 155 of 288 reports [53.8%].

Within the broader category of pancreatitis, both acute and chronic forms were documented. Acute pancreatitis occurred in 8 participants across four RCTs, ranging from 0.2% to 1.7%: 3/1,896 [0.2%] in SURMOUNT-1,⁷¹ 2/119 [1.7%] in SURMOUNT-OSA Trial 2,⁵⁷ 2/623 [0.3%] in SURMOUNT-2⁷² and 1/364 [0.3%] in SUMMIT.⁷³ In the FAERS database, acute pancreatitis accounted for 79/288 reports [27.4%] within the “Gastrointestinal Disorders” SOC, while case reports described one individual case.⁶⁵ Chronic pancreatitis was less frequently observed, with 3/288 reports [1.0%] identified only in FAERS.

Pancreatic enzyme elevations were consistently reported across five RCTs, affecting six participants. Study-specific proportions ranged from 0.3% to 0.9%: 2/623 [0.3%] participants in SURMOUNT-2,⁵⁰ 1/335 [0.3%] during the double-blind period of SURMOUNT-4,⁵² 1/150 [0.7%] in SURMOUNT-J,⁷⁴ 1/142 [0.7%] in SYNERGY-NASH⁷⁰ and 1/114 [0.9%] in SURMOUNT-OSA.⁷⁵ An additional seven reports were identified in FAERS.

Rare but notable findings included pancreatic neoplasms in FAERS. Within 16 reports classified under the “Neoplasms benign, malignant and unspecified (including cysts and polyps)” SOC, Pancreatic carcinoma accounted for 10/16 reports [62.5%], while pancreatic neoplasm and pancreatic carcinoma stage IV were each reported once (1/16 [6.3%]). Pancreatic carcinoma metastatic was reported in one RCT (SURMOUNT-1),⁷¹ involving 1/1,896 participants [0.1%], and in 1/16 FAERS reports [6.3%]. All pancreatic-related AEs with study-specific denominators and proportions are summarized and presented in Table 2 and visualized in Supplementary Figure 1A–D.

Table 2.

Distribution of Pancreatic-Related Adverse Events Across Studies

Adverse Events, n (%)		Study Types
		RCTs	Observational Studies	Pharmacovigilance Analysis of Individual Case Safety Reports (ICSRs) Using EudraVigilance Database,39	Case Reports	FAERS Database●
SOC Gastrointestinal disorders 10017947
HLGT Exocrine pancreas conditions 10015674
HLT Acute and chronic pancreatitis 10033646
1	Autoimmune pancreatitis 10069002	0	0	0	0	1/288 (0.3%)
2	Haemorrhagic necrotic pancreatitis 10076058	0	0	0	0	1/288 (0.3%)
3	Idiopathic pancreatitis 10088882	0	0	0	0	1/288 (0.3%)
4	Obstructive pancreatitis 10079822	1. SUMMIT (NCT04847557):73 2/364 (0.5%)***	0	0	0	22/288 (7.6%)
5	Oedematous pancreatitis 10052400	0	0	0	167,◇	6/288 (2.1%)
6	Pancreatitis 10033645	1. SURMOUNT-1 (NCT04184622), 72 weeks:48 3/1,896 (0.2%)&,○ SURMOUNT-1 (NCT04184622), 176 weeks:49 3/762 (0.4%)& 2. SURMOUNT-3 NCT04657016:51 1/287 (0.3%)& 3. SURMOUNT-2 (NCT04657003):50 2/623 (0.3%)& 4. SYNERGY-NASH (NCT04166773):70 1/142 (0.7%)** 5.NCT04081337:59 1/27 (3.7%)***	1. Retrospective cohort:60 3.84 events per 1,000 person-years /9,193 2. Retrospective analysis:61 63/10,002 (0.6%)$ 3. Retrospective cohort:64 1/110 (0.9%)$$ 4. Retrospective cohort:63 5,882/280,094 (2.1%)	363/5,774 (6.3%)39 363/1,873 (19.4%),39 #	0	155/288 (53.8%)
7	Pancreatitis acute 10033647	1. SURMOUNT-1 (NCT04184622), baseline up to week 193:71 3/1,896 (0.2%)*** 2. SURMOUNT-OSA (NCT05412004), Trial 2:57 2/119 (1.7%)& 3. SURMOUNT-2 (NCT04657003), Baseline through Week 76:72 2/623 (0.3%)*** 4. SUMMIT (NCT04847557):73 1/364 (0.3%)***	0	0	165,***	79/288 (27.4%)
8	Pancreatitis chronic 10033649	0	0	0	0	3/288 (1%)
9	Pancreatitis necrotising 10033654	0	0	0	166,◆	10/288 (3.5%)
HLT Pancreatic disorders NEC 10027692
10	Pancreatic atrophy 10033603	0	0	0	0	2/288 (0.7%)
11	Pancreatic disorder 10033616	0	0	0	0	7/288 (2.4%)
12	Pancreatic duct dilatation 10064858	0	0	0	0	1/288 (0.3%)
13	Pancreatic enlargement 10055024	0	0	0	0	1/288 (0.3%)
14	Pancreatic failure 10079281	0	0	0	0	3/288 (1%)
15	Pancreatic mass 10049082	0	0	0	0	2/288 (0.7%)
16	Pancreatolithiasis 10051252	0	0	0	0	2/288 (0.7%)
17	Pancreatic steatosis 10072787	0	0	0	0	1/288 (0.3%)
HLT Cystic pancreatic disorders 10011774
18	Pancreatic cyst 10033615	1. SURMOUNT-J (NCT04844918):74 2/150 (1.3%)**	0	0	0	3/288 (1%)
19	Pancreatic pseudocyst 10033635	0	1. Retrospective cohort:62 737/516,978 (0.1%)	0	0	2/288 (0.7%)
SOC Investigations 10022891
HLGT Gastrointestinal investigations 10017971
HLT Digestive enzymes 10012981
20	Pancreatic enzymes increased 10061900	1. SURMOUNT-2 (NCT04657003):50 2/623 (0.3%) 2. SURMOUNT-4 (NCT04660643), Double-blind:52 1/335 (0.3%) 3. SURMOUNT-J (NCT04844918):74 1/150 (0.7%)** 4. SYNERGY-NASH (NCT04166773):70 1/142 (0.7%)** 5. SURMOUNT-OSA (NCT05412004):75 1/114 (0.9%)**	0	0	0	7/7 (100%)
SOC Neoplasms benign, malignant and unspecified (incl cysts and polyps) 10029104
HLGT Gastrointestinal neoplasms malignant and unspecified 10017991
HLT Gastrointestinal neoplasms malignancy unspecified NEC 10017994
21	Pancreatic neoplasm 10061902	0	0	0	0	1/16 (6.3%)
HLT Pancreatic neoplasms malignant (excl islet cell and carcinoid) 10033633
22	Pancreatic carcinoma 10033609	0	0	0	0	10/16 (62.5%)
23	Pancreatic carcinoma metastatic 10033610	1. SURMOUNT-1 (NCT04184622), baseline up to week 193:71 1/1,896 (0.1%)***	0	0	0	1/16 (6.3%)
24	Pancreatic carcinoma stage IV 10059326	0	0	0	0	1/16 (6.3%)
SOC Infections and infestations 10021881
HLGT Infections - pathogen unspecified 10021879
HLT Abdominal and gastrointestinal infections 10017967
25	Pancreas infection 10051741	0	0	0	0	1/14 (7.1%)

Notes: ^◆Adverse Events were reported as severe. ◇ Adverse Events were reported as mild. ^○Adverse Events were not reported as severe. *Adverse Events were reported as severe or serious adverse events. **Adverse Events were reported as other (not including serious) adverse events. ***Adverse Events were reported as serious adverse events. ^$Reported for Mounjaro, Tirzepatide and Zepbound. ^$Reported for Dulaglutide, Exenatide, Liraglutide, Semaglutide, or Tirzepatide. ^#Among reported Gastrointestinal disorders. ^&Adjudication-Confirmed. ^{^}Acute gallbladder diseases. ^●FAERS:⁶⁹ Counts from the FAERS pharmacovigilance database reflect submitted reports rather than unique individuals and should not be interpreted as incidence estimates. The total number of reports for each SOC (SOC “Gastrointestinal disorders”, SOC “Hepatobiliary disorders”, SOC “Investigations”, SOC “Neoplasms benign, malignant and unspecified (incl cysts and polyps)”, SOC “Infections and infestations”, SOC “Surgical and medical procedures”) were considered in relation to adverse events related to the pancreas and gallbladder, rather than all reported adverse events in each group. MedDRA Version 28.1. Counts are not directly comparable across data sources and reflect different reporting units.

Overall, pancreatitis and its subtypes were reported across all evidence sources, with acute pancreatitis consistently observed in clinical studies, while a broader spectrum of rare pancreatic disorders was predominantly identified in post-marketing reports. Differences in reporting structures across study designs limit direct comparison but collectively support the importance of integrated pharmacovigilance and clinical trial evidence in characterizing tirzepatide’s pancreatobiliary safety profile.

Gallbladder-Related Adverse Events

Gallbladder-related AEs were identified across 11 RCTs, including SURMOUNT-1 (NCT04184622),^48,49,71 SURMOUNT-2 (NCT04657003),^50,72 SURMOUNT-3 (NCT04657016),^51,77 SURMOUNT-4 (NCT04660643),^52,78 SURMOUNT-5 (NCT05822830),^53,76 SURMOUNT-J (NCT04844918),^54,74 SYNERGY-NASH (NCT04166773),^55,70 SURMOUNT-CN (NCT05024032),⁵⁶ SURMOUNT-OSA (NCT05412004),^57,75 SUMMIT (NCT04847557)^58,73 and NCT04081337,⁵⁹ 2 observational studies,^60,61 1 EudraVigilance pharmacovigilance analysis,³⁹ 2 case reports^67,68 and 266 reports from FAERS.⁶⁹ Within the FAERS database, these 266 reports were classified as 249 within the “Hepatobiliary Disorders” SOC, 13 within “Infections and infestations” SOC, 3 within “Neoplasms benign, malignant and unspecified (including cysts and polyps)” SOC, and 1 within “Surgical and medical procedures” SOC.

Cholelithiasis was the most common gallbladder-related AE, reported in 73 participants across 11 RCTs (SURMOUNT-1,^48,49 SURMOUNT-2,⁵⁰ SURMOUNT-3,⁵¹ SURMOUNT-4,⁵² SURMOUNT-5,⁷⁶ SURMOUNT-J,⁵⁴ SYNERGY-NASH,⁷⁰ SURMOUNT-CN,⁵⁶ SURMOUNT-OSA,⁷⁵ SUMMIT⁷³ and NCT04081337),⁵⁹ with study-specific proportions ranging from 0.3% to 3.7% (Table 3). The lowest proportion was 0.3% in both SURMOUNT-4⁵² during the double-blind period (1 of 335) and SUMMIT⁷³ (1 of 364), while the highest was 3.7% in NCT04081337⁵⁹. In the FAERS database, cholelithiasis was documented in 143/249 reports, accounting for 57.4% of gallbladder-related AE reports, and 1 retrospective cohort study⁶⁰ reported an incidence of 11.89 events per 1,000 person-years/9,193.

Table 3.

Distribution of Gallbladder-Related Adverse Events Across Studies

Adverse Events, n (%)		Study types
		RCTs	Observational Studies	Pharmacovigilance Analysis of Individual Case Safety Reports (ICSRs) Using EudraVigilance Database,39	Case Reports	FAERS Database●
SOC Hepatobiliary disorders 10019805
HLGT Bile duct disorders 10004606
HLT Bile duct infections and inflammations 10004607
1	Biliary colic 10004663	1. SYNERGY-NASH (NCT04166773):70 1/142 (0.7%)** 2. SURMOUNT-2 (NCT04657003), Baseline through Week 76:72 1/623 (0.2%)***	0	0	0	28/249 (11.2%)
2	Cholangitis 10008604	1. SYNERGY-NASH (NCT04166773):70 1/142 (0.7%)**	0	0	0	9/249 (3.6%)
3	Cholangitis acute 10008605	1. SURMOUNT-1 (NCT04184622), baseline up to week 193:71 1/1,896 (0.1%)***	0	0	0	2/249 (0.8%)
HLT Obstructive bile duct disorders (excl neoplasms) 10029976
4	Bile duct stone 10004637	1. SURMOUNT-1 (NCT04184622), baseline up to week 193:71 2/1,896 (0.1%)***	0	0	0	10/249 (4%)
5	Biliary obstruction 10084247	1. SURMOUNT-1 (NCT04184622), baseline up to week 193:71 1/1,896 (0.1%)***	0	0	0	8/249 (3.2%)
HLT Structural and other bile duct disorders 10042257
6	Biliary dilatation 10057202	0	0	0	0	2/249 (0.8%)
7	Biliary tract disorder 10061008	0	0	0	0	2/249 (0.8%)
HLGT Gallbladder disorders 10017628
HLT Cholecystitis and cholelithiasis 10008616
8	Acute cholecystitis necrotic 10090325	0	0	0	0	2/249 (0.8%)
9	Cholecystitis 10008612	1. SURMOUNT-1 (NCT04184622), 72 weeks:48 7/1,896 (0.4%) SURMOUNT-1 (NCT04184622), 176 weeks:49 3/762 (0.4%) 2. SYNERGY-NASH (NCT04166773):55 1/142 (0.7%) 3. SUMMIT (NCT04847557):73 1/364 (0.3%)*** 4. SURMOUNT-5 (NCT05822830):76 1/374 (0.3%)***	1. Retrospective cohort:60 6.50 events per 1,000 person-years/9,193	0	0	41/249 (16.5%)
10	LLT Acalculous cholecystitis 10000347	0	0	0	143	0
11	Cholecystitis acute 10008614	1. SURMOUNT-1 (NCT04184622), 72 weeks:48 6/1,896 (0.3%) SURMOUNT-1 (NCT04184622), 176 weeks:49 5/762 (0.7%) 2. SURMOUNT-3 (NCT04657016):51 1/287 (0.3%) 3. SURMOUNT-4 (NCT04660643), Led-in:52 4/783 (0.5%) 4. SURMOUNT-CN (NCT05024032):56 1/141 (0.7%) 5. SURMOUNT-2 (NCT04657003):50 4/623 (0.6%) 6. SUMMIT (NCT04847557):58 6/364 (1.6%) 7. SURMOUNT-5 (NCT05822830):76 1/374 (0.3%)***	0	0	0	17/249 (6.8%)
12	Cholecystitis chronic 10008617	1. SURMOUNT-1 (NCT04184622), 72 weeks:48 5/1,896 (0.3%) SURMOUNT-1 (NCT04184622), 176 weeks:49 5/762 (0.7%) 2. SURMOUNT-4 (NCT04660643), Led-in:52 1/783 (0.1%) 3. SURMOUNT-CN (NCT05024032):56 1/141 (0.7%) 4. SUMMIT (NCT04847557):58 1/364 (0.3%)	0	0	0	1/249 (0.4%)
13	Cholelithiasis 10008629	1. SURMOUNT-1 (NCT04184622), 72 weeks:48 20/1,896 (1.1%) SURMOUNT-1 (NCT04184622), 176 weeks:49 22/762 (2.9%) 2. SURMOUNT-3 (NCT04657016):51 4/287 (1.4%) 3. SURMOUNT-4 (NCT04660643), Led-in:52 7/783 (0.9%) SURMOUNT-4 (NCT04660643), Double-blind:52 1/335 (0.3%) 4. SURMOUNT-J (NCT04844918):54 1/150 (0.7%) 5. SYNERGY-NASH (NCT04166773):70 1/142 (0.7%)** 6. SURMOUNT-CN (NCT05024032):56 3/141 (2.1%) 7. SURMOUNT-OSA (NCT05412004), Trial 2:75 2/119 (1.7%)** 8. SURMOUNT-2 (NCT04657003):50 8/623 (1.3%) 9. SUMMIT (NCT04847557):73 1/364 (0.3%)*** 10. NCT04081337:59 1/27 (3.7%)*** 11. SURMOUNT-5 (NCT05822830):76 2/374 (0.5%)***	1. Retrospective cohort:60 11.89 events per 1,000 person-years/9,193	0	0	143/249 (57.4%)
14	LLT Gallstones 10017650	0	0	0	167	0
15	Cholelithiasis Obstructive 10008630	0	0	0	0	1/249 (0.4%)
HLT Gallbladder disorders NEC 10017629
16	Gallbladder cholesterolosis 10017624	0	0	0	0	1/249 (0.4%)
17	Gallbladder disorder 10017626	1. SURMOUNT-5 (NCT05822830):76 1/374 (0.3%)*** 2. SURMOUNT-4 (NCT04660643), Led-in:52 6/783 (0.8%)* 3. SYNERGY-NASH (NCT04166773):55 4/142 (2.8%)	1. Retrospective analysis:61 5/10,002 (0.05%)$	0	0	30/249 (12%)
18	LLT Gallbladder disease 10017641	1. SURMOUNT-1 (NCT04184622), 72 weeks:48 22/1,896 (1.2%)* SURMOUNT-1 (NCT04184622), 176 weeks:49 22/762 (2.9%)* 2. SURMOUNT-3 (NCT04657016):51 2/287 (0.7%)*^ 3. SURMOUNT-4 (NCT04660643), Led-in:52 7/783 (0.9%)*^ 4. SURMOUNT-J (NCT04844918):54 1/150 (0.7%)^ 5. SURMOUNT-5 (NCT05822830):53 4/374 (1.1%) 6. SURMOUNT-CN NCT05024032:56 3/141 (2.1%) 7. SURMOUNT-2 (NCT04657003):50 6/623 (1%)*^	0	0	0	0
19	Gallbladder enlargement 10062693	0	0	0	0	6/249 (2.4%)
20	LLT Gallbladder wall thickening 10064076	0	0	0	167	0
21	Gallbladder hypofunction 10080936	0	0	0	0	1/249 (0.4%)
22	Gallbladder Obstruction 10017636	0	0	0	0	2/249 (0.8%)
23	Gallbladder polyp 10049704	1. SURMOUNT-OSA (NCT05412004), Trial 1:75 1/114 (0.9%)**	0	0	0	4/249 (1.6%)
24	Gallbladder rupture 10081151	0	0	0	0	5/249 (2%)
25	Hydrocholecystis 10056972	0	0	0	0	1/249 (0.4%)
HLGT Hepatic and hepatobiliary disorders 10019654
HLT Cholestasis and jaundice 10008636
26	Cholestasis 10008635	0	0	0	0	1/249 (0.4%)
27	Deficiency of bile secretion 10071634	0	0	0	0	1/249 (0.4%)
28 HLT Hepatic and hepatobiliary disorders NEC 10027681		0	0	169/5,774 AE (2.9%)39	0	0
29	Hepatobiliary disease 10062000	0	0	0	0	1/249 (0.4%)
SOC Neoplasms benign, malignant and unspecified (incl cysts and polyps) 10029104
HLGT Hepatobiliary neoplasms malignant and unspecified 10019815
HLT Bile duct neoplasms malignant 10004613
30	Cholangiocarcinoma 10008593	1. SUMMIT (NCT04847557):73 1/364 (0.3%)***	0	0	0	2/16 (12.5%)
HLT Gallbladder neoplasms malignant 10017634
31	Gallbladder cancer 10017614	0	0	0	0	1/16 (6.3%)
32	Gallbladder cancer stage III 10058282	1. SURMOUNT-1 (NCT04184622), baseline up to week 193:71 1/1,896 (0.1%)***	0	0	0	0
HLT Hepatobiliary neoplasms malignancy unspecified 10019814
33	Biliary neoplasm 10004668	1. SUMMIT (NCT04847557):73 1/364 (0.3%)***	0	0	0	0
SOC Surgical and medical procedures 10042613
HLGT Hepatobiliary therapeutic procedures 10019818
HLT Biliary tract and gallbladder therapeutic procedures 10004675
34	Cholecystectomy 10008611	1. SURMOUNT-1 (NCT04184622), baseline up to week 193:71 1/1,896 (0.1%)*** 2. SURMOUNT-CN (NCT05024032):56 1/141 (0.7%) 3. SURMOUNT-2 (NCT04657003):50 1/623 (0.2%)	0	0	0	1/1 (100%)
SOC Infections and infestations 10021881
HLGT Infections - pathogen unspecified 10021879
HLT Hepatobiliary and spleen infections 10064462
35	Cholecystitis infective 10062631	0	0	0	0	11/14 (78.6%)
36	Biliary sepsis 10057847	0	0	0	0	2/14 (14.3%)

Notes: *Adverse Events were reported as severe or serious adverse events. **Adverse Events were reported as other (not including serious) adverse events. ***Adverse Events were reported as serious adverse events. ^$Reported for Mounjaro, Tirzepatide and Zepbound. ^$Reported for Dulaglutide, Exenatide, Liraglutide, Semaglutide, or Tirzepatide. ^#Among reported Gastrointestinal disorders. ^&Adjudication-Confirmed. ^{^}Acute gallbladder diseases. ^●FAERS:⁶⁹ Counts from the FAERS pharmacovigilance database reflect submitted reports rather than unique individuals and should not be interpreted as incidence estimates. The total number of reports for each SOC (SOC “Gastrointestinal disorders”, SOC “Hepatobiliary disorders”, SOC “Investigations, SOC Neoplasms benign, malignant and unspecified (incl cysts and polyps)”, SOC “Infections and infestations”, SOC “Surgical and medical procedures”) were considered in relation to adverse events related to the pancreas and gallbladder, rather than all reported adverse events in each group. MedDRA Version 28.1. Counts are not directly comparable across data sources and reflect different reporting units.

Cholecystitis represented the second most frequently reported gallbladder event, predominantly in its acute form. Acute cholecystitis was observed in 28 participants across seven RCTs (SURMOUNT-1,^48,49 SURMOUNT-2,⁵⁰ SURMOUNT-3,⁵¹ SURMOUNT-4,⁵² SURMOUNT-5,⁷⁶ SURMOUNT-CN⁵⁶ and SUMMIT⁵⁸ with study-specific proportions ranging from 0.3% to 1.6% (6/1,896 [0.3%] in SURMOUNT-1,⁴⁸ at 72 weeks, 1/287 [0.3%] in SURMOUNT-3,⁵¹ 1/374 [0.3%] in SURMOUNT-5,⁷⁶ and 6/364 [1.6%] in SUMMIT)⁵⁸ and in 17/249 FAERS reports [6.8%]. Unspecified cholecystitis was identified in 13 participants across four RCTs: SURMOUNT-1, 7/1,896 [0.4%] at 72 weeks and 3/762 [0.4%] at 176 weeks,^48,49 SURMOUNT-5,⁷⁶ 1/374 [0.3%], SYNERGY-NASH,⁵⁵ 1/142 [0.7%] and SUMMIT⁷³ 1/364 [0.3%]. It was also reported in 41/249 FAERS reports [16.5%] and at an incidence rate of 6.50 events per 1,000 person-years/9,193 in one cohort study.⁶⁰

Chronic cholecystitis was infrequently reported in 13 participants across four RCTs: SURMOUNT-1, 5/1,896 [0.3%] at 72 weeks and 5/762 [0.7%] at 176 weeks,^48,49 SURMOUNT-4,⁵² 1/783 [0.1%] during the lead-in period, SURMOUNT-CN, 1/141 [0.7%]⁵⁶ and SUMMIT, 1/364 [0.3%].⁷³ In the FAERS database, 1/249 reports [0.4%] were classified as cholecystitis chronic.

Gallbladder disease occurred in seven RCTs (SURMOUNT-1,^48,49 SURMOUNT-2,⁵⁰ SURMOUNT-3,⁵¹ SURMOUNT-4,⁵² SURMOUNT-5,⁵³ SURMOUNT-J⁵⁴ and SURMOUNT-CN),⁵⁶ involving 67 participants, with study-specific proportions ranging from 0.7% to 2.9%: 1/150 [0.7%] in SURMOUNT-J⁵⁴ and 2/287 [0.7%] in SURMOUNT-3⁵¹ to 22/762 [2.9%] in SURMOUNT-1 at 176 weeks,⁴⁹ while gallbladder disorder was noted in 11 participants across three RCTs: 1/374 [0.3%] in SURMOUNT-5;⁷⁶ 6/783 [0.8%] during the lead-in period of SURMOUNT-4;⁵² and 4/142 [2.8%] in SYNERGY-NASH.⁵⁵ It was also reported in one retrospective analysis⁶¹ (5 /10,002 [0.05%]) and 30/249 FAERS reports [12.0%].

Biliary colic was documented in two RCTs, with 1/623 [0.2%] in SURMOUNT-2 from baseline through week 76⁷² and 1/142 [0.7%] in SYNERGY-NASH.⁷⁰ It was additionally reported in 28/249 FAERS reports [11.2%].

Neoplastic gallbladder- and biliary-related events were rare, including cholangiocarcinoma, reported in one participant in SUMMIT⁷³ (1/364 [0.3%]) and 2/16 FAERS reports [12.5%]; gallbladder cancer, reported in 1/16 FAERS reports [6.3%]; and isolated RCT reports of gallbladder cancer stage III in SURMOUNT-1⁷¹ and biliary neoplasm in SUMMIT⁷³ (1/1,896 [0.1%] and 1/364 [0.3%], respectively). Case reports^67,68 highlighted rare events such as acalculous cholecystitis, gallstones, and gallbladder wall thickening. All gallbladder-related AEs with study-specific denominators and proportions are summarized and presented in Table 3 and visualized in Supplementary Figure 2A–D.

In summary, gallbladder-related AEs were predominantly represented by cholelithiasis and cholecystitis across all evidence sources, whereas rare biliary, neoplastic, and complex inflammatory conditions were largely confined to post-marketing pharmacovigilance data. This pattern highlights greater heterogeneity and breadth of reporting in the FAERS database compared with clinical trials, while direct comparison of frequencies across data sources remains limited due to differences in reporting methodology.

Severe or Serious Adverse Events

Information on AE severity was limited across the included evidence sources. In the RCT publications, one study (SURMOUNT-1)⁴⁸ specified that 3/1,896 [0.2%] participants with adjudication-confirmed pancreatitis were not reported as severe. Among case reports, edematous pancreatitis was described as mild,⁶⁷ whereas pancreatitis necrotizing was described as severe.⁶⁶ For gallbladder disease and gallbladder disorder, several RCTs reported events as severe or serious, although the reported information did not permit classification of individual events as either severe or serious: SURMOUNT-1 at 72 weeks,⁴⁸ 22/1,896 [1.2%] and SURMOUNT-1, at 176 weeks,⁴⁹ 22/762 [2.9%], SURMOUNT-2,⁵⁰ 6/623 [1.0%], SURMOUNT-3,⁵¹ 2/287 [0.7%] and SURMOUNT-4 lead-in,⁵² 7/783 [0.9%] for gallbladder disease and SURMOUNT-4 lead-in,⁵² 6/783 [0.8%] for gallbladder disorder. One observational study reported moderate-to-severe cholecystitis, cholelithiasis, and pancreatitis.⁶⁰ The remaining observational evidence, EudraVigilance pharmacovigilance analysis, and the FAERS database did not provide information on AE severity.

Seriousness was reported more frequently than severity in RCT publications, with serious AEs occurring in 1/27 participants [3.7%] to 116/1,896 participants [6.1%] and study-specific proportions ranging from 2.0% (16/783) to 26.4% (96/364). Given that study publications generally reported seriousness for overall AEs rather than specifically for pancreatobiliary AEs, these figures are not directly attributable to pancreatic or gallbladder events. Detailed study-level data, including the corresponding denominators, are provided in Supplementary Table 11.

Except for gallbladder disease events described as severe or serious, the publications did not provide study-specific information on the seriousness of pancreatic- or gallbladder-related AEs. In accordance with the review methodology, ClinicalTrials.gov records were used to supplement information on pancreatic- and gallbladder-related AEs, including seriousness, only when these events were not reported in the corresponding publications. These additional data are presented in Table 4.

Table 4.

Pancreatic and Gallbladder Adverse Events Reported as Serious or Reported as Other (Not Including Serious) Adverse Events in RCTs

Serious AEs/Study ID	Other (Not Including Serious) AEs/Study ID
Pancreatic-related adverse events
Obstructive pancreatitis 10079822 SUMMIT (NCT04847557):73 2/364 (0.5%)	Pancreatitis 10033645 SYNERGY-NASH (NCT04166773):70 1/142 (0.7%)
Pancreatitis 10033645 NCT04081337:59 1/27 (3.7%)	Pancreatic cyst 10033615 SURMOUNT-J (NCT04844918):74 2/150 (1.3%)
Pancreatitis acute 10033647 SURMOUNT-1 (NCT04184622), baseline up to week 193:71 3/1,896 (0.2%) SURMOUNT-2 (NCT04657003), Baseline through Week 76:72 2/623 (0.3%) SUMMIT (NCT04847557):73 1/364 (0.3%)	Pancreatic enzymes increased 10061900 SURMOUNT-J (NCT04844918):74 1/150 (0.7%) SYNERGY-NASH (NCT04166773):70 1/142 (0.7%) SURMOUNT-OSA (NCT05412004):75 1/114 (0.9%)
Pancreatic carcinoma metastatic 10033610 SURMOUNT-1 (NCT04184622), baseline up to week 193:71 1/1,896 (0.1%)
Gallbladder-related adverse events
Biliary colic 10004663 SURMOUNT-2 (NCT04657003), Baseline through Week 76:72 1/623 (0.2%)	Biliary colic 10004663 SYNERGY-NASH (NCT04166773):70 1/142 (0.7%)
Cholangitis acute 10008605 SURMOUNT-1 (NCT04184622), baseline up to week 193:71 1/1,896 (0.1%)	Cholangitis 10008604 SYNERGY-NASH (NCT04166773):70 1/142 (0.7%)
Bile duct stone 10004637 SURMOUNT-1 (NCT04184622), baseline up to week 193:71 2/1,896 (0.1%)	Cholelithiasis 10008629 SYNERGY-NASH (NCT04166773):70 1/142 (0.7%) SURMOUNT-OSA (NCT05412004), Trial 2:75 2/119 (1.7%)
Biliary obstruction 10084247 SURMOUNT-1 (NCT04184622), baseline up to week 193:71 1/1,896 (0.1%)	Gallbladder polyp 10049704 SURMOUNT-OSA (NCT05412004), Trial 1:75 1/114 (0.9%)
Cholecystitis 10008612 SUMMIT (NCT04847557):73 1/364 (0.3%) SURMOUNT-5 (NCT05822830):76 1/374 (0.3%)
Cholecystitis acute 10008614 SURMOUNT-5 (NCT05822830):76 1/374 (0.3%)
Cholelithiasis 10008629 SUMMIT (NCT04847557):73 1/364 (0.3%) NCT04081337:59 1/27 (3.7%) SURMOUNT-5 (NCT05822830):76 2/374 (0.5%)
Gallbladder disorder 10017626 SURMOUNT-5 (NCT05822830):76 1/374 (0.3%)
Cholangiocarcinoma 10008593 SUMMIT (NCT04847557):73 1/364 (0.3%)
Gallbladder cancer stage III 10058283 SURMOUNT-1 (NCT04184622), baseline up to week 193:71 1/1,896 (0.1%)
Biliary neoplasm 10004668 SUMMIT (NCT04847557):73 1/364 (0.3%)
Cholecystectomy 10008611 SURMOUNT-1 (NCT04184622), baseline up to week 193:71 1/1,896 (0.1%)

In the FAERS database, serious reports predominated in the “Gastrointestinal Disorders” SOC (287/288, 99.7%), “Hepatobiliary Disorders” SOC (229/249, 92.0%), and “Investigations” SOC (4/7, 57.1%), while all reports in the other SOCs were classified as serious. Non-serious reports occurred in 1/288 [0.3%], 20/249 [8.0%], and 3/7 [42.9%] reports in these three SOCs, respectively. Detailed SOC-specific serious and non-serious outcomes are provided in Supplementary Table 14. The EudraVigilance pharmacovigilance analysis³⁹ similarly showed a predominance of serious reports (4,914/5,774, 85.1%) (Supplementary Table 12), whereas seriousness was not reported in the observational studies. Among case reports, one pancreatitis acute event was classified as serious⁶⁵ (Supplementary Table 13).

Population Characteristics

Supplementary Tables 11–14 provide detailed information across studies, tirzepatide dosages, denominators and comparators. For descriptive purposes, participant characteristics were summarized across all tirzepatide treatment arms to characterize the overall tirzepatide population. Detailed characteristics of the included RCTs, observational studies, and case reports, including study titles, NCT number and acronyms where available, are presented in Table 1.

Age

Across RCTs, mean participant age ranged from the mid-30s to mid-60s (Figure 2), with age eligibility criteria generally set at ≥18 years (9 RCTs),^{48–53,55–57,59} >20 (1 RCT),⁵⁴ ≥40 (1 RCT),⁵⁸ and upper age limits ranging from 78 to 84 years. The lowest mean age was 35.3 years in SURMOUNT-CN,⁵⁶ while the highest was 65.5 years in SUMMIT,⁵⁸ which specifically enrolled older adults (≥40 years). In observational studies, age reporting was limited, with a mean age of approximately 52 years documented in only two studies.^60,64 In the EudraVigilance pharmacovigilance analysis (2,102 ICSRs), most AEs occurred in adults aged 18–64 years.³⁹ Similarly, FAERS data showed the majority of events in this age group across all reaction categories except for neoplasms, which were more common among those aged 65–85 years. Some studies reported age groups, with most participants being younger adults under 50 and under 65 years (Supplementary Figure 3). Mean age and age-group data were not consistently available across studies. While most participants were ≥18 years, only the EudraVigilance pharmacovigilance analysis specified age subgroups, with 2 ICSRs involving individuals aged 12–17 years [0.1%] and 19 ICSRs involving individuals aged >85 years [0.9%].³⁹

Figure 2.

Mean participant age across randomized controlled trials, observational studies, and case reports.

Gender

The sex distribution across the included data sources is shown in (Supplementary Figures 4A–C). Female participants or reports predominated across several data sources. Across the RCTs, female participation ranged from 22 to 1,278 participants, while study-specific proportions ranged from 27.5% (33/120) to 81.5% (22/27). In the EudraVigilance pharmacovigilance analysis, 51.6% (1,085/2,102) of ICSRs involved females, 29.0% (609/2,102) involved males, and 19.4% (408/2,102) had unspecified sex.³⁹ Gender distribution was reported in only two observational studies, both demonstrating female predominance, at 70.5% (6,484/9,193) and 65.5% (72/110), respectively.^60,64 Case reports included three female patients^65–67 alongside a single male patient.⁶⁸ FAERS data corroborated this trend, with females representing 68.1% (196/288) of SOC “Gastrointestinal disorders”, 72.7% (181/249) of SOC “Hepatobiliary disorders”, and 71.4% (5/7) of SOC “Investigations” reaction groups. Conversely, reports within the SOC “Neoplasms Benign, Malignant and Unspecified (Incl Cysts and Polyps)” were more common among males (50.0%, 8/16) than females (31.3%, 5/16), whereas the SOCs “Surgical and Medical Procedures” and “Infections and Infestations” were documented exclusively among females.

Ethnicity and Race

Ethnicity and race were inconsistently reported. Studies showed substantial representation of non-Hispanic participants across RCTs, with study-specific proportions ranging from 37.9% (236/623) to 75.1% (281/374), and the number of non-Hispanic participants from 63/114 [55.3%] to 992/1,896 [52.3%]. In the observational cohort, 80.0% of participants were non-Hispanic (7,355/9,193).⁶⁰ Hispanic/Latino participants comprised 24.9% (93/374) to 59.9% (373/623) across RCTs, with the number of Hispanic/Latino participants spanning from 38/120 [31.7%] to 904/1,896 [47.7%]. In the observational cohort, 15.4% of participants were Hispanic/Latino (1,414/9,193).⁶⁰ White participants consistently formed the majority racial group across RCTs, with study-specific proportions of 51.9% (14/27) to 85.7% (246/287), while the number of White participants varied from 14/27 [51.9%] to 1,342/1,896 [70.8%] across the 9 RCTs.^{48–53,55,57–59} In two observational studies, White participants accounted for 89/110 [80.9%] and 7,097/9,193 [77.2%], respectively.^60,64 Asian and Japanese participants were variably represented, reaching 100% in some studies (141/141 Asian and 150/150 Japanese).^54,56 Black/African American participants represented 0.7% (1/142) to 40.7% (11/27) across RCTs, with participant counts of 1/142 [0.7%] to 146/1,896 [7.7%].^{48–53,55,57,59,73} In two observational studies, these participants formed 11.4% (1,050/9,193) and 17.3% (19/110), respectively.^60,64 Other racial groups were represented in smaller proportions. Race and ethnicity were not reported in case reports or the FAERS database.

Comorbidities

RCTs reported comorbidities most comprehensively compared with observational studies and case reports. The predominant comorbidities were metabolic: prediabetes (3 RCTs,^48,49,53,57 40.2% (762/1,896) − 100% (762/762); participant counts of 69/120 [57.5%] − 762/1,896 [40.2%]); diabetes (3 RCTs,^50,55,58 47.8% (174/364) − 100% (623/623); participant counts of 82/142 [57.7%] − 623/623 [100%], with 1 case report of gestational diabetes,⁶⁷ 1 observational study⁶⁰ (4,773/9,193 [51.9%]), and the EudraVigilance pharmacovigilance analysis³⁹ 802/2,205 ICSRs [36.4%]; dyslipidemia (8 RCTs,^{48–54,56,57} 23% (86/374) − 90.0% (135/150); participant counts of 71/287 [24.7%] − 571/1,896 [30.1%]); and nonalcoholic fatty liver disease (NAFLD)/nonalcoholic steatohepatitis (NASH) (7 RCTs,^{48–53,55,56} 2.9% (11/374) − 42.6% (60/141); participant counts of 9/287 [3.1%] − 134/1,896 [7.1%]). Cardiovascular comorbidities were also frequent, with hypertension reported in 8 RCTs^{48–54,56,57} (23.4% (33/141) − 75.8% (91/120); participant counts of 33/141 [23.4%] − 620/1,896 [32.7%]) and in 1 observational study⁶⁰ (62.4%, 5,741/9,193), while atherosclerotic cardiovascular disease (ASCVD) was documented in 5 RCTs^48–52,56 (1.7% (5/287) − 8.5% (53/623); participant counts of 5/287 [1.7%] − 57/1,896 [3.0%]).

Respiratory disorders included obstructive sleep apnea (8 RCTs,^{48–54,56,57} 5.0% (7/141) − 100% (114/114); participant counts of 7/141 [5.0%] − 138/1,896 [7.3%]) and asthma/chronic obstructive pulmonary disease (COPD) (6 RCTs,^48–53,56 2.8% (4/141) − 11.2% (42/374); participant counts of 4/141 [2.8%] − 189/1,896 [10.0%]), with 1 observational study⁶⁰ reporting asthma in 1,689/9,193 [18.4%] and COPD in 416/9,193 [4.5%]. Psychiatric conditions, including anxiety/depression, were captured in 6 RCTs^48–53,56 (0.7% (1/141) − 30.7% (115/374); participant counts of 1/141 [0.7%] − 314/1,896 [16.6%]) and major depressive disorder in 1 observational study⁶⁰ (2,026/9,193 [22.0%]). Musculoskeletal disorders included osteoarthritis, identified across 7 RCTs,^48–54,56 7.3% (11/150) − 20.9% (70/335), with participant counts of 11/150 [7.3%] − 250/1,896 [13.2%], and gout, reported in 4 RCTs,^48–52 2.1% (6/287) − 6.8% (52/762) with participant counts of 6/287 [2.1%] − 101/1,896 [5.3%], with 1 case report.⁶⁸ Selected hepatic/gastrointestinal, renal/urological, reproductive/gynecological and dermatologic/ophthalmologic disorders were also documented.

Multimorbidity was common, with the majority of participants experiencing at least one obesity-related comorbidity. Participants without comorbidities accounted for 27.3% (102/374) − 36.9% (699/1,896) across 4 RCTs,^{48,49,51–53} with participant counts of 96/287 [33.4%] − 699/1,896 [36.9%]; 1–2 comorbidities, 39.8% (248/623) − 72.7% (109/150) across 7 RCTs,^48–54,56 with participant counts of 75/141 [53.2%] − 873/1,896 [46.0%]; 3–4 comorbidities, 12.5% (36/287) − 46.9% (292/623) across 7 RCTs,^48–54,56 with participant counts of 36/287 [12.5%] − 292/623 [46.9%]; and ≥5 comorbidities, 1.7% (5/287) − 13.3% (83/623) across 6 RCTs,^48–53,56 with participant counts of 5/287 [1.7%] − 83/623 [13.3%]. One study reported multiple obesity‑related complications in 50% (187/374) of participants.⁵³ Detailed study-level comorbidity data are provided in Supplementary Tables 11–14. In contrast, comorbidity information was limited or inconsistently reported in observational studies and the EudraVigilance pharmacovigilance analysis, while the FAERS database did not capture comorbidity data.

Study Characteristics

A total of 11 RCTs (including 1 study - Phase 1: NCT04081337,⁵⁹ 1 study - Phase 2: SYNERGY-NASH,⁵⁵ and 9 studies - Phase 3: SURMOUNT-1;^48,49 SURMOUNT-2;⁵⁰ SURMOUNT-3;⁵¹ SURMOUNT-4;⁵² SURMOUNT-5;⁵³ SURMOUNT-J;⁵⁴ SURMOUNT-CN;⁵⁶ SURMOUNT-OSA⁵⁷ and SUMMIT⁵⁸), 4 case reports,^65–68 1 EudraVigilance pharmacovigilance analysis,³⁹ 4 retrospective cohort studies,^60,62–64 and 1 retrospective analysis⁶¹ were identified (Supplementary Tables 15–16).The design characteristics of the RCTs and observational studies and the key features of the EudraVigilance analysis and FAERS database are summarized in Supplementary Table 15, while the characteristics of the case reports are presented in Supplementary Table 16.

Comparator

RCTs compared tirzepatide with placebo (10 studies: SURMOUNT-1,^48,49 SURMOUNT-2,⁵⁰ SURMOUNT-3,⁵¹ SURMOUNT-4,⁵² SURMOUNT-J,⁵⁴ SURMOUNT-CN,⁵⁶ SURMOUNT-OSA,⁵⁷ SYNERGY-NASH,⁵⁵ SUMMIT⁵⁸ and NCT04081337⁵⁹), or Semaglutide (1 study: SURMOUNT-5),⁵³ while observational studies used Semaglutide (2 studies),^60,63 Dulaglutide/Liraglutide/Lixisenatide/Exenatide/Albiglutide (1 study),⁶² Semaglutide/Ozempic/Wegovy (1 study),⁶¹ and Dulaglutide/Exenatide/Liraglutide/Semaglutide (2 studies).^39,64

Surveillance

Active surveillance was observed in 15 studies, primarily through trial monitoring (11 RCTs: SURMOUNT-1,^48,49 SURMOUNT-2,⁵⁰ SURMOUNT-3,⁵¹ SURMOUNT-4,⁵² SURMOUNT-5,⁵³ SURMOUNT-J,⁵⁴ SYNERGY-NASH,⁵⁵ SURMOUNT-CN,⁵⁶ SURMOUNT-OSA,⁵⁷ SUMMIT⁵⁸ and NCT04081337⁵⁹), and registry/EHR data (4 studies),^60,62–64 while passive surveillance was identified across 8 sources, including 4 case reports,^65–68 EudraVigilance pharmacovigilance analysis,³⁹ the FAERS database⁶⁹ and social media posts (1 study).⁶¹

Sample Size

Across the included RCTs, sample sizes ranged from 55 to 2,539 participants, with tirzepatide groups comprising between 27 and 1,896 individuals. Observational studies enrolled between 110 and 2,666,800 participants. In addition, the EudraVigilance pharmacovigilance analysis identified 2,102 ICSRs associated with tirzepatide, reporting a total of 5,774 AEs (Supplementary Figure 5).

Study Duration and Reporting Period

Treatment periods in RCTs ranged from 18–72 weeks, with follow-up lasting 4–104 weeks and reporting extending 22–193 weeks. Observational studies reported treatment or follow-up ranging from 6 months to several years, while 2 retrospective analyses covered reporting periods ranging from 29 months to 10 years,^61,64 though the longest duration encompassed multiple agents beyond tirzepatide. FAERS data covered the period from 2022 to September 28, 2025, while the EudraVigilance pharmacovigilance analysis reported outcomes over a 22-month timeframe.

Population Types and Settings

Across studies, populations ranged from obesity without T2D (8 RCTs: SURMOUNT-1,⁴⁸ SURMOUNT-3,⁵¹ SURMOUNT-4,⁵² SURMOUNT-5,⁵³ SURMOUNT-J,⁵⁴ SURMOUNT-CN,⁵⁶ SURMOUNT-OSA⁵⁷ and NCT04081337⁵⁹) to mixed cohorts with obesity and T2D (3 RCTs: SURMOUNT-2,⁵⁰ SYNERGY-NASH⁵⁵ and SUMMIT⁵⁸), one enrolling exclusively all participants with diabetes SURMOUNT-2;⁵⁰ 1 case report with gestational diabetes;⁶⁷ 1 observational study;⁶⁰ EudraVigilance pharmacovigilance analysis),³⁹ or prediabetes (3 RCTs: SURMOUNT-1,^48,49 SURMOUNT-5⁵³ and SURMOUNT-OSA,⁵⁷ including 1 RCT: SURMOUNT-1,⁴⁹ that exclusively enrolled participants with prediabetes).

Some studies specifically recruited participants with obstructive sleep apnea,⁵⁷ metabolic dysfunction-associated steatohepatitis (MASH) with fibrosis,⁵⁵ chronic heart failure,⁵⁸ or inflammatory bowel disease (IBD).⁶⁴ In addition, three RCTs included participants with at least one weight-related complication,^48,51,52 while one RCT required two or more complications as inclusion criteria.⁵⁴

Study settings were inconsistently documented in observational studies and case reports, with the US predominating across study designs. Most RCTs were multinational, with substantial representation from the US (9 RCTs)^{48–53,55,57,59,73} and Latin America (7 RCTs).^{48–52,55,57,58} US participation spanned 22.8% (83/364) to 100% (374/374), while the number of US participants ranged from 27/27 [100%] to 853/1,896 [45%]. Latin American participation varied from 4.9% (7/142) to 53.0% (193/364), with participant counts of 7/142 [4.9%] to 776/1,896 [40.9%]. Specifically, one RCT was conducted in Japan,⁵⁴ one in China,⁵⁶ and two in the US.^53,59 Representation was lower in Asia (8 RCTs),^{48–50,52,54–57,73} with participation varying from 5.7% (38/670) to 100% (150/150 and 141/141), while the number of Asian participants ranged from 15/120 [12.5%] to 181/1,896 [9.5%]; European participants were included in 5 RCTs,^{48–50,55,57,73} with participation ranging from 3.6% (13/364) to 26.8% (38/142) and participant counts of 12/114 [10.5%] to 86/1,896 [4.5%]. Other regions contributed a small number of participants, with 1/114 [0.9%] and 10/120 [8.3%] reported in SURMOUNT-OSA⁵⁷ Trial 1 and Trial 2 respectively. Real-world pharmacovigilance data (FAERS and the EudraVigilance pharmacovigilance analysis) captured a broad geographic distribution, with reports predominantly originating from the US and non-EEA (European Economic Area) regions (Supplementary Figure 6).

Collectively, the included evidence reflects heterogeneous populations across metabolic status, comorbidity burden, and geographic settings, which may influence the generalizability of reported safety outcomes.

Study Aims

The primary endpoints in most RCTs focused on efficacy rather than safety. In contrast, several observational studies specifically addressed gastrointestinal safety,⁶³ rare pancreatic pseudocyst formation,⁶² AE trends in social media,⁶¹ and GLP-1 RA safety profiles in pharmacovigilance database.³⁹ The majority of trials were short-term, with only SURMOUNT-1 (Jastreboff et al, 2025) evaluating safety over a three-year period.⁴⁹

Dropouts Due to Adverse Events and Death

AE-related dropout and death reporting lacked consistency. Across the RCTs,^48–59 0.3% (1/364) to 10.5% (30/287) of participants discontinued treatment due to AEs, with participant counts of 1/364 [0.3%] to 111/1,896 [5.9%], in the observational studies, AE-related dropout was reported in 14/110 [12.7%] and 5,140/9,193 [55.9%] participants, respectively.^60,64 These discontinuations referred to AEs overall in the tirzepatide arms and were not specifically associated with pancreatic or gallbladder events. Deaths occurred in 0.1% (1/783) to 5.2% (19/364) of participants across the reported RCTs,^48–52,58 whereas observational studies provided no mortality data. These figures represent all reported deaths in the tirzepatide arms and should not be interpreted as treatment-emergent mortality attributable to pancreatic or gallbladder AEs. The EudraVigilance pharmacovigilance analysis³⁹ reported 75/5,774 fatal AEs [1.3%], while an isolated fatal event was described in one case report.⁶⁶ In the FAERS database, reports with a death outcome were identified in SOCs “Gastrointestinal disorders” − 4.9% (14/288), “Hepatobiliary disorders” − 1.6% (4/249), “Neoplasms Benign, Malignant and Unspecified (Incl Cysts and Polyps)” − 18.8% (3/16), and “Infections and Infestations” − 7.1% (1/14).

Evidence Gaps Across Data Sources

Across the included evidence, several important gaps and inconsistencies were identified across the domains examined in this scoping review. Table 5 summarizes the major evidence gaps across data sources.

Table 5.

Evidence Gaps

Evidence Gaps	Description
Population representation:	Underrepresented: Adolescents and older adults; Males; Black or African American, American Indian or Alaska native, Native Hawaiian or Other Pacific Islander;
Reporting practices:	Inconsistently reported demographic characteristics: RCTs: Age groups; Observational studies: Age mean, age groups, gender, race, ethnicity, geographical region; Case reports: Race, ethnicity and geographical region.
	Variation in AE reporting across data sources: EudraVigilance: Provides broader aggregated safety categories (HLT terms); FAERS and clinical studies: Offer more precise diagnostic detail (PT terms); Case reports: Capture highly specific observations (LLT terms).
Attrition details:	Inconsistently documented: Dropouts due to AEs; Deaths.
Severity classification:	Limited and inconsistently reported in RCT publications, observational studies and case reports; Not reported in ClinicalTrials.gov records, the EudraVigilance pharmacovigilance analysis, or the FAERS database; Difficulty distinguishing reported severity from seriousness in some RCT publications.
Seriousness classification:	• Limited event-specific reporting for pancreatobiliary AEs; • Not reported across observational studies.
Comorbidity data:	Not reported in: Spontaneous reporting systems; Some observational studies.
Geographic settings:	Unevenly distributed, with dominance of US-based studies; Limited representation from other regions; Underrepresentation of Africa, the Middle East, and Latin America (in FAERS); Inconsistent documentation in observational studies and case reports.
Real-world evidence:	Limited obesity-only cohorts.
Safety studies:	Lack of long-term safety studies.
Comparative evidence:	Paucity of head-to-head studies.

Discussion

This scoping review explored tirzepatide’s safety in obesity management, focusing on pancreatic and gallbladder AEs by mapping evidence from clinical trials, real-world studies, and pharmacovigilance data. Our findings demonstrate that pancreatic and gallbladder complications were reported across diverse data sources, with recurrent patterns observed across multiple sources, supporting continued surveillance of these AEs.

Pancreatitis accounted for most pancreatic AEs, occurring consistently but at low frequencies in RCTs, with acute forms predominating. Real-world data demonstrated broader reporting, with acute pancreatitis remaining the most frequent event. Literature indicates tirzepatide’s safety profile is broadly comparable to the GLP-1 RA class,^16,79,80 in which pancreatitis has been reported as an uncommon AE^15,31,81 or with low incidence not statistically significant.^32,34,82 However, recent systematic review findings indicate a modestly increased risk of pancreatitis,⁸³ while pharmacovigilance data suggest risk may rise with cumulative GLP-1 RA exposure,⁸⁴ supported by retrospective evidence of higher incidence within the first six months of therapy.⁸⁵ FAERS case series documented events across all GLP-1 RAs, but tirzepatide showed the lowest association.⁴⁰ Conversely, TriNetX analyses of comorbidity-free US patients found no evidence of elevated risk.⁸⁶ Reports of necrotizing pancreatitis and pseudocyst formation highlight progression to severe, complicated disease, underscoring the seriousness of these AEs.^87–90 Notably, both were not reported in RCTs, suggesting that these severe complications occur only rarely and primarily in real-world settings. Documented enzyme elevations, while often considered a markers of acute pancreatitis,⁹¹ are insufficient for diagnosis^92,93 and may reflect alternative etiologies such as IBD⁹⁴ or gallstone obstruction mimicking pancreatitis.⁹⁵ Evidence indicates that acute pancreatitis can occur with normal amylase levels,⁹⁶ while clinically evident pancreatitis remains rare despite occasional enzyme elevations.^36,97 Importantly, one RCT found that acute GLP-1 infusion did not increase serum lipase levels, although a small increase was observed following an ad libitum meal. Pancreatic lipase activity was strongly correlated with plasma lipase protein levels, suggesting that the observed lipase activity reflected circulating lipase protein.⁹⁸ These findings suggest that modest postprandial increases in serum lipase may reflect physiological pancreatic enzyme release rather than pancreatic injury. A 12-week RCT found no increases in lipase or amylase and no cases of pancreatitis,³⁰ whereas a large long-term randomized trial reported increases in serum lipase and amylase of 28% and 7%, respectively, but similar rates of confirmed acute pancreatitis with active treatment and placebo (0.4% vs 0.5%).⁹⁷ Overall, pancreatic enzyme elevations can occur during GLP-1 RA treatment without an increased incidence of clinically adjudicated pancreatitis. Such observations emphasize the need for cautious attribution of enzyme abnormalities in obese patients with comorbidities.

Compared with gallbladder and biliary events, the mechanistic basis of pancreatic AEs remains less well established. Experimental studies have demonstrated GLP-1 receptor signaling in pancreatic acinar cells and effects on pancreatic enzyme release. However, GLP-1 did not acutely increase amylase or lipase levels, and the observed increase in constitutive enzyme release was associated with mild c-Src-dependent acinar cell proliferation. Significantly, elevation of pancreatic enzyme levels during GLP-1 therapy was not considered an indicator of subclinical pancreatitis.²⁷ Other experimental study showed GLP-1 can regulate amylase secretion through cyclic adenosine monophosphate (cAMP)-dependent signaling, providing a potential biological basis for pancreatic effects of GLP-1 receptor activation.²⁸ However, these physiological effects do not establish a direct mechanism for pancreatitis. Consistent with this, a RCT found no changes in pancreatic exocrine secretion, maximum secretion speed, time to maximum secretion, or pancreatic duct diameter with GLP-1 RA treatment, suggesting that these changes are unlikely to contribute to pancreatic adverse effects.^29,30 Thus, available clinical and experimental evidence provides biological plausibility for pancreatic effects of GLP-1 receptor activation but does not establish a clear pathway leading to pancreatic injury. Additional mechanistic evidence indicates that chronic GLP-1 RA exposure may promote proliferation or chronic activation of pancreatic stellate cells, which may affect pancreatic function,⁹⁹ while pharmacological GLP-1 receptor stimulation has been associated with enhanced protein synthesis and pancreatic enlargement.¹⁰⁰ Yet, the clinical relevance of these findings remains uncertain.

Among gallbladder events, cholelithiasis emerged as the most common AE, reported across all RCTs, one observational study, and slightly over half of FAERS reports. This pattern is consistent with meta-analyses linking tirzepatide to increased risk of gallbladder and biliary disease, particularly cholelithiasis.³³ Cholecystitis, both acute and chronic, was observed in trials and real-world settings. Some studies reported low rates of cholelithiasis/cholecystitis across doses, concluding tirzepatide’s profile resembles GLP-1RAs.¹⁵ Tirzepatide trials showed no significant difference in cholecystitis,²⁰ cholelithiasis, or biliary risk vs comparators,³² however composite outcome of gallbladder or biliary disease was significantly elevated compared with placebo.³⁴ Studies of GLP-1RAs as a class showed increased risks of gallbladder and biliary disease, particularly cholelithiasis and cholecystitis, at higher doses and longer treatment durations and for weight-loss use.³⁵ In contrast, Chiang et al found GLP-1RAs linked to cholelithiasis but not other biliary AEs,¹⁰¹ with nonsignificant subgroup trends in obesity, steatohepatitis, weight-loss trials, and high-dose use. Xie et al reported higher cholelithiasis incidence with GLP-1RAs than tirzepatide,⁸² whereas Ismaiel et al and Verma et al found no significant increase in cholelithiasis or cholecystitis risk for GLP-1RAs as a class.^31,102 Real-world data provide additional context, showing that GLP-1RA therapy was linked to gallbladder and biliary disease, though not statistically significant,¹⁰³ whereas Caruso et al observed no signal apart from biliary colic with tirzepatide.²²

FAERS reports of pancreas infection, infective cholecystitis, and biliary sepsis, though uncommon, highlight the value of post-marketing surveillance in capturing rare but clinically significant events not seen in trials. The scarcity of cholecystectomy across RCTs and the FAERS database suggests gallbladder complications requiring surgery remain rare, while chronic conditions were more frequent in real-world data, reinforcing the complementary role of pharmacovigilance.

These findings also warrant consideration in the context of the biological and physiological effects of incretin-based therapies on the gallbladder and biliary system. For gallbladder events, activation of the GLP-1 receptor suppresses postprandial CCK secretion from intestinal enteroendocrine cells, thereby reducing gallbladder contractility and delaying bile emptying.¹⁰⁴ However, evidence from longer-term studies regarding the associated clinical risk remains inconsistent.¹⁰⁵ This mechanism is supported by randomized controlled studies conducted in healthy individuals, which demonstrated reduced CCK-induced gallbladder emptying with GLP-1 RAs compared with placebo.^106,107 Consistent with this mechanistic evidence, a systematic review and meta-analysis of 76 randomized clinical trials found an increased risk of gallbladder and biliary diseases with GLP-1 RAs, particularly in trials evaluating weight loss.³⁵ He et al specifically discussed impaired gallbladder motility and delayed gallbladder emptying, potentially related to suppression of CCK secretion, as a plausible mechanism underlying this increased risk. Importantly, this mechanistic rationale may also be relevant to tirzepatide because it is a dual GIP/GLP-1 RA, and both GIP and GLP-1 signaling have been implicated in gallbladder motility and biliary function. Zeng et al described a potential role of GLP-1-mediated effects on CCK in reducing gallbladder contractility and suggested that GIP may also contribute to gallbladder relaxation, potentially promoting gallbladder or biliary complications.³⁴ In contrast, Gasbjerg et al investigated the physiological roles of endogenous GIP and GLP-1 using selective receptor antagonism in a RCT in healthy men and found that GIP, but not GLP-1, influenced gallbladder motility.¹⁰⁸ This finding supports a physiological role for GIP in gallbladder function. However, the extent to which these effects contribute specifically to tirzepatide-associated outcomes remains uncertain. Nevertheless, the observed findings are biologically plausible in the context of incretin-based therapy, with multiple pathways potentially contributing. Several studies have explored the potential mechanisms underlying the association between GLP-1 RAs and gallbladder or biliary disease, providing additional mechanistic insights. Proposed pathways include suppression of CCK secretion, disruption of farnesoid X receptor (FXR) and Takeda G protein-coupled bile acid receptor 1 (TGR5) signaling, alterations in bile acid composition, and prolonged gallbladder refilling, which may impair gallbladder motility and promote gallstone formation.²⁶ Prolonged refilling, in particular, has been highlighted in prior literature as a plausible explanation for gallbladder events observed with GLP-1 receptor agonists.²⁵ Taken together, these proposed mechanisms provide a potential explanation for the observed gallbladder and biliary events, supporting their biological plausibility in the context of incretin-based therapy. Rapid and substantial weight loss may represent an additional pathway contributing to gallbladder-related events during tirzepatide treatment. Rapid weight reduction is an established risk factor for gallstone formation and may promote biliary stasis, thereby contributing to cholelithiasis and subsequent gallbladder complications.¹⁰⁹ However, the relative contribution of weight loss vs direct incretin-mediated effects cannot be established from the available evidence. The observed pancreatobiliary events may therefore reflect a combination of treatment-related physiological effects, weight-loss-associated changes, and underlying patient risk factors rather than a single direct drug effect.

Neoplastic events were also observed in both trial and real-world data. Published studies show no direct increase in gallbladder or biliary cancers with GLP-1 RAs,^35,103 and tirzepatide appears to demonstrate a more favorable neoplasm safety profile compared with GLP-1 mono-agonists.⁸² Regarding pancreatic cancer, systematic reviews and meta-analyses show no overall elevated risk, although slight associations have been noted in the context of background medications.⁸³ Observational findings remain mixed, with some studies reporting no differences,⁸⁵ and others suggesting a modest risk compared with comparators.³⁷ In addition, a large cohort study found a reduced 5-year incidence of pancreatic cancer among participants with chronic pancreatitis treated with GLP-1RAs.¹¹⁰ In the present review, neoplastic events were reported only rarely across clinical trials and pharmacovigilance data sources, and the available evidence does not support a causal association between tirzepatide and pancreatic or biliary malignancies. Nevertheless, the identification of isolated reports of pancreatic cysts, pancreatic neoplasms, and biliary neoplasms highlights the importance of continued monitoring as longer-term safety data emerge.

Although current evidence is generally reassuring regarding the short-term safety of tirzepatide, continued consideration of patient-specific factors and long-term surveillance remain important. While recent systematic reviews and meta-analyses suggest that tirzepatide use does not increase overall or site-specific cancer risk during relatively short follow-up,⁴³ the detection of gallbladder and biliary events in our study underscores the relevance of established risk factors such as gallstone disease, chronic inflammation, gallbladder polyps, obesity, female sex, cholecystitis, and primary sclerosing cholangitis, that may contribute to gallbladder and biliary tract cancers.^111–114 Moreover, the presence of pancreatic cysts confers a higher risk of pancreatic cancer,¹¹⁵ and our findings of scarce gallbladder polyps and pancreatic cysts, female predominance, and chronic comorbid conditions merit attention given their potential association with gallbladder or pancreatic malignancies. Given the chronic nature of obesity treatment and the limited duration of many available studies, ongoing pharmacovigilance and prospective observational research are warranted to further characterize long-term neoplastic outcomes in diverse patient populations.

Variation in AEs patterns between RCTs and pharmacovigilance sources likely reflect fundamental methodological differences rather than representing a true discrepancy in the occurrence of AEs. RCTs involve selected populations, strict inclusion/exclusion criteria, limited follow-up, and structured AEs reporting, whereas spontaneous reporting systems capture real-world patients with multimorbidity, polypharmacy, and longer cumulative exposure, making rare or delayed events more detectable. Nonetheless, spontaneous reporting systems carry inherent limitations, including under- or overreporting, notoriety bias, lack of denominators, and confounding by indication. Previous analyses of the FAERS database have suggested that most reporting is not substantially affected by FDA safety alerts, although stimulated reporting may occur for some drugs, while another analysis found no overall evidence of a significant impact of FDA safety alerts on reporting or disproportionality estimates.^116,117 Inconsistencies in MedDRA terminology (HLT, PT, LLT) further complicate comparisons across studies, while selective reporting of dropouts or mortality in observational data may obscure the overall safety profile and limit the identification of meaningful safety patterns. Reported deaths in pharmacovigilance databases should be interpreted as temporal associations rather than proof of causality.

Moreover, as most gallstones are asymptomatic and only ~8% develop symptoms within 24 months,¹¹⁸ trial incidence may underestimate the broader burden evident in pharmacovigilance data, potentially contributing to differences in AE patterns between RCTs and pharmacovigilance sources.

Pancreatic and gallbladder AEs occurred across heterogeneous groups with multiple comorbidities, reflecting real-world diversity but complicating attribution specifically to tirzepatide. Notably, NAFLD, present in many RCT participants, may contribute to an increased risk of acute pancreatitis¹¹⁹ and is potentially associated with gallstone disease and asymptomatic gallstones, particularly among women and younger individuals.^120,121 Furthermore, three included RCTs^50,55,58 reported participants with T2D, which may have influenced the observed occurrence of pancreatobiliary AEs. T2D itself has been associated with an increased risk of pancreatitis and biliary disease, with one population-based study reporting a 2.83-fold higher risk of pancreatitis (95% CI, 2.61–3.06) and a 1.91-fold higher risk of biliary disease (95% CI, 1.84–1.99) among individuals with T2D compared with those without diabetes.¹²² Therefore, the presence of diabetes in some trial populations represents a potential confounding factor when interpreting pancreatobiliary events associated with tirzepatide. Bariatric surgery is another possible risk factor for gallstones, largely due to rapid weight loss.^118,123,124 Cholecystectomy, although generally preferred by experts, has been reported to carry higher complication rates in some studies.¹²⁵ In contrast to RCTs and observational cohorts, the FAERS database does not capture comorbidities, thereby limiting the contextual interpretation of reported safety findings.

The overrepresentation of females, White participants, adults aged 18–64, and those from the US and Latin America in RCTs highlights the need for more inclusive recruitment to strengthen tirzepatide safety evidence. In real-world settings, female predominance may reflect epidemiology, healthcare-seeking behavior, and reporting practices, as women are more likely to seek care and report AEs.^126,127 Women also pursue obesity treatments more often, including bariatric surgery,¹²⁸ and severe obesity is projected to become the most common among women.¹²⁹ Furthermore, women are at higher risk of gallstone disease than men, particularly under 45,¹³⁰ with premenopausal status further increasing susceptibility.¹³¹ The concentration of reports among adults aged 18–64 likely reflects the limitations of FDA approvals by age. However, recent prescribing GLP-1RA trends show a rising tendency among adolescents, young adults and females.¹³² Higher utilization, affordability, and greater exposure in the US and Europe may partly reflect the early regulatory approval of tirzepatide in these regions, which facilitated the usage.

Inconsistent demographic reporting across multiple sources, particularly in observational studies and real-world settings, impedes pattern detection. Although safety may differ among racial and ethnic groups,¹³³ some studies report no significant racial impact on AEs occurrence.⁸² Severe obesity is projected to be most prevalent among non-Hispanic Black adults in the US,¹²⁹ Asians demonstrate greater weight loss and gastrointestinal AEs with tirzepatide,¹³⁴ and gallstone disease is more frequent in Caucasian, elderly, and female patients.^120,121 Risk profiles also vary by age and sex, with incidental gallbladder cancer more often diagnosed in older women¹³⁵ and hypertriglyceridemia-induced acute pancreatitis typically presenting in individuals in their early 40s,¹³⁶ illustrating the need for closer monitoring in these groups. Collectively, these findings point to the value of standardized demographic reporting and vigilant AEs surveillance.

Wide variation in study designs, populations, and follow-up periods complicates collation and trend identification. Most RCTs were placebo-controlled and short-term, focused on efficacy, with less emphasis on safety endpoints, emphasizing the importance for more head-to-head and long-term studies to contextualize safety findings. By contrast, observational studies with varied comparators yielded broader insights, with some studies focusing primarily on safety outcomes.

Obesity complicates attribution of AEs, as it is a recognized risk factor for gallstone formation, biliary complications, and acute pancreatitis.^{125,131,137–142} Untreated obesity itself is associated with increased risk of pancreatitis and gallbladder disease, and these overlapping pathways reinforce its role as a key confounder in interpreting pancreatic and gallbladder AEs. Close monitoring of biliary and pancreatic symptoms, particularly in patients with underlying conditions that may influence pancreatobiliary outcomes, may help inform individualized clinical decision-making regarding the long-term use of tirzepatide.⁸⁵ Interpretation of adverse-event patterns should consider broader contextual factors. Variations in prescribing practices, healthcare access, pharmacovigilance reporting systems, regulatory approval timelines, and healthcare-seeking behavior may influence the distribution of reported events across regions and populations. Consequently, observed geographic heterogeneity may reflect reporting practices and medication exposure patterns as much as underlying biological differences between populations.

Strengths

This scoping review ensured methodological rigor and transparency through established frameworks and OSF registration. Comprehensive searches across multiple databases, pharmacovigilance sources, and gray literature minimized publication bias, while PCC criteria, systematic screening, and MedDRA coding ensured consistency. Mapping evidence from RCTs, observational studies, and real-world data provided a comprehensive overview of the available evidence while acknowledging heterogeneity across study designs and populations.

By focusing specifically on individuals with obesity and excluding studies conducted exclusively in diabetes populations, the review provided clinically relevant insights into tirzepatide safety in non-diabetic patients. It further strengthens its contribution by exploring underrepresented groups, identifying dominant approaches and overlooked areas in study design, and examining AEs occurrence across both clinical trial and real-world settings. This comprehensive scope exposes key knowledge gaps and helps researchers and clinicians better understand pancreatic and gallbladder safety findings in the context of obesity management.

Limitations

This scoping review has several limitations that should be acknowledged. The restriction to English-language publications and free full-text articles may have excluded relevant evidence from non-English sources or subscription-based journals and may have underrepresented evidence from low- and middle-income countries. In addition, the exclusion of diabetes-only studies strengthened the obesity focus but may have constrained generalizability, since tirzepatide had been extensively evaluated in diabetic populations where pancreatic and gallbladder AEs had also been observed. Although the search strategy was comprehensive, indexing constraints and variability in AE reporting across study designs may have resulted in missed studies, particularly within gray literature and case reports beyond the targeted journals. Furthermore, the quality and completeness of gray literature and post-marketing data often differed, which may have impacted the robustness of conclusions. The scoping review methodology also did not include formal critical appraisal of study quality, which restricted the ability to assess the reliability of the included evidence.

Reliance on secondary data sources introduced further limitations, as underreporting and missing demographic details and comorbidities were common in pharmacovigilance databases and some observational studies. These gaps may have affected the completeness of the summaries and the identification of trends in AE reporting across patient subgroups.

In addition, overlapping publications of the same trials with different reporting periods,^48,49 reports of distinct phases within a single study (eg, SURMOUNT-4 lead-in and SURMOUNT-4 double-blind),⁵² and two separate RCTs conducted under a single master protocol (eg, SURMOUNT-OSA Trial 1 and Trial 2),⁵⁷ together with potential duplication of AEs in the FAERS database, may have led to double counting. Moreover, considerable heterogeneity in study designs, study populations, obesity-related comorbidities, dosing regimens, and follow-up duration complicated the summarization of AEs, while inconsistent demographic reporting hindered identification of dominant patterns. Finally, dependence on FAERS as the sole database searched directly, with the EudraVigilance pharmacovigilance analysis included only via published reports and limited case reports constrained the scope of pharmacovigilance evidence. Pooled dose data without comparators might have obscured dose–response relationships and constrained contextualization against active treatment. An additional limitation relates to the focus on studies reporting pancreatic or gallbladder AEs. Although this approach was necessary to address the review objectives, it may have preferentially captured evidence sources documenting relevant AEs while excluding studies in which such outcomes were absent or not reported. Consequently, the frequency and diversity of reported AEs within the included evidence base should not be interpreted as reflecting their true incidence or prevalence among all populations exposed to tirzepatide.

Recommendations

Future research should prioritize prospective, long-term safety studies in diverse populations to strengthen the evidence base for tirzepatide. Subsequent studies ought to examine age- and gender-stratified cohorts to more precisely identify AE patterns across demographic subgroups. Improvements in pharmacovigilance reporting are needed, with greater inclusion of demographic details to enable more meaningful subgroup analyses. Additional observational studies focusing specifically on pancreatic and gallbladder AEs are warranted, as these outcomes are of particular clinical relevance.

Attention should also be given to AEs that are frequently reported in real-world settings but not captured in RCTs, to ensure a more comprehensive understanding of pancreatic and gallbladder safety. Finally, the establishment and expansion of real-world registry studies are encouraged, as they can provide valuable longitudinal data and complement findings from clinical trials and pharmacovigilance sources.

Clinical Implications

Baseline predisposition to gallstone disease should be assessed before initiating tirzepatide therapy, with early monitoring of abdominal symptoms. Special attention is required for patients with gallbladder disease, pancreatitis, cardiovascular disease,^143,144 NAFLD¹¹⁹ or hypertriglyceridemia.^136,137 In addition, vigilance for pancreatic manifestations, including persistent abdominal pain or elevated lipase/amylase, is important, particularly in individuals with a prior history of pancreatitis. Strategies to minimize excessive or overly rapid weight reduction should be prioritized to mitigate the established risk for gallstone formation and biliary events.^{125,139,140,145,146} Of particular relevance, clinicians should acknowledge that evidence gaps remain for adolescents and older adults, males vs females, and underrepresented ethnic groups, thereby underscoring the importance of considering these limitations when prescribing tirzepatide and weighing patient-specific risk factors against the potential benefits of treatment.

Conclusion

Pancreatic and gallbladder complications, most commonly pancreatitis, gallbladder disease, cholelithiasis, and cholecystitis, were observed in RCTs, observational studies, case reports, and pharmacovigilance databases. Our findings highlighted the diverse range of pancreatobiliary events, spanning acute to chronic inflammatory processes. Notably, RCTs generally reported low frequencies of pancreatic and gallbladder AEs, whereas observational studies and pharmacovigilance data captured a broader and more heterogeneous spectrum of reported events. Despite these disparities, recurrent findings across multiple sources emphasized pancreatic and gallbladder disorders as important safety concerns that warrant careful interpretation. The biological plausibility of pancreatobiliary events in the context of incretin-based therapy provides additional context for the observed findings, although the available evidence does not establish causality. Differences in study designs, patient characteristics, adverse-event definitions and reporting practices, together with the scarcity of long-term data, constrained direct comparison and hindered the identification of clear patterns across evidence sources. The lack of long-term follow-up further limited understanding of tirzepatide’s safety profile, particularly given the chronic nature of obesity and the potential influence of overlapping comorbidities on AEs attribution. The complex interplay between obesity, comorbidities, treatment interventions, therapy duration, and demographic factors underscores the need for careful clinical consideration when prescribing tirzepatide, particularly in patients with underlying pancreatic or gallbladder vulnerability.

This scoping review provides an obesity-specific mapping of the available evidence and highlights important gaps in the current safety literature. Future studies should prioritize longer-term safety follow-up and more consistent reporting of AEs, including standardized AE terminology and coding, treatment exposure and duration, demographic characteristics, relevant comorbidities, AE severity, and attrition due to AEs. Inclusive representation of older adults, adolescents, males, individuals from underrepresented racial and ethnic groups, and those with multiple chronic conditions, across diverse geographic regions, is needed.

Furthermore, evidence from obesity-specific cohorts and comparative studies would strengthen the current evidence base. In parallel, continued post-marketing surveillance is warranted to further characterize the pancreatobiliary safety profile of tirzepatide in diverse populations.

Funding Statement

No funding was received.

Artificial Intelligence (AI)

No AI-related technologies were used in the preparation of this article.

Data Sharing Statement

FAERS data analyzed in this study were extracted by the authors from the publicly available FAERS database on September 28, 2025. Details of the FAERS search and data extraction procedures are provided in the Supplementary Materials and the associated OSF record. The review protocol and its successive versions are available at https://osf.io/zw4ug (DOI: 10.17605/OSF.IO/PDZM6).

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work. Tetiana Kutsyba: Conceptualization, Methodology, Investigation, Data curation, Formal analysis, Project administration, Visualization, Writing – original draft. Sonika Raj: Conceptualization, Methodology, Investigation, Data curation, Validation, Supervision, Project administration, Writing – Review & Editing. Malisha van der Berg: Investigation, Data curation, Writing – Review & Editing.

Disclosure

The authors declare no competing interests. Tetiana Kutsyba is employed by Macleods Pharmaceuticals Ltd. The company had no role in the design, conduct, analysis, interpretation, writing, or publication of this study. The author reports no other conflicts of interest in relation to this work.