Semaglutide · research summary

Pancreatic and biliary safety signals of GLP-1 receptor agonists and tirzepatide: a disproportionality analysis of FAERS reports.

Frontiers in endocrinology · 2026. Summary drawn from the study's full text. Each statement is followed by the exact sentence it comes from. This is a research summary, not medical advice.

Study type

Retrospective disproportionality analysis of individual case safety reports
“Retrospective disproportionality analysis of individual case safety reports.”

Subjects

Individual case safety reports from the US Food and Drug Administration Adverse Event Reporting System (FAERS)
“FAERS data from Q3–2009 through Q3–2025 were extracted, and deduplicated ICSRs were used as the unit of analysis.”

Sample size

18,281,897 unique individual case safety reports after deduplication
“A total of 21,645,384 raw records were retrieved, of which 3,363,487 duplicate records were removed using a global cross-quarter deduplication scheme (the highest PRIMARYID was retained for each CASEID; for the AERS format [2009–2012], the highest ISR per CASE was retained), yielding 18,281,897 unique ICSRs.”

Duration

Third quarter of 2009 through the third quarter of 2025
“Individual case safety reports (ICSRs) are derived from quarterly files of the US Food and Drug Administration Adverse Event Reporting System (FAERS) and cover the third quarter of 2009 through the third quarter of 2025.”

Reported findings

Disproportionality signals for pancreatic adverse events were observed for all four study drugs and control drugs
“A disproportionality signal for pancreatic adverse events was observed for all four study drugs, but the control drugs metformin and SGLT2i also showed positive signals.”
Disproportionality signals for biliary events were observed for all four study drugs but not for control drugs
“For biliary events, all four study drugs showed signals, whereas the two control drugs showed no signals.”
Pancreatic signals remained for diabetes-indication brands but were weaker for obesity-indication brands
“After stratification by brand, the pancreatic signal remained for brands with diabetes indications, while the signal was weaker for brands with obesity indications.”

Limitations stated by the authors

FAERS is subject to underreporting and selective reporting, and estimates cannot determine incidence or causality
“As a spontaneous reporting system, FAERS is inevitably subject to underreporting, selective reporting, and provocative reporting induced by media attention and regulatory alerts, which may alter adverse-event reporting patterns over time.”

Source: PubMed 42865361 · DOI 10.3389/fendo.2026.1941948 · all study summaries

Sample size

18,281,897 unique individual case safety reports after deduplication

“A total of 21,645,384 raw records were retrieved, of which 3,363,487 duplicate records were removed using a global cross-quarter deduplication scheme (the highest PRIMARYID was retained for each CASEID; for the AERS format [2009–2012], the highest ISR per CASE was retained), yielding 18,281,897 unique ICSRs.”

Reported findings

Disproportionality signals for pancreatic adverse events were observed for all four study drugs and control drugs

“A disproportionality signal for pancreatic adverse events was observed for all four study drugs, but the control drugs metformin and SGLT2i also showed positive signals.”

Disproportionality signals for biliary events were observed for all four study drugs but not for control drugs

“For biliary events, all four study drugs showed signals, whereas the two control drugs showed no signals.”

Pancreatic signals remained for diabetes-indication brands but were weaker for obesity-indication brands

“After stratification by brand, the pancreatic signal remained for brands with diabetes indications, while the signal was weaker for brands with obesity indications.”

Limitations stated by the authors

FAERS is subject to underreporting and selective reporting, and estimates cannot determine incidence or causality

“As a spontaneous reporting system, FAERS is inevitably subject to underreporting, selective reporting, and provocative reporting induced by media attention and regulatory alerts, which may alter adverse-event reporting patterns over time.”