129 findings across 17 drug records. Domain pages are descriptive indexes, not advice.
The grade rates how strong the evidence is, not how good or safe the drug is. A low or very-low grade means the evidence is thin, not that an effect is harmful or ruled out.
Most represented records
GLP-1-based therapies, multi-agent (62)
Semaglutide (21)
Tirzepatide (13)
Retatrutide (8)
Tirzepatide and semaglutide comparison (8)
Liraglutide (5)
Dulaglutide (2)
Aleniglipron (1)
Evidence spread
High evidence 23Moderate evidence 38Low evidence 44Very low evidence 24
CagriSemaINVESTIGATIONAL
safety-otherIncrease
High evidence
In REDEFINE 1, GI AEs common but AE-related discontinuation modest; only a small fraction discontinued specifically for GI events.
SourceGarvey WT, Blüher M, et al. N Engl J Med 2025 (REDEFINE 1)Source
PopulationREDEFINE 1 phase 3, adults with overweight/obesity without T2D
Real-world data associate tirzepatide with higher osteoporosis/fragility-fracture risk than other GLP-1 RAs - a greatest-weight-loss/dual-agonist signal.
SourceTirzepatide and osteoporosis/fragility fracture (TriNetX cohort), Diabetes Res Clin Pract...Source
A 2018 Bayesian network meta-analysis of 54 RCTs (49,602 T2D patients) found GLP-1 RAs were associated with decreased fracture risk versus placebo - an independent...
SourceZhang YS, Weng WY, Xie BC, et al. Glucagon-like peptide-1 receptor agonists and fracture ...Source
Full findingA 2018 Bayesian network meta-analysis of 54 RCTs (49,602 T2D patients) found GLP-1 RAs were associated with decreased fracture risk versus placebo - an independent net-neutral-to-protective fracture pole, separate from the contested anabolic-BMD claim.
PopulationGLP-1 RA fracture network meta-analysis (Zhang et al., Osteoporos Int 2018; 54 RCTs, T2D, PROSPERO CRD42018094433).
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsplacebo; other anti-hyperglycaemic drugs
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNot directional
Low evidence
GLP-1 RA-associated hair loss is markedly female-predominant (men about two-thirds less likely to report it), and more severe with tirzepatide.
SourceGLP-1 RA hair-loss sex-skew (cross-sectional, n=254), J Cosmet Dermatol 2026Source
In a Danish nationwide register of people with type-2 diabetes and hereditary haemochromatosis, each year of GLP-1 receptor agonist exposure was associated with...
SourceBain SC, Hansen TK et al. Glucose-lowering drugs and serum ferritin in haemochromatosis a...Source
Full findingIn a Danish nationwide register of people with type-2 diabetes and hereditary haemochromatosis, each year of GLP-1 receptor agonist exposure was associated with roughly 30% lower serum ferritin - i.e. GLP-1 RAs are associated with LOWER iron stores, not higher.
PopulationDanish nationwide register (observational cohort), 439 people with hereditary haemochromatosis + type-2 diabetes, 6,529 ferritin measurements 2008-2021; only ~36 GLP-1-exposed contributing ~366 samples.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsno GLP-1 RA treatment; SGLT-2 inhibitor
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Low evidence
In the same register, SGLT-2 inhibitors RAISED serum ferritin, so the headline GLP-1-vs-SGLT-2 gap (~40% at HH year 4) is partly a comparator artefact: part of the...
SourceBain SC, Hansen TK et al. Glucose-lowering drugs and serum ferritin in haemochromatosis a...Source
Full findingIn the same register, SGLT-2 inhibitors RAISED serum ferritin, so the headline GLP-1-vs-SGLT-2 gap (~40% at HH year 4) is partly a comparator artefact: part of the apparent GLP-1 'lowering' is relative to a drug that raises ferritin, which overstates GLP-1 lowering versus no treatment.
PopulationDanish nationwide register (observational cohort), people with hereditary haemochromatosis + type-2 diabetes; SGLT-2 inhibitor exposure arm.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsGLP-1 RA; no treatment
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNot directional
Low evidence
The register study did NOT establish a mechanism for the lower ferritin; the authors flag possible mediation via glycaemic control and body weight, and the model did...
SourceBain SC, Hansen TK et al. Glucose-lowering drugs and serum ferritin in haemochromatosis a...Source
Full findingThe register study did NOT establish a mechanism for the lower ferritin; the authors flag possible mediation via glycaemic control and body weight, and the model did not adjust for weight loss, CRP/inflammation or HbA1c - so the association cannot yet be attributed to a direct iron-handling effect.
PopulationDanish nationwide register (observational cohort), people with hereditary haemochromatosis + type-2 diabetes.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsno treatment
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherDecrease
Low evidence
A narrative review reports that GLP-1 RA users have 26-30% lower ferritin than SGLT2i comparators with frequent iron depletion, attributing it to appetite suppression,...
SourceUrbina et al., GLP-1 receptor agonists and iron status, Clin Obes 2026;16:e70070Source
Full findingA narrative review reports that GLP-1 RA users have 26-30% lower ferritin than SGLT2i comparators with frequent iron depletion, attributing it to appetite suppression, delayed gastric emptying and altered absorption.
PopulationNarrative review (Urbina et al., Clin Obes 2026) summarising GLP-1 RA iron-status literature.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsSGLT-2 inhibitor
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
Low evidence
The lowering effect is NOT universal: a 700-patient retrospective GLP-1 analogue cohort found no significant change in ferritin (haemoglobin fell 0.2 g/dL and 8.4%...
SourceEffect of GLP-1 receptor agonists on haematological and iron indices: a retrospective coh...Source
Full findingThe lowering effect is NOT universal: a 700-patient retrospective GLP-1 analogue cohort found no significant change in ferritin (haemoglobin fell 0.2 g/dL and 8.4% developed anaemia), which is counter-evidence to the class-wide ferritin-lowering claim.
Population700-patient retrospective observational cohort of GLP-1 analogue users (Saudi Med J 2025); ferritin a secondary endpoint.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsbaseline
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNot directional
Low evidence
Interpretive key bounding all direction findings: ferritin is a positive acute-phase reactant, so an elevated value may reflect inflammation not iron stores, and a...
SourceUrbina et al., GLP-1 receptor agonists and iron status, Clin Obes 2026;16:e70070Source
Full findingInterpretive key bounding all direction findings: ferritin is a positive acute-phase reactant, so an elevated value may reflect inflammation not iron stores, and a fall during weight loss / reduced steatosis on these drugs may partly reflect resolving inflammation rather than only true iron depletion.
PopulationNarrative review (Urbina et al., Clin Obes 2026) plus acute-phase-reactant physiology (AASLD Liver Fellow Network; PMC2893236/PMC5878890/PMC5223018).
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsnot applicable
DulaglutideMARKETED
safety-otherNo change
High evidence
Short-term dulaglutide did not alter sexual desire or the reproductive axis in healthy lean men.
SourceDulaglutide on male sexual function and HPG axis (RCT), EBioMedicine 2024Source
In a PCOS RCT, semaglutide improved menstrual-cycle recovery - a weight-loss-mediated effect.
SourceSemaglutide+metformin in PCOS (open-label RCT, n=100), Reprod Biol Endocrinol 2025Source
PopulationPCOS open-label RCT (Reprod Biol Endocrinol 2025): 100 overweight/obese women with PCOS (Rotterdam criteria), prospective randomised controlled open-label trial of semaglutide 1 mg weekly plus metformin vs metformin alone, 16 weeks.
Fundingacademic/government - Chinese government and institutional grants (Chongqing Health Commission/Science & Technology Bureau; NSFC; not industry)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherDecrease
Low evidence
The same narrative review reports vitamin-D deficiency as the most common micronutrient abnormality on GLP-1 RA therapy (7.5% at 6 mo / 13.6% at 12 mo in an n=69...
SourceUrbina et al., GLP-1 receptor agonists and iron status, Clin Obes 2026;16:e70070Source
Full findingThe same narrative review reports vitamin-D deficiency as the most common micronutrient abnormality on GLP-1 RA therapy (7.5% at 6 mo / 13.6% at 12 mo in an n=69 sub-study) and intakes below the DRI for calcium and iron in >60% of patients - but these are intake/small-substudy data, not serum status.
PopulationNarrative review (Urbina et al., Clin Obes 2026) summarising GLP-1 RA micronutrient-status literature; the vitamin-D-deficiency rates from an n=69 sub-study.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsSGLT-2 inhibitor
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Low evidence
In a large nationwide cohort, periconceptional GLP-1 exposure raised preterm-birth risk only among diabetes-indication users, not weight-management users - implicating...
Full findingIn a large nationwide cohort, periconceptional GLP-1 exposure raised preterm-birth risk only among diabetes-indication users, not weight-management users - implicating the underlying disease rather than the drug.
NET VERDICT (acute kidney injury) - a LOW/very-low-grade pharmacovigilance signal that is mostly VOLUME-DEPLETION-mediated, i.e. secondary to dehydration from GI...
SourceConsolidation of GLP-1RA AKI pharmacovigilance (volume-depletion mechanism) + comparative...
Full findingNET VERDICT (acute kidney injury) - a LOW/very-low-grade pharmacovigilance signal that is mostly VOLUME-DEPLETION-mediated, i.e. secondary to dehydration from GI adverse events (vomiting/diarrhoea/poor intake), NOT a primary nephrotoxic effect. Tirzepatide reported less than semaglutide (reporting, not incidence). Higher risk in pre-existing renal impairment. This is the AE signal; longer-term renal OUTCOMES are favourable and live in the renal domain.
PopulationFAERS pharmacovigilance + case literature; comparative tirz-vs-sema FAERS.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
GLP-1RA-associated acute kidney injury reported in pharmacovigilance/case literature, proposed mechanism dehydration from vomiting/diarrhoea/poor intake; renal...
SourceDong & Sun, Front Endocrinol 2022 (GLP-1RA AKI FAERS disproportionality)Source
Full findingGLP-1RA-associated acute kidney injury reported in pharmacovigilance/case literature, proposed mechanism dehydration from vomiting/diarrhoea/poor intake; renal disorders the second-largest AE category after GI in one case series.
PopulationPost-marketing pharmacovigilance + case reports/review
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Moderate evidence
Peri-operative aspiration meta-analysis: GLP-1 RA use associated with sharply increased residual gastric contents despite appropriate fasting, but NOT with increased...
Full findingPeri-operative aspiration meta-analysis: GLP-1 RA use associated with sharply increased residual gastric contents despite appropriate fasting, but NOT with increased pulmonary aspiration; withholding at least one dose lowered residual-contents odds.
Fundingacademic/independent - investigator (Australian Government Research Training Program Scholarship); no industry study sponsor
Comparatorsnon-GLP-1 RA exposure
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Moderate evidence
Peri-procedural meta-analysis: pre-procedural GLP-1 RA increased pre-procedural GI symptoms and elevated residual gastric content vs control, with no significant...
Sourcedo Nascimento TS et al., Perioper Med (Lond) 2024Source
Full findingPeri-procedural meta-analysis: pre-procedural GLP-1 RA increased pre-procedural GI symptoms and elevated residual gastric content vs control, with no significant difference in PONV or 30-day mortality.
Fundingacademic/independent - self-funded (no external sponsorship)
Comparatorscontrol (no GLP-1 RA)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Low evidence
Anaesthesia narrative reviews: GLP-1 RAs and tirzepatide delay gastric emptying and raise aspiration considerations; data preliminary/conflicting; gastric ultrasound...
SourceDoroba O et al., Anaesthesiol Intensive Ther 2025; Kwak HJ et al., Korean J Anesthesiol 2026Source
Full findingAnaesthesia narrative reviews: GLP-1 RAs and tirzepatide delay gastric emptying and raise aspiration considerations; data preliminary/conflicting; gastric ultrasound recommended; guidance is multi-society expert consensus.
PopulationNarrative reviews of peri-operative management of novel antidiabetic/anti-obesity drugs
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsSGLT2 inhibitors (contrast)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherDecrease
High evidence
FOLK-CLAIM VERDICT ('reduced inflammation / joint pain relief / skin improvements'): MIXED, and the mechanism-split is the whole story. (1) Knee-OA pain relief = TRUE...
SourceBliddal H et al. Semaglutide in knee osteoarthritis with obesity (STEP-9). N Engl J Med 2...Source
Full findingFOLK-CLAIM VERDICT ('reduced inflammation / joint pain relief / skin improvements'): MIXED, and the mechanism-split is the whole story. (1) Knee-OA pain relief = TRUE but predominantly WEIGHT-LOSS-mediated (less joint load; STEP-9 real and large, but tracks weight and the placebo arm also improved - not proven direct chondroprotection). (2) Systemic inflammation / CRP reduction = TRUE and PARTLY DIRECT (CRP falls somewhat beyond weight loss in SELECT), though magnitude still mostly adiposity-coupled. (3) Skin (psoriasis, hidradenitis) = UNPROVEN (small/uncontrolled, weight-confounded).
PopulationSemaglutide knee-OA RCT (STEP-9) + reta knee-OA phase-3 topline (TRIUMPH-4) + CRP data (SELECT/HFpEF) + small skin series.
Fundingindustry - Novo Nordisk (trial sponsor; disclosed)
Comparatorsplacebo
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Moderate evidence
NET VERDICT (peri-operative aspiration) - the most clinically live procedural signal: by delaying gastric emptying these agents leave residual gastric contents despite...
Full findingNET VERDICT (peri-operative aspiration) - the most clinically live procedural signal: by delaying gastric emptying these agents leave residual gastric contents despite guideline fasting, raising a theoretical aspiration risk under anaesthesia/sedation. Meta-analysis robustly demonstrates the RETAINED-CONTENTS surrogate but does NOT demonstrate increased actual pulmonary aspiration. Operationally live because multi-society/ASA guidance now recommends mitigation.
PopulationPatients on GLP-1RAs undergoing anaesthesia/endoscopy/sedation; meta-analyses of peri-procedural cohorts + gastric-ultrasound studies; semaglutide/liraglutide-era data; no reta-direct.
Fundingacademic/independent - investigator (Australian Government Research Training Program Scholarship); no industry study sponsor
Comparatorsnon-GLP-1-RA controls; fasted controls; dose-withheld vs continued
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNot directional
Moderate evidence
Multi-society clinical practice guidance for safe peri-operative use of GLP-1 RAs: individualised risk-stratified management rather than fixed cessation, balancing...
SourceKindel TL, Wang AY, Wadhwa A et al. (multisociety), Clin Gastroenterol Hepatol, 2024Source
Full findingMulti-society clinical practice guidance for safe peri-operative use of GLP-1 RAs: individualised risk-stratified management rather than fixed cessation, balancing aspiration-risk mitigation against the harms of interrupting therapy. Operationalises the earlier ASA October-2023 consensus (which advised holding GLP-1 RAs pre-procedure).
PopulationAdults on GLP-1 RAs presenting for elective procedures/anaesthesia/endoscopy.
Fundingprofessional-society guidance - no external/commercial funding (AGA and co-sponsoring societies; author consulting COIs disclosed)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Low evidence
In fasting patients undergoing endoscopy under anaesthesia, GLP-1RA therapy associated with markedly higher residual gastric contents, posing additional...
Full findingIn fasting patients undergoing endoscopy under anaesthesia, GLP-1RA therapy associated with markedly higher residual gastric contents, posing additional pulmonary-aspiration risk.
PopulationRetrospective cohort, single US institution (MGH) 2019-2023, EGD under anaesthesia, 90 vs 102
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorspatients starting GLP-1RA after procedure (control)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Moderate evidence
NET VERDICT (gallbladder / biliary) - one of the FEW high-confidence real GI safety signals: randomised-trial meta-analysis shows GLP-1-based therapy increases the...
Full findingNET VERDICT (gallbladder / biliary) - one of the FEW high-confidence real GI safety signals: randomised-trial meta-analysis shows GLP-1-based therapy increases the composite of gallbladder/biliary disease, concentrated at higher doses, longer durations and in weight-loss trials. Leading mechanism is rapid weight loss + reduced gallbladder motility, so it is expected to track with weight-loss magnitude.
PopulationPooled randomised participants across GLP-1RA trials (T2D and obesity) + CV-outcomes RCT-meta + obesity head-to-head meta. No retatrutide-direct biliary endpoint.
Fundingacademic-Chinese public (NSFC / Beijing NSF / CAMS)
Comparatorsplacebo; active comparators; semaglutide; tirzepatide
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Moderate evidence
GLP-1 RA randomisation associated with increased composite of gallbladder/biliary disease (cholelithiasis, cholecystitis, biliary disease); higher at higher doses,...
Full findingGLP-1 RA randomisation associated with increased composite of gallbladder/biliary disease (cholelithiasis, cholecystitis, biliary disease); higher at higher doses, longer durations, and in weight-loss trials.
Fundingacademic-Chinese public (NSFC / Beijing NSF / CAMS)
Comparatorsplacebo; non-GLP-1 RA drugs
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Moderate evidence
In a CV-outcomes meta-analysis, GLP-1 RAs increased gallbladder disorders by 26% but showed no difference in pancreatitis or neoplasm.
SourceGalli M et al., J Am Coll Cardiol 2025Source
Population21 RCTs, 99,599 patients, mean follow-up 2.4 yr; meta-analysis
Fundingacademic / independent - no industry sponsor (investigator grant: Sapienza University of Rome)
Comparatorsplacebo; controls
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
Low evidence
NET VERDICT (bone / fracture) - NEUTRAL-to-mildly-negative and LOW-grade: GLP-1R agonists show neutral-to-mildly-negative (non-significant) BMD/turnover effects, but...
SourceConsolidation of GLP-1RA/incretin bone-health appraisal with the rapid-weight-loss/sarcop...
Full findingNET VERDICT (bone / fracture) - NEUTRAL-to-mildly-negative and LOW-grade: GLP-1R agonists show neutral-to-mildly-negative (non-significant) BMD/turnover effects, but fracture risk does NOT appear increased at clinically relevant doses; dual/triple incretin/amylin analogues may even be bone-positive (mostly preclinical). The live concern is RAPID weight loss driving sarcopenia and downstream fracture risk, not a direct osteotoxic drug effect.
PopulationBone-health appraisal/review across GLP-1RA + incretin analogues; preclinical for dual/triple bone-positive signals.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsplacebo; post-bariatric-surgery populations; lifestyle weight loss
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
Low evidence
Bone-health appraisal: GLP-1R agonists show neutral-to-mildly-negative (non-significant) effects on bone turnover/BMD, but fracture risk does NOT appear increased at...
SourceAnastasilakis AD et al., Diabetes Obes Metab 2025Source
Full findingBone-health appraisal: GLP-1R agonists show neutral-to-mildly-negative (non-significant) effects on bone turnover/BMD, but fracture risk does NOT appear increased at clinically relevant doses; dual/triple incretin and amylin analogues may be bone-positive (mostly preclinical).
PopulationCritical-appraisal review of anti-obesity medications and bone metabolism (clinical + preclinical)
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherDecrease
Low evidence
Vs insulin, GLP-1RA initiation associated with LOWER liver and pancreatic cancer risk and no increased risk for most other major cancers, in a large T2D cohort.
SourceRashid Z et al., J Gen Intern Med 2026Source
PopulationRetrospective cohort, IBM-MarketScan, T2D 2013-2021, N=106,088, weighted vs insulin
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsinsulin
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Low evidence
In obesity, GLP-1RA use associated with LOWER overall cancer risk and reduced endometrial/ovarian/meningioma cancer, but a marginally non-significant INCREASED kidney...
Full findingIn obesity, GLP-1RA use associated with LOWER overall cancer risk and reduced endometrial/ovarian/meningioma cancer, but a marginally non-significant INCREASED kidney cancer risk.
PopulationTarget-trial emulation, OneFlorida+ EHR 2014-2024, 86,632 matched adults with obesity
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsnon-users eligible for anti-obesity medication
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Low evidence
In T2D, GLP-1RAs associated with lower risk of 10/13 obesity-associated cancers vs insulin; NO reduction for postmenopausal breast or thyroid cancer; vs metformin,...
Full findingIn T2D, GLP-1RAs associated with lower risk of 10/13 obesity-associated cancers vs insulin; NO reduction for postmenopausal breast or thyroid cancer; vs metformin, INCREASED kidney cancer.
PopulationRetrospective cohort, US nationwide EHR (113M), 1,651,452 T2D, 2005-2018, 15-yr follow-up, matched
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsinsulin; metformin
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Low evidence
Mechanistic 'double-edged sword' review: incretin therapies linked in some studies to increased pancreatic/thyroid/cholangiocarcinoma/colorectal cancer (esp....
Full findingMechanistic 'double-edged sword' review: incretin therapies linked in some studies to increased pancreatic/thyroid/cholangiocarcinoma/colorectal cancer (esp. genetically predisposed), while showing anticancer effects in prostate, breast, ovarian and others.
PopulationNarrative review of clinical cohorts + in vivo/in vitro
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Very low evidence
NET VERDICT (broader cancer) - observational cohorts (confounded by indication) lean reassuring, with a protective pattern for obesity-associated malignancies vs...
SourceConsolidation of GLP-1RA cancer-association cohorts (vs insulin/metformin) with FAERS tum...
Full findingNET VERDICT (broader cancer) - observational cohorts (confounded by indication) lean reassuring, with a protective pattern for obesity-associated malignancies vs insulin (lower liver/pancreatic and 10/13 obesity-associated cancers; lower overall incl. endometrial/ovarian/meningioma in obesity). This is NOT a protective claim and NOT a net-harm verdict: it sits against a contested marginal kidney-cancer signal, no reduction for postmenopausal breast/thyroid, residual pancreatic pharmacovigilance noise, and a mechanistic double-edged-sword hypothesis.
Multi-database active-comparator emulation (GLP-1RA vs SGLT2i): GLP-1RA associated with ~85% higher presumed-NAION risk, absolute risk small.
SourceTesfaye H, Patorno E et al., Diabetes Obes Metab 2025Source
Population482,912 propensity-matched pairs GLP-1RA vs SGLT2i, three US claims databases 2016-2024
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsSGLT2 inhibitors
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
Moderate evidence
REFUTING/NULL pole at class level: meta-analysis of 28 observational studies found GLP-1 RAs did NOT increase NAION, glaucoma, or retinopathy vs other antidiabetics.
SourceAnatriello A et al., Front Pharmacol 2026Source
PopulationMeta-analysis, 28 observational studies (6 semaglutide, 22 all GLP-1 RA) in T2D, 2006-2025
Fundingacademic/independent - no financial support received
Comparatorsother antidiabetic treatments
Tirzepatide and semaglutide comparisonSTATUS UNKNOWN
safety-otherIncrease
Very low evidence
FAERS disproportionality: strong post-marketing reporting signal for optic ischaemic neuropathy with semaglutide; weaker but significant for tirzepatide and...
SourceKakkar AK, Payra S, Charaya A, Diabetes Obes Metab 2026Source
Full findingFAERS disproportionality: strong post-marketing reporting signal for optic ischaemic neuropathy with semaglutide; weaker but significant for tirzepatide and liraglutide; none for dulaglutide/exenatide/lixisenatide.
FOLK-CLAIM VERDICT ('hair loss/shedding around month 3-4'): TRUE that it happens / OVERSTATED as a drug toxicity. Alopecia is a real, RCT-replicated signal (RR ~3.25),...
SourceCheng & Chang. GLP-1 receptor agonists and hair loss: a meta-analysis. Diabetes Res Clin ...Source
Full findingFOLK-CLAIM VERDICT ('hair loss/shedding around month 3-4'): TRUE that it happens / OVERSTATED as a drug toxicity. Alopecia is a real, RCT-replicated signal (RR ~3.25), and the month 3-4 timing is textbook TELOGEN EFFLUVIUM. But the weight of evidence attributes it predominantly to RAPID WEIGHT LOSS + caloric/nutritional deficit (the classic trigger), NOT a direct drug-on-follicle toxicity - i.e. the drug is largely the vehicle for rapid weight loss. It is REVERSIBLE, dose-/weight-loss-magnitude-dependent and female-biased. Mitigation overlaps the muscle-loss fix (adequate protein + micronutrients: iron, zinc, B12, vitamin D).
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsplacebo
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Very low evidence
NET VERDICT (gastroparesis / ileus / intestinal obstruction) - a WEAK, low-grade signal: the class delays gastric emptying (on-target, prominent in tolerability) and...
SourceGI Safety Assessment of GLP-1 RAs from FAERS, Diagnostics (MDPI) 2024Source
Full findingNET VERDICT (gastroparesis / ileus / intestinal obstruction) - a WEAK, low-grade signal: the class delays gastric emptying (on-target, prominent in tolerability) and spontaneous reports surface ileus/obstruction events, but randomised-trial-level evidence does NOT confirm elevated mechanical-obstruction risk and dedicated FAERS work found only weak/non-significant signals. Honest net: low-grade, trial-unconfirmed.
PopulationFAERS spontaneous-report databases + a randomised-trials network meta-analysis for the obstruction endpoint. No reta-direct.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
FAERS disproportionality found only WEAK / non-significant intestinal-obstruction signals across GLP-1RAs; no clear class signal for mechanical obstruction.
SourceGI Safety Assessment of GLP-1 RAs from FAERS, Diagnostics (MDPI) 2024Source
PopulationFAERS 2007-2023, US reports, 16,568 GI AE reports (tirzepatide not included)
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsFAERS background
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Very low evidence
FOLK-CLAIM VERDICT ('libido changes, both up and down'): UNPROVEN / MIXED - no controlled evidence of a direct GLP-1 libido effect in either direction. The only...
SourceNarrative review of GLP-1 RAs and sexual function (men and women): dulaglutide crossover ...Source
Full findingFOLK-CLAIM VERDICT ('libido changes, both up and down'): UNPROVEN / MIXED - no controlled evidence of a direct GLP-1 libido effect in either direction. The only RCT-level human datapoint (dulaglutide crossover) was null; FAERS was non-significant; the plausible 'up' direction is confounded by weight-loss-driven testosterone normalisation and improved body image, and the 'down' direction lacks a significant controlled signal. Bidirectional anecdotes are genuine experiences but do not establish a pharmacological effect.
Full findingFOLK-CLAIM VERDICT ('menstrual changes - intermenstrual/heavy/irregular bleeding'): MIXED / EMERGING - no longer pure self-report. Two independent low-grade signals now converge (online self-report + a semaglutide-specific FAERS reproductive signal), but it is NOT trial-grade: FAERS cannot establish incidence/causation (reporting bias, no denominator, and weight loss itself shifts cycles). Notably the pharmacovigilance signal is AGENT-SPECIFIC (semaglutide), with tirzepatide/dulaglutide trending the opposite way.
PopulationOnline self-report cohort + FAERS disproportionality; no dedicated RCT endpoint.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
High evidence
EMA PRAC concluded its class review and found no evidence of a causal association with suicidal/self-injurious thoughts/actions; no product-information update...
SourceEMA PRAC, Meeting highlights 8-11 April 2024Source
Full findingEMA PRAC concluded its class review and found no evidence of a causal association with suicidal/self-injurious thoughts/actions; no product-information update warranted. Triggered by Icelandic case reports (initially 3), ~150 reports analysed.
PopulationRegulatory review of non-clinical, clinical-trial, post-marketing and observational data; Jul 2023-Apr 2024
Fundingregulatory agency (FDA/EMA/MHRA/WHO)
Tirzepatide and semaglutide comparisonSTATUS UNKNOWN
safety-otherNo change
High evidence
FDA, after a comprehensive review incl. a meta-analysis of 91 trials, found no increased suicidality risk and REQUESTED REMOVAL of the suicidal-behaviour-and-ideation...
SourceFDA Drug Safety Communication, 'FDA Requests Removal of Suicidal Behavior and Ideation Wa...Source
Full findingFDA, after a comprehensive review incl. a meta-analysis of 91 trials, found no increased suicidality risk and REQUESTED REMOVAL of the suicidal-behaviour-and-ideation warning from labels for liraglutide, semaglutide 2.4 mg and tirzepatide.
PopulationFDA pooled clinical-trial meta-analysis + post-marketing/observational review
Fundingregulatory agency (FDA/EMA/MHRA/WHO)
Comparatorsplacebo
Tirzepatide and semaglutide comparisonSTATUS UNKNOWN
safety-otherMixed
Very low evidence
VigiBase (WHO) disproportionality: significantly increased ROR for suicidal IDEATION with semaglutide, liraglutide, tirzepatide, and for 'depression/suicidal' and...
SourceMcIntyre RS et al., J Affect Disord 2024;369:922-927Source
Full findingVigiBase (WHO) disproportionality: significantly increased ROR for suicidal IDEATION with semaglutide, liraglutide, tirzepatide, and for 'depression/suicidal' and suicidal behaviour with semaglutide/liraglutide; BUT significantly DECREASED ROR for suicide attempts and completed suicide.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsfull database background
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
Very low evidence
FAERS disproportionality (2018-2022): no signal of disproportionate reporting for suicidal/self-injurious behaviours; co-medication analysis flagged...
Full findingFAERS disproportionality (2018-2022): no signal of disproportionate reporting for suicidal/self-injurious behaviours; co-medication analysis flagged psychiatric-comorbidity proxies.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsfull database background
Tirzepatide and semaglutide comparisonSTATUS UNKNOWN
safety-otherMixed
Very low evidence
FAERS unmasking analysis (2005-2023) with metformin/orlistat comparators: per-agent RORs for psychiatric ADRs elevated for semaglutide (2.03), tirzepatide (1.76),...
SourceGuirguis A et al., Eur Neuropsychopharmacol 2024;82:82-91Source
Full findingFAERS unmasking analysis (2005-2023) with metformin/orlistat comparators: per-agent RORs for psychiatric ADRs elevated for semaglutide (2.03), tirzepatide (1.76), liraglutide (1.64) but lower than metformin; 42 deaths incl. 13 completed suicides; no causal link inferable.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsmetformin; orlistat
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
Very low evidence
FAERS disproportionality (2005-2023): no overall disproportionate increase in suicide/self-injury; a marginal ROR elevation in children was NOT confirmed by EBGM05.
SourceChen C et al., Eur Psychiatry 2023;66(1):e99Source
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsfull database background
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
Low evidence
Propensity-weighted population-based cohort (Diabetologia): assessed GLP-1 RA association with suicidal ideation/self-injury in diabetes and obesity; part of the...
Full findingPropensity-weighted population-based cohort (Diabetologia): assessed GLP-1 RA association with suicidal ideation/self-injury in diabetes and obesity; part of the cohort evidence body informing regulators.
PopulationIndividuals with diabetes and obesity; propensity-weighted population-based cohort
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsactive comparators
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Low evidence
NET VERDICT (thyroid C-cell / MTC) - a RODENT-based FDA boxed warning and class contraindication (personal/family MTC history, or MEN 2) coexists with a...
SourceConsolidation of the FDA semaglutide/tirzepatide boxed warning with human cohort + pharma...
Full findingNET VERDICT (thyroid C-cell / MTC) - a RODENT-based FDA boxed warning and class contraindication (personal/family MTC history, or MEN 2) coexists with a NULL-to-unconfirmed HUMAN signal. The regulatory contraindication stands as a labelling FACT while no human MTC excess is confirmed - the rodent C-cell tumourigenic effect does NOT translate (human calcitonin screening negative; null cohorts dominate over reporting-driven signal poles).
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsplacebo; DPP-4 inhibitors; other antidiabetics; no-treatment cohorts
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Low evidence
SIGNAL pole: French SNDS nested case-control found GLP-1 RA use for 1-3 years associated with increased all-thyroid-cancer AND medullary thyroid cancer risk.
Population2562 thyroid-cancer cases vs 45,184 controls; T2D on second-line antidiabetics 2006-2018, France
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsGLP-1 RA non-users among T2D
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
Low evidence
REFUTING pole: Scandinavian three-country active-comparator cohort found no substantially increased thyroid cancer risk vs DPP-4 inhibitors; MTC estimate...
Full findingREFUTING pole: Scandinavian three-country active-comparator cohort found no substantially increased thyroid cancer risk vs DPP-4 inhibitors; MTC estimate imprecise/null on point estimate.
Population145,410 GLP-1 RA vs 291,667 DPP-4i initiators; Denmark/Norway/Sweden 2007-2021; mean follow-up 3.9 yr
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsDPP-4 inhibitors; SGLT2 inhibitors
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
Low evidence
Second NULL pole: Korean nationwide cohort found no association with new-onset cancers incl. pancreatic and MTC; thyroid point estimate >1 but NS.
Population7827 propensity-matched (GLP-1 RA / DM control / non-DM control); Korean NHIS 2004-2021; median follow-up 8 yr
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsDM controls; non-DM controls
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Very low evidence
FAERS tumour disproportionality (2004-2021): NO overall increase at system-organ-class level, but significant signals for MTC, papillary thyroid cancer, malignant...
Full findingFAERS tumour disproportionality (2004-2021): NO overall increase at system-organ-class level, but significant signals for MTC, papillary thyroid cancer, malignant pancreatic neoplasms and islet-cell neoplasms.
Population8718 GLP-1 RA tumour reports; FAERS Q1 2004-Q2 2021; disproportionality
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsFAERS background
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherDecrease
Moderate evidence
Network meta-analysis (body composition): potent agents produce greatest fat-mass loss but significant ABSOLUTE lean-mass loss; RELATIVE (% baseline) lean mass...
Full findingNetwork meta-analysis (body composition): potent agents produce greatest fat-mass loss but significant ABSOLUTE lean-mass loss; RELATIVE (% baseline) lean mass unchanged. Tirzepatide/semaglutide most effective for weight/fat, among least effective at preserving lean mass.
PopulationNetwork meta-analysis, 22 RCTs, 2258 participants (diabetes and/or overweight/obesity), through Nov 2024
Fundingacademic / independent - no specific grant funding
Tirzepatide and semaglutide comparisonSTATUS UNKNOWN
safety-otherDecrease
Low evidence
Narrative review framing: incretin agents cause rapid significant lean-mass loss (~10% of body weight, ~6 kg), characterised as comparable to a decade+ of ageing,...
SourceLocatelli JC et al., Diabetes Care 2024Source
Full findingNarrative review framing: incretin agents cause rapid significant lean-mass loss (~10% of body weight, ~6 kg), characterised as comparable to a decade+ of ageing, raising sarcopenia/frailty concern; resistance exercise proposed to preserve muscle.
PopulationNarrative review across liraglutide, semaglutide, tirzepatide, retatrutide (~15-24% weight loss)
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Tirzepatide and semaglutide comparisonSTATUS UNKNOWN
safety-otherMixed
Very low evidence
Opposite-pole mechanistic view: emerging preclinical/translational data suggest possible muscle-specific BENEFICIAL effects (attenuated atrophy, improved...
SourceKoceva A, Janez A, Jensterle M, Medicina (Kaunas) 2025Source
Full findingOpposite-pole mechanistic view: emerging preclinical/translational data suggest possible muscle-specific BENEFICIAL effects (attenuated atrophy, improved myogenesis/mitochondrial function, reduced myosteatosis), alongside clinical proportional fat/lean loss.
PopulationNarrative review of preclinical (animal) and clinical evidence on skeletal muscle
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Moderate evidence
NET VERDICT (NAION, non-arteritic anterior ischaemic optic neuropathy) - a REAL but DEBATED and rare class signal: observational cohorts split (some ~doubled risk,...
SourceNatividade GR et al., JAMA Ophthalmol, 2025 (RCT-level ocular meta) consolidating observa...Source
Full findingNET VERDICT (NAION, non-arteritic anterior ischaemic optic neuropathy) - a REAL but DEBATED and rare class signal: observational cohorts split (some ~doubled risk, others null/protective); the 2025 RCT-level meta-analysis found a significant NAION Peto OR 3.92 but TSA judged the data insufficient for a definitive conclusion; EMA/PRAC added NAION to the semaglutide label (frequency 'unknown'/very rare). Absolute incidence is small throughout.
PopulationAdults with T2D and overweight/obesity; signal strongest in T2D, longer follow-up (>=2 yr) and with hypertension. RCT meta n=73,640; observational cohorts range single-centre to nationwide.
Fundingacademic / investigator-initiated; no industry sponsor found in indexed metadata (Europe PMC fundingList empty); full JAMA COI disclosures paywalled
Comparatorsplacebo; non-GLP-1RA antidiabetics; SGLT2i; active comparators
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
Moderate evidence
NET VERDICT (pancreatitis / pancreatic cancer) - largely REASSURED at trial level: randomised and meta-analytic evidence shows no clear increase in pancreatitis (any...
SourceWen J et al., Endocrinol Diabetes Metab 2025Source
Full findingNET VERDICT (pancreatitis / pancreatic cancer) - largely REASSURED at trial level: randomised and meta-analytic evidence shows no clear increase in pancreatitis (any nominal excess attenuates on stratification) and no overall pancreatic-cancer association. Against this sits a persistent FAERS reporting signal (reporting != incidence). Net: not a confirmed real signal at trial level, with the FAERS pole retained as a low-grade caveat.
PopulationPooled randomised trial participants (the Wen 2025 meta reportedly INCLUDES retatrutide); CV-outcomes trial populations; FAERS.
Fundingnone-declared (independent meta-analysis)
Comparatorsplacebo; active comparators; background-medication strata
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Moderate evidence
Significantly increased pancreatitis risk overall, but NOT significant when stratified by background medication; pancreatic cancer not associated overall, but a signal...
SourceWen J et al., Endocrinol Diabetes Metab 2025Source
Full findingSignificantly increased pancreatitis risk overall, but NOT significant when stratified by background medication; pancreatic cancer not associated overall, but a signal in the with-background-medication stratum.
Population62 RCTs, 66,232 patients, mean follow-up 43.5 wk; meta-analysis
Fundingnone-declared (independent meta-analysis)
Comparatorsplacebo; active comparators
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
Low evidence
Periconceptional GLP-1 RA exposure NOT associated with a large increased major-congenital-malformation risk vs insulin, in a multinational cohort of pregnant women...
Full findingPericonceptional GLP-1 RA exposure NOT associated with a large increased major-congenital-malformation risk vs insulin, in a multinational cohort of pregnant women with T2D.
PopulationCohort, 4 Nordic + US MarketScan + Israeli Maccabi, 2009-2021; T2D pregnancies, infants to 1 yr
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsinsulin; sulfonylureas; DPP-4i; SGLT2i
Tirzepatide and semaglutide comparisonSTATUS UNKNOWN
safety-otherMixed
Low evidence
Animal reproductive-tox studies show adverse offspring outcomes (decreased fetal growth, skeletal/visceral anomalies, embryonic death); no consistent pattern of human...
Full findingAnimal reproductive-tox studies show adverse offspring outcomes (decreased fetal growth, skeletal/visceral anomalies, embryonic death); no consistent pattern of human congenital anomalies to date.
PopulationReview of animal repro-tox + human case reports/cohort/population studies
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsinsulin
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Low evidence
GLP-1 RAs may improve ovulatory function/menstrual regularity (esp. obesity/PCOS), potentially increasing conception and unintended pregnancy; human pregnancy-exposure...
Full findingGLP-1 RAs may improve ovulatory function/menstrual regularity (esp. obesity/PCOS), potentially increasing conception and unintended pregnancy; human pregnancy-exposure safety data remain limited.
PopulationNarrative review 2020-2025, reproductive-age women
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Very low evidence
FAERS: liraglutide carried the greatest pancreatitis reporting signal of the class; semaglutide greatest for nausea/vomiting/diarrhoea/constipation.
PopulationFAERS Jan 2018-Sep 2022; 21,281 GI reports; disproportionality
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsFAERS background
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Low evidence
NET VERDICT (reproductive) - three threads, distinct grades. (1) ORAL hormonal contraceptive efficacy is REDUCED by tirzepatide (delayed gastric emptying) - a real...
SourceConsolidation of the tirzepatide oral-contraceptive labelled interaction with GLP-1RA fer...
Full findingNET VERDICT (reproductive) - three threads, distinct grades. (1) ORAL hormonal contraceptive efficacy is REDUCED by tirzepatide (delayed gastric emptying) - a real LABELLED interaction (barrier/non-oral advised), labelled for tirzepatide but not injectable semaglutide. (2) FERTILITY: GLP-1RAs improve ovulatory function in obesity/PCOS ('Ozempic babies'), raising conception/unintended-pregnancy risk. (3) PREGNANCY: contraindicated - animal repro-tox shows adverse offspring outcomes, while human cohorts show NO large major-malformation excess vs insulin (reassuring but confounded, not exonerating).
PopulationTirzepatide label (PK interaction); GLP-1RA ovulation/PCOS literature; animal reproductive-toxicology + a multinational human pregnancy-exposure cohort (T2D).
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
NET VERDICT (early-worsening diabetic retinopathy): SUSTAIN-6 found retinopathy complications significantly MORE frequent with semaglutide than placebo (HR 1.76),...
SourceMarso SP et al. (SUSTAIN-6), NEJM, 2016Source
Full findingNET VERDICT (early-worsening diabetic retinopathy): SUSTAIN-6 found retinopathy complications significantly MORE frequent with semaglutide than placebo (HR 1.76), widely attributed to RAPID glucose-lowering ('early worsening') in those with pre-existing retinopathy - a KNOWN phenomenon, not necessarily a direct drug effect. At RCT level the 2025 meta found NO overall DR increase (OR 1.04, TSA-sufficient), and a post-hoc RCT showed CRT change explained by glycaemic change rather than drug-specific toxicity.
PopulationT2D at high CV risk with variable baseline retinopathy (SUSTAIN-6, n=3297, 104 wk); pooled RCT n=73,640; post-hoc mechanistic substudy n=40.
Fundingindustry - Novo Nordisk
Comparatorsplacebo; empagliflozin; active comparators
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Very low evidence
CROSS-SIGNAL STRUCTURE (the dossier spine) - the class safety signals ranked by EVIDENCE GRADE, not by alarm. HIGH/REAL (RCT or consistent replication):...
SourceStructural synthesis over the safety-other corpus (the per-signal consolidations); the do...
Full findingCROSS-SIGNAL STRUCTURE (the dossier spine) - the class safety signals ranked by EVIDENCE GRADE, not by alarm. HIGH/REAL (RCT or consistent replication): gallbladder/biliary (weight-loss-coupled); rapid-glucose-lowering early diabetic-retinopathy worsening (SUSTAIN-6 HR 1.76; note steady-state pooled DR is NULL, OR 1.04, TSA-sufficient); peri-operative retained-gastric-contents SURROGATE (the surrogate is robust ~6-fold but actual pulmonary aspiration is unproven). CLEARED/NULL in humans: medullary thyroid carcinoma (rodent-basis boxed warning, no human MTC signal); suicidality (regulatory-cleared). DEBATED: NAION (cohorts split; RCT meta OR 3.92 but TSA-insufficient; label 'very rare'). WEAK/pharmacovigilance-only (reporting != incidence): pancreatitis, ileus, acute kidney injury (mostly volume-depletion-mediated).
PopulationClass-wide across GLP-1 mono, GIP/GLP-1 dual, triple, and amylin co-agonists; strength/population vary by signal.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsplacebo; other antidiabetics; SGLT2i; non-GLP-1RA
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherNo change
Moderate evidence
Pooled meta-analyses found NO significant increase in DKA or severe hypoglycaemia with GLP-1 RAs in T1D - but the DKA estimate is VERY LOW certainty and this null does...
SourceKateel R, Parida A, Chogtu B, Holla SN. Safety of GLP-1 receptor agonists in type 1 diabe...Source
Full findingPooled meta-analyses found NO significant increase in DKA or severe hypoglycaemia with GLP-1 RAs in T1D - but the DKA estimate is VERY LOW certainty and this null does NOT overturn the dose-dependent ketosis signal in the individual ADJUNCT trials.
PopulationGLP-1 RA T1D safety meta-analysis (Kateel et al., 25 studies [23 RCTs + 2 observational]; corroborated by Tan et al. PMC10797415).
Fundingacademic/government (Kasturba Medical College, Manipal; Dept of Health Research, India)
Comparatorsplacebo (insulin background)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Moderate evidence
A class-level meta-analysis found GLP-1 RAs reduce serum uric acid before-versus-after treatment but NOT significantly versus placebo, and less than active comparators...
SourceNajafi S, Bahrami M, Butler AE, Sahebkar A. The effect of glucagon-like peptide-1 recepto...Source
Full findingA class-level meta-analysis found GLP-1 RAs reduce serum uric acid before-versus-after treatment but NOT significantly versus placebo, and less than active comparators including SGLT2 inhibitors - so the apparent urate-lowering is not an established placebo-controlled drug effect at the class level.
PopulationGLP-1 RA class meta-analysis (Najafi et al., 17 studies; RCTs, observational and uncontrolled studies pooled).
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
In a large real-world cohort, GLP-1 RAs were gout-neutral on clinical OUTCOMES while SGLT2 inhibitors reduced incident gout and outperformed GLP-1 RAs head-to-head -...
SourcePreston FG et al. SGLT2 inhibitors, but not GLP-1 receptor agonists, reduce incidence of ...Source
Full findingIn a large real-world cohort, GLP-1 RAs were gout-neutral on clinical OUTCOMES while SGLT2 inhibitors reduced incident gout and outperformed GLP-1 RAs head-to-head - the biochemical urate-lowering did not translate into a gout-incidence benefit for GLP-1 RAs.
PopulationTriNetX federated EHR cohort (Preston et al.), T2D on metformin/insulin +/- SGLT2i or GLP-1 RA, propensity-matched, 5-yr gout incidence.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsmetformin; insulin; SGLT-2 inhibitor
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Low evidence
A 4-patient case series reported paradoxical transient hyperuricaemia and acute gout flares during rapid GLP-1/GIP-driven weight loss, including a recurrent flare...
SourceChaaya JA et al. Transient increase in serum uric acid and gout attacks after weight-loss...Source
Full findingA 4-patient case series reported paradoxical transient hyperuricaemia and acute gout flares during rapid GLP-1/GIP-driven weight loss, including a recurrent flare despite urate-lowering therapy and normal uric acid - a hypothesis-generating counter-signal to the net urate-lowering RCT data.
PopulationAACE Endocrinology & Diabetes 2026 observational case series / uncontrolled cohort (Chaaya et al.); 4 adults on tirzepatide or semaglutide with substantial weight loss.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsnone (uncontrolled observational case series)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherIncrease
Low evidence
A real-world cohort found a small but statistically significant HIGHER gout/colchicine signal in GLP-1 users versus non-users - not protective - consistent with a...
SourceLam DCH et al. Assessing gout risk associated with GLP-1 therapy in obese patients with t...Source
Full findingA real-world cohort found a small but statistically significant HIGHER gout/colchicine signal in GLP-1 users versus non-users - not protective - consistent with a transient-flare or detection-bias effect rather than causal urate-raising.
PopulationTriNetX observational cohort (Lam et al.), overweight/obese T2D, GLP-1 users vs non-users.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsnon-users
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Low evidence
In mechanistic renal-physiology trials, acute exenatide transiently raised renal urate excretion without lowering plasma urate, and prolonged liraglutide/lixisenatide...
SourceTonneijck L et al. Effect of immediate and prolonged GLP-1 receptor agonist administratio...Source
Full findingIn mechanistic renal-physiology trials, acute exenatide transiently raised renal urate excretion without lowering plasma urate, and prolonged liraglutide/lixisenatide produced no urate change - indicating only a marginal, transient tubular effect, not chronic urate-lowering.
PopulationGLP-1 RA renal-physiology trials (Tonneijck et al., post hoc of 4 controlled phase-4 trials; Study-A n=9 healthy, Study-B n=52, Study-C n=36, Study-D n=35 T2D).
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsplacebo; insulin glulisine
DulaglutideMARKETED
safety-otherMixed
Moderate evidence
In a REWIND exploratory substudy, dulaglutide did NOT significantly reduce incident substantive cognitive impairment on the prespecified primary (unadjusted) analysis;...
SourceCukierman-Yaffe T, Gerstein HC, Colhoun HM, et al. Effect of dulaglutide on cognitive imp...Source
Full findingIn a REWIND exploratory substudy, dulaglutide did NOT significantly reduce incident substantive cognitive impairment on the prespecified primary (unadjusted) analysis; significance appeared only in a post-hoc baseline-score-adjusted analysis.
PopulationREWIND cognition substudy: N=8828 with baseline plus follow-up MoCA/DSST (of 9901); type 2 diabetes; double-blind placebo-controlled randomised trial; median 5.4 years; NCT01394952.
Fundingindustry - Eli Lilly (trial sponsor; inferred from registration trial)
Comparatorsplacebo
LiraglutideMARKETED
safety-otherDecrease
Very low evidence
In an HFE-knockout haemochromatosis mouse model, liraglutide lowered both serum and hepatic iron through a hepcidin-independent mechanism (Hamp not differentially...
SourceBozadjieva-Kramer N et al. Liraglutide lowers iron in HFE-knockout mice independent of he...Source
Full findingIn an HFE-knockout haemochromatosis mouse model, liraglutide lowered both serum and hepatic iron through a hepcidin-independent mechanism (Hamp not differentially expressed), arguing for a genuine iron-handling effect rather than only an acute-phase change.
PopulationHFE-knockout (HFE-KO) haemochromatosis mice; hepatic and serum iron, Hamp/hepcidin expression measured. Murine model; no human read-across.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsvehicle-treated HFE-KO mice
LiraglutideMARKETED
safety-otherNo change
High evidence
Rodent C-cell tumour/calcitonin signal does not translate: human sequential calcitonin screening over up to 2 yr in >5000 liraglutide/control subjects showed no...
SourceHegedus L et al., J Clin Endocrinol Metab 2011Source
Full findingRodent C-cell tumour/calcitonin signal does not translate: human sequential calcitonin screening over up to 2 yr in >5000 liraglutide/control subjects showed no consistent dose/time-dependent rise and no between-group difference.
Population>5000 subjects with T2D or non-diabetic obesity; calcitonin at 3-month intervals up to 2 yr
Fundingindustry - Novo Nordisk (trial sponsor)
Comparatorscontrol therapy
LiraglutideMARKETED
safety-otherNot directional
High evidence
The pivotal-trial investigators themselves concluded that, despite the metabolic benefits, the increased hypoglycaemia and hyperglycaemia-with-ketosis limit the...
SourceMathieu C, Zinman B, Hemmingsson JU, et al. Efficacy and Safety of Liraglutide Added to I...Source
Full findingThe pivotal-trial investigators themselves concluded that, despite the metabolic benefits, the increased hypoglycaemia and hyperglycaemia-with-ketosis limit the clinical use of liraglutide as an insulin adjunct in T1D - the honest net verdict.
Fundingindustry - Novo Nordisk (sponsor; NCT01836523)
Comparatorsplacebo (insulin background)
LiraglutideMARKETED
safety-otherIncrease
High evidence
Liraglutide added to insulin significantly increased the rate of symptomatic hypoglycaemia in type-1 diabetes - a disease-specific harm that accompanies the modest...
SourceAhren B, Hirsch IB, Pieber TR, et al. Efficacy and Safety of Liraglutide Added to Capped ...Source
Full findingLiraglutide added to insulin significantly increased the rate of symptomatic hypoglycaemia in type-1 diabetes - a disease-specific harm that accompanies the modest efficacy.
Fundingindustry - Novo Nordisk (sponsor; NCT02098395)
Comparatorsplacebo (insulin background)
LiraglutideMARKETED
safety-otherIncrease
High evidence
Both pivotal trials showed a dose-dependent increase in hyperglycaemia-with-ketosis at the highest liraglutide dose (event rate roughly doubled) - the real,...
SourceMathieu C, Zinman B, Hemmingsson JU, et al. Efficacy and Safety of Liraglutide Added to I...Source
Full findingBoth pivotal trials showed a dose-dependent increase in hyperglycaemia-with-ketosis at the highest liraglutide dose (event rate roughly doubled) - the real, disease-specific DKA-adjacent safety signal that, with the hypoglycaemia, limits clinical use.
Fundingindustry - Novo Nordisk (sponsor; NCT01836523)
Comparatorsplacebo (insulin background)
OrforglipronMARKETED
safety-otherIncrease
High evidence
AE-related discontinuation rose with orforglipron dose; GI events the commonest AEs, mostly mild-to-moderate, in escalation.
SourceLilly ATTAIN-1 (NEJM 2025) and ATTAIN-2 resultsSource
PopulationATTAIN-1 (obesity, no diabetes) and ATTAIN-2 (obesity + T2D) phase 3 trials
Fundingindustry - Eli Lilly
Comparatorsplacebo
RetatrutideINVESTIGATIONAL
safety-otherDecrease
Moderate evidence
In the JCEM metabolite substudy of the two phase-2 retatrutide trials, retatrutide lowered circulating uric acid dose-dependently versus baseline and placebo,...
SourcePearson SM et al., J Clin Endocrinol Metab 2026 (metabolomics); PubMed null result 2026 (...Source
Full findingIn the JCEM metabolite substudy of the two phase-2 retatrutide trials, retatrutide lowered circulating uric acid dose-dependently versus baseline and placebo, consistent with the weight-mediated class pattern.
PopulationRetatrutide phase-2 obesity + T2D metabolomics substudy (Pearson et al., post hoc of NCT04881760 / NCT05929066).
Fundingindustry - Eli Lilly and Company
Comparatorsplacebo
RetatrutideINVESTIGATIONAL
safety-otherNot directional
Moderate evidence
EXPLICIT ABSENCE: the retatrutide phase-2a MASLD trial (Sanyal 2024) did not report any uric-acid or gout endpoint.
SourceSanyal AJ et al. Triple hormone receptor agonist retatrutide for metabolic dysfunction-as...Source
PopulationRetatrutide phase-2a MASLD substudy (Sanyal et al., NCT04881760); reviewed for any uric-acid/gout endpoint - none present.
Fundingindustry-Eli Lilly
Comparatorsplacebo
RetatrutideINVESTIGATIONAL
safety-otherNot directional
Moderate evidence
EXPLICIT ABSENCE: the retatrutide phase-2 obesity trial (Jastreboff 2023) did not report any uric-acid or gout endpoint; the only retatrutide urate data come from the...
SourceJastreboff AM et al. (retatrutide phase 2), NEJM, 2023;389:514-526Source
Full findingEXPLICIT ABSENCE: the retatrutide phase-2 obesity trial (Jastreboff 2023) did not report any uric-acid or gout endpoint; the only retatrutide urate data come from the separate JCEM metabolite substudy.
PopulationRetatrutide phase-2 obesity trial (Jastreboff et al., NCT04881760); reviewed for any uric-acid/gout endpoint - none present in the trial report.
Fundingindustry-Eli Lilly
Comparatorsplacebo
RetatrutideINVESTIGATIONAL
safety-otherNot directional
Moderate evidence
EXPLICIT ABSENCE: the retatrutide phase-2a MASLD trial did not report any iron, ferritin, hepcidin, anaemia or haemoglobin endpoint; its biomarker panel was...
SourceSanyal AJ et al. Triple hormone receptor agonist retatrutide for metabolic dysfunction-as...Source
Full findingEXPLICIT ABSENCE: the retatrutide phase-2a MASLD trial did not report any iron, ferritin, hepcidin, anaemia or haemoglobin endpoint; its biomarker panel was ALT/AST/FIB-4/K-18/ELF/pro-C3 and no iron variable appears anywhere, including the adverse-event table.
PopulationRetatrutide phase-2a MASLD substudy (Sanyal et al., NCT04881760); reviewed for iron/ferritin/hepcidin/anaemia/haemoglobin reporting - none present.
Fundingindustry-Eli Lilly
Comparatorsplacebo
RetatrutideINVESTIGATIONAL
safety-otherNot directional
Moderate evidence
EXPLICIT ABSENCE: the retatrutide phase-2 obesity trial did not report any iron, ferritin, hepcidin or anaemia endpoint; its adverse events were gastrointestinal plus...
SourceJastreboff AM et al. (retatrutide phase 2), NEJM, 2023;389:514-526Source
Full findingEXPLICIT ABSENCE: the retatrutide phase-2 obesity trial did not report any iron, ferritin, hepcidin or anaemia endpoint; its adverse events were gastrointestinal plus a dose-dependent heart-rate increase only, with no iron variable reported.
PopulationRetatrutide phase-2 obesity trial (Jastreboff et al., NCT04881760); reviewed for iron/ferritin/hepcidin/anaemia reporting - none present.
Fundingindustry-Eli Lilly
Comparatorsplacebo
RetatrutideINVESTIGATIONAL
safety-otherNot directional
Moderate evidence
EXPLICIT ABSENCE: the retatrutide phase-2 obesity trial did not report any non-iron micronutrient (vitamin D/B12/folate/magnesium/calcium) endpoint - no retatrutide...
SourceJastreboff AM et al. (retatrutide phase 2), NEJM, 2023;389:514-526Source
Full findingEXPLICIT ABSENCE: the retatrutide phase-2 obesity trial did not report any non-iron micronutrient (vitamin D/B12/folate/magnesium/calcium) endpoint - no retatrutide micronutrient effect can be asserted or excluded.
PopulationRetatrutide phase-2 obesity trial (Jastreboff et al., NCT04881760); reviewed for any non-iron micronutrient (vitamin D/B12/folate/Mg/Ca) endpoint - none present.
Fundingindustry-Eli Lilly
Comparatorsplacebo
RetatrutideINVESTIGATIONAL
safety-otherMixed
Moderate evidence
RETATRUTIDE DIRECT SAFETY position: across the two phase-2 trials the profile is GI-DOMINANT + a DOSE-DEPENDENT HEART-RATE rise, with NO MTC/pancreatitis/NAION/biliary...
SourceJastreboff AM et al. (retatrutide phase 2), NEJM, 2023;389:514-526Source
Full findingRETATRUTIDE DIRECT SAFETY position: across the two phase-2 trials the profile is GI-DOMINANT + a DOSE-DEPENDENT HEART-RATE rise, with NO MTC/pancreatitis/NAION/biliary events reported - but these are SHORT (36-48 wk), small per-arm trials, UNDERPOWERED for rare events, so the absence of a signal is NOT clearance. Reta has NO long-term safety data (phase-3 TRIUMPH ongoing).
FOLK-CLAIM VERDICT ('retatrutide-specific dysesthesia / skin tingling-buzzing, in 5-15% of users, not in the other two'): TRUE / CONFIRMED - and it OVERTURNS the prior...
SourceJastreboff AM et al. (retatrutide phase 2), NEJM, 2023;389:514-526Source
Full findingFOLK-CLAIM VERDICT ('retatrutide-specific dysesthesia / skin tingling-buzzing, in 5-15% of users, not in the other two'): TRUE / CONFIRMED - and it OVERTURNS the prior assumption that this was baseless folk chatter. Dysesthesia is a NAMED, DOSE-DEPENDENT, placebo-differentiated, retatrutide-CLASS-SPECIFIC sensory signal (no GLP-1/GIP equivalent), surfaced by the folk-claims loop and NOT previously in the safety dossier (D6/Thread 14). The lay '5-15%' figure if anything UNDERSTATES the 12 mg rate (~21%). Mechanism is UNPROVEN (a glucagon-receptor effect is the leading hypothesis given dose-dependence + class-specificity; rapid weight loss / B-vitamin / electrolyte contributions are partial confounders at best).
PopulationRetatrutide phase-2 obesity trial (skin hyperesthesia ~7%) + phase-3 TRIUMPH-4 topline (dysesthesia 8.8/20.9% at 9/12 mg); no comparable signal for semaglutide or tirzepatide.
Fundingindustry-Eli Lilly
Comparatorsplacebo
SemaglutideMARKETED
safety-otherNot directional
High evidence
Semaglutide is contraindicated in the US for personal/family history of MTC or MEN 2 (also a boxed warning) and for serious hypersensitivity; the EU contraindication...
SourceOzempic (semaglutide) US Prescribing Information, Novo NordiskSource
Full findingSemaglutide is contraindicated in the US for personal/family history of MTC or MEN 2 (also a boxed warning) and for serious hypersensitivity; the EU contraindication is hypersensitivity only (the MTC/MEN2 boxed warning is a US construct).
In 51 people with type-2 diabetes, 10 weeks of semaglutide attenuated (did not abolish) the post-oral-iron-load serum iron rise, consistent with reduced intestinal...
SourceMelis M et al. Effect of semaglutide on intestinal iron absorption in type-2 diabetes: a ...Source
Full findingIn 51 people with type-2 diabetes, 10 weeks of semaglutide attenuated (did not abolish) the post-oral-iron-load serum iron rise, consistent with reduced intestinal iron absorption on the drug.
Population51 people with type-2 diabetes; single-arm uncontrolled observational pre-post study (no control arm), 10 weeks; post-oral-iron-load serum-iron and TSAT measurement (NCT06629688).
Fundingindustry - Novo Nordisk (trial sponsor; inferred from registration trial)
Comparatorswithin-subject baseline (no control arm)
SemaglutideMARKETED
safety-otherDecrease
Low evidence
The authors attribute the reduced iron absorption (author-hedged 'may') to delayed gastric emptying / altered GI motility, warn it could contribute to iron deficiency...
SourceMelis M et al. Effect of semaglutide on intestinal iron absorption in type-2 diabetes: a ...Source
Full findingThe authors attribute the reduced iron absorption (author-hedged 'may') to delayed gastric emptying / altered GI motility, warn it could contribute to iron deficiency or anaemia in susceptible patients, and suggest monitoring iron status.
Population51 people with type-2 diabetes; single-arm uncontrolled observational pre-post study (no control arm), 10 weeks; post-oral-iron-load serum-iron and TSAT measurement (NCT06629688).
Fundingindustry - Novo Nordisk (trial sponsor; inferred from registration trial)
Comparatorswithin-subject baseline (no control arm)
SemaglutideMARKETED
safety-otherMixed
Moderate evidence
In obesity without diabetes, semaglutide increased gallbladder disorders (esp. cholelithiasis) >2.6-fold, whereas tirzepatide showed no significant biliary risk;...
Full findingIn obesity without diabetes, semaglutide increased gallbladder disorders (esp. cholelithiasis) >2.6-fold, whereas tirzepatide showed no significant biliary risk; neither significantly increased hepatic or pancreatic AEs.
Population13 RCTs, 26,894 obese participants without diabetes; meta-analysis
Fundingacademic/independent - no funding (None)
Comparatorsplacebo
SemaglutideMARKETED
safety-otherIncrease
Very low evidence
VigiBase disproportionality identified a potential signal between semaglutide and bile-duct cancer (cholangiocarcinoma); pancreatic cancer was the most frequent...
SourceKaur RJ et al., Indian J Gastroenterol 2026Source
Full findingVigiBase disproportionality identified a potential signal between semaglutide and bile-duct cancer (cholangiocarcinoma); pancreatic cancer was the most frequent neoplasm reported.
PopulationVigiBase ICSRs 1 Jan 2009-31 Jul 2023; disproportionality
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsVigiBase background
SemaglutideMARKETED
safety-otherIncrease
Low evidence
Index signal-raising cohort: patients on semaglutide had markedly higher cumulative NAION incidence than those on non-GLP-1 RA medications, in both T2D and...
SourceHathaway JT et al., JAMA Ophthalmol 2024Source
Full findingIndex signal-raising cohort: patients on semaglutide had markedly higher cumulative NAION incidence than those on non-GLP-1 RA medications, in both T2D and overweight/obese populations.
PopulationRetrospective matched cohort, single neuro-ophthalmology registry (Mass Eye and Ear); 710 T2D, 979 overweight/obese; Dec 2017-Nov 2023
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsnon-GLP-1 RA antidiabetics; non-GLP-1 RA weight-loss meds
PopulationTriNetX; 174,584 semaglutide vs 174,584 non-GLP-1 RA with diabetes; Oct 2019-Dec 2023
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsnon-GLP-1 RA antidiabetics
SemaglutideMARKETED
safety-otherDecrease
Low evidence
REFUTING/NULL pole: in a large racially diverse Military Health System cohort, semaglutide associated with LOWER odds of NAION in T2D and no significant difference in...
Full findingREFUTING/NULL pole: in a large racially diverse Military Health System cohort, semaglutide associated with LOWER odds of NAION in T2D and no significant difference in overweight/obesity.
PopulationMilitary Health System beneficiaries; 973,529 T2D + 239,246 overweight/obese; Dec 2017-Sep 2023
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsnon-GLP-1 RA medications; PDE-5 inhibitors (subset)
SemaglutideMARKETED
safety-otherIncrease
Moderate evidence
SIGNAL-confirming meta-analysis (T2D-restricted): semaglutide associated with increased NAION risk vs non-GLP-1RAs, persisting across follow-up windows.
SourceLampsas S et al., Graefes Arch Clin Exp Ophthalmol 2026Source
PopulationMeta-analysis; 8 retrospective cohorts (14,255,247 participants), mean age 59.3 yr
Fundingacademic / investigator-initiated; no industry sponsor found in indexed metadata (Europe PMC fundingList empty)
Comparatorsnon-GLP-1 RA medications
SemaglutideMARKETED
safety-otherIncrease
High evidence
Regulatory: EMA PRAC concluded its review and recommended adding NAION to semaglutide product information as a 'very rare' side effect; WHO issued a confirmatory...
SourceEMA PRAC highlights 2-5 June 2025; WHO safety communication 27 June 2025Source
Full findingRegulatory: EMA PRAC concluded its review and recommended adding NAION to semaglutide product information as a 'very rare' side effect; WHO issued a confirmatory signal.
PopulationEU regulatory review of all NAION data for semaglutide
Fundingregulatory agency (FDA/EMA/MHRA/WHO)
GLP-1-based therapies, multi-agentSTATUS UNKNOWN
safety-otherMixed
Moderate evidence
A randomised-trials network meta-analysis found intestinal-obstruction risk not elevated for most GLP-1 agents (liraglutide possibly protective), corroborating the...
SourceChen JJ et al. Risk of intestinal obstruction with GLP-1 receptor agonists: a randomised-...Source
Full findingA randomised-trials network meta-analysis found intestinal-obstruction risk not elevated for most GLP-1 agents (liraglutide possibly protective), corroborating the weak/very-low post-marketing spontaneous-report obstruction signal.
PopulationRCT network meta-analysis (Chen et al., Int J Mol Sci 2026; the anchor source) of intestinal-obstruction risk across GLP-1 agents, read against the weaker post-marketing spontaneous-report signal.
Fundingacademic/independent - no external funding
Comparatorsplacebo/active comparators in RCTs; post-marketing spontaneous-report background
SemaglutideMARKETED
safety-otherDecrease
Low evidence
Real-world propensity-matched cohort (TriNetX EHR): semaglutide associated with LOWER risk of incident and recurrent suicidal ideation vs non-GLP1R anti-obesity meds;...
SourceWang W, Volkow ND, Berger NA, Davis PB, Kaelber DC, Xu R. Nat Med 2024;30(1):168-176Source
Full findingReal-world propensity-matched cohort (TriNetX EHR): semaglutide associated with LOWER risk of incident and recurrent suicidal ideation vs non-GLP1R anti-obesity meds; replicated in T2DM.
Across the STEP programme (RCT data), depression-symptom endorsement on PHQ-9 similar/lower for semaglutide 2.4 mg vs placebo; pooled analysis found no clinically...
SourceWadden TA et al., JAMA Intern Med 2024 (STEP 1/2/3/5 psychiatric post hoc)Source
Full findingAcross the STEP programme (RCT data), depression-symptom endorsement on PHQ-9 similar/lower for semaglutide 2.4 mg vs placebo; pooled analysis found no clinically meaningful difference in incident significant depressive symptoms and no PHQ-9 worsening.
PopulationPost-hoc analysis of the pooled STEP 1/2/3/5 placebo-controlled RCTs in adults with obesity (PHQ-9 psychiatric-safety endpoints).
Fundingindustry-Novo Nordisk
Comparatorsplacebo
SemaglutideMARKETED
safety-otherIncrease
High evidence
FDA boxed warning: in mice and rats semaglutide caused dose- and treatment-duration-dependent thyroid C-cell tumours (adenomas/carcinomas) at clinically relevant...
SourceOzempic/Wegovy (semaglutide) US Prescribing Information, FDASource
Full findingFDA boxed warning: in mice and rats semaglutide caused dose- and treatment-duration-dependent thyroid C-cell tumours (adenomas/carcinomas) at clinically relevant exposures; contraindicated with personal/family history of MTC or MEN 2. Human relevance not determined.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
ComparatorsFAERS background
SemaglutideMARKETED
safety-otherIncrease
Low evidence
One year after withdrawal of semaglutide 2.4 mg plus lifestyle intervention, participants regained about two-thirds of lost weight and cardiometabolic improvements...
SourceWilding JPH et al., Diabetes Obes Metab 2022Source
Full findingOne year after withdrawal of semaglutide 2.4 mg plus lifestyle intervention, participants regained about two-thirds of lost weight and cardiometabolic improvements reverted towards baseline.
Diabetic retinopathy complications significantly more frequent with semaglutide than placebo in SUSTAIN-6 (early-worsening signal).
SourceMarso SP et al. (SUSTAIN-6), NEJM, 2016Source
PopulationSUSTAIN-6 RCT; 3297 T2D at high CV risk; semaglutide 0.5/1.0 mg vs placebo, 104 weeks
Fundingindustry - Novo Nordisk
Comparatorsplacebo
SemaglutideMARKETED
safety-otherIncrease
High evidence
In the cardiovascular-outcomes trial, Wegovy-treated patients aged >=75 reported numerically more hip/pelvis fractures than placebo (2.4% vs 0.6%); small numerators,...
SourceWegovy (semaglutide) US Prescribing Information, Novo NordiskSource
Full findingIn the cardiovascular-outcomes trial, Wegovy-treated patients aged >=75 reported numerically more hip/pelvis fractures than placebo (2.4% vs 0.6%); small numerators, post-hoc, causality NOT established.
A network meta-analysis of randomised outcome trials found SGLT2 inhibitors significantly reduce incident gout while GLP-1 RAs and DPP-4 inhibitors are neutral - the...
SourceWang A, Shi W, Zhang N, Tang H, Feng X. Newer glucose-lowering drugs and risk of gout: a ...Source
Full findingA network meta-analysis of randomised outcome trials found SGLT2 inhibitors significantly reduce incident gout while GLP-1 RAs and DPP-4 inhibitors are neutral - the contrast pole showing a genuinely uricosuric, gout-protective class distinct from GLP-1 RAs.
Tirzepatide is contraindicated in the US for personal/family history of MTC or MEN 2 (also a boxed warning) and serious hypersensitivity; the EU contraindication is...
SourceMounjaro (tirzepatide) US Prescribing Information, Eli Lilly; FDA/MHRA/TGA advisoriesSource
Full findingTirzepatide is contraindicated in the US for personal/family history of MTC or MEN 2 (also a boxed warning) and serious hypersensitivity; the EU contraindication is hypersensitivity only.
In Zepbound trials the majority of patients developed anti-tirzepatide antibodies; neutralising antibodies were uncommon in the weight-management pool (2.7-2.8%) and...
SourceZepbound (tirzepatide) US Prescribing Information, Eli LillySource
Full findingIn Zepbound trials the majority of patients developed anti-tirzepatide antibodies; neutralising antibodies were uncommon in the weight-management pool (2.7-2.8%) and none were detected in the OSA pool; no clinically significant PK/efficacy effect.
Tirzepatide has no established link to elevated ferritin per trial data and the SmPC, and no known direct effect on iron metabolism; elevated ferritin observed in...
SourceUrbina et al., GLP-1 receptor agonists and iron status, Clin Obes 2026;16:e70070Source
Full findingTirzepatide has no established link to elevated ferritin per trial data and the SmPC, and no known direct effect on iron metabolism; elevated ferritin observed in users reflects underlying metabolic, inflammatory or liver disease (ferritin as an acute-phase reactant), not a drug effect.
PopulationSynthesised from the Urbina narrative review plus non-measurement of iron in SURPASS/SURMOUNT; tirzepatide iron metabolism not directly studied.
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorsnot applicable
Tirzepatide and semaglutide comparisonSTATUS UNKNOWN
safety-otherMixed
Very low evidence
FAERS comparative disproportionality found acute kidney injury reported LESS frequently with tirzepatide than semaglutide.
SourceComparative Renal Safety of Tirzepatide and Semaglutide: FAERS Disproportionality Study 2025Source
PopulationFAERS Jan 2022-Sep 2025, 133,872 reports (92,807 tirzepatide; 41,065 semaglutide)
Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)
Comparatorssemaglutide
TirzepatideMARKETED
safety-otherMixed
Moderate evidence
Composite gallbladder/biliary disease significantly associated with tirzepatide; pancreatitis and individual biliary endpoints NOT significantly increased.
Tirzepatide may reduce ORAL hormonal contraceptive efficacy via delayed gastric emptying; US labelling advises switching to a non-oral method or adding a barrier...
SourceMounjaro (tirzepatide) US Prescribing Information, Eli Lilly; FDA/MHRA/TGA advisoriesSource
Full findingTirzepatide may reduce ORAL hormonal contraceptive efficacy via delayed gastric emptying; US labelling advises switching to a non-oral method or adding a barrier method for 4 weeks after initiation and after each dose escalation.
PopulationRegulatory label (FDA Mounjaro/Zepbound PI; EU SmPC); PK interaction basis
Fundingregulatory agency (FDA/EMA/MHRA/WHO)
TirzepatideMARKETED
safety-otherNo change
Moderate evidence
Post hoc pooled analysis of SURMOUNT-1/-2/-3 (RCT data): no clinically meaningful PHQ-9 difference vs placebo over 72 weeks; suicidal ideation/behaviour (C-SSRS) low...
SourceWadden TA et al., 'Psychiatric Safety of Tirzepatide... Post Hoc Analysis of SURMOUNT'. O...Source
Full findingPost hoc pooled analysis of SURMOUNT-1/-2/-3 (RCT data): no clinically meaningful PHQ-9 difference vs placebo over 72 weeks; suicidal ideation/behaviour (C-SSRS) low and similar, though a numerically greater percentage of tirzepatide participants reported moderate-risk ideation.
PopulationPooled SURMOUNT-1/-2/-3, obesity, no known major psychopathology; 72 weeks; placebo-controlled RCTs
Fundingindustry-Eli Lilly
Comparatorsplacebo
TirzepatideMARKETED
safety-otherIncrease
High evidence
FDA boxed warning: in a 2-year rat study tirzepatide caused dose- and treatment-duration-dependent thyroid C-cell tumours at clinically relevant exposures;...
SourceMounjaro (tirzepatide) US Prescribing Information, Eli Lilly; FDA/MHRA/TGA advisoriesSource
Full findingFDA boxed warning: in a 2-year rat study tirzepatide caused dose- and treatment-duration-dependent thyroid C-cell tumours at clinically relevant exposures; contraindicated with personal/family history of MTC or MEN 2. Human relevance not determined.
Population2-year rat carcinogenicity study; regulatory label
Fundingregulatory agency (FDA/EMA/MHRA/WHO)
Comparatorsvehicle control
TirzepatideMARKETED
safety-otherDecrease
Moderate evidence
In a network meta-analysis, tirzepatide produced the greatest reduction in both fat mass and (absolute) lean body mass.
SourceRakhsha SS et al., Diabetes Obes Metab 2026Source
PopulationNetwork meta-analysis, 41 RCTs, 2906 participants, through Mar 2025
Fundingacademic - Tehran University of Medical Sciences (Chronic Diseases Research Center; grant 1403-04-75722-221); no industry sponsor
Withdrawing tirzepatide after a 36-week lead-in led to substantial weight REGAIN, whereas continued treatment maintained/augmented loss (randomized-withdrawal design).
SourceAronne LJ et al., JAMA 2024 (SURMOUNT-4)Source
PopulationSURMOUNT-4 (NCT04660643), phase 3 randomized-withdrawal, 670 adults with obesity/overweight (no diabetes), 52-wk double-blind after open-label lead-in
Fundingindustry-Eli Lilly
Comparatorsplacebo (withdrawal)
TirzepatideMARKETED
safety-otherDecrease
Moderate evidence
In a post hoc analysis of the SURMOUNT-1 obesity RCT, tirzepatide lowered serum uric acid dose-dependently versus placebo over 72 weeks, with the effect independent of...
SourceSattar N et al. Tirzepatide and change in uric acid and its association with weight reduc...Source
Full findingIn a post hoc analysis of the SURMOUNT-1 obesity RCT, tirzepatide lowered serum uric acid dose-dependently versus placebo over 72 weeks, with the effect independent of baseline uric-acid level or BMI.
PopulationTirzepatide phase-3 obesity RCT post hoc (Sattar et al., SURMOUNT-1, NCT04184622); 2,539 adults with obesity/overweight + >=1 weight-related complication.
Fundingindustry - Eli Lilly (trial sponsor; inferred from registration trial)
Comparatorsplacebo
TirzepatideMARKETED
safety-otherDecrease
Moderate evidence
A mediation analysis in the same SURMOUNT-1 post hoc found that about 73% of tirzepatide's serum-uric-acid reduction was explained by weight loss - implicating weight...
SourceSattar N et al. Tirzepatide and change in uric acid and its association with weight reduc...Source
Full findingA mediation analysis in the same SURMOUNT-1 post hoc found that about 73% of tirzepatide's serum-uric-acid reduction was explained by weight loss - implicating weight loss rather than a direct uricosuric drug effect.
PopulationTirzepatide phase-3 obesity RCT post hoc (Sattar et al., SURMOUNT-1, NCT04184622); mediation analysis of weight change vs SUA change.
Fundingindustry - Eli Lilly (trial sponsor; inferred from registration trial)
Comparatorsplacebo
AleniglipronINVESTIGATIONAL
safety-otherNo change
Moderate evidence
No drug-induced liver injury events reported in the phase 2b trial.
SourceRosenstock J et al. Oral small molecule GLP-1 receptor agonist aleniglipron in people wit...Source
The Echo Compendium