Skip to main content
NCBI home page
Preventive Medicine Reports logoLink to Preventive Medicine Reports
. 2026 Jun 20;68:103548. doi: 10.1016/j.pmedr.2026.103548

Glucagon-like peptide-1 receptor agonist use and risk of gynecologic cancers: a meta-analysis of multinational real-world cohort studies

Rongxia Li a,b,1, Xuewei Zhao c,1, Jinhai Shen d,e,f, Bo Luo d, Yongmei Wang g,⁎, Yancang Duan h,i,j,⁎⁎
PMCID: PMC13316655  PMID: 42381894

Abstract

Objective

Glucagon-like peptide-1 receptor agonists (GLP-1RAs) are widely used for the management of type 2 diabetes mellitus (T2DM) and obesity. Given the established links between metabolic dysfunction and gynecologic malignancies, the potential association between GLP-1RA use and gynecologic cancer risk remains clinically relevant but incompletely defined. We conducted a meta-analysis of real-world cohort studies to evaluate this association.

Methods

We systematically searched PubMed (from 1966) and Embase (from 1974) through February 15, 2026 for observational cohort studies assessing the association between (GLP-1 RA) use and gynecologic cancer outcomes. Hazard ratios (HRs) with 95% confidence intervals (CIs) were pooled using random-effects models.

Results

Nine retrospective cohort studies were included. GLP-1RA use was not significantly associated with gynecologic cancer risk (HR 0.80; 95% CI: 0.63, 1.01). In cancer subtype analyses, a reduced risk was observed for ovarian cancer (HR 0.68; 95% CI: 0.49, 0.93), whereas associations for endometrial, cervical, and other gynecologic cancers were not statistically significant.

Conclusions

GLP-1RA use was not significantly associated with overall gynecologic cancer risk, although potential reductions were observed in obesity populations and for ovarian cancer. Given substantial heterogeneity and the observational nature of the evidence, these findings should be interpreted cautiously and considered hypothesis-generating.

Keywords: Glucagon-like peptide-1 receptor agonists, Gynecologic cancers, Endometrial cancer, Ovarian cancer, Meta-analysis

Highlights

  • •

    No link between glucagon-like peptide-1 drugs and gynecologic cancer risk.

  • •

    Glucagon-like peptide-1 receptor drug use linked to reduced ovarian cancer risk.

  • •

    Risk reduction was observed in obese populations but not in diabetes cohorts.

1. Introduction

Obesity and type 2 diabetes mellitus (T2DM) are well-established risk factors for multiple malignancies, particularly gynecologic cancers such as endometrial and ovarian cancer (Pati et al., 2023; Espinoza et al., 2026). Among these, endometrial cancer exhibits one of the strongest associations with excess adiposity, hyperinsulinemia, and unopposed estrogen exposure (Kiesel et al., 2020; Huang et al., 2023). Obesity-related chronic inflammation, insulin resistance, and hormonal dysregulation are also implicated in ovarian carcinogenesis (Shea et al., 2023; Liu et al., 2023; Rubinstein et al., 2021). In parallel with the global rise in obesity and T2DM prevalence, the incidence of obesity-related cancers continues to increase, underscoring the importance of understanding how metabolic therapies may influence long-term cancer risk (Berrington de Gonzalez et al., 2025; Feng et al., 2025; Chen et al., 2024).

Glucagon-like peptide-1 receptor agonists (GLP-1RAs) have emerged as highly effective agents for glycemic control and weight reduction, with rapidly expanding indications in both T2DM and obesity management (Wong et al., 2025; Drucker, 2024). Beyond their metabolic effects, experimental studies suggest that glucagon-like peptide-1 signaling may influence cellular proliferation, apoptosis, oxidative stress, and inflammatory pathways—mechanisms potentially relevant to carcinogenesis (Lin et al., 2025; Lucente et al., 2025). These biological observations have generated substantial interest regarding the direction and magnitude of any association between GLP-1RA exposure and cancer risk.

Several meta-analyses of randomized controlled trials (RCTs) have evaluated the overall malignancy risk associated with GLP-1RAs and generally reported no significant increase in cancer incidence (Silverii et al., 2025; Ko et al., 2026). However, cancer outcomes were not primary endpoints in these trials. RCTs are inherently limited by relatively short follow-up durations, highly selected patient populations, and inadequate statistical power to detect differences in site-specific cancers, which occur infrequently within trial settings. Consequently, the ability of RCT data to inform long-term, real-world cancer risk remains constrained.

Real-world evidence derived from large healthcare databases and population-based cohorts offers complementary insights into long-term safety. Observational studies typically include broader and more heterogeneous populations, longer exposure periods, and substantially larger sample sizes, thereby enhancing the capacity to evaluate site-specific cancer outcomes. In recent years, multiple large-scale propensity score–matched cohort studies have examined the association between GLP-1RA use and obesity-related cancers, including endometrial and ovarian cancer (Dai et al., 2025; Ipaye et al., 2025; Wang et al., 2024). However, their findings have been inconsistent, with reported associations varying according to comparator selection, baseline patient characteristics, and analytic strategies.

To date, no comprehensive meta-analysis has systematically synthesized real-world cohort evidence to clarify the association between GLP-1RA exposure and the risk of gynecologic cancers while accounting for differences in study design and comparator groups. Given the rapidly expanding and long-term use of GLP-1RAs in both diabetes and obesity management, a rigorous evaluation of observational data is essential to inform clinical decision-making, pharmacovigilance, and regulatory assessment.

Therefore, we conducted a systematic review and meta-analysis of real-world cohort studies to evaluate the association between GLP-1RA use and the risk of gynecologic cancers, with particular emphasis on endometrial and ovarian cancer. We aimed to generate pooled site-specific risk estimates, explore sources of heterogeneity through pre-specified subgroup analyses, and assess the consistency of real-world findings with prior randomized evidence.

2. Methods

2.1. Study guideline and registration

This meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 statement (Page et al., 2021). The study protocol was prospectively registered in the International Prospective Register of Systematic Reviews (registration number: CRD420261333977).

2.2. Data sources and search strategy

A comprehensive literature search was performed in PubMed (from 1966) and Embase (from 1974) through February 15, 2026. The search strategy combined controlled vocabulary terms (MeSH and Emtree) and free-text keywords related to:“glucagon-like peptide-1 receptor agonist” and “gynecologic cancer” OR “endometrial cancer” OR “ovarian cancer” OR “cervical cancer”. The detailed search strategy for each database is provided in Supplementary Table 1.

2.3. Eligibility criteria and study selection

Observational cohort studies based on real-world data sources were considered eligible if they evaluated adult patients with T2DM and/or obesity and examined the association between GLP-1RA use and incident gynecologic cancers. Eligible studies were required to report adjusted hazard ratios (HRs) with corresponding 95% confidence intervals (CIs). RCTs, case–control studies, cross-sectional analyses, conference abstracts without full-text data, and duplicate reports from the same cohort were excluded. When multiple publications were derived from the same underlying database, the most comprehensive or recent study with the longest follow-up was retained.

Titles and abstracts were independently screened by two investigators, followed by full-text evaluation of potentially relevant articles. Discrepancies were resolved through discussion and, when necessary, consultation with a third reviewer.

2.4. Data extraction and quality assessment

Data were independently extracted by two investigators using a standardized form. Extracted information included study characteristics, population features, exposure definitions, comparator groups, and the most fully adjusted HRs along with covariates included in multivariable models. When multiple adjusted models were reported, the estimate with the most comprehensive adjustment was selected for analysis. In studies reporting multiple comparator groups for the same cancer outcome, only one effect estimate per outcome was included in the primary meta-analysis according to a pre-specified hierarchy of comparator selection to avoid double counting of participants and preserve statistical independence.

Study quality was assessed using the Newcastle–Ottawa Scale (NOS) for cohort studies (Stang, 2010). Studies scoring ≥7 points were considered high quality. Quality assessment was conducted independently by two reviewers, with discrepancies resolved by consensus.

2.5. Statistical analysis

Pooled HRs were calculated using a random-effects model (DerSimonian and Laird method) to account for between-study heterogeneity. Statistical heterogeneity was assessed using the Cochran Q test and quantified with the I2 statistic. I2 values of 25%, 50%, and 75% were interpreted as low, moderate, and high heterogeneity, respectively.

The primary analysis evaluated overall gynecologic cancer risk. Because gynecologic malignancies comprise biologically distinct disease entities, separate analyses were also performed according to cancer subtype whenever sufficient data were available.

Pre-specified subgroup analyses were performed to explore potential effect modification according to underlying population characteristics and specific GLP-1RA agents when sufficient data were available. Differences between subgroups were evaluated using interaction tests based on between-group heterogeneity. Sensitivity analyses were conducted by sequentially omitting individual studies to assess the robustness of pooled HRs. Publication bias was planned to be evaluated using funnel plots and Egger's regression test when at least ten studies were available for a given outcome.

All statistical analyses were performed using R software (version 4.5.1). A two-sided P value <0.05 was considered statistically significant.

2.6. Ethical approval

This study was based on publicly available anonymized data from previously published studies; therefore, institutional review board approval and informed consent were not required.

3. Results

3.1. Study selection and characteristics of included studies

The literature search identified 410 records, of which 56 were removed as duplicates. After title and abstract screening, 12 articles underwent full-text review, and 9 retrospective cohort studies met the eligibility criteria and were included in the meta-analysis (Dai et al., 2025; Ipaye et al., 2025; Wang et al., 2024; Abdelgadir et al., 2025; Chuang et al., 2025; Levy et al., 2024; Lu et al., 2025; Rothman et al., 2025; Yen et al., 2026). The study selection process is illustrated in Fig. 1.

Fig. 1.

Fig. 1

Open in a new tab

Flow diagram of study selection for multinational real-world cohort studies evaluating GLP-1 receptor agonist use and gynecologic cancer risk among adults with type 2 diabetes or obesity, 2024–2026.

All included studies were observational cohort studies based on real-world data sources, primarily from the United States, with one multinational database study and one study from the United Kingdom. Most studies applied propensity score matching, while one used propensity score–based weighting. Sample sizes ranged from 807 to 260,610 participants. The study populations included individuals with T2DM, overweight/obesity, or both, and all participants were cancer-free at baseline.

GLP-1RAs evaluated included liraglutide, semaglutide, tirzepatide, or GLP-1RAs as a drug class. Comparators varied across studies and included non-GLP-1RA use, insulin, metformin, sulfonylureas, sodium-glucose cotransporter 2 (SGLT2) inhibitors, dipeptidyl peptidase 4 (DPP4) inhibitors, and progestins. Outcomes of interest included endometrial cancer, ovarian cancer, uterine cancer, cervical cancer, and broader female genital organ cancers. Median or mean follow-up durations ranged from approximately 1.8 years to over 5 years, with some studies reporting maximum follow-up up to 15 years. The main characteristics of the included studies are summarized in Table 1.

Table 1.

Characteristics of real-world cohort studies of glucagon-like peptide-1 receptor agonist use and gynecologic cancer risk in adults with type 2 diabetes or obesity from the United States, the United Kingdom, and multinational databases, 2024–2026.

Study Design Country Sample size Glucagon-like peptide-1 receptor agonist Comparator Population Cancer type Median follow-up Hazard ratio
(95% confidence interval)
Abdelgadir et al., 2025 Retrospective cohort (1:2 propensity score matching) Unite States 807 Liraglutide No liraglutide use Women ≥65 years, obesity-related cancer-free at baseline Endometrial cancer 63.3 months 1.44 (0.79, 2.16)
Ovarian cancer 64.3 months 0.21 (0.03, 2.08)
Lu et al., 2025 Retrospective cohort (1:1 propensity score matching) Unite States 21,844 Glucagon-like peptide-1 receptor agonist Sodium-glucose cotransporter 2 inhibitor Adults ≥66 years with type 2 diabetes mellitus, cancer-free at baseline Endometrial cancer 1.91 years 1.21 (0.81, 1.81)
23,450 Dipeptidyl peptidase 4 inhibitor 1.95 years 1.55 (1.01, 2.37)
Levy et al., 2025 Retrospective cohort (1:1 propensity score matching) Unite States 260,610 Glucagon-like peptide-1 receptor agonist No glucagon-like peptide-1 receptor agonist use Adults 18–75 years with obesity (body mass index ≥30), cancer-free at baseline Female genital organs cancer 5 years 0.61 (0.53, 0.71)
Chuang et al., 2026 Retrospective cohort (1:1 propensity score matching) Unite States 196,112 Glucagon-like peptide-1 receptor agonist No glucagon-like peptide-1 receptor agonist use Adults ≥20 years with newly diagnosed type 2 diabetes mellitus (2016–2024), cancer-free at baseline Female genital organs cancer Not reported (follow-up period: 2016–2024) 0.87 (0.76, 1.00)
Dai et al., 2025 Retrospective cohort (1:1 time-dependent propensity score matching) Unite States 59,120 liraglutide, semaglutide, tirzepatide No glucagon-like peptide-1 receptor agonist use Adults ≥18 years with obesity (body mass index ≥30) or overweight (body mass index 27–29.9 and weight-related comorbidity), cancer-free at baseline Endometrial cancer and ovarian cancer Not reported (follow-up period: 2014–2024) 0.68 (0.52, 0.87)
Yen et al., 2026 Retrospective cohort (1:1 propensity score matching) Global 31,381 Glucagon-like peptide-1 receptor agonist Progestins Women ≥18 years with endometrial hyperplasia or benign uterine pathology, received progestins (2005–2022) Endometrial cancer Not reported (mean follow-up: ∼3.65–5.38 years) 0.34 (0.27, 0.44)
Wang et al., 2024 Retrospective cohort (1:1 propensity score matching) Unite States 25,750 Glucagon-like peptide-1 receptor agonist Insulin Patients with type 2 diabetes mellitus, no prior obesity-associated cancers, prescribed glucagon-like peptide-1 receptor agonist or insulins (2005–2018) Endometrial cancer Not reported (max follow-up: 15 years; mean: ∼5.4–5.7 years) 0.74 (0.60, 0.91)
25,739 Ovarian cancer 0.52 (0.37, 0.74)
17,168 Metformin Patients with type 2 diabetes mellitus, no prior obesity-associated cancers, prescribed glucagon-like peptide-1 receptor agonist or metformin (2005–2018) Endometrial cancer 1.11 (0.86, 1.43)
17,197 Ovarian cancer 0.99 (0.68, 1.44)
Rothman et al., 2025 Retrospective cohort (propensity score fine stratification weighting) United Kiddoms 89,325 Glucagon-like peptide-1 receptor agonist Sulfonylureas Women with type 2 diabetes mellitus, ≥40 years, new users of study drugs, no prior endometrial cancer (2007–2020) Endometrial cancer 1.8 years (glucagon-like peptide-1 receptor agonist); 2.7 years (Sulfonylureas) 1.11 (0.66, 1.88)
Ipaye et al., 2025 Retrospective cohort (1:1 propensity score matching) Unite States 128,356 Semaglutide Dipeptidyl peptidase 4 inhibitor Adults with type 2 diabetes mellitus and overweight / obesity (body mass index ≥25), no prior obesity-associated cancers, new users of study drugs (2005–2025) Ovarian cancer Semaglutide: 911 days
Dipeptidyl peptidase 4 inhibitor: 864 days		 0.97 (0.64, 1.46)
Uterine cancer 0.96 (0.75, 1.22)
39,364 Tirzepatide Ovarian cancer Tirzepatide: 435 days
Dipeptidyl peptidase 4 inhibitor: 439 days		 0.31 (0.10, 0.95)
Uterine cancer 0.51 (0.25, 1.02)
Open in a new tab

The methodological quality of the nine included studies was assessed using the NOS. Overall, the studies were of high quality, with one study receiving a score of 8 stars and the remaining eight studies receiving the maximum score of 9 stars. Detailed NOS scores for each study are presented in Supplementary Table 2.

3.2. GLP-1RA use and gynecologic cancers

In the overall analysis pooling all eligible HRs, GLP-1RA use was not significantly associated with the risk of gynecologic cancers (pooled HR 0.80; 95% CI: 0.63, 1.01; Fig. 2). However, substantial between-study heterogeneity was observed (I2 = 84.63%).

Fig. 2.

Fig. 2

Open in a new tab

Forest plot of the association between glucagon-like peptide-1 receptor agonist use and overall gynecologic cancer risk in adults with type 2 diabetes or obesity in multinational real-world cohort studies, 2024–2026. GLP-1RA: glucagon-like peptide-1 receptor agonist; EC: endometrial cancer; OC: ovarian cancer; UC: uterine cancer; SGLT2i: sodium-glucose cotransporter 2 inhibitor; DPP4i: dipeptidyl peptidase 4 inhibitor; HR: hazard ratio; CI: confidence interval.

When stratified by cancer subtype, no statistically significant association was observed for uterine /endometrial cancer (HR 0.84; 95% CI: 0.65, 1.08; I2 = 87.64%; Fig. 3) or cervical cancer (HR 0.79; 95% CI: 0.60, 1.05; I2 = 0%; Fig. 3). For vulvar/vaginal cancers, the pooled HR was 1.34 (95% CI: 0.80, 2.23; Fig. 3), based on limited data.

Fig. 3.

Fig. 3

Open in a new tab

Subgroup analysis of glucagon-like peptide-1 receptor agonist use and gynecologic cancer risk by cancer subtype in multinational real-world cohort studies, 2024–2026. GLP-1RA: glucagon-like peptide-1 receptor agonist; EC: endometrial cancer; OC: ovarian cancer; CC: cervical cancer; UC: uterine cancer; SGLT2i: sodium-glucose cotransporter 2 inhibitor; DPP4i: dipeptidyl peptidase 4 inhibitor; HR: hazard ratio; CI: confidence interval.

In contrast, GLP-1RA use was associated with a significantly reduced risk of ovarian cancer (HR 0.68; 95% CI: 0.49, 0.93; Fig. 3), with moderate heterogeneity (I2 = 50.92%). However, the test for subgroup differences across cancer types was not statistically significant (p = 0.16), suggesting no strong evidence that the association differed by cancer subtype.

3.3. Subgroup analyses

3.3.1. According to baseline population

In studies enrolling patients with T2DM, GLP-1RA use was not significantly associated with gynecologic cancer risk (HR 0.95; 95% CI: 0.75, 1.19; Fig. 4), with substantial heterogeneity (I2 = 72.00%).

Fig. 4.

Fig. 4

Open in a new tab

Subgroup analysis of glucagon-like peptide-1 receptor agonist use and gynecologic cancer risk by baseline population characteristics in multinational real-world cohort studies, 2024–2026. T2DM: type 2 diabetes mellitus; GLP-1RA: glucagon-like peptide-1 receptor agonist; EC: endometrial cancer; OC: ovarian cancer; UC: uterine cancer; SGLT2i: sodium-glucose cotransporter 2 inhibitor; DPP4i: dipeptidyl peptidase 4 inhibitor; HR: hazard ratio; CI: confidence interval.

In contrast, among populations with overweight or obesity, GLP-1RA use was associated with a significantly reduced risk (HR 0.63; 95% CI: 0.55, 0.72; Fig. 4), and no heterogeneity was observed (I2 = 0%). In studies including mixed T2DM and overweight/obesity populations, the association was not statistically significant (HR 0.72; 95% CI: 0.44, 1.20; I2 = 52.51%; Fig. 4).

The test for subgroup differences by baseline population was statistically significant (p = 0.01), indicating potential effect modification according to underlying metabolic status.

3.3.2. According to specific GLP-1RA agents

When stratified by specific agents, no significant association was observed for liraglutide (HR 0.94; 95% CI: 0.49, 1.80; I2 = 31.61%) or semaglutide (HR 0.71; 95% CI: 0.39, 1.27; I2 = 93.86%) (Supplementary Fig. 1).

For tirzepatide, a significant reduction in risk was observed (HR 0.44; 95% CI: 0.23, 0.84; Supplementary Fig. 1), with no detected heterogeneity (I2 = 0%). However, the test for subgroup differences across GLP-1RA agents was not statistically significant (p = 0.26).

3.4. Sensitivity analysis

Leave-one-out sensitivity analyses demonstrated that the direction of the pooled HR remained consistent after sequential omission of each study (Supplementary Fig. 2). Although statistical significance varied when individual studies were excluded, the overall HRs ranged from 0.77 to 0.86, indicating that no single study substantially altered the direction of the association. Between-study heterogeneity remained high across sensitivity analyses, suggesting that heterogeneity was not driven by any single study.

3.5. Publication bias

Visual inspection of the funnel plot did not suggest marked asymmetry (Supplementary Fig. 3). Because fewer than 10 studies were included, Egger's regression test were not performed, in accordance with current methodological recommendations.

4. Discussion

In this meta-analysis of real-world cohort studies, we found that GLP-1RA use was not significantly associated with an overall reduction in gynecologic cancer risk, although a borderline protective trend was observed. Subgroup analyses suggested a significantly reduced risk among individuals with overweight or obesity and for ovarian cancer specifically. In addition, a significant association was observed for tirzepatide in drug-specific analyses. However, substantial heterogeneity was present in several analyses, and subgroup differences were not consistently statistically significant, warranting cautious interpretation.

The overall pooled HR suggested a possible protective trend, but the association did not reach statistical significance and was accompanied by considerable between-study heterogeneity. This indicates that the relationship between GLP-1RA use and gynecologic cancer risk is unlikely to be uniform across populations or clinical contexts. The heterogeneity may reflect differences in comparator selection, baseline metabolic status, cancer definitions, follow-up duration, and residual confounding inherent in observational data.

Another notable source of clinical heterogeneity was the substantial variation in comparator groups across the included studies, including non-users, insulin, metformin, sulfonylureas, DPP-4 inhibitors, SGLT2 inhibitors, and progestins. These therapies differ considerably in their metabolic effects and have each been hypothesized to influence cancer risk through distinct biological pathways. Consequently, the pooled estimates may reflect not only the effects of GLP-1RAs but also differences in the underlying comparator therapies. We carefully considered comparator-specific subgroup analyses; however, because only nine studies were available and comparator categories were highly fragmented, most subgroups would contain only two to three studies, resulting in unstable and potentially misleading estimates. Therefore, comparator-based subgroup analyses were not performed. This limitation should be considered when interpreting the pooled results.

Our findings should also be interpreted in the context of recent meta-analytic evidence from RCTs. A recent network meta-analysis evaluating the risk of gynecologic tumors associated with GLP-1RAs and SGLT2 inhibitors reported that most GLP-1RAs were not associated with a significantly increased risk of gynecologic tumors, whereas high-dose tirzepatide was associated with an increased risk of overall gynecologic tumors and intra-uterine tumors (Tseng et al., 2025). Several methodological differences may explain the discrepancies between that analysis and the present study. First, RCTs typically have shorter follow-up durations and were not designed to detect long-latency cancer outcomes, whereas real-world cohort studies often include substantially larger populations and longer observation periods. Second, clinical trials enroll highly selected participants under strict eligibility criteria, while observational databases capture broader and more heterogeneous populations that better reflect routine clinical practice. In addition, many randomized trials evaluate specific agents and dose regimens, whereas observational studies frequently examine GLP-1RAs as a drug class. These differences in study design, exposure classification, and population characteristics may contribute to variations in estimated cancer risk. However, the apparent advantages of real-world evidence should be interpreted cautiously. While real-world cohort studies provide longer follow-up and more representative populations than many randomized trials, observational studies are inherently subject to important limitations, including residual confounding, confounding by indication, exposure misclassification, and time-related biases. Moreover, observational studies cannot establish causal relationships. These limitations may partly explain the discrepancies between our findings and those reported in previous RCT-based meta-analyses and should be considered when interpreting the present results.

A significant association was observed among populations with overweight or obesity, whereas no significant association was found in T2DM-only cohorts. One plausible explanation is that the magnitude of weight reduction achieved with GLP-1RAs may play a mediating role. Obesity is a well-established risk factor for several gynecologic malignancies, particularly endometrial and ovarian cancers, through mechanisms involving chronic inflammation, hyperinsulinemia, and estrogen excess derived from peripheral aromatization (Avgerinos et al., 2019). In populations with obesity, the metabolic and hormonal improvements associated with GLP-1RA therapy may translate into a more pronounced reduction in carcinogenic pathways.

In contrast, among T2DM populations, competing metabolic abnormalities, longer disease duration, or differential medication exposure histories may attenuate potential protective effects. Additionally, T2DM cohorts often involve comparisons against other active glucose-lowering therapies, which themselves may influence cancer risk, thereby narrowing between-group differences.

The observed association for ovarian cancer, but not endometrial or cervical cancer, is noteworthy but should be interpreted cautiously. Although ovarian cancer risk appeared significantly reduced in pooled analysis, the number of contributing studies was limited, and heterogeneity was moderate. Furthermore, the test for subgroup differences across cancer types was not statistically significant, suggesting insufficient evidence to conclude that the association differs meaningfully by cancer subtype.

Biologically, GLP-1 receptor signaling has been implicated in modulation of insulin sensitivity, inflammatory pathways, and potentially cellular proliferation (Chen et al., 2022; Alharbi, 2024; Zheng et al., 2024). However, direct mechanistic evidence linking GLP-1RA therapy to ovarian tumorigenesis remains limited. Therefore, the ovarian cancer finding should be considered hypothesis-generating rather than confirmatory.

Drug-level stratification suggested a significant association for tirzepatide, whereas no significant associations were observed for liraglutide or semaglutide. However, the number of studies contributing to individual drug estimates was small, and in some cases derived from the same database. Moreover, the formal test for subgroup differences across agents was not statistically significant. These results therefore do not provide robust evidence of differential oncologic effects among GLP-1RAs but may reflect variations in study design, population characteristics, or statistical power.

Substantial heterogeneity was observed in several analyses, particularly in the overall and endometrial cancer models. The persistence of high I2 values in sensitivity analyses suggests that heterogeneity was systemic rather than driven by a single outlying study. Potential contributors include differences in exposure definition (ever use vs. new use), duration of follow-up, outcome ascertainment methods, and comparator selection.

Although most studies applied propensity score–based methods to balance measured confounders, residual confounding remains possible. Confounding by indication, differences in obesity severity, duration of diabetes, prior medication exposure, and screening practices could influence observed associations. Screening or detection bias may also have influenced the observed associations. Patients receiving GLP-1RAs for obesity or diabetes often have more frequent healthcare encounters and closer clinical monitoring, which may increase opportunities for cancer detection independent of any true biological effect. Consequently, differences in surveillance intensity may have affected outcome ascertainment and influenced the estimated associations. In addition, immortal time bias and other time-related biases may affect certain retrospective designs, particularly if exposure was not modeled as time-dependent.

This meta-analysis synthesizes contemporary real-world evidence across diverse populations and comparator strategies. The inclusion of active-comparator designs and propensity score methods enhances internal validity relative to crude observational comparisons. Nevertheless, several limitations must be acknowledged. First, all included studies were observational, precluding causal inference. Second, heterogeneity was substantial in several analyses. Third, the number of studies for certain cancer subtypes and specific GLP-1RA was limited. Fourth, follow-up duration in some studies may have been insufficient to capture long-latency cancer outcomes. Finally, patient-level data were unavailable, preventing exploration of dose-response relationships, treatment duration effects, or mediation by weight change.

Current evidence does not support a definitive association between GLP-1RA use and overall gynecologic cancer risk. However, the observed associations in obesity populations and for ovarian cancer suggest that further investigation is warranted. Future studies with longer follow-up, consistent active-comparator new-user designs, and careful adjustment for metabolic and reproductive risk factors are needed. Mechanistic research examining whether weight loss, glycemic control, or direct receptor-mediated pathways contribute to potential cancer risk modification would also be valuable.

5. Conclusion

In summary, GLP-1RA use was not significantly associated with overall gynecologic cancer risk in pooled analysis, although potential reductions were observed in specific subgroups. Given the observational nature of the evidence and substantial heterogeneity, these findings should be interpreted cautiously and considered hypothesis-generating rather than definitive.

CRediT authorship contribution statement

Rongxia Li: Writing – original draft, Formal analysis, Data curation, Conceptualization. Xuewei Zhao: Writing – original draft, Formal analysis, Data curation, Conceptualization. Jinhai Shen: Writing – original draft, Software, Formal analysis, Data curation. Bo Luo: Visualization, Software, Methodology, Investigation, Formal analysis, Data curation. Yongmei Wang: Writing – review & editing, Supervision, Project administration, Conceptualization. Yancang Duan: Writing – review & editing, Supervision, Project administration, Conceptualization.

Consent for publication

Not applicable.

Ethics approval and consent to participate

Not applicable.

Funding information

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors expressed their gratitude to the databases that supplied the essential data for this investigation.

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.pmedr.2026.103548.

Contributor Information

Yongmei Wang, Email: wm110101@163.com.

Yancang Duan, Email: duanyancang@hebcm.edu.cn.

Appendix A. Supplementary data

Supplementary material

mmc1.docx (1.2MB, docx)

Data availability

Data will be made available on request.

References

  1. Abdelgadir O., Hussain M.R., Hollis-Hansen K., Barcenas C.H., Kuo Y.F., Skinner C.S., et al. Anti-obesity medications and the risk of obesity-related cancers in older women: a propensity score matching analysis of 2007-2015 SEER-Medicare data. Cancers (Basel) 2025;17(10) doi: 10.3390/cancers17101624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alharbi S.H. Anti-inflammatory role of glucagon-like peptide 1 receptor agonists and its clinical implications. Ther. Adv. Endocrinol. Metab. 2024;15 doi: 10.1177/20420188231222367. 20420188231222367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Avgerinos K.I., Spyrou N., Mantzoros C.S., Dalamaga M. Obesity and cancer risk: emerging biological mechanisms and perspectives. Metabolism. 2019;92:121–135. doi: 10.1016/j.metabol.2018.11.001. [DOI] [PubMed] [Google Scholar]
  4. Berrington de Gonzalez A., Brayley M., Frost R., Freedman N., Gunter M.J., Jackson I., et al. Trends in Cancer incidence in younger and older adults : an international comparative analysis. Ann. Intern. Med. 2025;178(12):1677–1687. doi: 10.7326/ANNALS-24-02718. [DOI] [PubMed] [Google Scholar]
  5. Chen J., Mei A., Wei Y., Li C., Qian H., Min X., et al. GLP-1 receptor agonist as a modulator of innate immunity. Front. Immunol. 2022;13 doi: 10.3389/fimmu.2022.997578. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chen J., Dalerba P., Terry M.B., Yang W. Global obesity epidemic and rising incidence of early-onset cancers. J. Glob. Health. 2024;14:04205. doi: 10.7189/jogh.14.04205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chuang C.H., Tsai P.K., Kao S.W., Wang Y.H., Yeh C.B. Early GLP-1 agonist use and Cancer risk in type 2 diabetes: a real-world data cohort study. Oncol. Res. 2025;34(1):12. doi: 10.32604/or.2025.072875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dai H., Li Y., Lee Y.A., Lu Y., George T.J., Donahoo W.T., et al. GLP-1 receptor agonists and Cancer risk in adults with obesity. JAMA Oncol. 2025;11(10):1186–1193. doi: 10.1001/jamaoncol.2025.2681. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Drucker D.J. Efficacy and safety of GLP-1 medicines for type 2 diabetes and obesity. Diabetes Care. 2024;47(11):1873–1888. doi: 10.2337/dci24-0003. [DOI] [PubMed] [Google Scholar]
  10. Espinoza C., Veas-Torres J., Miranda D.I., Owen G.I. Impact of type 2 diabetes on malignancies of the female reproductive system. Mol. Asp. Med. 2026;108 doi: 10.1016/j.mam.2026.101459. [DOI] [PubMed] [Google Scholar]
  11. Feng X., Li R., Yi H., Chen S., Liu M., Wu Y. Global cancer burden attributable to excess body weight, 1990 to 2021, decomposed by population size, aging, and epidemiological change. Obesity (Silver Spring) 2025;33(3):567–577. doi: 10.1002/oby.24219. [DOI] [PubMed] [Google Scholar]
  12. Huang P., Fan X., Yu H., Zhang K., Li H., Wang Y., et al. Glucose metabolic reprogramming and its therapeutic potential in obesity-associated endometrial cancer. J. Transl. Med. 2023;21(1):94. doi: 10.1186/s12967-022-03851-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Ipaye T., Goldney J., Wilkinson T.J., Zaccardi F., Yates T., Davies M.J., et al. Weight loss interventions and obesity-associated cancers in people with type 2 diabetes and overweight/obesity: a real-world observational study. Diabetes Obes. Metab. 2025;27(12):6914–6926. doi: 10.1111/dom.70090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kiesel L., Eichbaum C., Baumeier A., Eichbaum M. Obesity epidemic-the underestimated risk of endometrial Cancer. Cancers (Basel) 2020;12(12):3860. doi: 10.3390/cancers12123860. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Ko A., Chang Y.C., Bahar F., Wang T.H., Xanthavanij N., Yu C.C., et al. Risk for Cancer with glucagon-like Peptide-1 receptor agonists and dual agonists : a systematic review and Meta-analysis. Ann. Intern. Med. 2026;179(2):216–229. doi: 10.7326/ANNALS-25-02237. [DOI] [PubMed] [Google Scholar]
  16. Levy S., Attia A., Elshazli R.M., Abdelmaksoud A., Tatum D., Aiash H., et al. Differential effects of GLP-1 receptor agonists on cancer risk in obesity: a nationwide analysis of 1.1 million patients. Cancers (Basel) 2024;17(1) doi: 10.3390/cancers17010078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lin A., Ding Y., Li Z., Jiang A., Liu Z., Wong H.Z.H., et al. Glucagon-like peptide 1 receptor agonists and cancer risk: advancing precision medicine through mechanistic understanding and clinical evidence. Biomark. Res. 2025;13(1):50. doi: 10.1186/s40364-025-00765-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Liu Y., Yang J., Hilliard T.S., Wang Z., Johnson J., Wang W., et al. Host obesity alters the ovarian tumor immune microenvironment and impacts response to standard of care chemotherapy. J. Exp. Clin. Cancer Res. 2023;42(1):165. doi: 10.1186/s13046-023-02740-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Lu Y., Dai H., Tang H., Donahoo W.T., George T.J., Sun R.C., et al. Association of Glucagon-like Peptide-1 receptor agonists with Cancer risk in older adults with type 2 diabetes. Obesity (Silver Spring) 2025 doi: 10.1002/oby.24366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lucente D., Bellino S., La Salvia A. GLP-1 receptor agonists in solid tumour therapy: exploring their anticancer potential and underlying molecular pathways. Genes (Basel) 2025;16(11) doi: 10.3390/genes16111352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Page M.J., McKenzie J.E., Bossuyt P.M., Boutron I., Hoffmann T.C., Mulrow C.D., et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372 doi: 10.1136/bmj.n71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Pati S., Irfan W., Jameel A., Ahmed S., Shahid R.K. Obesity and Cancer: a current overview of epidemiology, pathogenesis, outcomes, and management. Cancers (Basel) 2023;15(2):485. doi: 10.3390/cancers15020485. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Rothman S.M., Yin H., Yu O.H.Y., Pollak M., Azoulay L. Incretin-based drugs and the incidence of endometrial Cancer among people with type 2 diabetes: active comparator new-user design. Drug Saf. 2025;48(9):1023–1033. doi: 10.1007/s40264-025-01551-8. [DOI] [PubMed] [Google Scholar]
  24. Rubinstein M.M., Brown K.A., Iyengar N.M. Targeting obesity-related dysfunction in hormonally driven cancers. Br. J. Cancer. 2021;125(4):495–509. doi: 10.1038/s41416-021-01393-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Shea A.A., Heffron C.L., Grieco J.P., Roberts P.C., Schmelz E.M. Obesity modulates the cellular and molecular microenvironment in the peritoneal cavity: implication for ovarian cancer risk. Front. Immunol. 2023;14:1323399. doi: 10.3389/fimmu.2023.1323399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Silverii G.A., Marinelli C., Bettarini C., Del Vescovo G.G., Monami M., Mannucci E. GLP-1 receptor agonists and the risk for cancer: a meta-analysis of randomized controlled trials. Diabetes Obes. Metab. 2025;27(8):4454–4468. doi: 10.1111/dom.16489. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur. J. Epidemiol. 2010;25(9):603–605. doi: 10.1007/s10654-010-9491-z. [DOI] [PubMed] [Google Scholar]
  28. Tseng P.T., Zeng B.Y., Hsu C.W., Sun C.K., Suen M.W., Carvalho A.F., et al. The gynecologic tumor risk related to GLP-1 receptor agonists and SGLT2 inhibitors use: a network meta-analysis of 91 randomized controlled trials. J. Hematol. Oncol. 2025;18(1):109. doi: 10.1186/s13045-025-01750-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Wang L., Xu R., Kaelber D.C., Berger N.A. Glucagon-like peptide 1 receptor agonists and 13 obesity-associated cancers in patients with type 2 diabetes. JAMA Netw. Open. 2024;7(7) doi: 10.1001/jamanetworkopen.2024.21305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Wong H.J., Sim B., Teo Y.H., Teo Y.N., Chan M.Y., Yeo L.L.L., et al. Efficacy of GLP-1 receptor agonists on weight loss, BMI, and waist circumference for patients with obesity or overweight: a systematic review, Meta-analysis, and Meta-regression of 47 randomized controlled trials. Diabetes Care. 2025;48(2):292–300. doi: 10.2337/dc24-1678. [DOI] [PubMed] [Google Scholar]
  31. Yen T.T., TYJ Hsieh, Lee G.Y., Toy E.P., Wei J.C., Tanner E.J. GLP-1 receptor agonists plus progestins and endometrial cancer risk in nonmalignant uterine diseases. JAMA Netw. Open. 2026;9(2):e2558205. doi: 10.1001/jamanetworkopen.2025.58205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Zheng Z., Zong Y., Ma Y., Tian Y., Pang Y., Zhang C., et al. Glucagon-like peptide-1 receptor: mechanisms and advances in therapy. Signal Transduct. Target. Ther. 2024;9(1):234. doi: 10.1038/s41392-024-01931-z. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material

mmc1.docx (1.2MB, docx)

Data Availability Statement

Data will be made available on request.

Articles from Preventive Medicine Reports are provided here courtesy of Elsevier

RESOURCES