Obesity, Type 2 Diabetes Mellitus, and Endometrial Cancer: Metabolic-Endocrine Mechanisms and the Evolving Evidence for Metformin Chemoprevention

Introduction

Endometrial cancer (EC) incidence has risen in parallel with the global increase in obesity and type 2 diabetes mellitus (T2DM), and an extensive body of mechanistic and epidemiologic data implicates a closely intertwined metabolic-endocrine pathogenesis. Metabolic syndrome—encompassing central obesity, dyslipidemia, hyperglycemia, and hypertension—confers a relative risk of approximately 1.89 (95% CI 1.34–2.67) for EC, with obesity alone carrying an even stronger signal (RR 1.95, 95% CI 1.80–2.11) 1. Among individual metabolic syndrome components, elevated fasting glucose (RR 1.36), high blood pressure (RR 1.31), and hypertriglyceridemia (RR 1.13) each contribute incrementally, underscoring that carcinogenic signaling arises from the convergence of multiple metabolic perturbations rather than any single factor 1. For clinicians managing patients with these conditions, understanding the mechanistic underpinnings can support risk-aware counseling and the critical appraisal of investigational prevention strategies such as metformin.

Mechanistic Framework: How Obesity and T2DM Drive Endometrial Carcinogenesis

Insulin Resistance, Hyperinsulinemia, and Growth Factor Signaling

Insulin resistance and compensatory hyperinsulinemia—hallmarks of both obesity and T2DM—directly activate mitogenic signaling in endometrial epithelium. The insulin receptor (IR) exists as two isoforms: IR-A (predominantly mitogenic) and IR-B (primarily metabolic). In endometrial adenocarcinoma, both IR-B and IGF-1R expression are 5- to 6-fold higher than in normal endometrium, while IR-A remains aberrantly elevated—a dual-receptor overexpression signature that provides malignant cells with multiple growth advantages in hyperinsulinemic states 2. Importantly, this receptor dysregulation appears intrinsic to the malignant phenotype, as it is independent of body mass index 2.

At the downstream signaling level, endometrial cancer tissue from women with T2DM demonstrates significantly increased IGF-1R-mediated signaling and downstream phosphoinositide 3-kinase (PI3K) pathway activation relative to diabetic controls without cancer—despite comparable serum IGF-1 and IGF-2 levels—suggesting that altered receptor expression and signaling patterns, rather than simply elevated circulating ligands, drive carcinogenesis in insulin-resistant states 4. Separately, high-glucose conditions (25 mM) directly increase endometrial cancer cell growth, adhesion, invasion, and glycolytic activity through AMPK/mTOR/S6 and MAPK pathway dysregulation 11, providing a mechanistic basis for the contribution of hyperglycemia independent of insulin. Adding a further layer of complexity, insulin epigenetically sensitizes endometrial cancer cells to estrogen by upregulating G-protein-coupled estrogen receptor (GPER) expression via TET1-mediated DNA hydroxymethylation—a mechanism operating independently of classical estrogen receptor changes and amplifying estrogenic signaling even without increased circulating estrogen levels 5.

Estrogen Excess and Sex Steroid Imbalance

Obesity drives endometrial carcinogenesis through a parallel estrogen-mediated pathway. Adipose tissue is the dominant source of circulating estrogen in postmenopausal women via aromatase-catalyzed conversion of androgens to estradiol. Increased adiposity elevates total and free estrogen levels while simultaneously suppressing hepatic sex hormone-binding globulin (SHBG) synthesis, compounding the effect by increasing the bioavailable estrogen fraction. Insulin resistance and T2DM independently suppress SHBG, further amplifying free estradiol availability 1. The resulting state of unopposed estrogen—acting via estrogen receptor-alpha to drive endometrial proliferation, inhibit apoptosis, stimulate angiogenesis through vascular endothelial growth factor, and generate genotoxic metabolites—constitutes one of the most established pathogenic mechanisms in type I endometrial carcinogenesis 1.

Adipokine Dysregulation and Chronic Inflammation

Adipose tissue in obesity functions as a dysregulated endocrine organ, producing excess leptin (which activates JAK-STAT and MAPK proliferative signaling in endometrial epithelium) and reduced adiponectin (an insulin-sensitizing, anti-inflammatory adipokine whose deficiency removes a key antiproliferative brake) 9. Chemerin, a novel adipokine elevated in both obesity and T2DM, further exemplifies this pathological adipokine milieu: its increased secretion activates inflammatory pathways, amplifies oxidative stress, and worsens insulin resistance through CMKLR-1 receptor signaling 6. The resulting chronic low-grade inflammatory state—characterized by elevated tumor necrosis factor-alpha, interleukin-6, and NF-kB activation—promotes genomic instability, angiogenesis, and evasion of apoptosis, creating an endometrial microenvironment permissive for malignant transformation 1.

Clinical Impact of T2DM on Prognosis

Beyond cancer initiation, metabolic dysfunction impairs outcomes in women with established EC. Diabetes has been associated with elevated recurrence rates and worse overall survival in patients treated for type I EC, indicating that hyperinsulinemia and its downstream effects remain operative throughout disease progression and treatment 1.

Metformin as a Candidate Chemopreventive Agent

Proposed Antineoplastic Mechanisms

Metformin's rationale for EC chemoprevention rests on a coherent, multilayered mechanistic framework. Its primary molecular action—activation of AMPK via mitochondrial complex I inhibition and, in part, via a novel duodenal AMPK-dependent gut-brain-liver pathway that reduces hepatic glucose production 8—leads to downstream mTOR inhibition and suppression of anabolic cellular processes 1617. In endometrial cancer cells, AMPK activation by metformin promotes FOXO1 nuclear translocation, a tumor-suppressor transcription factor whose expression is markedly reduced in EC tumors compared to normal endometrium; silencing FOXO1 with siRNA abolishes metformin's antiproliferative effect in vitro, confirming its role as a critical effector 19. Separately, metformin reverses palmitate-induced mTOR dysregulation by restoring AMPK phosphorylation in the face of lipid excess 13—particularly relevant in obese patients with elevated circulating free fatty acids.

In addition to direct AMPK-mTOR effects, metformin lowers circulating insulin, IGF-1, and leptin while increasing adiponectin 922, thereby indirectly reducing the ligand-level drivers of endometrial proliferation. In PCOS endometrium—a high-risk tissue—metformin induces GLUT4 expression, inhibits androgen receptor signaling, and suppresses the PI3K/Akt/mTOR axis directly in endometrial tissue 10. GLP-1 receptor agonist exenatide, which converges on AMPK by a different mechanism, also inhibits endometrial xenograft growth and promotes apoptosis 14, further supporting AMPK as a relevant therapeutic node.

Human Translational Evidence: Window-of-Opportunity Trials

Short-term presurgical window-of-opportunity trials provide the most direct human evidence of metformin's pharmacodynamic activity in endometrial tissue. In a presurgical window study of 40 women (28 metformin-treated, 12 controls) with atypical endometrial hyperplasia or endometrioid adenocarcinoma, metformin 850 mg twice daily for a median of 20 days was associated with a 12.9% reduction in tumour Ki-67 (95% CI 3.7–22.1, P = 0.008) after adjustment for age, BMI, baseline Ki-67, and Ki-67 change in controls 12. (The full publication of the same cohort in the British Journal of Cancer 2016 reported a 17.2% Ki-67 reduction, 95% CI 7.0–27.4, P = 0.002, using slightly different modelling.) Importantly, the larger PREMIUM phase III randomized placebo-controlled trial subsequently reported no overall difference in post-treatment Ki-67 between metformin and placebo arms (mean difference −0.57%, 95% CI −7.57 to 6.42, P = 0.87).A complementary cohort of 20 patients receiving metformin for a median of 9.5 days demonstrated decreased phospho-AKT in 90% of treated tumors (P = 0.0002), decreased phospho-S6rp in 70% (P = 0.057), and decreased phospho-p44/42MAPK in 83% (P = 0.0038), alongside significant serum reductions in IGF-1, insulin, C-peptide, leptin, and omentin 22. However, neither Ki-67 nor apoptosis markers changed significantly in this shorter window, suggesting that the duration of exposure may be insufficient to translate signaling changes into measurable proliferative arrest.

Critically, heterogeneity in response is substantial: 35–40% of patients show minimal Ki-67 reduction, and resistance appears to be predicted by tumor hypoxia and hyperglycemic microenvironments. In a mechanistic analysis of the Manchester cohort, high baseline HIF-1alpha expression in tumors was significantly associated with lower Ki-67 response to metformin (adjusted mean difference −2.5%, 95% CI −0.4 to −4.6%, P = 0.018) 26. Under hypoxic, high-glucose conditions, cancer cells shift to glycolytic metabolism, reducing their dependence on oxidative phosphorylation—precisely the pathway that metformin inhibits. This finding suggests that metformin's efficacy in EC prevention may be context-dependent, with normoglycemic, low-grade, normoxic tumors representing the most pharmacologically responsive subgroup 26.

Observational and Epidemiological Evidence

The largest observational evidence base comes from a Taiwanese retrospective cohort of 478,921 women with newly diagnosed T2DM, in which metformin ever-use was associated with a 32.5% reduction in EC incidence (propensity score-adjusted HR 0.675, 95% CI 0.614–0.742), with a significant dose-response relationship across tertiles of cumulative exposure (HR 0.313 in the highest tertile, P-trend <0.0001) 39. This dose-response gradient strengthens causal inference within an observational framework. In contrast, a 2024 systematic review and meta-analysis in Gynecologic Oncology reported substantial between-study heterogeneity (I² = 90%); after excluding the Tseng 2015 Taiwanese cohort as the principal source of heterogeneity, a fixed-effects pooled analysis of the remaining studies showed an increased EC incidence among metformin users (HR 1.17, 95% CI 1.09–1.26, P <0.0001) 35, highlighting the profound impact of study heterogeneity, confounding by indication, and variable definitions of metformin exposure in pooled analyses. Prognostic data from the same meta-analysis were more consistently favorable: metformin was associated with improved all-cause mortality (HR 0.62, 95% CI 0.52–0.74) and progression-free survival (HR 0.55, 95% CI 0.44–0.68) in women with established EC 35.

The only prospective randomized prevention trial identified in the retrieved literature—NCT01697566, a 2×2 factorial design in 26 obese postmenopausal women with prediabetes—found that 16 weeks of metformin (1,700 mg/day) produced modest weight loss (−3.43 kg) but no statistically significant changes in endometrial Ki-67 or serum biomarkers 28. Lifestyle intervention outperformed metformin for weight loss (−4.23 kg, P = 0.006) and fat mass reduction. The trial's small sample, stringent eligibility criteria (only 29 of 576 approached women were randomized), very low baseline endometrial proliferation (mean 7.1%), and short duration collectively limit its interpretive power, but this remains the sole available randomized prevention-context evidence.

Summary Evidence Table

Clinical Interpretation and Unanswered Questions

For medical professionals, the current evidence supports the following conclusions. First, the mechanistic case for metformin in EC chemoprevention is biologically coherent: metformin directly addresses the insulin resistance, mTOR hyperactivation, and adipokine dysregulation that drive obesity- and T2DM-associated endometrial carcinogenesis. Second, short-term human data confirm pharmacodynamic activity in tumor tissue, with reproducible reductions in Ki-67 and inhibitory phosphoprotein signatures. Third, the large Taiwanese population-based cohort, which demonstrated a dose-response relationship, provides the most compelling incidence-level observational evidence, but it is retrospective and cannot exclude residual confounding (notably the absence of BMI adjustment). Fourth, and critically, no incidence-powered randomized controlled trial has demonstrated that metformin reduces the development of endometrial hyperplasia or EC in high-risk populations.

Clinicians should therefore not prescribe metformin with a primary EC prevention indication outside of clinical trials. For women with T2DM and obesity who already require antidiabetic therapy, metformin remains a rational first-line agent, and its use can be contextualized by reassuring observational associations with reduced EC mortality. Women with PTEN-deficient tumors or those with highly hypoxic, high-grade disease may represent a biologically resistant subgroup in whom metformin is less likely to confer benefit 2326.

Key unanswered questions that should drive future research include: (1) whether longer-duration exposure (years rather than weeks) is necessary to produce meaningful prevention; (2) whether combination strategies targeting both glycolysis and oxidative phosphorylation can overcome metformin resistance in hypoxic tumors 26; (3) whether biomarker-driven patient selection (HIF-1alpha status, PTEN expression, insulin resistance indices) can identify the subpopulations most likely to benefit; and (4) whether metformin combined with structured weight loss programs—which independently reduce estrogen excess and insulin resistance—produces synergistic prevention effects 28. Ongoing randomized trials examining metformin in atypical hyperplasia (NCT01685762) and in combination with hormonal therapy in early EC (NCT01686126) may provide important data on these questions. Until incidence-powered randomized evidence is available, metformin's role in EC chemoprevention must remain hypothesis-generating rather than clinically established.