Use of Glucagon-Like Peptide-1 Receptor Agonists Does Not Increase the Risk of Cancer in Patients with Type 2 Diabetes Mellitus

Article information

Diabetes Metab J. 2024;.dmj.2024.0105

Publication date (electronic) : 2024 October 24

doi : https://doi.org/10.4093/dmj.2024.0105

Mijin Kim^1,2, Seung Chan Kim³, Jinmi Kim³, Bo Hyun Kim^,1,2

¹Department of Internal Medicine, Pusan National University School of Medicine, Yangsan, Korea

²Biomedical Research Institute, Pusan National University Hospital, Busan, Korea

³Department of Biostatistics, Clinical Trial Center, Biomedical Research Institute, Pusan National University Hospital, Busan, Korea

Corresponding author: Bo Hyun Kim https://orcid.org/0000-0001-9632-9457 Department of Internal Medicine, Pusan National University School of Medicine, 49 Busandaehak-ro, Mulgeum-eup, Yangsan 50612, Korea E-mail: pons71@hanmail.net

Received 2024 March 4; Accepted 2024 May 13.

Abstract

KEY FIGURE

Highlights

INTRODUCTION

Type 2 diabetes mellitus (T2DM) is currently a global epidemic, affecting 537 million individuals worldwide. Therefore, T2DM continues to burden global health systems and is associated with high mortality rates [1,2]. The higher mortality in T2DM occurs mostly owing to cardiovascular diseases; however, accumulating epidemiological evidence has shown a higher risk of incidence and mortality for some types of cancer in individuals with T2DM [3]. A comprehensive meta-analysis of T2DM and the risks of developing cancer and mortality showed that the presence of diabetes was associated with a 10% increase in the relative risk of developing cancer [4]. The mechanistic processes that link diabetes and cancer are not completely understood; however, both hyperglycemia and hyperinsulinemia in diabetes can elicit cell damage responses and induce neoplastic transformation and cancer progression [5]. Additionally, some antidiabetic therapies may affect cancer development through largely undefined mechanisms [5].

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have antihyperglycemic and pleiotropic effects that reduce cardiovascular mortality in individuals with T2DM [6]. Therefore, recent guidelines recommend GLP-1 RAs for patients with T2DM, established or high-risk atherosclerotic cardiovascular disease, and/or chronic kidney disease for glycemic management and comprehensive cardiovascular risk reduction [6,7]. GLP-1 RAs directly activate the GLP-1 receptor, which stimulates pancreatic insulin secretion in a glucose-dependent manner while also inhibiting glucagon secretion [8]. Since their approval, there has been a rising interest in GLP-1 RAs, resulting in a high prescription rate for this class of medications in patients with T2DM. However, concerns have arisen that their use may be associated with pancreatic and thyroid cancer development [9]. This concern has been heightened by an analysis of the U.S. Food and Drug Administration Adverse Events Database, where the spontaneous reporting rates of pancreatic and thyroid cancers were 2.9 and 4.7 times higher, respectively, with exenatide compared with other oral antidiabetic drugs [10]. Indeed, carcinogenicity studies in rats and mice have demonstrated a dose- and time-dependent increased risk of medullary thyroid carcinoma (MTC) with GLP-1 RAs [11,12].

Based on these findings, the USA has contraindicated the use of liraglutide, dulaglutide, extended-release exenatide, and semaglutide in patients with a personal or family history of MTC and multiple endocrine neoplasia type 2. However, randomized controlled trials (RCTs) have not demonstrated this association, although most of these trials were of short duration and none were designed or powered to assess the risk of cancer [13-15]. Consequently, the potential association between pancreatic and thyroid cancer risk and GLP-1 RA exposure remains unclear. This population-based, nationwide cohort study aimed to evaluate the site-specific incidence of cancer in patients with T2DM treated with GLP-1 RAs.

METHODS

Study design and participant selection

We conducted a retrospective cohort study using Korean NHIS data between 2004 and 2021, with a 5-year wash-out period. We identified a T2DM cohort of 294,148 participants from the Korean NHIS database from 2009 to 2018 (Fig. 1). T2DM was defined as a fasting plasma glucose level ≥126 mg/dL or assigned International Classification of Diseases 10th Revision (ICD-10) codes E11–E14 with taking glucose-lowering drugs or insulin. Age- and sex-matched non-diabetic participants (n=263,894) in the National Health Insurance cohort with at least one claim data between 2009 and 2018 were selected as controls. After excluding patients aged <19 years (n=4,015); with antecedent malignancies (n=33,719); without health screening examination data (n=62,007); treated with dipeptidyl peptidase 4 (DPP-4) inhibitors, including sitagliptin, vildagliptin, saxagliptin, linagliptin, gemigliptin, teneligliptin, alogliptin, evogliptin, and anagliptin (n=64,269); and who had died or been followed up for <1 year (n=8,582), the T2DM cohort (n=173,164) was divided into the GLP-1 RA (n=3,485) and diabetes mellitus (DM) control (n=169,679) groups based on the use of antidiabetic drugs. The GLP-1 RA group consisted of patients receiving GLP-1 RAs, including exenatide, dulaglutide, and liraglutide. The DM control group consisted of patients with T2DM who were prescribed insulin or glucose-lowering drugs but not incretin-based drugs (GLP-1 RAs or DPP-4 inhibitors). The exclusion of patients treated with DPP-4 inhibitors from this study is justified by the classification of both GLP-1 RAs and DPP-4 inhibitors as incretin-based therapies. This exclusion is consistent with RCTs that assess the cardiovascular outcomes of GLP-1 RAs [17,18]. By excluding DPP-4 inhibitors, we ensure that the observed effects can be attributed specifically to GLP-1 RAs, thus eliminating potential confounding influences from other incretin-based treatments. All cohorts were exclusively defined, and 212,286 participants were included in the non-DM control group (Supplementary Table 1). The participants were followed up until the development of cancer, death, or the end of the study period (December 31, 2021). The median follow-up duration for all participants was 8 years.

Fig. 1.

Study population flow diagram of included case and control participants. DPP-4, dipeptidyl peptidase 4; DM, diabetes mellitus; GLP-1 RA, glucagon-like peptide-1 receptor agonist.

RESULTS

Baseline characteristics

After PS matching, 7,827 participants were analyzed, and their baseline characteristics are presented in Table 1. In total, 5,218 patients had T2DM (DM group), and 2,609 individuals were included in the non-DM control group. Among the 5,218 patients in the DM group, 2,609 were classified into the GLP-1 RA and DM control groups. The mean age was 51.7±12.8 years, and the mean diabetes duration was 6 years at baseline. Overall, 3,539 (45.2%) patients were male, and the mean BMI was 28.3±4.8 kg/m². After PS matching, no differences were observed in age, sex, height, weight, BMI, eGFR, smoking status, alcohol consumption, or the number of health screening examinations among the three groups. Additionally, there were no significant differences in diabetes duration, fasting glucose levels, antidiabetic drug use (including insulin), or comorbidities between the GLP-1 RA group and the DM control group. After PS matching, the mean diabetes duration in the GLP-1 RA group was 6.5±4.5 years, and the mean duration of GLP-1 RA treatment was 2.4±3.0 years.

Table 1.

Baseline characteristics of case and control subjects after matching

Incidence rate ratios for cancers in patients with T2DM

The risk of cancer occurrence in patients with T2DM compared with that in individuals without DM is shown in Fig. 2 and Supplementary Table 2. The IRR of subsequent cancer was 1.73 (95% CI, 1.43 to 2.10; P<0.001), and it increased in both men (IRR, 1.95; 95% CI, 1.45 to 2.62; P<0.001) and women (IRR, 1.58; 95% CI, 1.23 to 2.03; P<0.001). In total, a significant increase in incidence was observed in malignancies of the liver (IRR, 2.05; 95% CI, 1.20 to 3.53; P=0.009), gallbladder and biliary tract (IRR, 5.16; 95% CI, 1.22 to 21.90; P=0.026), pancreas (IRR, 29.7; 95% CI, 4.13 to 214.2; P=0.001), breast (IRR, 2.36; 95% CI, 1.11 to 5.03; P=0.026), corpus uteri (IRR, 2.60; 95% CI, 1.01 to 6.73; P=0.048), prostate (IRR, 2.70; 95% CI, 1.21 to 6.00; P=0.015), and kidney (IRR, 8.98; 95% CI, 1.21 to 66.91; P=0.032). With stratification by sex, the analyses yielded similar results, with a significantly higher IRR of liver (2.52), pancreatic (33.84), and prostate (2.95) cancers in men and a significantly higher IRR of gallbladder and biliary tract (12.19), pancreatic (13.49), and breast (2.16) cancers in women.

Fig. 2.

Incidence rate ratios (IRRs) of malignancy in patients with type 2 diabetes mellitus (DM): DM group vs. non-DM group. (A) Total, (B) men, and (C) women. PY, person-year; CI, confidence interval.

Incidence rate ratios for cancers in patients with T2DM using GLP-1 RA

To evaluate the association between the use of GLP-1 RAs and cancer occurrence, we further analyzed the IRR of subsequent cancers in the GLP-1 RA and DM control groups (Fig. 3 and Supplementary Table 3). Analysis performed in patients with T2DM showed no increased cancer risk associated with the use of GLP-1 RAs (IRR, 0.94; 95% CI, 0.79 to 1.12; P=0.465), and similar results were observed in both men (IRR, 0.92; 95% CI, 0.70 to 1.20; P=0.532) and women (IRR, 0.95; 95% CI, 0.76 to 1.20; P=0.682). The IRRs of pancreatic cancer (0.74; 95% CI, 0.46 to 1.21), thyroid cancer (1.32; 95% CI, 0.79 to 2.20), and MTC (0.34; 95% CI, 0.04 to 3.23) did not significantly increase in the GLP-1 RA group compared with the DM control group.

Fig. 3.

Incidence rate ratios (IRRs) of malignancy in patients with type 2 diabetes mellitus (DM): glucagon-like peptide-1 receptor agonist (GLP-1 RA) group vs. DM control group. (A) Total, (B) men, and (C) women. PY, person-year; CI, confidence interval.

DISCUSSION

This nationwide cohort study evaluated the IRRs associated with cancer development in patients with T2DM treated with GLP-1 RAs in South Korea. There was a 73% higher risk of cancer in patients with T2DM compared with that in the general population. However, among patients with T2DM, there was no association between GLP-1 RAs and new-onset cancers, such as pancreatic cancer, thyroid cancer, and MTC.

GLP-1 RAs are increasingly used in the treatment of T2DM due to their multifaceted benefits, including improving glycemic control, reducing body weight, enhancing β-cell function, and yielding favorable cardiovascular and renal outcomes [22]. However, the broad distribution of GLP-1 receptors across various tissues, such as the pancreas, lungs, kidneys, central nervous system, cardiovascular system, gastrointestinal tract, skin, and thyroid gland, has raised concerns about the potential impact of GLP-1 RAs on cancer development and progression [23]. To date, no clear clinical evidence has suggested a tumorigenic effect of GLP-1 RAs [24-26], yet the steadily increasing use of these agents worldwide emphasizes the importance of thoroughly evaluating their potential association with cancer risk.

A major concern with the use of GLP-1 RAs has been their potential association with an increased risk of pancreatic cancer [10]. GLP-1 RAs exert trophic effects on pancreatic β-cells, promoting proliferation and differentiation [27]. While beneficial for glycemic control, this activity has raised concerns about potential uncontrolled cell growth and an increased risk of pancreatic cancer [27]. Additionally, GLP-1 RAs may reduce β-cell apoptosis, further contributing to the risk of neoplastic transformation [27]. A RCT of liraglutide reported numerically more pancreatic cancer cases in the liraglutide group than those reported in the placebo group (13/4,668 vs. 5/4,672), although this difference was not statistically significant [18]. However, this study was primarily designed to assess cardiovascular and diabetes outcomes, and its short follow-up duration may have limited its ability to accurately assess cancer risk [18]. Recent meta-analyses of RCTs have failed to demonstrate a significant increase in pancreatic cancer risk associated with GLP-1 RA treatment [24,25]. Nonetheless, these meta-analyses were also limited by the short follow-up periods of the included trials [18,24,25]. In our large, nationwide cohort study with a median follow-up of 8 years, we observed no increase in pancreatic cancer risk with GLP-1 RA use compared to other antidiabetic drug users. This finding is particularly remarkable given the large sample size and the extended follow-up duration of our study, which enhances the statistical power to detect potential associations.

The association between the use of GLP-1 RAs and thyroid cancer, particularly MTC, is an active area of research [9]. This concern arises from rodent studies indicating a link between GLP-1 and parafollicular C-cell proliferation, which could influence the risk of new-onset MTC [28]. However, it is important to note that humans have fewer C-cells and GLP-1 receptors in the thyroid gland compared to rodents, and no C-cell changes have been observed in clinical trials involving patients treated with GLP-1 RAs [26]. Moreover, given the rarity of MTC, a large sample size is pertinent to accurately evaluating the possibility of cancer development associated with GLP-1 RA use. In our extensive nationwide population study, we did not observe an increased risk of developing all thyroid cancers, including MTC, in patients with T2DM treated with GLP-1 RAs compared to those not receiving these agents.

Contrasting with concerns about pancreatic and thyroid cancers, emerging evidence suggests that GLP-1 RAs may inhibit the growth of several types of cancer, including ovarian, breast, cervical, prostate, and pancreatic cancers [29-31]. The proposed mechanisms underlying these protective effects include the inhibition of several signaling pathways implicated in cancer initiation and progression, such as the phosphoinositide 3 kinase (PI3K)/protein kinase B (AKT)/mammalian target of rapamycin (mTOR), extracellular signal-regulated kinase/microtubule-associated protein kinase, and nuclear factor kappa-light-chain-enhancer of activated B cell pathways [32-34]. Intriguingly, the combination of GLP-1 RAs with anticancer drugs may indirectly inhibit tumor migration, invasion, and growth by increasing the chemosensitivity of cancer cells [32,35]. For instance, Wenjing et al. [32] found that exendin-4 mitigated resistance to the androgen receptor inhibitor enzalutamide in prostate cancer via the PI3K/AKT/mTOR pathway, suggesting that a combination of exendin-4 with enzalutamide may be a more effective therapeutic option for advanced prostate cancer. Similarly, liraglutide has demonstrated significant antiproliferative and pro-apoptotic effects in gemcitabine-resistant human pancreatic cancer cells, enhancing their chemosensitivity in both in vitro and in vivo experiments [35]. These findings highlight the complex and potentially tissue-specific effects of GLP-1 RAs on cancer development and progression, underscoring the need for further research to elucidate the underlying mechanisms and clinical implications.

By employing a robust study design that included two control groups (DM control and non-DM control), we were able to comprehensively evaluate the increased cancer risk in patients with T2DM compared to individuals without diabetes. Consistent with previous studies [4,36], our findings confirm an elevated risk of developing cancers in several organs, including the liver, biliary tract, pancreas, breast, corpus uteri, prostate, and kidney, among patients with T2DM. Several mechanisms have been proposed to explain the association between diabetes and cancer, such as hyperglycemia, hyperinsulinemia, increased bioactivity of insulin-like growth factor 1, oxidative stress, sex hormone dysregulation, and chronic inflammation [37]. Additionally, certain factors associated with T2DM, including hyperglycemia, insulin resistance, adiposity, and specific medications used for treatment, have been implicated in either increasing or reducing the risk of cancer development [37].

The present study has some limitations. First, the study population was drawn from a single country, making it difficult to generalize the results to other racial or ethnic backgrounds. Second, due to the lack of information on pathology and disease severity, we could not evaluate the association between GLP-1 RA use and cancer stage. Third, given the relatively recent introduction of GLP-1 RAs as antidiabetic treatments, the median follow-up duration in this study may have been too brief to adequately assess the long-term risk of cancer development. Nonetheless, to the best of our knowledge, this is the first nationwide study to evaluate the risk of developing various cancers in patients with T2DM using GLP-1 RAs compared with users of other antidiabetic drugs and non-DM control individuals over a median follow-up period of 8 years. We selected appropriate confounders, such as index year, sex, age, diabetes duration, health screening variables, comorbidities, and antidiabetic drugs, for PS matching to minimize the influence of factors other than GLP-1 RA exposure.

In conclusion, the findings of our study, comprising a combined cohort of over 38 million individuals with T2DM and controls, suggest that the use of GLP-1 RAs is not associated with an increased risk of cancers, including pancreatic cancer and MTC, compared with the use of other antidiabetic drugs. As GLP-1 RAs are increasingly used in the treatment of T2DM, their potential carcinogenic and antineoplastic effects should be further investigated in future clinical studies.

SUPPLEMENTARY MATERIALS

Article information Continued