Glucagon-like peptide 1 receptor agonist: A potential game changer for cholangiocarcinoma

Abstract

Glucagon-like peptide-1 receptor (GLP-1R) agonist, a subgroup of incretin-based anti-diabetic therapies, is an emerging medication with benefits in reducing blood glucose and weight and increasing cardiovascular protection. Contrarily, concerns have been raised about GLP-1R agonists increasing the risk of particular cancers. Recently, several epidemiological studies reported contradictory findings of incretin-based therapy on the risk modification for cholangiocarcinoma (CCA). The first cohort study demonstrated that incretin-based therapy was associated with an increased risk of CCA. Later studies, however, showed a null effect of incretin-based therapy on CCA risk for dipeptidyl peptidase-4 inhibitor nor GLP-1R agonist. Mechanistically, glucagon-like peptide 1 receptor is multifunctional, including promoting cell growth. High GLP-1R expressions were associated with progressive phenotypes of CCA cells in vitro. Unexpectedly, the GLP-1R agonist showed anti-tumor effects on CCA cells in vitro and in vivo with unclear mechanisms. Our recent report also showed that GLP-1R agonists suppressed the expression of GLP-1R in CCA cells in vitro and in vivo, leading to the inhibition of CCA tumor growth. This editorial reviews recent evidence, discusses the potential effects of GLP-1R agonists in CCA patients, and proposes underlying mechanisms that would benefit from further basic and clinical investigation.

Key Words: Carcinogenesis; Cholangiocarcinoma; Diabetes mellitus; Incretin; Glucagon-like peptide 1 receptor

Core Tip: Glucagon-like peptide 1 receptor (GLP-1R) agonists, an anti-diabetic drug with other systemic benefits, have been reported for their association with increased risk of some cancers. Although the associations between GLP-1R agonists and the risk of cholangiocarcinoma (CCA) have not been consensus, the use of GLP-1R agonists showed the anti-tumor effects against CCA in vitro and animal models are evident. This editorial reviews and discusses recent studies of the effects of GLP-1R agonists both at epidemiological and molecular levels. The understanding of how GLP-1R agonist affects CCA will be beneficial for the management of patients with CCA and diabetes mellitus.

INTRODUCTION

Cholangiocarcinoma (CCA) is an aggressive cancer of the biliary tract. Approximately 70% of patients with CCA were diagnosed at an advanced stage[1]. The incidence of CCA varies enormously globally, from 0.4 per 100000 in Canada to 85 per 100000 in northeastern Thailand, the highest incidence in the world[1,2]. According to the 8^th Edition American Joint Committee on Cancer staging for Hepato-Pancreato-Biliary cancer, CCA is classified into 3 subtypes due to anatomical origin: Intrahepatic CCA, perihilar CCA, and distal CCA[3]. Several risk factors for the development of CCA have been identified, including liver fluke infections, biliary-tract disorders, hepatolithiasis, toxins, cirrhosis, chronic hepatitis B and hepatitis C infections, chronic alcohol consumption, diabetes mellitus (DM), and obesity[4].

Major risk factors for CCA differ greatly globally. At this point, DM is considered a minor risk. However, the DM burden is increasing significantly - the age-standardized prevalence of DM increased by 90.5% worldwide between 1990 and 2021[5]. Globally, the number of DM patients is expected to increase from 529 million in 2021 to 1.31 billion in 2050. In combination with carcinogenic liver fluke-Opisthorchis viverrini infection, DM has been shown to synergistically strengthen the association for the development of CCA[5,6]. Further, both CCA and DM exhibit high mortality rates in the CCA endemic regions in northeastern Thailand[7].

Previous studies have described molecular mechanisms linking DM to the development and progression of CCA[8]. The effects of insulin on increasing the risk of CCA are unclear, but the usage of exogenous insulin for DM treatment has shown an association with an increased risk of extrahepatic CCA[9]. The roles of insulin on CCA progression have yet to be clarified, although the expressions of insulin-like growth factor receptors[10] and insulin receptor substrate protein have been reported with prognostic roles in CCA[11,12]. On the other hand, our previous reports have clearly demonstrated that hyperglycemia or high glucose itself is a major contributor to CCA progression. High glucose activates multiple pro-tumorigenic signaling pathways in CCA cells, namely, Janus kinase 2/signal transducer and activator of transcription 3[13], nuclear factor-kappa B[13], glycogen synthase kinase-3β/β-catenin pathways[14]. Since glucose seems to be a key player in DM promoting CCA[15], anti-diabetic medications whose effects are to control blood glucose are therefore potentially involved in CCA development and progression, either promoting or retarding the cancer cells. One of the hypoglycemic agents that draw the attention of researchers in CCA biology is a glucagon-like peptide 1 receptor agonist[8].

GLUCAGON-LIKE PEPTIDE 1 RECEPTOR AGONIST: EMERGING ANTI-DIABETIC AGENTS WITH QUESTIONABLE EFFECTS ON CANCERS

Glucagon-like peptide 1 receptors (GLP-1R) are expressed in many organs or tissues, such as the pancreas, thyroid, brain, and gastrointestinal tract. A primary effect of hypoglycemia relies on the activation by its ligand glucagon-like peptide 1 (GLP-1) at the pancreatic β cells. The activation of GLP-1R signals via Akt pathways to stimulate insulin secretion. The primary hypoglycemia effects are thus dependent on the actions of insulin, and additional anti-diabetic effects have also been reported[16]. A native GLP-1 itself, however, has low bioavailability and is rapidly degraded by dipeptidyl peptidase-4 (DPP-4)[16]. DPP-4 inhibitors and analogs of GLP-1 have then been developed to increase the actions and the activation of GLP-1 and GLP-1R for the treatment of DM, called incretin-based therapy. They are globally recommended as a second or third-line treatment for type 2 DM as they possess some benefits for other systems, like weight control and cardiovascular benefits[17].

The primary effects of incretin-based therapy are via increasing insulin secretion. Since insulin is questioned for its roles in some cancer development, GLP-1R agonists have also been queried for their effects on tumor development and progression as well. In addition, the activation of GLP-1R mainly signals via the cascades of various pro-tumorigenic pathways, such as PI3K/Akt pathways[8]. These have led to investigations into the association between the use of GLP-1R agonists and the development of cancers. Nowadays, positive associations between using GLP-1R agonists and increased risk of thyroid cancers are widely reported[18-20], whereas other endocrine-related cancers, e.g., breast and pancreatic cancers, have shown no association[19,21-24]. On the other hand, GLP-1R agonists use has been associated with reduced risk of prostate[18,25] and lung cancer[18].

GLP-1R AND GLP-1R AGONISTS: FRIENDS OR FOES OF CCA?

A large retrospective cohort study in the United Kingdom showed that DPP-4 inhibitors were associated with an increased risk of CCA[26]. However, the same study did not find any association between using GLP-1R agonists and the increased risk of this cancer. Since the hypoglycemic effects of DPP-4 inhibitors are passed through the bioavailability of GLP-1 and insulin, this might suggest that the effects of DPP-4 inhibitors are possibly associated with the other mechanisms. Although this was a large cohort study, it has limitations due to being an observational study. Also, the discrepancy between the effects of DPP-4 inhibitor and GLP-1R agonist suggested that the association between incretin use and increased risk of CCA might not be a direct effect of incretin-based therapy. Later, case-control studies conducted in Scandinavia[27] and Italy[28] consistently reported null effects for GLP-1R agonists on CCA risk. However, again limited by being observational studies in relatively smaller sample sizes, these latter studies could not totally exclude the possible risk of using GLP-1R agonists and CCA development.

A recent systematic review and meta-analysis of randomized control trials also found that although using GLP-1R agonists is significantly associated with the increased risk of benign biliary diseases, the increased risk of malignancy development in the biliary tract in GLP-1R agonist users was not different from the control groups[29]. This meta-analysis of randomized control trials strengthens the suggestion that GLP-1R agonists are potentially safe from increasing the risk of CCA in patients who use this drug group. To date, all epidemiological studies suggest that GLP-1R agonists are less likely to be associated with increased CCA risk. However, most studies were conducted in Western countries where the incidence of CCA is rare and is not associated with liver fluke infection. These different genetic and environmental backgrounds might confound study results, and more investigations that cover other ethnicities and regions would help clarify the effects of GLP-1R agonists on CCA development and the generalizability of results. The findings from epidemiological studies on GLP-1R agonists and the risk of biliary tract cancer/CCA are summarized in Table 1.

Table 1 Epidemiological studies on associations of glucagon-like peptide-1 receptor agonist and risk of cholangiocarcinoma.

Ref.	Region of study	Study design	Results (95%CI)
Abrahami et al[26], 2018	United Kingdom	Cohort study	HR: 1.97 (0.83-4.66)
Giorda et al[28], 2020	Italy	Case-control study	OR: 1.09 (0.63-1.89)
Ueda et al[27], 2021	Scandinavia	Cohort study	HR: 1.25 (0.89-1.76)
He et al[29], 2022	Worldwide	Systematic review and meta-analysis	RR: 1.43 (0.80-2.56)

HR: Hazard ratio; OR: Odds ratio; RR: Risk ratio.

In addition to epidemiological evidence, molecular studies on GLP-1R and GLP-1R agonists indicate that using GLP-1R agonists may be beneficial for CCA treatment. GLP-1R is widely expressed in cholangiocytes and protects the biliary epithelium from cell apoptosis[30,31], which has led to the hypothesis that excessive activation of GLP-1R might promote the immortality of stimulated cells and increase the risk of cancer development[8]. Further, GLP-1R has also demonstrated pro-tumorigenic roles, as it controls the epithelial-mesenchymal transition and migration of CCA cells in vitro[32]. However, the roles of GLP-1R in CCA cells are not straightforward. An immunohistochemical study of GLP-1R expression in CCA tissues did not show any associations between expression levels and CCA patients’ survival or other prognostic factors[33]. In contrast, another study activating GLP-1R by exendin-4 showed the opposite results in CCA cell lines. Exendin-4, a GLP-1R agonist with approximately 53% analogous to the human native GLP-1 peptide, suppressed CCA proliferation and induced chemosensitivity in vitro and in vivo[34]. Exendin-4 also exerted inhibitory effects against CCA metastatic potential by suppressing the migratory activity of CCA cells in vitro with unclear mechanisms. Our recent study also supports the findings of exendin-4’s effects on CCA cells. Liraglutide, another GLP-1R agonist with a higher degree of analogy to native GLP-1, showed significant anti-tumor effects against intrahepatic CCA, partly by downregulation of GLP-1R[35]. In parallel to the downregulated GLP-1R expressions, our study also showed that Akt and STAT3 signaling pathways were suppressed in CCA after treatment with liraglutide in vitro and in vivo. Suppressing these signaling pathways thus resulted in the inhibition of growth and epithelial-mesenchymal transition of CCA cells. However, whether liraglutide exhibiting anti-tumor effects on CCA cells is GLP-1R dependent or independent and whether liraglutide affects other subtypes of CCA needs further investigation. Further, investigating the effects of other GLP-1R agonists in CCA patients with different genetic backgrounds would also help clarify existing epidemiological findings.

In summary, all the data from in vitro and in vivo studies suggest that even though GLP-1R has pro-tumorigenic roles, using GLP-1R agonists may not result in higher aggressive phenotypes of CCA cells. On the other hand, GLP-1R agonists seem beneficial for CCA treatment, but the underlying mechanisms are not fully understood. In addition, there is a lack of cohort or randomized control trial studies of the association between using GLP-1R agonists and the prognosis in patients with CCA and DM. These contradictions need further investigation, not only for the proper management of patients with CCA and DM but also for a possible repurposing of GLP-1R agonists for CCA add-on treatments.

CONCLUSION

At present, evidence indicating an association between GLP-1R agonist usage and the increased risk of biliary tract cancer and CCA is unclear. Pro-tumorigenic roles of GLP-1R in CCA have been reported; however, the benefits of using GLP-1R agonists in CCA treatment in vitro and in vivo are also evident. Based on previous studies, using GLP-1R agonists to treat DM might still be safe and beneficial for those who have underlying CCA.

ACKNOWLEDGEMENTS

RT is grateful for the general support of the Cho-Kalaphruek Excellent Research Project for Medical Students, Faculty of Medicine, Khon Kaen University. We also thank Professor John F Smith for editing the English of this manuscript via the Khon Kaen University Publication Clinic (PCO-1197).