Association between incretin mimetics and pancreas tumour: which evidence?

Incretin mimetics or GPL-1 mimetics (glucagon like peptide-1) are molecules used for the treatment of type 2 diabetes mellitus. These drugs stimulate the GLP-1 receptors by acting as exogenous agonist (i.e. exenatide and liraglutide) or like inhibitors of the dipeptidyl peptidase-4 (DPP4), the enzyme responsible for the degradation of the endogenous hormone (i.e. sitagliptin and vildagliptin).¹
Exenatide has been the first commercialized drug of this new class of anti-diabetics. The FDA, in fact, authorized exenatide in 2005 based on its demonstrated effectiveness in controlling glycaemia and on the offered advantages in respect to other oral anti-diabetics, such as reduced onset of hypoglycaemia and weight loss promotion.² However, not long after the marketing of this drug, the first doubts about long-term adverse effects on pancreas were raised. In October 2007, based on 30 spontaneous reports of suspected pancreatitis caused by exenatide, the FDA published the first alert about the risk of acute pancreatitis and updated the product information leaflet. In 2008, this communication was even more reinforced by the FDA based on 6 more reports for haemorrhagic/necrotising pancreatitis.³ Later, in 2009, sitagliptin - the second drug belonging to this new class of anti-diabetics and authorized by FDA in 20064 – followed the same destiny. Since then several studies, independent and not – tried to assess if the pancreatitis risk is really higher for patients treated with incretin mimetics. These studies show contrasting results, however confirming that diabetes is a relevant risk factor for the disease.⁵

In 2011, this class of drugs has been associated for the first time to pancreas tumor.⁶ Once again the alert came from the analysis of spontaneous reports received by FDA, whose results needed more studies for their validation. Later, in February 2013, a case-control study of good methodology quality reported that the use of exenatide and sitagliptin could double the risk of hospitalization for acute pancreatitis in diabetic patients treated with these drugs.⁵
Since pancreatitis is considered a relevant risk factor for pancreas tumour, these results inevitably reopened questions about possible carcinogenic effects on pancreas of incretin mimetics.⁷
Few days after the study publication, also because of the strong concerns raised within the international scientific community, the European Medicine Agency (EMA) announced a revision of the evidences about long-term effects of incretin mimetics⁹ on pancreas. The EMA revision particularly aimed at investigating the previously cited study, which had carried out by a group of independent US researchers by analysing 34 samples of pancreatic tissue from deceased diabetic patients (whose death had not been caused by diabetes).⁸ These samples came from diabetic patients treated with incretin mimetics (n=8) or other drugs (n=12) and from non-diabetic subjects. The results of the analysis showed that the pancreas of incretin mimetic-treated patients had a mass 40% higher than diabetic patients treated with other drugs, with a pronounced tissue enlargement both for endocrine and exocrine pancreas, the first one associated to hyperplasia of alpha cells - potentially evolving in neuroendocrine tumour - and the second one associated to increased proliferation and dysplasia. In fact, 7 out of 8 samples from incretin mimetics treated patients showed relevant alpha-cell hyperplasia, 3 out of 8 had microadenomas secreting glucagon and one of these 3 samples had a neuroendocrine tumour at initial stage. Besides the increased risk for pancreatitis, these results linked incretin mimetics therapy to the onset of pre-cancer lesions and to metaplasia of the pancreatic duct, with possible subsequent progression to pancreas tumour and therefore in accordance with the already existing concern about possible long-term effects of incretin mimetics on pancreas.

The mechanism by means of which GLP-1 agonists or DPP-4 inhibitors cause pancreatitis and promote the onset of pancreas carcinoma remains unclear. However, it has been hypothesised (on the base of preclinical studies) that incretin mimetics – by stimulating GLP-1 receptors – could induce the proliferation of pancreatic acinar and ductal cells and therefore cause, in some subjects only, the occlusion of the duct itself.1 This obstruction would produce enough pressure for stimulating the release of digestive enzymes from acinar cells toward the lumen of the occluded duct, therefore causing the onset of pancreatitis, which is acknowledged as important risk factor for the subsequent developing of pancreas carcinoma.
Another possible mechanism by means of which the duct obstruction might take place is the hyperplasia of periductal alpha cells. The suppression of glucagon release by incretin mimetics, in fact, would determine higher alpha cells proliferation, which might potentially develop in neuroendocrine tumour.
It is well known that human pre-cancerous pancreatic lesions, which generally precede pancreas tumour, express the GLP-1 receptors. These pre-cancerous lesions seem to be present in more than 50% of middle-aged population. Therefore, similarly to oestrogen therapy and mammary gland cancer onset, it could be hypothesised that incretin mimetics therapy might promote the development of pre-existing pre-cancerous lesions at pancreatic level, increasing the probability of progression toward pancreas tumour.¹
In 2009, during some experiments on diabetic rats treated with sitagliptin, cases of pancreatic haemorrhage and precancerous metaplasia at the pancreatic duct level were observed. Two other studies on animals, ordered by the FDA itself, actually provided reassuring conclusions on the effects of exenatide and liraglutide on the exocrine pancreas, even though, during these last years, a rat died for pancreatic necrosis and increased levels of pancreatic enzymes have been generally registered, besides some cases of metaplasia and hyperplasia of the pancreatic duct.

The EMA revision of pancreas tumour risk induced by incretin mimetics drugs for the treatment of diabetes type 2 has been concluded in just 4 months. The EMA expert committee communicated that no available data confirm the concerns at present and reinforced the utility of these drugs for the treatment of diabetes type 2.¹⁰
Particularly regarding the study the revision focused on, the EMA concluded that methodological limits of the analysis – such as the low number of pancreatic samples and some potential bias like differences between the evaluated groups in respect to age, sex, disease’s duration and treatment – prevented from accepting those as conclusive results.
Anyway, considering the action mechanism of this class of drugs, long-term effects on pancreas remain uncertain. Further evidences will be provided by two large independent studies, which are expected to be concluded in 2014, funded by the EMA for assessing the pancreatic toxicity not just of incretin mimetics, but also for the other anti-diabetic drugs currently used.
The table shows the number of adverse events related to pancreas by incretin mimetics reported by the Italian Pharmacovigilance Network up to December the 31st, 2012.

Table 1 – Adverse pancreatic events in association with incretin mimetics, reported by the Italian Pharmacovigilance Network (data up to 31/12/2012)