Advertisement

GLP-1–Based Therapies: The Dilemma of Uncertainty

      See “Pancreatitis, pancreatic, and thyroid cancer with glucagon-like peptide-1–based therapies,” by Elashoff M, Matveyenko AV, Gier B, et al, on page 150.
      In this issue of Gastroenterology, Elashoff et al
      • Elashoff M.
      • Matveyenko A.V.
      • Gier B.
      • et al.
      Pancreatitis, pancreatic and thyroid cancer with glucagon-like peptide-1–based therapies.
      present an analysis of case reports of pancreatitis, pancreatic cancer, and thyroid cancer in patients treated with exenatide or sitagliptin for type-2 diabetes mellitus (T2DM), from the US Food and Drug Administration (FDA) database containing spontaneous reports of adverse drug reactions.
      • Elashoff M.
      • Matveyenko A.V.
      • Gier B.
      • et al.
      Pancreatitis, pancreatic and thyroid cancer with glucagon-like peptide-1–based therapies.
      The data raise safety concerns about incretin-based therapies, which are promising novel antidiabetic drugs.
      Historically, oral glucose administration was shown to increase insulin secretion to a greater extent than isoglycemic intravenous infusion, which was named the “incretin” effect of glucose. Subsequently, the incretin hormones, glucagon-like peptide (GLP)-1 and glucose-dependent insulinotropic peptide, were identified, that partially explained the effect. The improvement of glucose-dependent insulin secretion, the suppression of glucagon release, and the modification of satiety are major characteristics of GLP-1.
      • Brubaker P.L.
      Minireview: update on incretin biology: focus on glucagon-like peptide-1.
      Generally, 2 pharmacologically different approaches have been established to modify the incretin system. The GLP-1 receptor agonist exenatide and the GLP-1 analog liraglutide directly activate the GLP-1R and have been approved for patients with T2DM. The second principle is based on inhibiting the enzyme that degrades both GLP-1 and glucose-dependent insulinotropic peptide, dipeptidylpeptidase IV (DPP-IV). DPP-IV inhibitors (sitagliptin, vildagliptin, and saxagliptin) prolong the physiologic increase of incretins, although incretin-independent effects may also exist. All incretin-based therapies effectively improve glycemic control and lower hemoglobin A1c levels in T2DM patients. Notably, whereas exenatide and liraglutide reduce body weight, the DPP-IV inhibitors have little effect on body weight or body composition. Despite all this promise, no randomized, controlled trials demonstrating that GLP-1–based therapies improve patient-relevant endpoints such as myocardial infarction or stroke have been reported.

      Risk of Pancreatitis

      Acute pancreatitis accompanying exenatide was first reported in 2006.
      • Denker P.S.
      • Dimarco P.E.
      Exenatide (exendin-4)-induced pancreatitis: a case report.
      The FDA drew attention to a potential pancreatitis risk associated with sitagliptin in 2009.
      • Cure P.
      • Pileggi A.
      • Alejandro R.
      Exenatide and rare adverse events.
      Existing human studies on pancreatitis and incretin-based therapy are conflicting. Two recent analyses involving either exenatide or sitagliptin treatment compared these treatments with other anti-diabetic drug regimen. No relationship between GLP-1 therapy and pancreatitis was reported.
      • Dore D.D.
      • Bloomgren G.L.
      • Wenten M.
      • et al.
      A cohort study of acute pancreatitis in relation to exenatide use.
      • Dore D.D.
      • Seeger J.D.
      • Arnold Chan K.
      Use of a claims-based active drug safety surveillance system to assess the risk of acute pancreatitis with exenatide or sitagliptin compared to metformin or glyburide.
      Notably, Amylin Pharmaceuticals sponsored these studies and one could argue that interest conflicts may have existed, although the company at least made the effort to look at the problem. Merck also investigated the relation between sitagliptin and pancreatitis. When analyzing 19 controlled clinical trials including about 10,000 patients with T2DM, they found no increased risk of pancreatitis.
      • Engel S.S.
      • Williams-Herman D.E.
      • Golm G.T.
      • et al.
      Sitagliptin: review of preclinical and clinical data regarding incidence of pancreatitis.
      Although this study provides reasonable evidence, 2 limitations should be considered. First, not only did Merck sponsor the study, but also most of the studies included in the meta-analysis were industry funded, which raises concerns on study biases. Second, the treatment protocols of the approval studies may differ from clinical practice because, in most of the studies, gliptins were combined with metformin whereas in clinical practice monotherapies may be used more frequently. Another report analyzed phase II and III trials and concluded that an association between GLP-1 agonist use and acute pancreatitis could exist.
      • Anderson S.L.
      • Trujillo J.M.
      Association of pancreatitis with glucagon-like peptide-1 agonist use.
      Hence, clinical data are inconclusive. Are preclinical data giving more insight? Exocrine pancreatic cells express the GLP-1 receptor and administering GLP-1 has biological effects on these cells. In animal experiments, 1 week of GLP-1R activation with liraglutide or exenatide increased pancreatic weight and changed pancreatitis-associated gene expression, although the DPP-IV inhibitor, sitagliptin had no such effects.
      • Koehler J.A.
      • Baggio L.L.
      • Lamont B.J.
      • et al.
      Glucagon-like peptide-1 receptor activation modulates pancreatitis-associated gene expression but does not modify the susceptibility to experimental pancreatitis in mice.
      Theoretically, the increased pancreas weight may reflect edema or similar changes, although the underlying cause of the weight increase was not further analyzed and histopathology results were not reported.
      • Koehler J.A.
      • Baggio L.L.
      • Lamont B.J.
      • et al.
      Glucagon-like peptide-1 receptor activation modulates pancreatitis-associated gene expression but does not modify the susceptibility to experimental pancreatitis in mice.
      Within the same study, liraglutide and exenatide did not modify pancreatitis susceptibility in a specific experimental model of chemically induced pancreatitis.
      • Koehler J.A.
      • Baggio L.L.
      • Lamont B.J.
      • et al.
      Glucagon-like peptide-1 receptor activation modulates pancreatitis-associated gene expression but does not modify the susceptibility to experimental pancreatitis in mice.
      Regulatory required preclinical toxicologic studies were performed before approval of GLP-1–based therapies without pathologic findings indicative of pancreatitis. However, in Sprague-Dawley rats it was demonstrated that exenatide could induce pancreatic acinar inflammation.
      • Nachnani J.S.
      • Bulchandani D.G.
      • Nookala A.
      • et al.
      Biochemical and histological effects of exendin-4 (exenatide) on the rat pancreas.
      Thus, a low-grade subclinical inflammatory response to exenatide may exist.

      Risk of Pancreatic Cancer

      Besides an unexpectedly high number of pancreatitis, Elashoff et al report on pancreatic cancer in patients having been treated with exenatide or sitagliptin. Although no clinical study has been performed addressing this issue, the results are in line with a recent analysis of spontaneous reports from Germany performed by the Drug Commission of the German Medical Association. In the German database 11 reports of pancreatic cancer in association with exenatide were identified,
      Arzneimittelkommission der Deutschen ärzteschaft (Drug Commission of the German Medical Association) “Aus der UAW-Datenbank“: Pankreaskarzinome im Zusammenhang mit Exenatid (Byetta®).
      which was an unusual frequency of reports compared with other anti-diabetic medication. Notably, Elashoff et al report an association between sitagliptin and pancreatic cancer, whereas no such signal was found for DPP-IV inhibitors in the German database. Eventually different therapeutic habits, for example, predominantly combination with metformin in Europe versus more monotherapy in the United States may explain those differences. Experimental data may also explain potential differences between exenatide/liraglutide and DPP-IV inhibitors. Exenatide and liraglutide permanently activate the GLP-1 R, whereas DPP-IV inhibitors solely booster the physiologic incretin response, which results in temporally but not permanent GLP-1R activation. Apparently, other mechanisms may also explain differences between direct GLP-1R activators and DPP-IV inhibitors. In any case, the lack of an association between pancreatic cancer and sitagliptin in the German database argues against reporting bias as an exclusive explanation for the link between pancreatic cancer and treatment with exenatide.
      Arzneimittelkommission der Deutschen ärzteschaft (Drug Commission of the German Medical Association) “Aus der UAW-Datenbank“: Pankreaskarzinome im Zusammenhang mit Exenatid (Byetta®).
      If the association resulted exclusively from reporting bias, one might wonder why that bias would not occur for sitagliptin, although that drug is more often prescribed than exenatide and warnings about pancreatitis have been provided about both drugs to prescribers and the public.
      Might GLP-1–based therapies induce or promote pancreatic cancer? Recent studies suggested that the time between tumor induction, tumor development, and metastasis/clinical diagnosis is >10 years.
      • Yachida S.
      • Jones S.
      • Bozic I.
      • et al.
      Distant metastasis occurs late during the genetic evolution of pancreatic cancer.
      The German cases were individually assessed and the exposure time to exenatide was consistently found to be short (2–33 months). Thus, given that exenatide and liraglutide have been on the market for <10 years, there is considerable doubt that exenatide induces tumor development. If the suspected relationship exists, exenatide is likely to promote tumor progression rather than initiation, which corresponds with data showing that exenatide promotes pancreatic ductal hyperplasia.
      • Matveyenko A.V.
      • Dry S.
      • Cox H.I.
      • et al.
      Beneficial endocrine but adverse exocrine effects of sitagliptin in the human islet amyloid polypeptide transgenic rat model of type 2 diabetes: interactions with metformin.
      Do experimental data provide more insight? GLP-1 agonists increase β-cell mass in rodents.
      • Brubaker P.L.
      • Drucker D.J.
      Minireview: glucagon-like peptides regulate cell proliferation and apoptosis in the pancreas, gut, and central nervous system.
      Although human trials do not directly support the rodent data, this characteristic could potentially slow the progression of diabetes. Particularly, 3 GLP-1–induced pathways have been proposed, namely enhancement of β-cell proliferation, inhibition of β-cell apoptosis, and enhanced differentiation of adult stem cells in the ductal pancreatic epithelium.
      • Brubaker P.L.
      • Drucker D.J.
      Minireview: glucagon-like peptides regulate cell proliferation and apoptosis in the pancreas, gut, and central nervous system.
      GLP-1–dependent effects on β-cell proliferation include activated signaling pathways, such as phosphatidylinositol-3 kinase, Akt, mitogen-activated protein kinase, protein kinase Cζ, the src kinase, and the epidermal growth factor receptor.
      • Brubaker P.L.
      • Drucker D.J.
      Minireview: glucagon-like peptides regulate cell proliferation and apoptosis in the pancreas, gut, and central nervous system.
      Although β-cell proliferation may be beneficial with regard to progression of T2DM, similar trophic mechanisms in other cell types might be detrimental. Thus, the epidermal growth factor receptor system and src have been implicated in the pathogenesis and progression of numerous malignant tumors, including pancreatic cancer; mechanisms increasing the activity of these pathways could promote tumor development or progression. Although there is no evidence yet that those systems are affected in non–β-cells by GLP-1, nagging doubts remain (Figure 1).
      Figure thumbnail gr1
      Figure 1GLP-1 agonists increase β-cell mass in rodents. GLP-1–dependent effects on β-cell proliferation include activation of phosphatidylinositol-3 (PI3) kinase, AKT, mitogen-activated protein kinase (MAPK), protein kinase Cζ, the src kinase, and the epidermal growth factor receptor.
      • Brubaker P.L.
      • Drucker D.J.
      Minireview: glucagon-like peptides regulate cell proliferation and apoptosis in the pancreas, gut, and central nervous system.
      Similar trophic mechanisms have been implicated in the pathogenesis of pancreatic cancer and activation of those mechanisms in cancer progenitor cells might be detrimental.
      Sitagliptin induced metaplasia of pancreatic ductal cells in an animal model, an effect that was ameliorated by the simultaneous administration of metformin.
      • Matveyenko A.V.
      • Dry S.
      • Cox H.I.
      • et al.
      Beneficial endocrine but adverse exocrine effects of sitagliptin in the human islet amyloid polypeptide transgenic rat model of type 2 diabetes: interactions with metformin.
      Accordingly, metformin may counter the actions of GLP-1 activation on the exocrine pancreas. Given that DPP-4 inhibitors may be used in combination with metformin, any adverse actions of DPP-4 inhibitors could theoretically be suppressed by metformin. Notably, none of the preclinical chronic studies required by drug regulatory bodies have reported an increased frequency of malignancies in the pancreas for exenatide or sitagliptin alone.

      Risk of Thyroid Cancer

      Liraglutide was associated with medullary thyroid cancer in rodents.
      Victoza ® (liraglutide injection): Human Relevance of Rodent Thyroid C-cell Tumors.
      Despite those findings, liraglutide was approved considering data that suggested a non-genotoxic mechanism and also indicated that the GLP-1R expression in rodent thyroid C cells is higher than in human cells. Marketing approval was given, and linked to specific postmarketing surveillance activities to address this question. The link between exenatide and thyroid cancer presented by Elashoff et al raises some concern. However, medullary cancer was not distinguished from other types of thyroid cancer in their study. Thus, there is currently no adequately powered study providing sufficiently robust evidence about whether or not GLP-1–based therapies are associated with thyroid carcinoma.
      Spontaneous reporting databases were established to support postmarketing surveillance, specifically considering rare adverse events that are missed in preapproval clinical trials. Obviously, such databases have serious limitations. First, reporting bias is likely to exist. In cases where warnings have been issued, physicians are probably aware of possible drug-related problems, which results in increased surveillance and reporting. Safety information on pancreatitis might have increased the frequency in reporting suspected adverse reactions affecting the pancreas in association with exenatide and sitagliptin as compared with other antidiabetic drugs. On the other hand, some studies demonstrated that because of underreporting only 5%–10% of serious adverse reactions can be detected by spontaneous reporting systems.
      • Hazell L.
      • Shakir S.A.
      Under-reporting of adverse drug reactions: a systematic review.
      Whether similar underreporting exists with respect to incretin-based therapies is unclear. Analysis of spontaneous reports does not allow adjustment for known risk factors for pancreatic cancer such as obesity, smoking, family history, or chronic pancreatitis, even if this information is available. This is a major drawback of Elashoff et al's analysis, given that T2DM is a common sequelae of obesity and that T2DM itself is associated with solid tumors, including pancreatic cancer. Furthermore, databases on spontaneous adverse drug reports generally do not provide detailed information on patient characteristics and cannot quantify the number of patients being exposed to the drug. Thus, incidence rates cannot be calculated, and although an association with an adverse event might be detected by analyzing spontaneous reports, the nature and extent of the problem cannot be addressed with certainty.
      The study by Elashoff et al is faced with these limitations and should therefore be interpreted with caution. Cynics may argue that the pharmaceutical industry has an incentive to downplay the relevance of adverse findings from spontaneous reports. On the other hand, the weaknesses inherent in Elashoff et al's study should not be ignored. The conundrum must be viewed against the background of the fact that drug regulatory agencies are unlikely to receive data on drug safety that are independent of industry ties. Moreover, university-based medicine institutions have not viewed the problem of drug surveillance as a worthy academic pursuit. Until surveillance tools devoid of industry influence have been established to provide more robust data, such dilemmas of uncertainty regarding adverse effects will remain unsolved.

       Quo vadis?

      Elashoff et al draw attention to an association between incretin-based therapies and pancreatitis, pancreatic cancer. and thyroid cancer. The observations raise safety concerns particularly considering that potential mechanisms have been suggested in preclinical studies. Data from spontaneous reports, including this study, are hampered by certain biases. Thus, more reliable studies must be performed. Until then, the questions about the potential link between GLP-1–based therapies and pancreatitis and tumorigenesis remain open. The fact that GLP-1–based therapies have no record of lowering clinical endpoints (ie, mortality, stroke, myocardial infarction) does not inspire confidence and more informative prospective randomized trials with such patient-relevant clinical endpoints are urgently required. The past serves us with uncomfortable reminders. Some years ago, a small report from the World Health Organization adverse events database drew attention to the fact that rosiglitazone, an accepted diabetes treatment of the time, was associated with an increased number of cardiovascular events.
      • Cohen D.
      Rosiglitazone: what went wrong?.
      Notably, rosiglitazone therapy was also devoid of clinical endpoint data, such as cardiovascular morbidity and mortality. When effects on such clinically relevant endpoints were sought, none were found other than the fact that rosigliatzone increased bone fractures, body weight, and heart failure in patients with diabetes.
      • Nissen S.E.
      • Wolski K.
      Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes.
      • Home P.D.
      • Pocock S.J.
      • Beck-Nielsen H.
      • et al.
      JJ; RECORD Study Team Rosiglitazone evaluated for cardiovascular outcomes in oral agent combination therapy for type 2 diabetes (RECORD): a multicentre, randomised, open-label trial.
      For the sake of patients, history will hopefully not repeat itself in the case of GLP-1–based drugs. Two thoughts remain: primum non nocere and “vigilance equals avoidance.”

      Acknowledgment

      The authors thank Friedrich Luft for his thoughtful comments.

      References

        • Elashoff M.
        • Matveyenko A.V.
        • Gier B.
        • et al.
        Pancreatitis, pancreatic and thyroid cancer with glucagon-like peptide-1–based therapies.
        Gastroenterology. 2011; 141: 150-156
        • Brubaker P.L.
        Minireview: update on incretin biology: focus on glucagon-like peptide-1.
        Endocrinology. 2010; 151: 1984-1989
        • Denker P.S.
        • Dimarco P.E.
        Exenatide (exendin-4)-induced pancreatitis: a case report.
        Diabetes Care. 2006; 29: 471
        • Cure P.
        • Pileggi A.
        • Alejandro R.
        Exenatide and rare adverse events.
        N Engl J Med. 2008; 358: 1969-1970
        • Dore D.D.
        • Bloomgren G.L.
        • Wenten M.
        • et al.
        A cohort study of acute pancreatitis in relation to exenatide use.
        Diabetes Obes Metab. 2011; 13: 559-566
        • Dore D.D.
        • Seeger J.D.
        • Arnold Chan K.
        Use of a claims-based active drug safety surveillance system to assess the risk of acute pancreatitis with exenatide or sitagliptin compared to metformin or glyburide.
        Curr Med Res Opin. 2009; 25: 1019-1027
        • Engel S.S.
        • Williams-Herman D.E.
        • Golm G.T.
        • et al.
        Sitagliptin: review of preclinical and clinical data regarding incidence of pancreatitis.
        Int J Clin Pract. 2010; 64: 984-990
        • Anderson S.L.
        • Trujillo J.M.
        Association of pancreatitis with glucagon-like peptide-1 agonist use.
        Ann Pharmacother. 2010; 44: 904-909
        • Koehler J.A.
        • Baggio L.L.
        • Lamont B.J.
        • et al.
        Glucagon-like peptide-1 receptor activation modulates pancreatitis-associated gene expression but does not modify the susceptibility to experimental pancreatitis in mice.
        Diabetes. 2009; 58: 2148-2161
      1. Scientific discussion.
        (Accessed April 26, 2011)
        • Nachnani J.S.
        • Bulchandani D.G.
        • Nookala A.
        • et al.
        Biochemical and histological effects of exendin-4 (exenatide) on the rat pancreas.
        Diabetologia. 2010; 53: 153-159
      2. Arzneimittelkommission der Deutschen ärzteschaft (Drug Commission of the German Medical Association).
        Dtsch Arztebl. 2011; 108 (A–1080)
        • Yachida S.
        • Jones S.
        • Bozic I.
        • et al.
        Distant metastasis occurs late during the genetic evolution of pancreatic cancer.
        Nature. 2010; 467: 1114-1117
        • Matveyenko A.V.
        • Dry S.
        • Cox H.I.
        • et al.
        Beneficial endocrine but adverse exocrine effects of sitagliptin in the human islet amyloid polypeptide transgenic rat model of type 2 diabetes: interactions with metformin.
        Diabetes. 2009; 58: 1604-1615
        • Brubaker P.L.
        • Drucker D.J.
        Minireview: glucagon-like peptides regulate cell proliferation and apoptosis in the pancreas, gut, and central nervous system.
        Endocrinology. 2004; 145: 2653-2659
      3. Scientific discussion.
        (Accessed May 03, 2011)
      4. Victoza ® (liraglutide injection): Human Relevance of Rodent Thyroid C-cell Tumors.
        (Accessed April 26, 2011)
        • Hazell L.
        • Shakir S.A.
        Under-reporting of adverse drug reactions: a systematic review.
        Drug Saf. 2006; 29: 385-396
        • Cohen D.
        Rosiglitazone: what went wrong?.
        BMJ. 2010; 6 (341): 530-534
        • Nissen S.E.
        • Wolski K.
        Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes.
        N Engl J Med. 2007; 356: 2457-2471
        • Home P.D.
        • Pocock S.J.
        • Beck-Nielsen H.
        • et al.
        JJ; RECORD Study Team.
        Lancet. 2009; 373: 2125-2135

      Linked Article