Editorials EDITORIAL (SEE DRUCKER ET AL., P. 428)

GLP-1–Based Therapy for Diabetes: What You Do Not Know Can Hurt You

ccording to the Oxford Dictionary of oped as novel approaches for the treat- tential signals have already emerged in the Proverbs, the oldest written version ment of type 2 diabetes building on case of GLP-1 mimetic therapy, one is A of the saying “What you don’t know elegant studies of basic physiology. There pancreatitis (12–14) and another, which can’t hurt you” comes from Petit Palace, was a clear and rational initial therapeutic is currently confined to rodents studies, is written in 1576 by G. Pettie: “So long as I target with both classes of drugs: insulin thyroid cancer (11). know it not, it hurteth mee not.” resistance for PPAR-␥ agonists and en- Pancreatitis first emerged as a poten- In this issue of Diabetes Care, Drucker hanced glucose-mediated insulin secre- tial side effect of therapy with exenatide, et al. (1) conclude that the safety profile of tion for the GLP-1 class of drugs. initially reported as case reports (12–14) the newly available glucagon-like peptide Before either class of drugs reached and subsequently by numerous reports 1 (GLP-1) class of drugs is favorable in market, possible additional attractive at- made through the U.S. Food and Drug comparison to their benefits as therapy, tributes were identified mostly through Administration (FDA) adverse reporting and the class of drugs might be consid- rodent studies. In the case of the PPAR-␥ mechanism. The Amylin Corporation’s ered as next in line after metformin for agonist drugs, the most widely antici- response to this putative link has been to treatment for type 2 diabetes. The pur- pated additional benefit was decreased suggest that it was a consequence of guilt pose of this counterpoint is to suggest vascular disease because of favorable ef- by association rather than a drug effect such a conclusion is premature. History fects of the drug class on risk factors for since pancreatitis is more common in in- has taught us that enthusiasm for new vascular disease supported by murine dividuals with obesity and type 2 diabetes classes of drugs, heavily promoted by the studies reporting protection against isch- (15). The Amylin Corporation also sug- pharmaceutical companies that market emic heart disease (3). Not until the Eu- gested that since no mechanism is known them, can obscure the caution that should ropean regulatory authorities required to link GLP-1 mimetic therapy to pancre- be exercised when the long-term conse- appropriately powered studies to demon- atitis, the association is unlikely causal. quences are unknown. Of perhaps great- strate vascular benefit to support these Pancreatitis was also seen in clinical stud- est concern in the case of the GLP-1– claims were such studies undertaken ies of the GLP-1 agonist liraglutide (16). based drugs, including GLP-1 agonists (4,5). The results, despite optimistic in- More recently, the FDA has reported and dipeptidyl peptidase-4 (DPP-4) in- terpretation by the sponsors, showed lit- more than 80 documented cases of pan- hibitors, is preliminary evidence to sug- tle if any cardiovascular benefit that could creatitis in patients treated with sitaglip- gest the potential risks of asymptomatic not have been a consequence of glucose tin, a DPP-4 inhibitor (17). It is also chronic pancreatitis and, with time, pan- lowering with some suggestion that the Merck’s position that the reported pan- creatic cancer. net effects of some agents might be harm- creatitis with sitagliptin therapy is due to The GLP-1–related drugs arrived in ful on vascular disease (6). the increased risk of pancreatitis in type 2 clinical practice with much fanfare and an- History may be repeating itself with diabetes rather than a consequence of ticipation. As summarized in the article by the GLP-1 class of drugs. Putative benefits drug therapy (18), mimicking the Amylin Drucker et al., it is a class of drugs that has of GLP-1 mimetic therapy, in addition to Corporation position. potential benefits in the treatment of type 2 enhanced insulin secretion, have been In post-marketing studies sponsored diabetes. The concept of gut-related factors proposed and often arise from rodent by the marketing companies, no in- that enhance glucose-mediated insulin se- studies. These benefits include cardiovas- creased signal for acute pancreatitis has cretion, the incretin effect, has been recog- cular protection against ischemia and pre- been identified (19,20). However, the du- nized for many years (2). Once it was vention and/or reversal of the defect in ration of treatment in those studies is typ- demonstrated that an intravenous infusion ␤-cell mass that is characteristic of type 2 ically short, the quality of the patient of GLP-1 could decrease blood glucose con- diabetes (7,8). While these attributes follow-up is questionable, and evidence centrations in patients with type 2 diabetes, would be highly desirable, there is no cur- that prescriptions were actually taken is the race was on to exploit the properties of rent data available to support either of absent. Nonetheless, on the basis of the this action. Many millions of dollars have these claims in humans, and recent stud- available clinical information, we agree been invested by the pharmaceutical indus- ies imply that the beneficial effects on with the conclusions of Drucker et al. (1) try in developing products, the first of ␤-cell mass in part may be an artifact of that the data required to link GLP-1 ther- which are now in clinical practice. Many studies in juvenile rodents (9–11). apy and acute pancreatitis is currently in- millions of dollars therefore are now also What is the risk profile of GLP-1 complete. However, in the context of a invested to market the new agents, reminis- drugs? Perhaps the parallel with the new class of medical therapy, the proverb cent of the period that followed the launch PPAR-␥ receptor agonists is again worth “What you do not know cannot hurt you” of the most recent new class of drugs for considering. The receptors targeted by clearly does not apply. We feel that type 2 diabetes, the peroxisome prolifera- each drug—the PPAR-␥ receptor and the enough preliminary evidence has accu- tor–activated receptor-␥ (PPAR-␥) agonists. GLP-1 receptor, respectively—are widely mulated to suggest that there is a plausible The parallels with the launch of the distributed in numerous tissues with as risk that long-term recipients of GLP-1– PPAR-␥ agonist and GLP-1 mimetic class yet ill-defined roles. As such, it is not sur- based therapy may develop asymptomatic of drugs is worthy of comparison. GLP-1 prising when unintended consequences chronic pancreatitis (Fig. 1), and worse, and PPAR-␥ agonist therapies were devel- of chronic receptor activation emerge. Po- subsequently a minority of individuals care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 2, FEBRUARY 2010 453 Editorial

frequency of ductal replication. Therefore, arguably the HIP rat successfully predicts both the increased risk of pancreatitis with sitagliptin and the decreased risk of pancre- atic cancer in individuals with type 2 diabe- tes treated with metformin (28). Exenatide therapy given over 75 days to rats induced low-grade chronic pancreatitis (26). Again, as in the case of the sitagliptin- treated HIP rats, there was no discernable clinical manifestation of the low-grade pan- creatitis induced by exenatide, with the rats in no apparent pain. If GLP-1 mimetic ther- apy with either GLP-1 mimetic therapy or DPP-4 inhibition induces asymptomatic chronic pancreatitis in rats, how do we know that a similar effect is not present in humans using these therapies? If GLP-1– based therapy causes low chronic pancre- atitis, why was this not established in toxicology studies? One possibility is that since ductal replication is increased with obesity or type 2 diabetes (24), and GLP-1 may amplify this, studies in lean nondia- betic animals may have had a limited pro- pensity to GLP-1–induced pancreatitis. Also, most toxicology studies are carried out Figure 1—Theoretical model to explain currently available observations with increased risks for in juvenile mice in which the pancreas is pancreatic cancer in individuals with obesity and type 2 diabetes, a risk that is decreased by still growing. Enhanced ductal replication metformin treatment and theoretically may be increased by GLP-1–based treatment. under these circumstances may simply lead to pancreas growth as observed by Koehler treated by this class of drugs may develop able in rodents. Koehler et al. (25) re- et al. (25) rather than distortion of the archi- pancreatic cancer. ported no evidence of GLP-1–induced tecture of the acinar to duct relationship, The incidence of both pancreatitis pancreatitis based on RNA levels in mice, thus predisposing to chronic pancreatitis. and pancreatic cancer is increased in in- but histology was not provided and num- While low-grade asymptomatic pan- dividuals with obesity and/or type 2 dia- bers of mice in most experimental groups creatitis in and of itself as a result of GLP- betes (21–23), although the underlying (n ϳ5) were perhaps small to conclude a 1–based therapy would not be a cause for mechanisms are not well understood (Fig. negative finding. On the other hand, both major concern, the problem is that it rep- 1). One potential link is the frequency of sitagliptin and exenatide have been resents a risk for pancreatic cancer (21). pancreatic duct replication, which is in- shown to induce pancreatitis in rats The risk for developing pancreatic cancer creased in humans with obesity and/or (26,27). Sitagliptin administered to the increases with the duration of chronic type 2 diabetes (24). It is not known why high-fat–fed human islet amyloid pancreatitis (22). Because medications for ductal turnover is increased with obesity polypeptide (HIP) rat model of type 2 di- type 2 diabetes may be taken for many and type 2 diabetes. One of the conse- abetes amplified the increased pancreatic years, if GLP-1–based therapy did induce quences of chronically increased ductal duct cell replication present in that model low-grade asymptomatic pancreatitis, replication can be distortion of small pan- (27). Moreover, sitagliptin therapy in- there is a real concern that such a therapy creatic ducts with subsequent outflow duced acinar to ductal metaplasia in might increase the risk for pancreatic can- obstruction of pancreatic enzymes pro- ϳ30% of treated animals. Acinar to ductal cer. Even if this is a relatively small risk viding a plausible mechanistic link be- metaplasia follows increased ductal repli- (which we do not know), how many of us tween obesity and/or diabetes and the cation in the morphological progression practicing physicians would choose a increased risk for pancreatitis. Moreover, of chronic pancreatitis to pancreatic ade- therapeutic strategy for ourselves with in- increased ductal replication and chronic nocarcinoma (Fig. 1) (15). sight into the potential for this risk? Since pancreatitis are both risk factors for pan- Was this finding a quirk of the HIP rat metformin has been shown to decrease creatic cancer (21). Given the apparent model of type 2 diabetes? Perhaps, but it is the risk of pancreatic cancer, at the least signal of occasional acute pancreatitis in of interest to note that increased ductal rep- we would suggest that GLP-1–based patients treated with GLP-1–based ther- lication in the HIP rat model of type 2 dia- medications should be reserved for pa- apy, how do we reassure ourselves that betes compared with wild-type rats tients taking metformin. asymptomatic chronic pancreatitis is not reproduces that which was observed in hu- In conclusion, we believe it is prema- also induced in some patients? mans with type 2 diabetes compared with ture to conclude that the GLP-1 class of The most significant challenge is lim- nondiabetic individuals (24). Moreover, drugs has been established as having a good ited access to the human pancreas. To metformin therapy in the HIP rat had the safety profile and is appropriate for a rela- date there are also limited studies avail- opposite effect of sitagliptin, decreasing the tively early choice of therapy for type 2 di-

454 DIABETES CARE, VOLUME 33, NUMBER 2, FEBRUARY 2010 care.diabetesjournals.org Butler, Dry, and Elashoff abetes. There are grounds for concern that active Study (PROspective pioglitAzone Available from http://www.fda.gov/Safety/ the GLP-1 class of drugs may induce Clinical Trial In macroVascular Events): a MedWatch/SafetyInformation/SafetyAlerts asymptomatic pancreatitis and, over time in randomised controlled trial. Lancet 2005; forHumanMedicalProducts/ucm183800. some individuals, induce pancreatic cancer. 366:1279–1289 htm. Accessed 12 October 2009 At present, these concerns are based on lim- 5. Home PD, Pocock SJ, Beck-Nielsen H, 18. Heavey S. UPDATE 3-US FDA sees Curtis PS, Gomis R, Hanefeld M, Jones pancreatitis link with Merck’s Januvia ited data. However, the implications of the NP, Komajda M, McMurray JJ, RECORD [article online], Reuters, 25 Septem- data are sufficiently serious that continuing Study Team. Rosiglitazone evaluated for ber 2009. Available from http://www. to promote this class of drugs without es- cardiovascular outcomes in oral agent reuters.com/article/companyNewsAndPR/ tablishing clear experimental evidence to combination therapy for type 2 diabetes idUSN2550919420090925. Accessed 12 permit the concern to be rejected is irre- (RECORD): a multicentre, randomised, October 2009 sponsible. Moreover, arguably patients pre- open-label trial. Lancet 2009;373:2125– 19. Williams-Herman D, Round E, Swern AS, scribed these drugs should be made aware 2135 Musser B, Davies MJ, Stein PP, Kaufman of the potential risks of pancreatic cancer. 6. Nissen SE, Wolski K. Effect of rosiglita- KD, Amatruda JM. Safety and tolerability Otherwise, we collectively subscribe to the zone on the risk of myocardial infarction of sitagliptin in patients with type 2 dia- proverb “What you do not know cannot and death from cardiovascular causes. betes: a pooled analysis. BMC Endocr Dis- N Engl J Med 2007;356:2457–2471 ord 2008;8:14 hurt you” and, in the case of new drug ther- 7. Noyan-Ashraf MH, Momen MA, Ban K, 20. Dore DD, Seeger JD, Arnold Chan K. Use apy, this proverb has already been shown to Sadi AM, Zhou YQ, Riazi AM, Baggio LL, of a claims-based active drug safety sur- be flawed. Henkelman RM, Husain M, Drucker DJ. veillance system to assess the risk of acute GLP-1R agonist liraglutide activates cyto- pancreatitis with exenatide or sitagliptin PETER C. BUTLER, MD protective pathways and improves out- compared to metformin or glyburide. SARAH DRY, MD comes after experimental myocardial Curr Med Res Opin 2009;25:1019–1027 ROBERT ELASHOFF, PHD infarction in mice. Diabetes 2009;58: 21. Jura N, Archer H, Bar-Sagi D. Chronic 975–983 pancreatitis, pancreatic adenocarcinoma From the Larry L. Hillblom Islet Research Center, 8. Buteau J. GLP-1 receptor signaling: effects and the black box in-between. Cell Res Department of Pathology and Department of Bi- on pancreatic beta-cell proliferation and 2005;15:72–77 omathematics, University of California, Los An- survival. Diabete Metab 2008;34(Suppl 22. Rebours V, Boutron-Ruault MC, Schnee geles, Los Angeles, California. 2):S73–S77 M, Fe´rec C, Le Mare´chal C, Hentic O, Corresponding author: Peter Butler, pbutler@ 9. Tschen SI, Dhawan S, Gurlo T, Bhushan Maire F, Hammel P, Ruszniewski P, Le´vy mednet.ucla.edu. DOI: 10.2337/dc09-1902 A. Age-dependent decline in beta cell pro- P. The natural history of hereditary pan- © 2010 by the American Diabetes Association. liferation restricts the capacity of beta cell creatitis: a national series. Gut 2009;58: Readers may use this article as long as the work is regeneration in mice. Diabetes 2009 97–103 properly cited, the use is educational and not for 10. Rankin MM, Kushner JA. Adaptive ␤-cell 23. Li D, Yeung SC, Hassan MM, Konopleva profit, and the work is not altered. See http:// proliferation is severely restricted with M, Abbruzzese JL. Antidiabetic therapies creativecommons.org/licenses/by-nc-nd/3.0/ for advanced age. Diabetes 2009;58:1365– affect risk of pancreatic cancer. Gastroen- details. 1372 terology 2009;137:482–488 11. Parnaud G, Bosco D, Berney T, Pattou F, 24. Butler AE, Galasso R, Matveyenko AV, Kerr-Conte J, Donath MY, Bruun C, Man- Rizza RA, Dry S, Butler PC. Pancreatic Acknowledgments— No potential conflicts of drup-Poulsen T, Billestrup N, Halban PA. duct replication is increased with obesity interest relevant to this article were reported. Proliferation of sorted human and rat beta and type 2 diabetes in humans. Diabeto- cells. Diabetologia 2008;51:91–100 logia, In Press, 2009 ●●●●●●●●●●●●●●●●●●●●●●● 12. Denker PS, Dimarco PE. Exenatide (ex- 25. Koehler JA, Baggio LL, Lamont BJ, Ali S, References endin-4)-induced pancreatitis: a case re- Drucker DJ. Glucagon-like peptide-1 re- 1. Drucker DJ, Sherman SI, Gorelick FS, port. Diabetes Care 2006;29:471 ceptor activation modulates pancreatitis- Bergenstal RM, Sherwin RS, Buse JB. In- 13. Cure P, Pileggi A, Alejandro R. Exenatide associated gene expression but does not cretin-based therapies for the treatment and rare adverse events. N Engl J Med modify the susceptibility to experimental of type 2 diabetes: evaluation of the risks 2008;358:1969–1970 pancreatitis in mice. Diabetes 2009; and benefits. Diabetes Care 2010;33: 14. Tripathy NR, Basha S, Jain R, Shetty S, 58:2148–2161 428–433 Ramachandran A. Exenatide and acute 26. Nachnani JS, Bulchandani DG, Nookala 2. Holst JJ, Vilsbøll T, Deacon CF. The incre- pancreatitis. J Assoc Physicians India A, Herndon B, Molteni A, Pandya P, Tay- tin system and its role in type 2 diabetes 2008;56:987–988 lor R, Quinn T, Weide L, Alba LM. Bio- mellitus. Mol Cell Endocrinol 2009;297: 15. Whitcomb DC. Mechanisms of disease: chemical and histological effects of 127–136 advances in understanding the mecha- exendin-4 (exenatide) on the rat pan- 3. Staels B, Fruchart JC. Therapeutic roles of nisms leading to chronic pancreatitis. Nat creas. Diabetologia, 2009 [Epub ahead of peroxisome proliferator-activated receptor Clin Pract Gastroenterol Hepatol 2004;1: print] agonists. Diabetes 2005;54:2460–2470 46–52 27. Matveyenko AV, Dry S, Cox HI, Mosh- 4. Dormandy JA, Charbonnel B, Eckland DJ, 16. Buse JB, Rosenstock J, Sesti G, Schmidt taghian A, Gurlo T, Galasso R, Butler AE, Erdmann E, Massi-Benedetti M, Moules IK, WE, Montanya E, Brett JH, Zychma M, Butler PC. Beneficial endocrine but ad- Skene AM, Tan MH, Lefe`bvre PJ, Murray Blonde L, the LEAD-6 Study Group. Lira- verse exocrine effects of sitagliptin in the GD, Standl E, Wilcox RG, Wilhelmsen L, glutide once a day versus exenatide twice human islet amyloid polypeptide trans- Betteridge J, Birkeland K, Golay A, Heine RJ, a day for type 2 diabetes: a 26-week ran- genic rat model of type 2 diabetes: inter- Kora´nyi L, Laakso M, Moka´n M, Norkus A, domised, parallel-group, multinational, actions with metformin. Diabetes 2009; Pirags V, Podar T, Scheen A, Scherbaum W, open-label trial (LEAD-6). Lancet 2009; 58:1604–1615 Schernthaner G, Schmitz O, Skrha J, Smith 374:39–47 28. Currie CJ, Poole CD, Gale EA. The influ- U, Taton J, the PROactive investigators. Sec- 17. U.S. Food and Drug Administration. Sitaglip- ence of glucose-lowering therapies on ondary prevention of macrovascular events tin (marketed as Januvia and Janumet)–acute cancer risk in type 2 diabetes. Diabetolo- in patients with type 2 diabetes in the PRO- pancreatitis [Internet], 25 September 2009. gia 2009;52:1766–1777 care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 2, FEBRUARY 2010 455