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

A ccording to the Oxford Dictionary ofProverbs, the oldest written versionof the saying “What you don’t know

can’t hurt you” comes from Petit Palace,written in 1576 by G. Pettie: “So long as Iknow it not, it hurteth mee not.”

In this issue of Diabetes Care, Druckeret al. (1) conclude that the safety profile ofthe newly available glucagon-like peptide1 (GLP-1) class of drugs is favorable incomparison to their benefits as therapy,and the class of drugs might be consid-ered as next in line after metformin fortreatment for type 2 diabetes. The pur-pose of this counterpoint is to suggestsuch a conclusion is premature. Historyhas taught us that enthusiasm for newclasses of drugs, heavily promoted by thepharmaceutical companies that marketthem, can obscure the caution that shouldbe exercised when the long-term conse-quences are unknown. Of perhaps great-est concern in the case of the GLP-1–based drugs, including GLP-1 agonistsand dipeptidyl peptidase-4 (DPP-4) in-hibitors, is preliminary evidence to sug-gest the potential risks of asymptomaticchronic pancreatitis and, with time, pan-creatic cancer.

The GLP-1–related drugs arrived inclinical practice with much fanfare and an-ticipation. As summarized in the article byDrucker et al., it is a class of drugs that haspotential benefits in the treatment of type 2diabetes. The concept of gut-related factorsthat enhance glucose-mediated insulin se-cretion, the incretin effect, has been recog-nized for many years (2). Once it wasdemonstrated that an intravenous infusionof GLP-1 could decrease blood glucose con-centrations in patients with type 2 diabetes,the race was on to exploit the properties ofthis action. Many millions of dollars havebeen invested by the pharmaceutical indus-try in developing products, the first ofwhich are now in clinical practice. Manymillions of dollars therefore are now alsoinvested to market the new agents, reminis-cent of the period that followed the launchof the most recent new class of drugs fortype 2 diabetes, the peroxisome prolifera-tor–activated receptor-� (PPAR-�) agonists.

The parallels with the launch of thePPAR-� agonist and GLP-1 mimetic classof drugs is worthy of comparison. GLP-1and PPAR-� agonist therapies were devel-

oped as novel approaches for the treat-ment of type 2 diabetes building onelegant studies of basic physiology. Therewas a clear and rational initial therapeutictarget with both classes of drugs: insulinresistance for PPAR-� agonists and en-hanced glucose-mediated insulin secre-tion for the GLP-1 class of drugs.

Before either class of drugs reachedmarket, possible additional attractive at-tributes were identified mostly throughrodent studies. In the case of the PPAR-�agonist drugs, the most widely antici-pated additional benefit was decreasedvascular disease because of favorable ef-fects of the drug class on risk factors forvascular disease supported by murinestudies reporting protection against isch-emic heart disease (3). Not until the Eu-ropean regulatory authorities requiredappropriately powered studies to demon-strate vascular benefit to support theseclaims were such studies undertaken(4,5). The results, despite optimistic in-terpretation by the sponsors, showed lit-tle if any cardiovascular benefit that couldnot have been a consequence of glucoselowering with some suggestion that thenet effects of some agents might be harm-ful on vascular disease (6).

History may be repeating itself withthe GLP-1 class of drugs. Putative benefitsof GLP-1 mimetic therapy, in addition toenhanced insulin secretion, have beenproposed and often arise from rodentstudies. These benefits include cardiovas-cular protection against ischemia and pre-vention and/or reversal of the defect in�-cell mass that is characteristic of type 2diabetes (7,8). While these attributeswould be highly desirable, there is no cur-rent data available to support either ofthese claims in humans, and recent stud-ies imply that the beneficial effects on�-cell mass in part may be an artifact ofstudies in juvenile rodents (9–11).

What is the risk profile of GLP-1drugs? Perhaps the parallel with thePPAR-� receptor agonists is again worthconsidering. The receptors targeted byeach drug—the PPAR-� receptor and theGLP-1 receptor, respectively—are widelydistributed in numerous tissues with asyet ill-defined roles. As such, it is not sur-prising when unintended consequencesof chronic receptor activation emerge. Po-

tential signals have already emerged in thecase of GLP-1 mimetic therapy, one ispancreatitis (12–14) and another, whichis currently confined to rodents studies, isthyroid cancer (11).

Pancreatitis first emerged as a poten-tial side effect of therapy with exenatide,initially reported as case reports (12–14)and subsequently by numerous reportsmade through the U.S. Food and DrugAdministration (FDA) adverse reportingmechanism. The Amylin Corporation’sresponse to this putative link has been tosuggest that it was a consequence of guiltby association rather than a drug effectsince pancreatitis is more common in in-dividuals with obesity and type 2 diabetes(15). The Amylin Corporation also sug-gested that since no mechanism is knownto link GLP-1 mimetic therapy to pancre-atitis, the association is unlikely causal.Pancreatitis was also seen in clinical stud-ies of the GLP-1 agonist liraglutide (16).More recently, the FDA has reportedmore than 80 documented cases of pan-creatitis in patients treated with sitaglip-tin, a DPP-4 inhibitor (17). It is alsoMerck’s position that the reported pan-creatitis with sitagliptin therapy is due tothe increased risk of pancreatitis in type 2diabetes rather than a consequence ofdrug therapy (18), mimicking the AmylinCorporation position.

In post-marketing studies sponsoredby the marketing companies, no in-creased signal for acute pancreatitis hasbeen identified (19,20). However, the du-ration of treatment in those studies is typ-ically short, the quality of the patientfollow-up is questionable, and evidencethat prescriptions were actually taken isabsent. Nonetheless, on the basis of theavailable clinical information, we agreewith the conclusions of Drucker et al. (1)that the data required to link GLP-1 ther-apy and acute pancreatitis is currently in-complete. However, in the context of anew class of medical therapy, the proverb“What you do not know cannot hurt you”clearly does not apply. We feel thatenough preliminary evidence has accu-mulated to suggest that there is a plausiblerisk that long-term recipients of GLP-1–based therapy may develop asymptomaticchronic pancreatitis (Fig. 1), and worse,subsequently a minority of individuals

E d i t o r i a l sE D I T O R I A L ( S E E D R U C K E R E T A L . , P . 4 2 8 )

care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 2, FEBRUARY 2010 453

treated by this class of drugs may developpancreatic cancer.

The incidence of both pancreatitisand pancreatic cancer is increased in in-dividuals with obesity and/or type 2 dia-betes (21–23), although the underlyingmechanisms are not well understood (Fig.1). One potential link is the frequency ofpancreatic duct replication, which is in-creased in humans with obesity and/ortype 2 diabetes (24). It is not known whyductal turnover is increased with obesityand type 2 diabetes. One of the conse-quences of chronically increased ductalreplication can be distortion of small pan-creatic ducts with subsequent outflowobstruction of pancreatic enzymes pro-viding a plausible mechanistic link be-tween obesity and/or diabetes and theincreased risk for pancreatitis. Moreover,increased ductal replication and chronicpancreatitis are both risk factors for pan-creatic cancer (21). Given the apparentsignal of occasional acute pancreatitis inpatients treated with GLP-1–based ther-apy, how do we reassure ourselves thatasymptomatic chronic pancreatitis is notalso induced in some patients?

The most significant challenge is lim-ited access to the human pancreas. Todate there are also limited studies avail-

able in rodents. Koehler et al. (25) re-ported no evidence of GLP-1–inducedpancreatitis based on RNA levels in mice,but histology was not provided and num-bers of mice in most experimental groups(n �5) were perhaps small to conclude anegative finding. On the other hand, bothsitagliptin and exenatide have beenshown to induce pancreatitis in rats(26,27). Sitagliptin administered to thehigh-fat–fed human islet amyloidpolypeptide (HIP) rat model of type 2 di-abetes amplified the increased pancreaticduct cell replication present in that model(27). Moreover, sitagliptin therapy in-duced acinar to ductal metaplasia in�30% of treated animals. Acinar to ductalmetaplasia follows increased ductal repli-cation in the morphological progressionof chronic pancreatitis to pancreatic ade-nocarcinoma (Fig. 1) (15).

Was this finding a quirk of the HIP ratmodel of type 2 diabetes? Perhaps, but it isof interest to note that increased ductal rep-lication in the HIP rat model of type 2 dia-betes compared with wild-type ratsreproduces that which was observed in hu-mans with type 2 diabetes compared withnondiabetic individuals (24). Moreover,metformin therapy in the HIP rat had theopposite effect of sitagliptin, decreasing the

frequency of ductal replication. Therefore,arguably the HIP rat successfully predictsboth the increased risk of pancreatitis withsitagliptin 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 torats induced low-grade chronic pancreatitis(26). Again, as in the case of the sitagliptin-treated HIP rats, there was no discernableclinical manifestation of the low-grade pan-creatitis induced by exenatide, with the ratsin no apparent pain. If GLP-1 mimetic ther-apy with either GLP-1 mimetic therapy orDPP-4 inhibition induces asymptomaticchronic pancreatitis in rats, how do weknow that a similar effect is not present inhumans using these therapies? If GLP-1–based therapy causes low chronic pancre-atitis, why was this not established intoxicology studies? One possibility is thatsince ductal replication is increased withobesity or type 2 diabetes (24), and GLP-1may 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 outin juvenile mice in which the pancreas isstill growing. Enhanced ductal replicationunder these circumstances may simply leadto pancreas growth as observed by Koehleret al. (25) rather than distortion of the archi-tecture of the acinar to duct relationship,thus predisposing to chronic pancreatitis.

While low-grade asymptomatic pan-creatitis in and of itself as a result of GLP-1–based therapy would not be a cause formajor concern, the problem is that it rep-resents a risk for pancreatic cancer (21).The risk for developing pancreatic cancerincreases with the duration of chronicpancreatitis (22). Because medications fortype 2 diabetes may be taken for manyyears, if GLP-1–based therapy did inducelow-grade asymptomatic pancreatitis,there is a real concern that such a therapymight increase the risk for pancreatic can-cer. Even if this is a relatively small risk(which we do not know), how many of uspracticing physicians would choose atherapeutic strategy for ourselves with in-sight into the potential for this risk? Sincemetformin has been shown to decreasethe risk of pancreatic cancer, at the leastwe would suggest that GLP-1– basedmedications should be reserved for pa-tients taking metformin.

In conclusion, we believe it is prema-ture to conclude that the GLP-1 class ofdrugs has been established as having a goodsafety profile and is appropriate for a rela-tively early choice of therapy for type 2 di-

Figure 1—Theoretical model to explain currently available observations with increased risks forpancreatic cancer in individuals with obesity and type 2 diabetes, a risk that is decreased bymetformin treatment and theoretically may be increased by GLP-1–based treatment.

Editorial

454 DIABETES CARE, VOLUME 33, NUMBER 2, FEBRUARY 2010 care.diabetesjournals.org

abetes. There are grounds for concern thatthe GLP-1 class of drugs may induceasymptomatic pancreatitis and, over time insome individuals, induce pancreatic cancer.At present, these concerns are based on lim-ited data. However, the implications of thedata are sufficiently serious that continuingto promote this class of drugs without es-tablishing clear experimental evidence topermit the concern to be rejected is irre-sponsible. Moreover, arguably patients pre-scribed these drugs should be made awareof the potential risks of pancreatic cancer.Otherwise, we collectively subscribe to theproverb “What you do not know cannothurt you” and, in the case of new drug ther-apy, this proverb has already been shown tobe flawed.

PETER C. BUTLER, MD

SARAH DRY, MD

ROBERT ELASHOFF, PHD

From the Larry L. Hillblom Islet Research Center,Department of Pathology and Department of Bi-omathematics, University of California, Los An-geles, Los Angeles, California.

Corresponding author: Peter Butler, pbutler@mednet.ucla.edu.

DOI: 10.2337/dc09-1902© 2010 by the American Diabetes Association.

Readers may use this article as long as the work isproperly cited, the use is educational and not forprofit, and the work is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ fordetails.

Acknowledgments— No potential conflicts ofinterest relevant to this article were reported.

● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●

References1. Drucker DJ, Sherman SI, Gorelick FS,

Bergenstal RM, Sherwin RS, Buse JB. In-cretin-based therapies for the treatmentof type 2 diabetes: evaluation of the risksand benefits. Diabetes Care 2010;33:428 – 433

2. Holst JJ, Vilsbøll T, Deacon CF. The incre-tin system and its role in type 2 diabetesmellitus. Mol Cell Endocrinol 2009;297:127–136

3. Staels B, Fruchart JC. Therapeutic roles ofperoxisome proliferator-activated receptoragonists. Diabetes 2005;54:2460–2470

4. Dormandy JA, Charbonnel B, Eckland DJ,Erdmann E, Massi-Benedetti M, Moules IK,Skene AM, Tan MH, Lefebvre PJ, MurrayGD, Standl E, Wilcox RG, Wilhelmsen L,Betteridge J, Birkeland K, Golay A, Heine RJ,Koranyi L, Laakso M, Mokan M, Norkus A,Pirags V, Podar T, Scheen A, Scherbaum W,Schernthaner G, Schmitz O, Skrha J, SmithU, Taton J, the PROactive investigators. Sec-ondary prevention of macrovascular eventsin patients with type 2 diabetes in the PRO-

active Study (PROspective pioglitAzoneClinical Trial In macroVascular Events): arandomised controlled trial. Lancet 2005;366:1279–1289

5. Home PD, Pocock SJ, Beck-Nielsen H,Curtis PS, Gomis R, Hanefeld M, JonesNP, Komajda M, McMurray JJ, RECORDStudy Team. Rosiglitazone evaluated forcardiovascular outcomes in oral agentcombination therapy for type 2 diabetes(RECORD): a multicentre, randomised,open-label trial. Lancet 2009;373:2125–2135

6. Nissen SE, Wolski K. Effect of rosiglita-zone on the risk of myocardial infarctionand death from cardiovascular causes.N Engl J Med 2007;356:2457–2471

7. Noyan-Ashraf MH, Momen MA, Ban K,Sadi AM, Zhou YQ, Riazi AM, Baggio LL,Henkelman RM, Husain M, Drucker DJ.GLP-1R agonist liraglutide activates cyto-protective pathways and improves out-comes after experimental myocardialinfarction in mice. Diabetes 2009;58:975–983

8. Buteau J. GLP-1 receptor signaling: effectson pancreatic beta-cell proliferation andsurvival. Diabete Metab 2008;34(Suppl2):S73–S77

9. Tschen SI, Dhawan S, Gurlo T, BhushanA. Age-dependent decline in beta cell pro-liferation restricts the capacity of beta cellregeneration in mice. Diabetes 2009

10. Rankin MM, Kushner JA. Adaptive �-cellproliferation is severely restricted withadvanced age. Diabetes 2009;58:1365–1372

11. Parnaud G, Bosco D, Berney T, Pattou F,Kerr-Conte J, Donath MY, Bruun C, Man-drup-Poulsen T, Billestrup N, Halban PA.Proliferation of sorted human and rat betacells. Diabetologia 2008;51:91–100

12. Denker PS, Dimarco PE. Exenatide (ex-endin-4)-induced pancreatitis: a case re-port. Diabetes Care 2006;29:471

13. Cure P, Pileggi A, Alejandro R. Exenatideand rare adverse events. N Engl J Med2008;358:1969–1970

14. Tripathy NR, Basha S, Jain R, Shetty S,Ramachandran A. Exenatide and acutepancreatitis. J Assoc Physicians India2008;56:987–988

15. Whitcomb DC. Mechanisms of disease:advances in understanding the mecha-nisms leading to chronic pancreatitis. NatClin Pract Gastroenterol Hepatol 2004;1:46–52

16. Buse JB, Rosenstock J, Sesti G, SchmidtWE, Montanya E, Brett JH, Zychma M,Blonde L, the LEAD-6 Study Group. Lira-glutide once a day versus exenatide twicea day for type 2 diabetes: a 26-week ran-domised, parallel-group, multinational,open-label trial (LEAD-6). Lancet 2009;374:39–47

17. U.S. Food and Drug Administration. Sitaglip-tin (marketed as Januvia and Janumet)–acutepancreatitis [Internet], 25 September 2009.

Available from http://www.fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/ucm183800.htm. Accessed 12 October 2009

18. Heavey S. UPDATE 3-US FDA seespancreatitis link with Merck’s Januvia[article online], Reuters, 25 Septem-ber 2009. Available from http://www.reuters.com/article/companyNewsAndPR/idUSN2550919420090925. Accessed 12October 2009

19. Williams-Herman D, Round E, Swern AS,Musser B, Davies MJ, Stein PP, KaufmanKD, Amatruda JM. Safety and tolerabilityof sitagliptin in patients with type 2 dia-betes: a pooled analysis. BMC Endocr Dis-ord 2008;8:14

20. Dore DD, Seeger JD, Arnold Chan K. Useof a claims-based active drug safety sur-veillance system to assess the risk of acutepancreatitis with exenatide or sitagliptincompared to metformin or glyburide.Curr Med Res Opin 2009;25:1019–1027

21. Jura N, Archer H, Bar-Sagi D. Chronicpancreatitis, pancreatic adenocarcinomaand the black box in-between. Cell Res2005;15:72–77

22. Rebours V, Boutron-Ruault MC, SchneeM, Ferec C, Le Marechal C, Hentic O,Maire F, Hammel P, Ruszniewski P, LevyP. The natural history of hereditary pan-creatitis: a national series. Gut 2009;58:97–103

23. Li D, Yeung SC, Hassan MM, KonoplevaM, Abbruzzese JL. Antidiabetic therapiesaffect risk of pancreatic cancer. Gastroen-terology 2009;137:482–488

24. Butler AE, Galasso R, Matveyenko AV,Rizza RA, Dry S, Butler PC. Pancreaticduct replication is increased with obesityand type 2 diabetes in humans. Diabeto-logia, In Press, 2009

25. Koehler JA, Baggio LL, Lamont BJ, Ali S,Drucker DJ. Glucagon-like peptide-1 re-ceptor activation modulates pancreatitis-associated gene expression but does notmodify the susceptibility to experimentalpancreatitis in mice. Diabetes 2009;58:2148–2161

26. Nachnani JS, Bulchandani DG, NookalaA, Herndon B, Molteni A, Pandya P, Tay-lor R, Quinn T, Weide L, Alba LM. Bio-chemical and histological effects ofexendin-4 (exenatide) on the rat pan-creas. Diabetologia, 2009 [Epub ahead ofprint]

27. Matveyenko AV, Dry S, Cox HI, Mosh-taghian A, Gurlo T, Galasso R, Butler AE,Butler PC. Beneficial endocrine but ad-verse exocrine effects of sitagliptin in thehuman islet amyloid polypeptide trans-genic rat model of type 2 diabetes: inter-actions with metformin. Diabetes 2009;58:1604–1615

28. Currie CJ, Poole CD, Gale EA. The influ-ence of glucose-lowering therapies oncancer risk in type 2 diabetes. Diabetolo-gia 2009;52:1766–1777

Butler, Dry, and Elashoff

care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 2, FEBRUARY 2010 455