00b7d526687272df11000000

Effects of Exenatide (Exendin-4) onGlycemic Control Over 30 Weeks inPatients With Type 2 Diabetes TreatedWith Metformin and a SulfonylureaDAVID M. KENDALL, MD1

MATTHEW C. RIDDLE, MD2

JULIO ROSENSTOCK, MD3

DONGLIANG ZHUANG, PHD4

DENNIS D. KIM, MD4

MARK S. FINEMAN, BS4

ALAIN D. BARON, MD4

OBJECTIVE This study evaluated the effects of exenatide, a novel incretin mimetic, inhyperglycemic patients with type 2 diabetes unable to achieve glycemic control with metformin-sulfonylurea combination therapy.

RESEARCH DESIGN AND METHODS A 30-week, double-blind, placebo-controlled study was performed in 733 subjects (aged 55 10 years, BMI 33.6 5.7 kg/m2,A1C 8.5 1.0%; means SD) randomized to 5 g subcutaneous exenatide b.i.d. (arms A andB) or placebo for 4 weeks. Thereafter, arm A remained at 5 g b.i.d. and arm B escalated to 10g b.i.d. Subjects continued taking their dose of metformin and were randomized to eithermaximally effective (MAX) or minimum recommended (MIN) doses of sulfonylurea.

RESULTS Week 30 A1C changes from baseline (SE) were0.8 0.1% (10g),0.60.1% (5 g), and 0.2 0.1% (placebo; adjusted P 0.0001 vs. placebo), yielding placebo-adjusted reductions of 1.0% (10 g) and 0.8% (5 g). In the evaluable population, ex-enatide-treated subjects were more likely to achieve A1C 7% than placebo-treated subjects(34% [10 g], 27% [5 g], and 9% [placebo]; P 0.0001). Both exenatide arms demonstratedsignificant weight loss (1.6 0.2 kg from baseline each exenatide arm,0.9 0.2 kg placebo;P 0.01 vs. placebo). Mild or moderate nausea was the most frequent adverse event. Theincidence of mild/moderate hypoglycemia was 28% (10g), 19% (5g), and 13% (placebo) andappeared lower with MIN than with MAX sulfonylurea treatment.

CONCLUSIONS Exenatide significantly reduced A1C in patients with type 2 diabetesunable to achieve adequate glycemic control with maximally effective doses of combinedmetformin-sulfonylurea therapy. This improvement in glycemic control was associated with noweight gain and was generally well tolerated.

Diabetes Care 28:10831091, 2005

In most individuals with type 2 diabe-tes, hyperglycemia results from a fail-ure of -cell insulin secretory capacityto adequately compensate for insulin re-sistance in peripheral tissues (1,2). Re-sults from the U.K. Prospective DiabetesStudy indicate that -cell failure is a pro-gressive defect even in the setting of effec-tive glucose-lowering therapy with diet,metformin, sulfonylureas, or insulin (35). Despite the use of traditional thera-pies, it is often difficult to both achieveand maintain adequate glycemic controlin the majority of patients with type 2 dia-betes. Indeed, population-based glycemiccontrol has not improved substantially inthe past decade, with current A1C values8% in many patients (6,7). In addition,many of the currently available therapeuticagents have treatment-limiting side ef-fects, such as weight gain, hypoglycemia,gastrointestinal intolerance, and periph-eral edema, any of which may limit theuse of some agents, reduce treatmentcompliance, and make it difficult toachieve glycemic treatment targets (8,9).

Despite the introduction of a numberof new classes of medication for managingtype 2 diabetes, recent studies (6,8,10,11)suggest that even currently available mul-tidrug therapies often do not allow pa-tients to achieve adequate glycemiccontrol. The most common combinationof oral therapies used for patients withtype 2 diabetes is metformin with a sul-fonylurea. Both drugs are safe and effec-tive when used either as monotherapy orin combination, albeit with a risk of sig-nificant, drug-associated side effects(6,8,10,11). Clinicians and researcherscurrently debate the question of whichtherapies are best when combinationtherapy no longer maintains acceptableglycemic control. Addition of insulin or athiazolidinedione to the existing oraltherapies are leading choices, but each isassociated with characteristic side effectsincluding weight gain (8,12,13).

Glucoregulatory peptides, such as in-

From the 1International Diabetes Center and University of Minnesota, Minneapolis, Minnesota; the 2Sectionof Diabetes, Oregon Health & Science University, Portland, Oregon; the 3Dallas Diabetes and EndocrineCenter, Dallas, Texas; and 4Amylin Pharmaceuticals, San Diego, California.

Address correspondence and reprint requests to Alain D. Baron, MD, Senior Vice President, Research,Amylin Pharmaceuticals, 9360 Towne Centre Dr., Suite 110, San Diego, CA 92121. E-mail: [email protected].

Received for publication 27 September 2004 and accepted in revised form 31 January 2005.D.M.K. has served on an advisory panel for Amylin/Lilly; has received honoraria from Eli Lilly, Amylin,

Novo Nordisk, and Merck; and has received research/grant support from Eli Lilly, Amylin, Novo Nordisk,Merck, and GlaxoSmithKline. M.C.R. has received honoraria/consulting fees from Amylin and Amylin/Lilly.J.R. has served on advisory boards for and has received honoraria/consulting fees from Pfizer, Sanofi-Aventis,Novo Nordisk, GlaxoSmithKline, Takeda, Centocor, Johnson & Johnson, and Amylin and has received grantsupport from Merck, Pfizer, Sanofi-Aventis, Novo Nordisk, Eli Lilly, GlaxoSmithKline, Takeda, Novartis,AstraZeneca, Amylin, Sankyo, and MannKind.

Abbreviations: AUC, area under the curve; GLP-1, glucagon-like peptide 1; ITT, intent to treat.A table elsewhere in this issue shows conventional and Syste`me International (SI) units and conversion

factors for many substances. 2005 by the American Diabetes Association.The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby

marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

E m e r g i n g T r e a t m e n t s a n d T e c h n o l o g i e sO R I G I N A L A R T I C L E

DIABETES CARE, VOLUME 28, NUMBER 5, MAY 2005 1083

cretins, are currently being investigated aspotential therapies for type 2 diabetes(14). Exenatide (exendin-4) is a 39aminoacid peptide incretin mimetic that exhib-its glucoregulatory activities similar to themammalian incretin hormone glucagon-like peptide 1 (GLP-1) (1519). The pres-ent study evaluated the ability of exena-tide to improve glycemic control in pa-tients with type 2 diabetes unable toachieve adequate glycemic control de-spite treatment with combined metfor-min-sulfonylurea therapy.

RESEARCH DESIGN ANDMETHODS Subjects were 2277years of age with type 2 diabetes treatedwith metformin and a sulfonylurea. Gen-eral inclusion criteria were a screeningfasting plasma glucose concentration13.3 mmol/l, BMI 2745 kg/m2, inclu-sive, and an A1C value of 7.511.0%. Themetformin dose was1,500 mg/day, andthe sulfonylurea dose was at least themaximally effective dose for 3 months be-fore screening. In addition, subjects wereweight stable (10%) for 3 months be-fore screening and had no clinically rele-vant (for a type 2 diabetes population)abnormal laboratory test values (25%outside normal laboratory values). Fe-male subjects were postmenopausal, sur-gically sterile, or using contraceptives forat least 3 months before screening andcontinuing throughout the study. Sub-jects were excluded if they had evidenceof other clinically significant medical con-ditions or had used thiazolidinediones,meglitinides, -glucosidase inhibitors,exogenous insulin, or weight loss drugswithin the prior 3 months. Further ex-clusion criteria included therapy withcorticosteroids, drugs known to affectgastrointestinal motility, transplantationmedications, or any investigational drug.

Seven hundred thirty-three adultswith type 2 diabetes participated at 91sites in the U.S. (May 2002 to August2003). A common clinical protocol wasapproved for each site by an institutionalreview board and in accordance with theprinciples described in the Declaration ofHelsinki, including all amendmentsthrough the 1996 South Africa revision(20). All subjects provided written in-formed consent before participation.

This was a balanced, randomized,double-blind, placebo-controlled, paral-lel-group clinical study designed afterconsultation with the U.S. Food and Drug

Administration to evaluate glycemic con-trol, primarily as assessed by A1C, andsafety. The study commenced with a4-week, single-blind, lead-in period withinjection of placebo twice daily. Thereaf-ter, subjects were randomized to one offour treatment arms. As nausea had beenthe most frequent adverse event in earlierclinical trials (21) and gradual dose esca-lation attenuated this side effect, the cur-rent study included an acclimation periodof 4 weeks at a lower fixed dose of ex-enatide (5 g b.i.d.) in the active treat-ment arms (A and B) before the fixed doseof exenatide was either increased to 10gb.i.d. (arm B) or left at 5 g b.i.d. (arm A)for the duration of the study. Placebo in-jections with volumes equal those used inarms A and B were administered in treat-ment arms C and D. Study medicationwas self-injected subcutaneously in theabdomen within 15 min before meals inthe morning and evening.

All subjects continued their prestudymetformin regimen. To standardize sul-fonylurea use in the clinical trial, subjectswere randomized (one for one) to eithermaximally effective sulfonylurea dose(MAX group; 4 mg/day glimepiride, 20mg/day glipizide, 10 mg/day glipizide XL,10 mg/day glibenclamide [glyburide], 6mg/day micronized glibenclamide, 350mg/day chlorpropamide, 500 mg/day to-lazamide, or 1,500 mg/day tolbutamide)or to minimum recommended dose (MINgroup; 1 mg/day glimepiride, 5 mg/dayglipizide, 5 mg/day glipizide XL, 1.25 mg/day glibenclamide, 0.75 mg/day micron-i zed g l ibenc l amide , 100 mg/day chlorpropamide, 100 mg/day tolaz-amide, or 250 mg/day tolbutamide).This approach allowed assessment of theinfluence of concurrent sulfonylurea dos-ing on hypoglycemia risk. The assign-ment to the sulfonylurea managementgroup was not blinded.

During the blinded phase of thestudy, sulfonylurea dose could be re-duced by 50%, regardless of the subjectsassigned sulfonylurea managementgroup, in the event of one documentedhypoglycemic event (blood glucose con-centration3.3 mmol/l) or two undocu-mented suspected hypoglycemic events.Further 50% reductions, including com-plete cessation of sulfonylurea dose, wereallowed upon repetition of the previouscriteria. In subjects assigned to the MINgroup, if a majority of fasting plasmaglucose readings exceeded 6.9 mmol/l

during weeks 4 6, 6 9, or 9 12, theinvestigator was to consider doubling thesulfonylurea dose. No further escalationof sulfonylurea dose was permitted afterweek 12. Any subject with either an A1Cchange of 1.5% from baseline at anyclinic visit or an A1C11.5% at week 18or 24 could be withdrawn from the study.Similarly, subjects could be withdrawn ifthey had fasting plasma glucose values13.3 mmol/l on two consecutive studyvisits during weeks 1824 or if a subjectconsistently recorded finger-stick fastingblood glucose values14.4 mmol/l for atleast 2 weeks during weeks 1824, notsecondary to a readily identified illness orpharmacological treatment.

A subset of subjects (meal cohort) un-derwent a standardized meal tolerancetest on day 1 (placebo administered to allsubjects), at week 4, and at week 30. Afteran overnight fast of 8 h, subjects re-ceived their morning dose of metforminand sulfonylurea within 1 h of their clinicvisit. Exenatide or placebo was injected15 min before a standardized breakfast.The size of the meal was calculated indi-vidually at screening to provide 20% of asubjects total daily caloric requirements,with a macronutrient composition of 55%carbohydrate, 15% protein, and 30% fatbased on body weight and activity level.The size of the standardized breakfast wasthe same on each test day for each indi-vidual subject.

Study end pointsThe primary outcome measures were gly-cemic control, as assessed by change inA1C, and safety. Secondary objectives in-cluded examining the effects of exenatideon fasting and postprandial (meal cohortonly) plasma glucose concentrations,body weight, fasting plasma lipids, andexenatide pharmacokinetics (meal cohortonly).

Statistical analysisRandomization was stratified accordingto screening A1C values (9.0 and9.0%) to achieve a balanced distribu-tion of subjects across treatment arms.Within each arm, subjects were random-ized in a one-for-one ratio to either remainon their maximally effective sulfonylureadose or to decrease their sulfonylurea tothe minimum recommended dose. Asample size of 600 subjects who were ex-pected to complete treatment through week30 were estimated to provide at least 90%

Exenatide and glycemic control

1084 DIABETES CARE, VOLUME 28, NUMBER 5, MAY 2005

power to detect a difference of 0.5% in thechange from baseline in A1C values be-tween at least one exenatide treatmentarm and placebo ( 0.05; Fishers pro-tected testing procedure). Placebo arms Cand D were combined for all analyses.

All inferential statistical tests wereconducted at the significance level of 0.05(two sided). A general linear model wasused to test for differences in the changefrom baseline to each visit in A1C andweight across treatments (22,23). Factorsin the model included treatment, strata ofbaseline A1C, sulfonylurea managementgroup, and aggregate study site as fixedeffects. Before data analysis, sites werepooled according to geographic locationto prevent loss of too many degrees offreedom in the model. Pairwise compari-sons of the treatment effects were per-formed using Fishers protected testingprocedure to control type I errors due tomultiple comparisons (24). Similar anal-yses were performed for each fasting met-abolic parameter and for postprandialplasma glucose concentrations withoutadjusting for the multiple comparisons.

The intent-to-treat (ITT) populationwas defined as all randomized subjectswho received at least one injection of ran-domized medication starting from theevening of day 1. All efficacy and safetyanalyses were performed on the ITT pop-ulation, with the exception of the percent-age of subjects achieving A1C 7%. Forthe latter analysis, the more clinically rel-evant population of evaluable subjects (allrandomized subjects who completedtreatment through week 30 and receivedat least 80% of the study medication in-jections) with baseline A1C 7% wasused. For ITT subjects who had recordedvalues for at least one scheduled visit sub-sequent to the baseline measurement,missing data were imputed from sched-uled visits using the last-observation-carried-forward method. Results aregiven as means SE for the ITT popula-tion, unless otherwise indicated. The pro-portion of subjects achieving A1C targetvalues were compared using the Cochran-Mantel-Haenszel test. Post hoc evaluation ofchange in body weight versus duration ofnausea was performed using regressionanalysis.

Safety analysisAll safety analyses were performed usingthe ITT population. Safety end points in-cluded adverse events, clinical laboratory

tests, physical examination, 12-lead elec-trocardiogram, vital signs, and titering ofanti-exenatide antibodies. Treatment-emergent adverse events were defined asthose occurring upon or after receiving thefirst randomized dose. The intensity of hy-poglycemic episodes was defined as mild/moderate or severe. For mild/moderatehypoglycemia, subjects reported symp-toms consistent with hypoglycemia thatmay have been documented by a plasmaglucose concentration value (3.33mmol/l). For severe hypoglycemia, sub-jects required the assistance of anotherperson to obtain treatment for their hypo-glycemia, including intravenous glucoseor intramuscular glucagon.

AssaysPlasma analytes and A1C were quanti-tated by Quintiles Laboratories (Smyrna,GA) or Esoterix Endocrinology (Cala-basas Hills, CA) using standard meth-ods. A1C was measured using a high-performance liquid chromatographymethodology (25,26). Plasma exenatideand anti-exenatide antibodies were mea-sured as described previously (16).

RESULTS

Baseline characteristics of subjectsBaseline demographics demonstratingbalanced randomization are given in Fig.1. Seven hundred thirty-three subjectswere randomized to treatment and re-ceived at least one dose of study medica-tion (ITT population); 593 subjects(81%) completed the entire study. Allsubjects were treated with metformin anda sulfonylurea during the course of thestudy (43% glipizide, 42% glibenclamide,14% glimepiride, 3% glibenclamide com-bination with metformin, 1% tolaz-amide, and 1% chlorpropamide). Fiftypercent of ITT subjects were treated withan ACE inhibitor, 38% with an anti-thrombotic agent, and 53% with a serumlipidreducing agent.

A1C and plasma glucoseThere was a modest decline in A1C valuesin all treatment arms during the placeborun-in period and the initial 2 weeks ofthe study after randomization (ITT popu-lation; Fig. 2). Thereafter, A1C values inthe exenatide arms continued to declinefor another 12 weeks of observation,while A1C values in the placebo arm re-mained stable and then gradually in-

creased such that these values returned tobaseline by the end of the study. Ex-enatide treatment arms had significant re-ductions in A1C from baseline comparedwith placebo (P 0.0001; Fig. 1A).

When stratified by baseline A1C9%, significant reductions in A1C frombaseline in each exenatide arm comparedwith placebo were observed (adjustedP 0.0002 for pairwise comparisons).Similarly, for subjects with baseline A1C9%, exenatide treatment at 5 or 10 gresulted in significant reductions frombaseline in A1C, compared with an in-crease in the placebo arm, at week 30(adjusted P 0.0001 for pairwise com-parisons).

Of ITT subjects at week 30 with base-line A1C7%, 30% (70 of 230 subjects)in the 10-g exenatide arm and 24% (56of 235 subjects) in the 5-g exenatidearm reached an A1C 7%, and this pro-portion was significantly greater than thatobserved in the placebo arm (7% [16 of238 subjects]; P 0.0001 for pairwisecomparisons). Of subjects evaluable atweek 30 with baseline A1C 7% (n 554), 34% (60 of 179 subjects) in the10-g exenatide arm and 27% (54 of 197subjects) in the 5-g exenatide armreached an A1C 7%, and this propor-tion was significantly greater than in theplacebo arm (9% [16 of 174 subjects];P 0.0001 for pairwise comparisons).

At baseline, fasting plasma glucoseconcentrations were similar among treat-ment arms (Fig. 1). At week 30, fastingplasma glucose concentrations were re-duced 0.6 0.2 mmol/l in the 10-gexenatide arm and0.5 0.2 mmol/l inthe 5-g arm compared with an increaseof0.8 0.2 mmol/l in the placebo arm(P 0.0001 for pairwise comparisons;ITT population).

Postprandial plasma glucose concen-trations were evaluated in the subset ofsubjects who underwent a standardizedmeal tolerance test (Fig. 1). At baseline,all subjects received a placebo injec-tion, and postprandial plasma glucosegeometric mean area under the curve(AUC)(15180 min) values were similaracross treatment arms (Fig. 3A). At week4, there was a significant decrease in post-prandial plasma glucose geometric meanAUC(15180 min) in both exenatide armscompared with the placebo arm (P 0.001), and this pattern was sustained atweek 30 (Fig. 3B; P 0.001 for pairwisecomparisons). Exenatide also reduced in-

Kendall and Associates


cremental postprandial plasma glucoseconcentrations at weeks 4 and 30, as mea-sured by incremental AUC(15180 min) andincremental Cave (average concentration;P 0.01 vs. placebo). At week 30,changes from baseline for AUC(15180 min)were474 87 mmol min 11 (10-garm),318 49 mmol min 11 (5-garm), and3 72 mmol min 11 (pla-cebo arm) and for Cave were 2.9 0.76

mmol/l (10-g-arm), 1.9 0.3 mmol/l(5-g arm), and 0.0 0.4 mmol/l (placeboarm). Thus, incremental plasma glucoseAUC values were reduced by 59% (5-garm) and 87% (10-g arm) compared witha1% decrease in the placebo arm.

Body weightSubjects in the exenatide arms had pro-gressive weight loss over the entire 30-

week treatment period: week 30 reduc-tion of 1.6 0.2 kg from baseline ineach exenatide arm compared with0.9 0.2 kg from baseline in the pla-cebo arm (P 0.01 vs. placebo; ITT pop-ulation; Fig. 2C).

SafetyThere was no evidence of cardiovascular,pulmonary, hepatic, or renal toxicity with

Figure 1Protocol flow chart. One subject was randomized twice. Both screening numbers were counted in the randomized population, but only onewas counted in the ITT population. Subjects in the 10-g exenatide b.i.d. arm received 5 g exenatide b.i.d. during weeks 04. Subjects in alltreatment arms were maintained on metformin and a sulfonylurea. Data are means SD.



exenatide treatment or of drug-relatedidiosyncratic side effects associated with itsuse. The incidence of serious (5% in 10-garm, 6% in 5-g arm, and 6% in the pla-cebo arm) and severe (12% in 10-g arm,14% in 5-g arm, and 8% in the placeboarm) adverse events was low and evenly dis-tributed across treatment arms. The mostfrequent adverse events were mostly mild ormoderate and gastrointestinal in nature (Ta-ble 1). Nausea was the most frequent severeadverse event (3% in 10-g arm, 5% in5-g arm, and 1% in the placebo arm),with a low incidence of withdrawals due tonausea (4% in 10-g arm, 2% in 5-g arm,and1% in the placebo arm). Nausea wasreported at a higher incidence during theinitial weeks of therapy (weeks 08) and

declined thereafter (Fig. 4). There was nocorrelation between change in body weightand duration of nausea. Post hoc analysis ofnausea and body weight change showed alack of correlation between change inbody weight and nausea duration (10gexenatide: Y 0.013 1.625; R2 0.056; 5 g exenatide: Y 0.003

1.273; R2 0.003; placebo: Y 0.013 1.596; R2 0.028). Subjectswho never experienced nausea still lostweight: 1.1 0.3 kg (10-g arm) and1.7 0.2 kg (5-g arm).

The overall incidence of hypoglyce-mia was higher in each exenatide treat-ment arm compared with the placebo arm(Table 1). One case of hypoglycemia re-quiring assistance from another individ-

ual (severe hypoglycemia), but no medi-cal intervention, occurred in the 5-g arm.All other hypoglycemic events were mildor moderate in intensity, and there wereno withdrawals due to hypoglycemia.

At week 30, 49% of exenatide-treatedsubjects (193 of 398) had detectable anti-exenatide antibody titers, with the major-ity of titers in the low range (1/5 to 1/125titer). The anti-exenatide antibodies hadno predictive effect on the magnitude ofan individuals glycemic response or theincidence of adverse events.

Management of sulfonylurea dosingIn the MAX group, all treatment armsmaintained relatively constant dosage lev-els of sulfonylurea throughout the study.

Figure 2Glycemic control in subjects with type 2 diabetes treatedwithmetformin and a sulfonylurea plus exenatide or placebo (ITT population)A: A1Cvalues over the course of the study (raw data). B: Change in A1C over 30 weeks. *Adjusted P 0.0001 compared with placebo. Week 30 changes in A1Cvalues from baseline were0.77 0.08% (10-g arm; adjusted P 0.0001 vs. placebo),0.55 0.07% (5-g arm; adjusted P 0.0001 vs. placebo),and0.23 0.07% (placebo arm).C: Week 30 change in A1C stratified by baseline A1C. For subjects with baseline A1C9%, baseline A1C values were7.92 0.04% (n 169), 7.91 0.04% (n 172), and 7.94 0.04% (n 172) for the 10-g exenatide, 5-g exenatide, and placebo arms, respectively.The corresponding values for subjectswith baselineA1C9%were 9.860.07%(n72), 9.750.07%(n73), and9.750.07%(n75).D: Effectsof exenatide on bodyweight. Subjects in the 10-g exenatide b.i.d. treatment arm received 5g exenatide b.i.d. during weeks 04. Subjects in all treatmentarms were maintained on metformin-sulfonylurea therapy. *P 0.001 compared with placebo treatment. Data are means SE.



Figure 3Postprandial plasma glucose concentrationsin the meal tolerance test subpopulation. A: Postpran-dial plasma glucose concentrations after a standardizedmeal at day 1. Subjects in all treatment arms receivedplacebo. Postprandial plasma glucose geometric meanAUC(15180 min) values were 2,033 mmol min l

1 inthe 10-g exenatide arm, 2,089 mmol min l1 in the5-g exenatide arm, and 2,090 mmol min l1 in theplacebo arm. B: Postprandial plasma glucose concentra-tions after a standardized meal at week 30. Geometricmean AUC(15180 min) values were 1,539 mmol min l1 in the 10-g exenatide arm (P 0.0004 vs. pla-cebo), 1,584 mmol min l1 in the 5-g exenatide arm(P 0.0009 vs. placebo), and 2,087 mmol min l1 inthe placebo arm. Exenatide or placebo were adminis-tered at time zero. Evaluable population: 10 g ex-enatide, n 27; 5 g exenatide, n 27; placebo, n 23. Subjects in the 10-g exenatide b.i.d. arm received5 g exenatide b.i.d. during weeks 04. Subjects in alltreatment arms were maintained on metformin-sulfonylurea therapy. Data are means SE.



In the MIN group, sulfonylurea dose was64% of MAX sulfonylurea dose acrossall treatment arms at study outset (base-line). By week 2, MIN subjects reducedthe dose of sulfonylurea to a nadir of 30%of MAX dose across treatment arms. Thislow dose was maintained for severalweeks, then sulfonylurea doses graduallyincreased throughout the remainder ofthe study. At week 30, subjects on pla-cebo reached 94% of MAX dose com-pared with 79% of MAX dose in theexenatide arms. For the two sulfonylureadosing groups, there were similar overalleffects on A1C when comparing ex-enatide treatment arms with placebo, butthe MAX group had a slightly greater re-duction in A1C from baseline (P 0.0001 for pairwise comparisons; Table2). However, the overall incidence of hy-poglycemia was lower in the MIN group,with a small attenuation of the effects onglycemic control.

CONCLUSIONS This study, con-ducted in a large subject population,demonstrated that exenatide therapy im-proved glycemic control in patients withtype 2 diabetes inadequately controlledon maximal doses of combined met-formin-sulfonylurea therapy. Exenatideadded to metformin-sulfonylurea therapywas associated with weight loss and wasgenerally well tolerated. The addition ofexenatide resulted in similar beneficial ef-fects in patients whose sulfonylurea dosewas initially reduced to minimally effec-tive doses, with later titration, when com-pared with patients using fixed maximallyeffective doses of sulfonylurea through-out the study.

The glucoregulatory activities of thenovel incretin mimetic exenatide includeglucose-dependent enhancement of insu-lin secretion, suppression of inappropri-

ately high glucagon secretion, andslowing of gastric emptying (1518,2730). These effects together may explainthe reduction in postmeal glucose excur-sions observed in the meal challenge testspresented in our report. The ability of ex-enatide to enhance glucose-dependentinsulin secretion may be mediated by ex-enatide binding to the pancreatic GLP-1receptor (31). In animal models of diabe-tes and in insulin secretory cell lines, ex-enatide and GLP-1 reportedly improve-cell function by increasing the expres-sion of key genes involved in -cell func-tion, by increasing insulin biosynthesisand processing, and by augmenting-cellmass through multiple mechanisms (18).Data obtained in animal and human stud-ies also indicate that exenatide and GLP-1

reduce food intake, cause weight loss, andmay have an insulin-sensitizing effect(18,27,32,33). The latter effect has beenshown in some but not all studies, andthere is some question as to whether theapparent insulin sensitizing effect, if ob-served, is an indirect consequence ofoverall improved metabolic control andtherefore reduced glucotoxicity or an ac-tual direct effect.

In the current trial, the observation ofa modest reduction in fasting plasma glu-cose, in keeping with the pharmacoki-netic profile of exenatide, and yet asignificant drop in A1C strongly suggestsa robust effect of exenatide on postpran-dial plasma glucose concentrations. Thisis confirmed by the sustained reductionsin postprandial glucose concentrationsobserved in the meal challenge test atweeks 4 and 30. Incremental plasma glu-cose AUC and average concentration dur-ing the postprandial period were reducedby 60% (5-g arm) and 90% (10-garm) by exenatide treatment.

Because the magnitude of any fall inA1C is dependent upon a number of fac-tors (e.g., baseline A1C, background ther-apy, and endogenous -cell function), itis not possible to directly compare the ef-ficacy of exenatide in this population to theeffects of other agents, as no comparabletrial data are currently available. Perhaps abetter measure of efficacy is to examine the

Figure 4Time-dependent incidence of subjects experiencing treatment-emergent nausea (ITTpopulation).

Table 1 Most frequent adverse events with incidence >10% in any treatment arm (ITT)*

Preferred term Placebo 5 g exenatide 10 g exenatide

n 247 245 241Nausea 51 (20.6) 96 (39.2) 117 (48.5)Hypoglycemia 31 (12.6) 47 (19.2) 67 (27.8)Upper respiratory tract infection 48 (19.4) 28 (11.4) 42 (17.4)Vomiting 11 (4.5) 36 (14.7) 33 (13.7)Diarrhea 16 (6.5) 25 (10.2) 42 (17.4)Feeling jittery 17 (6.9) 21 (8.6) 28 (11.6)Headache 12 (4.9) 27 (11.0) 18 (7.5)

Data are n (%). *Treatment emergent.



proportion of subjects in each specific treat-ment arm who were able to achieve an A1C7% at the 10- and 5-g fixed doses ofexenatide, which was 34 and 27% of sub-jects who completed the study, respec-tively, compared with 9% of subjects whoreceived placebo.

These results are consistent withthose reported in another 30-week, pla-cebo-controlled, phase 3 study (34) of theeffects of exenatide on glycemic controland safety in subjects with type 2 diabetesfailing to achieve glycemic control withsulfonylureas. In that study, at week 30,the 10-g exenatide b.i.d. arm had signif-icant placebo-adjusted reductions of1.0% in A1C and 1.0 kg in weight. Inaddition, a reduction in the proinsulin-to-insulin ratio in the 10-g exenatide b.i.d.arm indicated that exenatide had a bene-ficial effect on the -cell (34). In a parallel30-week, placebo-controlled, phase 3study in subjects with type 2 diabetes fail-ing to achieve glycemic control with met-formin, the 10-g exenatide b.i.d. armhad significant placebo-adjusted reduc-tions of 0.9% in A1C and 2.5 kg inweight at week 30 (35).

Importantly, exenatide treatment wasnot associated with the weight gain ordi-narily seen with the glycemic improve-ment achieved by other therapies(6,8,10,11), even though no program ofbehavior modification was included inthe protocol. The progressive weight lossobserved is consistent with the known ac-tion of exenatide to reduce food intake(29,32). The most common adverse eventassociated with exenatide treatment wasdose-dependent mild to moderate nau-sea. Nausea was reported most commonlyduring the initial 8 weeks of therapy andwas reported less frequently as the studyprogressed. Although nausea was re-ported most frequently in the initialweeks after starting therapy, weight losswas progressive over 30 weeks, thus sup-porting the dissociation of the two effects.

Across all three treatment arms, there

was a significant incidence of hypoglyce-mia, but the vast majority of events weremild or moderate in severity. Occurenceof hypoglycemia appeared to be influ-enced by three factors: background sul-fonylurea dose, exenatide dose, andambient level of glycemia. For example, ahigher incidence of hypoglycemia wasobserved for the MAX group taking thehigher dose of exenatide with an ambientglycemia closer to 7% A1C. Associatedstudies suggest that exenatides action isglucose dependent and would thereforepresent minimal hypoglycemic risk. Un-der controlled clamped condition, ex-enatides insulinotropic effect on the-cell is not observed once plasma glu-cose is lowered to3.9 mmol/l (36) Also,when exenatide is administered to met-formin-treated patients, where you wouldexpect minimal background hypoglyce-mia risk, no increase in hypoglycemia isobserved despite a lowering of A1C (35).In the present study, the hypoglyemia ob-served seems most likely the result of anexenatide-induced improvement in gly-cemia superimposed upon the nonglu-cose-dependent actions of sulfonylureatreatment. The findings further suggestthat a proactive approach to sulfonylureadose management will likely limit the in-cidences of hypoglycemia in exenatide-treated patients. That is, initial reductionof dosage of a sulfonylurea may limit therisk of hypoglycemia associated with suchtherapies upon initiation of exenatidetreatment.

In summary, exenatide therapy sig-nificantly improved glycemic control inpatients with type 2 diabetes and was as-sociated with significant sustained weightloss when added to the commonly usedcombination of metformin and sulfonyl-urea. The main adverse events were dose-dependent nausea and mild to moderatehypoglycemia that was ameliorated by ap-propriate adjustments of sulfonylureadosage. This novel therapy may offer an-other potential treatment option when

two-drug oral therapy fails to maintainadequate glycemic control.

Acknowledgments This report was sup-ported by Amylin Pharmaceuticals, San Diego,California, and Eli Lilly, Indianapolis, Indiana.

The authors thank the Exenatide-115 Clin-ical Study Group for their excellent assistancein the conduct, reporting, and quality controlof the study, and all the patients who volun-teered to participate. The data reported herewere analyzed by all authors, and all authorscontributed to and reviewed the final manu-script. The following are gratefully acknowl-edged for their valuable contributions to theconduct, reporting, and quality control of thestudy, and to the development of the manu-script: Maria Aisporna, Thomas Bicsak, JennyHan, John Holcombe, Orville Kolterman,Leigh MacConell, David Maggs, LorettaNielsen, Terri Poon, James Ruggles, AnnaMarie Rasmussen, Larry Shen, Michael Sier-zega, Kristin Taylor, Michael Trautmann,Amanda Varns, Barbara Wilkinson, MatthewWintle, and Liping Xie.

APPENDIXPrincipal investigators in the Exenatide-115 Clinical Study Group were L. Abbott,A. Ahmann, J. Albu, R. Arakaki, L.Blonde, J. Buse, E. Busick, P. Casner, L.Chaykin, R. Cherlin, M. Chertman, A.J.Cohen, J. Cohen, G. Collins, L. Corn, J.Cyrus, L. Davis, R. DeFronzo, C. de laGarza, S. Duckor, J. Durden, T. Doty, J.Farrell, M.H. Farooqi, N. Fishman, L. Fo-gelfeld, W. Gaman, S. Garg, W.T. Garvey,L. Gavin, J. Gilbert, B. Goldstein, A.Gupta, W. Harvey, I. Hartman, C. Her-ring, B. Horowitz, W. Isley, D. James, R.Juneja, R. Kaplan, D. Kayne, D. Kendall,E. Klein, T. Knecht, J. Kopin, S. Leichter,M. Magee, S. Magill, D. Martinez, R.McInroy, J. Miller, S. Miller, R. Mills, T.Moretto, P. Moriarty, S. Mudaliar, L. My-ers, D. Normandin, P. Norwood, L. Olan-sky, K. Osei, J. Pappas, A. Philis-Tsimikas, J. Pullman, A. Radparvar, P.Reith, L.R. Reynolds, D. Riff, J. Robinson,R. Rood, J. Rosenstock, S. Schwartz, D.

Table 2 Change in A1C from baseline to week 30 and incidence of hypoglycemia stratified by sulfonylurea management group (ITT)

Sulfonylureamanagementgroup

Change in A1C (%) Incidence of hypoglycemia (%)

Placebo 5 g exenatide 10 g exenatide Placebo 5 g exenatide 10 g exenatide

MAX 0.2 0.1 0.7 0.1; P 0.0001* 0.9 0.1; P 0.0001* 15% 22% 35%MIN 0.3 0.1 0.4 0.1; P 0.0001* 0.6 0.1; P 0.0001* 10% 16% 21%

Data are means SE, unless otherwise noted. *A1C change comparison for exenatide treatment versus placebo.



Schumacher, J. Shapiro, G. Silver, W.Smith, J. Snyder, J. Sullivan, B. Troupin,L. Tuttle, C. Vance, M. Weerasinghe, R.Weinstein, D. Weiss, R. Weiss, P. Weiss-man, F. Whitehouse, K. Williams, N.Winer, M. Wofford, C. Wysham, A.Zayed, L. Zemel.

References1. Weyer C, Bogardus C, Mott DM, Pratley

RE: The natural history of insulin secre-tory dysfunction and insulin resistance inthe pathogenesis of type 2 diabetes melli-tus. J Clin Invest 104:787794, 1999

2. Haffner SM, Mykkanen L, Festa A, BurkesJP, Stern MP: Insulin-resistant prediabeticsubjects have more atherogenic risk fac-tors than insulin-sensitive prediabeticsubjects: implications for preventing cor-onary heart disease during the prediabeticstate. Circulation 101:975980, 2000

3. UK Prospective Diabetes Study Group:UK Prospective Diabetes Study 16. Over-view of 6 years therapy of type II diabetes:a progressive disease. Diabetes 44:12491258, 1995

4. UK Prospective Diabetes Study Group:Intensive blood-glucose control with sul-phonylureas or insulin compared withconventional treatment and risk of com-plications in subjects with type 2 diabetes(UKPDS 33). Lancet 352:837853, 1998

5. UK Prospective Diabetes Study Group:UK Prospective Diabetes Study 28. A ran-domized trial of efficacy of early additionof metformin in sulfonylurea-treated type2 diabetes. Diabetes Care 21:8792, 1998

6. Koro CM, Bowlin SJ, Bourgeois N, FedderDO: Glycemic control from 1988 to 2000among U.S. adults diganosed with type 2diabetes. Diabetes Care 27:1720, 2004

7. Harris MI, Eastman RC, Cowie CC, FlegalKM, Eberhardt MS: Racial and ethnic dif-ferences in glycemic control of adults withtype 2 diabetes. Diabetes Care 22:403408,1999

8. DeFronzo RA: Pharmacologic therapy fortype 2 diabetes mellitus. Ann Intern Med131:281303, 1999

9. Inzucchi SE: Oral antihyperglycemictherapy for type 2 diabetes: scientific re-view. JAMA 287:360372, 2002

10. Kirpichnikov D, McFarlane SI, Sowers JR:Metformin: an update. Ann Intern Med137:2533, 2002

11. Nathan DM: Initial management of glyce-mia in type 2 diabetes mellitus. N EnglJ Med 347:13421349, 2002

12. Riddle MC, Rosenstock J, Gerich J: TheTreat-to-Target Trial: randomized addi-tion of glargine or human NPH insulin tooral therapy of type 2 diabetic patients.Diabetes Care 26:30803086, 2003

13. Dailey GE, Noor MA, Park JS, Bruce S,

Fiedorek FT: Glycemic control with gly-buride/metformin tablets in combinationwith rosiglitazone in patients with type 2diabetes: a randomized, double-blind trial.Am J Med 116:223229, 2004

14. Drucker DJ: Enhancing incretin action forthe treatment of type 2 diabetes. DiabetesCare 26:29292940, 2003

15. Kolterman OG, Buse JB, Fineman MS,Gaines E, Heintz S, Bicsak TA, Taylor K,Kim D, Aisporna M, Wang Y, Baron AD:Synthetic exendin-4 (exenatide) signifi-cantly reduces postprandial and fastingplasma glucose in subjects with type 2 di-abetes. J Clin Endocrinol Metab 88:30823089, 2003

16. Fineman MS, Bicsak TA, Shen LZ, TaylorK, Gaines E, Varns A, Kim D, Baron AD:Effect on glycemic control of exenatide(synthetic exendin-4) additive to existingmetformin and/or sulfonylurea treatmentin patients with type 2 diabetes. DiabetesCare 26:23702377, 2003

17. Nielsen LL, Baron AD: Pharmacology ofexenatide (synthetic exendin-4) for thetreatment of type 2 diabetes. Curr OpinionInvestig Drugs 4:401405, 2003

18. Nielsen LL, Young AA, Parkes D: Pharma-cology of exenatide (synthetic exendin-4): a potential therapeutic for improvedglycemic control of type 2 diabetes. RegulPept 117:7788, 2004

19. Holst JJ, Gromada J: Role of incretin hor-mones in the regulation of insulin secre-tion in diabetic and nondiabetic humans.Am J Physiol Endocrinol Metab 287:E199E206, 2004

20. World Medical Association Declarationof Helsinski: Recommendations guidingphysicians in biomedical research involv-ing human subjects (Letter). JAMA 277:925926, 1997

21. Fineman MS, Shen LZ, Taylor K, Kim DD,Baron AD: Effect of progressive dose-es-calation of exenatide (synthetic exendin-4) on dose-limiting side-effects in subjectswith type 2 diabetes. Diabetes Metab ResRev 20:411417, 2004

22. Neter J, Wasserman W, Kutner MH: Ap-plied Linear Statistical Models. 3rd Ed.Homewood, IL, Irwin, 1990

23. Milliken GA, Johnson DE: Analyses ofMessy Data. Vol I. Designed Experiments.New York, Van Nostrand Reinhold, 1984

24. Hayter AJ: The maximum familywise errorrate of Fishers least significant differencetest. J Am Stat Assoc 396:10001004, 1986

25. Davis JE, McDonald JM, Jarret L: Ahigh performance liquid chromatogra-phy method for hemoglobin A1c. Diabe-tes 27:102107, 1978

26. Cole RA, Soeldner JS, Dunn PJ, Bunn HF:A rapid method for the determination ofglycosylated hemoglobins using highpressure liquid chromatography. Metabo-lism 27:289301, 1978

27. Young AA, Gedulin BR, Bhavsar S, BodkinN, Jodka C, Hansen B, Denaro M: Glu-cose-lowering and insulin-sensitizingactions of exendin-4: studies in obesediabetic (ob/ob, db/db) mice, diabetic fattyZucker rats, and diabetic rhesus monkeys(Macaca mulatta). Diabetes 48:1026 1034, 1999

28. Parkes DG, Pittner R, Jodka C, Smith P,Young A: Insulinotropic actions of ex-endin-4 and glucagon-like peptide-1 invivo and in vitro. Metabolism 50:583589,2001

29. Edwards CM, Stanley SA, Davis R, BrynesAE, Frost GS, Seal LJ, Ghatei MA, BloomSR: Exendin-4 reduces fasting and post-prandial glucose and decreases energy in-take in healthy volunteers. Am J PhysiolEndocrinol Metab 281:E155E161, 2001

30. Egan JM, Clocquet AR, Elahi D: The insu-linotropic effect of acute exendin-4 ad-ministered to humans: comparison ofnondiabetic state to type 2 diabetes. J ClinEndocrinol Metab 87:12821290, 2002

31. Goke R, Fehmann HC, Linn T, SchmidtH, Krause M, Eng J, Goke B: Exendin-4 isa high potency agonist and truncated ex-endin-(939)-amide an antagonist at theglucagon-like peptide 1-(736)-amide re-ceptor of insulin-secreting -cells. J BiolChem 268:1965019655, 1993

32. Szayna M, Doyle ME, Betkey JA, Hollo-way HW, Spencer RGS, Greig NH, EganJM: Exendin-4 decelerates food intake,weight gain, and fat deposition in Zuckerrats. Endocrinology 141:19361941, 2000

33. Gedulin BR, Nikoulina SE, Smith PA,Gedulin G, Nielsen LL, Baron AD, ParkesD, Young AA: Exenatide (exendin-4) im-proves insulin sensitivity and -cell massin insulin-resistant obese fa/fa Zuckerrats independent of glycemia and bodyweight. Endocrinology. 146:20692076,2005

34. Buse JB, Henry RR, Han J, Kim DD, Fine-man MS, Baron AD: Effects of exenatide(exendin-4) on glycemic control over 30weeks in sulfonylurea-treated patientswith type 2 diabetes mellitus. DiabetesCare 27:26282635, 2004

35. DeFronzo R, Ratner R, Han J, Kim D, Fine-man M, Baron A: Effects of exena-tide (exendin-4) on glycemic controland weight over 30 weeks in metformin-treated patients with type 2 diabetes(Abstract).Diabetologia 47 (Suppl. 1):A280,2004

36. Degn KB, Brock B, Juhl CB, Djurhuus CB,Grubert J, Kim D, Han J, Taylor K, Fine-man M, Schmitz O: Effect of intravenousinfusion of exenatide (synthetic exendin-4) on glucose-dependent insulin secre-tion and counterregulation during hypo-glycemia. Diabetes 53:23972403, 2004



00b7d526687272df11000000

Documents