OBJECTIVE—AC2993 (synthetic exendin-4; exenatide) is a peptide that enhances glucose-dependent insulin secretion, suppresses inappropriately elevated glucagon secretion, and slows gastric emptying. AC2993 also promotes β-cell proliferation and neogenesis in vitro and in animal models. This study examines the activity and safety of subcutaneously injected AC2993 in patients with type 2 diabetes currently treated with diet and/or oral antidiabetic agents (OAAs).

RESEARCH DESIGN AND METHODS—A total of 109 patients treated with diet and a sulfonylurea and/or metformin were enrolled in a blinded study. Patients were randomly assigned to one of three subcutaneously (SC) injected regimens of AC2993 (0.08 μg/kg) or placebo for 28 days.

RESULTS—All three AC2993 regimens led to significant reductions in serum fructosamine relative to placebo (P ≤ 0.004). Mean reductions ranged from 39 to 46 μmol/l. All AC2993 groups had reductions in HbA1c ranging from 0.7 to 1.1% (P ≤ 0.006). An end-of-study HbA1c <7% was achieved by 15% of AC2993 patients versus 4% of placebo patients, confirming AC2993 effects on fasting and postprandial glycemia. On days 14 and 28, the β-cell index (homeostasis model assessment) for patients treated with AC2993 was 50–100% higher than baseline, contrasting with unchanged levels for placebo. The most common adverse event was transient mild-to-moderate nausea.

CONCLUSIONS—AC2993 is a promising therapeutic for patients with type 2 diabetes. In this study, it had significant effects on HbA1c levels in patients not currently achieving optimal glucose control with diet and/or OAAs.

The natural history of type 2 diabetes is characterized by the emergence of postprandial and, subsequently, fasting hyperglycemia (1). In most individuals, hyperglycemia results from a failure of β-cell insulin-secretory capacity to adequately compensate for insulin resistance in peripheral tissues (2). Results from the U.K. Prospective Diabetes Study (UKPDS) indicate that β-cell failure is progressive, despite therapy with diet, insulin, sulfonylurea, or metformin (3,4). In addition, a reduction in glycosylated hemoglobin A1c (HbA1c) lowered the risk of vascular complications, suggesting that any reduction in HbA1c may be beneficial to subjects with type 2 diabetes. As β-cell dysfunction is progressive, many type 2 diabetic patients eventually require insulin as primary therapy to achieve glycemic goals (5,6). Glucose control in this population remains inadequate, as average HbA1c values reported in most epidemiologic studies are well above 8% as compared with normal levels <6%, despite the availability of a number of therapeutic agents (7,8).

AC2993 (synthetic exendin-4; exenatide), a 39–amino acid peptide, exhibits many desirable features of a novel antidiabetic therapy, including glucose-dependent enhancement of insulin secretion (911), suppression of inappropriately high glucagon secretion (12), and slowing of gastric-emptying rate (13), which may be paradoxically accelerated in people with diabetes (14). At least some of these antidiabetic actions (e.g., enhancement of insulin secretion) may be mediated by binding of AC2993 to the pancreatic glucagon-like peptide-1 (GLP-1) receptor (15). Also, in animal and in vitro models, AC2993 and GLP-1 have been shown to promote β-cell proliferation and neogenesis from precursor cells (1618). Data obtained in animal models also indicate that AC2993 reduces food intake, causes weight loss, and has an insulin-sensitizing effect (9,19,20). The present study was undertaken to evaluate the effectiveness of AC2993 in improving glycemic control among patients with type 2 diabetes who are treated with sulfonylureas or metformin, alone or in combination, over a 28-day period.

Study subjects

A total of 116 patients with type 2 diabetes were recruited from 24 sites throughout the U.S. Males, or females who were surgically sterile or postmenopausal, aged 18–65 years were eligible for enrollment. Patients were to be treated with diet and a regimen of sulfonylurea or metformin (alone or in combination) that was stable for at least the prior 6 months. The entry oral antidiabetic agent regimen was continued throughout the study. Planned entry HbA1c was between 8.0 and 11.0%. Actual enrolled HbA1c ranges are given in Table 1. The target BMI at enrollment was 27–40 kg/m2, inclusive. The following institutional review boards approved this study: The Western IRB, Olympia, Washington; University of North Carolina, Chapel Hill, North Carolina; Henry Ford Hospital, Detroit, Michigan; Washington Hospital, Freemont, California; Medstar Research Institute, Washington, D.C.; and Scripps Clinic, San Diego, California. All patients provided written informed consent.

Study design

This was a randomized, triple-blind (with respect to subject, investigator, and sponsor), parallel-group, placebo-controlled study designed to assess glucose control and evaluate safety in patients receiving subcutaneously (SC) injected AC2993 (0.08 μg · kg−1 · injection−1) or placebo for 28 days. After a 2-week, single-blind, placebo lead-in, patients were randomly assigned to one of three AC2993 treatment groups: twice daily (BID) [breakfast (b) and dinner (d)]; BID [b and bedtime (s)]; thrice daily (TID) [bds]; or placebo TID [bds]. To maintain the study blind, patients treated BID received a third injection of placebo. Appropriate volumes of study medication [0.1 mg/ml AC2993 (Star Biochemicals, now Mallinckrodt, Inc., Torrance, CA)] or placebo, both manufactured according to current good manufacturing practices, were administered using an insulin syringe.

Activity assessments included changes from baseline to day 28 in serum fructosamine and postprandial plasma glucose, HbA1c, fasting plasma glucose, body weight, and fasting and postprandial lipids. Baseline was defined as day –1 for glucose and lipid measures and as day 1 for all other measures. Subjects were required to fast each day after midnight. After at least 14 days of single-blind placebo lead-in, subjects returned for a standardized meal tolerance test (day –1). The test was performed in the morning after the all-night fast, 15 min after injection of placebo. Blood samples were collected at 15 and 5 min before the injection, and at 0.5, 1, 1.5, 2, 3, 4, and 6 h after the injection. The 6-h sample was only assayed for AC2993. The following morning (day 1), after another all-night fast, subjects were randomized to placebo or one of three AC2993 treatments. Subjects also underwent identical standardized meal tolerance tests with timed blood sampling on days 1, 14, and 28. The standardized meal had a composition of 55% carbohydrate, 15% protein, and 30% fat. Patients consumed one bagel, one cheese slice, orange juice, soft corn margarine, and 2% milk. The quantity of liquids and margarine were adjusted based on body weight to maintain equal caloric intake per kilogram and activity level across the patient cohort.

Quantitation of plasma glucose, fructosamine, insulin, HbA1c, lipids, and cortisol was performed by Esoterix Laboratory (Calabasas Hills, CA) according to well-established methods. Quantitation of AC2993 and anti-AC2993 antibodies was done by Amylin Pharmaceuticals, Inc. (San Diego, CA). Glucose was measured using the Roche Glucose/HK Reagent Set (Roche Diagnostics, Indianapolis, IN) for glucose analysis. Fructosamine was measured using a colorimetric assay (21). Insulin was measured using a two-site immunochemiluminometric assay. Total hemoglobin and HbA1c were measured using high-performance liquid chromatography (HPLC), with a normal reference range of 4.2–5.9%. The assay provides values comparable to those obtained in the Diabetes Control and Complications Trial (22). Cortisol was measured by radioimmunoassay. Plasma AC2993 concentrations were measured using an immunoenzymetric assay. Briefly, microtiter plates were coated with an anti-AC2993 monoclonal antibody (Amylin Pharmaceuticals) at 4°C for 1–3 days. Nonspecific binding sites were blocked using 1% nonfat dried milk diluted in 0.05 mol/l carbonate buffer (pH 9.5). Samples were added to wells and incubated for 1–2 h at room temperature. A biotinylated secondary monoclonal antibody (Amylin Pharmaceuticals) preincubated with streptavidin-alkaline phosphatase was added to each well and incubated at room temperature for 1–2 h. Wells were washed with Tris-buffered saline/Tween (0.05 mol/l Tris, 0.15 mol/l sodium chloride, 0.02% sodium azide, and 0.1% Tween-20) between incubation steps. The assay was developed using 50 μl of 0.1 mg/ml 4-methylumbelliferyl phosphate solution. The reaction was stopped using 0.1 mol/l sodium phosphate/1.5 mol/l sodium chloride.

Measurement of anti-AC2993 antibodies was performed using a solid-phase enzyme-linked immunosorbent assay. Briefly, AC2993 (5 μg/ml in 0.05 mol/l carbonate buffer, pH 9.5) was added to wells of microtiter plates, except for control wells coated with anti-human Ig (10 μg/ml in the same buffer). The plates were incubated for 1–3 days at 4°C, then incubated with MegaBloc3 (1:500; Cell Associates, The Sea Ranch, CA) for 1 h at room temperature. Samples were added and incubated for 1 h. Between incubation steps, wells were washed with PBS/0.1% Tween-20. Anti-human Ig/horseradish peroxidase was added and the mixture was incubated for 1 h. The assay was then developed using O-phenylenediamine (Sigma, St. Louis, MO) in citrate buffer (0.055 mol/l phosphate/0.024 mol/l citrate, pH 5.0). The reaction was stopped by 4N sulfuric acid. Absorbance was read at 490 nm.

Homeostasis model assessment (HOMA) (23) was conducted to assess β-cell function at baseline and at days 14 and 28. The HOMA scores (24) were calculated as follows:

Safety was evaluated using spontaneous adverse event reporting and monitoring of clinical laboratory measures and vital signs. Nausea was assessed as part of the routine adverse event–gathering process.

Statistical analysis

All summaries and analyses were based on the intent-to-treat population. A total of 109 patients were randomized and received at least one dose of study medication. Twelve subjects withdrew from the study [three in placebo, four in AC2993 BID (bd), two in AC2993 BID (bs), and three in AC2993 TID], but were included in the statistical analyses up until the time of their withdrawal. Efficacy end points were analyzed using a one-way ANOVA to test the null hypothesis of no difference among treatment groups versus the alternative hypothesis of a difference among treatment groups. No data imputation was employed. Pairwise comparisons were also conducted between the AC2993 arms and placebo. P values ≤0.05 were considered significant. Specific P values are given in the text.

The study was powered to detect a statistically significant difference (α = 0.05) of 30 μmol/l in the change in fructosamine from baseline between one of the AC2993 treatment groups and placebo. The sample size also provided ∼80% power to detect a statistically significant difference (α = 0.05) of a 1.7 mmol/l change in average postprandial plasma glucose from baseline between one of the AC2993 groups and placebo.

Time-weighted averages over all postmeal time points were computed for days −1, 1, and 28 postprandial glucose concentration values. The time-weighted average was computed as the area under the concentration–time curve (AUC) value obtained using the trapezoidal rule divided by the sampling duration. Time-weighted averages were also calculated for postprandial lipids.

Study subjects

A total of 116 patients with type 2 diabetes were recruited from 24 sites throughout the U.S., resulting in 109 subjects randomized to the intent-to-treat population. Of the seven excluded subjects, four did not meet the inclusion/exclusion criteria after screening, two withdrew consent, and one had an adverse event before randomization (motor vehicle accident). Baseline and demographic characteristics were well balanced across the four treatment groups (Table 1).

Serum fructosamine and HbA1c

AC2993 treatment led to statistically significant (P ≤ 0.004) reductions in serum fructosamine at day 28 [45, 39, and 46 μmol/l for BID (bd), BID (bs), and TID (bds), respectively] relative to placebo (5 μmol/l; Fig. 1A). Similarly, statistically significant (P ≤ 0.006) reductions in HbA1c were observed after all AC2993 regimens [1.1, 0.7, and 1.0% for BID (bd), BID (bs), and TID, respectively] compared with placebo (0.3%) during the same treatment period (Fig. 1B). The overall reduction for the three AC2993 groups combined was ∼0.9%. End-of-study HbA1c of <7% was achieved by 15% of all AC2993 patients compared with 4% of placebo. The proportion of patients with entry HbA1c ≥8% achieving HbA1c <8% was 43% and 5% for AC2993 and placebo patients, respectively.

β-Cell function

AC2993 treatment appeared to enhance β-cell function, as assessed by HOMA analysis (Fig. 1C). The β-cell index for all regimens ranged from 50 to 100% greater at days 14 and 28 of AC2993 treatment compared with baseline (day –1) and day 1. The β-cell index for placebo-treated patients remained unchanged.

Plasma glucose

Statistically significant reductions (P ≤ 0.004) from baseline to day 28 in mean postprandial plasma glucose concentration were observed with AC2993 treatment [4.4, 3.2, and 3.4 mmol/l for BID (bd), BID (bs), and TID, respectively] compared with placebo (0.6 mmol/l). These results were similar to the reductions seen on the first day of treatment (Fig. 2). Mean reductions in fasting plasma glucose after AC2993 treatment at day 28 (range of means 1.2–1.7 mmol/l) were not significantly different from placebo (1.1 mmol/l), although there was a trend toward reduced fasting glucose throughout the study. Compared with day 1, the AC2993 plasma glucose profile on day 28 showed reduced fasting glucose concentrations. On day 28, there was a modest rise in postprandial glycemia after AC2993 treatment; however, this change was markedly lower than in the placebo treatment arm on day 28.

Body weight

There was no significant effect of AC2993 treatment on change in body weight from day 1 to day 28 (range of means −0.8 to +0.1 kg) compared with placebo treatment arm (+0.9 kg). There was no detectable influence of baseline HbA1c, baseline body weight, or concomitant antidiabetic therapy on weight loss; however, the overall size of the cohort and the duration of the trial were insufficient to definitively state that these were not covariates.

Lipids

There were no notable differences among the treatment groups in fasting concentrations of triglycerides, HDL cholesterol, LDL cholesterol, or apolipoprotein B at day 28. While not statistically significant, incremental (baseline-adjusted) time-weighted average postprandial triglyceride concentrations tended to be decreased at day 28 compared with day –1 in the AC2993 treatment groups (change from day –1 to day 28, range −0.25 to –0.35 mmol/l) compared with placebo (change from day –1 to day 28, 0.14 mmol/l).

AC2993 pharmacokinetics

Plasma AC2993 for all patients increased steadily until reaching peak concentrations 2–3 h after administration of 0.08 μg/kg SC and was still detectable 6 h postdose. For subjects with undetectable anti-AC2993 antibodies, t1/2 was 202 ± 182 min on day 1 and 226 ± 170 min on day 28 with corresponding Cmax values of 163 ± 86 and 159 ± 81 pg/ml, respectively (n = 63; mean ± SD). For subjects with anti-AC2993 antibodies at any time during the study, t1/2 was 125 ± 42 min on day 1 and 373 ± 250 min on day 28 with corresponding Cmax values of 172 ± 57 and 357 ± 215 pg/ml, respectively (n = 18 for all AC2993 arms combined).

Vital signs and clinical laboratory assessments

No notable differences were observed for the change in vital signs (blood pressure and heart rate) from day –1 to day 28 among any treatment groups, which included a substantial proportion of patients (∼50%) receiving concomitant medications for preexisting hypertension. No patient withdrew prematurely from the study due to a vital sign abnormality.

There were no clinically relevant changes in hematology, clinical chemistry, or urinalysis analyte values from day –1 to day 28 in any of the treatment groups.

There were no statistically significant effects of AC2993 on postprandial plasma cortisol on days 1 and 28 (data not shown). The range of means was 246–480 nmol/l. There was a small, acute, transient increase in serum cortisol concentrations after dosing with AC2993 on day 1, as compared with placebo. There was no rise in postdose mean cortisol values in any of the treatment groups at day 28.

Fifteen (19%) of the 81 AC2993-treated patients developed low-titer anti-AC2993 antibodies during the study. There was no evidence that patients with an antibody response had a diminished glycemic response.

Adverse events

The most common treatment-emergent adverse events reported were nausea (31% overall incidence) and hypoglycemia (15% overall incidence). The normal range for plasma glucose is 3.9–5.8 mmol/l. Values <3.3 mmol/l were considered hypoglycemic. The majority (91%) of the reported nausea and all hypoglycemic cases were mild or moderate in intensity, with no reports of hypoglycemia requiring the assistance of another individual. Hypoglycemia only occurred in patients who were taking a sulfonylurea.

Nausea was only observed in the AC2993 arms and was most pronounced during the initial days of treatment. Thereafter, the incidence of nausea declined to ∼13% by the end of the 28-day treatment period. Four (3.7%) of the 109 patients withdrew due to nausea, all within the first 12 days of the study.

There was no evidence of adverse events that could be associated with an allergic reaction to study medication for patients with anti-AC2993 antibodies.

Exendin-4 was originally isolated from the salivary secretions of the lizard Heloderma suspectum (Gila monster), in which it circulates after ingestion of a meal (25), and may have endocrine functions related to metabolic control. Exendin-4 has a 53% amino acid sequence overlap with mammalian GLP-1. However, exendin-4 is the product of a separate gene distinct from the proglucagon gene from which GLP-1 is expressed (26). Unlike GLP-1, which is degraded within 1–2 min by dipeptidyl peptidase-IV (DPP-IV) when administered SC, exendin-4 is resistant to DPP-IV degradation and, thus, is more readily available for exerting beneficial metabolic effects similar to those of GLP-1 (27). These properties confer upon AC2993 (synthetic exendin-4, exenatide) a very high (>1,000-fold) in vivo potency relative to GLP-1 (10,28). However, not all actions of exendin-4 are predictable based on the known pharmacology of GLP-1. For example, GLP-1, but not exendin-4, suppresses gastric acid secretion (29). Moreover, while intraportal GLP-1 infusion triggers the hepatic vagal afferents, exendin-4 does not (30).

The current study indicates that 28 days of AC2993 treatment reduces HbA1c by ∼0.9% compared with baseline in patients with type 2 diabetes not attaining HbA1c goals with oral agent therapy or diet. In addition, the proportion of patients achieving the American Diabetes Association target HbA1c <7% (31) was fourfold greater in the AC2993 arms. Given that HbA1c only fully reflects a change in glycemia 3 months after a sustained change has occurred, this reduction in HbA1c and enhanced ability to achieve clinically relevant HbA1c target values over 1 month are highly clinically significant.

Glucose profiles during ingestion of a mixed meal demonstrated a marked acute effect of AC2993 to reduce postprandial glycemia that was sustained over the 28-day observation period. This postprandial effect is likely mediated via three key actions of AC2993 that have been observed in animal models of diabetes: 1) increased release of insulin and amylin from the β-cell (10); 2) suppression of the paradoxically high glucagon secretion observed in animals with diabetes (12); and 3) slowing of the rate of gastric emptying (13). While no statistically demonstrable effect on fasting plasma glucose was observed at day 28, day 14 fasting plasma glucose in the AC2993 group was significantly reduced compared with placebo (data not shown). The ability of AC2993 to reduce fasting plasma glucose has previously been reported to be secondary to enhanced insulin secretion (10) and suppression of inappropriately elevated glucagon secretion (12). In patients with type 2 diabetes, GLP-1 has also been shown in a 6-week study to lower both fasting and postprandial glucose (32). However, in contrast to AC2993, GLP-1 needed to be continuously infused (32).

The effects of AC2993 on postprandial and, to a lesser extent, fasting glycemia were the key factors leading to the changes in fructosamine and HbA1c (indicators of average glycemic control over the prior 2 weeks and 3 months, respectively). It is important to note that the study was not designed to assess differences among the various treatment groups. Across the different regimens, 28 days of AC2993 resulted in fructosamine concentrations approaching the upper limit of normal (285 μmol/l) and HbA1c reductions of ∼0.9%. This clinically significant improvement in glycemic indexes over 28 days in patients not previously achieving glycemic control with metformin and/or sulfonylurea treatment is remarkable, as the magnitude of improvement is difficult to achieve with the simple addition of a second or third oral agent (33). Moreover, while insulin therapy can be used to achieve this outcome, a vast literature documents that this approach is generally associated with significant weight gain (4), increased hypoglycemia (4), and attendant morbidities (34). Interestingly, AC2993 treatment was associated with no weight gain in the face of improved glycemic control. Consistent with this observation, exendin-4 has been reported to acutely reduce food intake in healthy human subjects (35) and cause weight loss in animal models of obesity (9).

HOMA analysis revealed improved β-cell secretory function following AC2993 therapy. It is noteworthy that fasting values of plasma glucose and insulin, which were used to calculate HOMA, were obtained before the morning dose of AC2993 when plasma concentrations of AC2993 were negligible, suggesting a fundamental alteration in β-cell function following AC2993 exposure. These data are consistent with the extensive literature documenting enhanced β-cell function after treatment with exendin-4 in animal models of diabetes (1618).

AC2993 treatment was also associated with a strong trend toward reduced postprandial serum triglyceride concentrations, as was also seen in previous, short-term clinical studies (36). While fasting lipid parameters tended to improve, there were no differences compared with placebo.

Safety was evaluated using spontaneous adverse event reporting and monitoring of clinical laboratory measures and vital signs. The most common adverse events encountered were mild-to-moderate nausea and hypoglycemia. Nausea tended to occur mainly upon initiation of therapy and subsided over the first week. Hypoglycemia was mostly undocumented, mild to moderate, and did not require assistance from another person. Importantly, there were no reports of hypoglycemia in AC2993-treated patients receiving metformin alone—consistent with the notion that sulfonylureas are inherently hypoglycemic independent of the prevailing glucose concentration. Although one mechanism of AC2993 action is to slow gastric emptying to better match the rate of nutrient inflow with the rate of glucose disposal, there was no clinical evidence from this study that AC2993 treatment resulted in the induction or exacerbation of conditions such as gastroparesis.

While a small, acute, and transient rise in serum cortisol was observed on the first day of AC2993 treatment, similar to that seen with GLP-1 (37), the assessment on day 28 revealed no such rise in any study patient. There were no clinically relevant effects of AC2993 on other clinical laboratory analytes, blood pressure, or heart rate.

In conclusion, these data demonstrate for the first time in a randomized, triple-blinded trial that AC2993 administered BID or TID for 28 days in patients with type 2 diabetes failing oral agent therapy causes a marked reduction in HbA1c.

Figure 1—

A: Mean (±SE) change in fructosamine from day 1 to day 28 of treatment with 0.08 μg/kg AC2993 or placebo, summarized for each treatment group separately (intent-to-treat population). B: Mean (±SE) change in HbA1c from baseline (day –1) to day 28 of treatment with 0.08 μg/kg AC2993 or placebo, summarized for each treatment group separately (intent-to-treat population). C: Mean (±SE) β-cell index calculated by HOMA at baseline (day −1), day 1, day 14, and day 28 of treatment with 0.08 μg/kg AC2993 or placebo, summarized for each treatment group separately (intent-to-treat population). b, breakfast; d, dinner; s, bedtime.

Figure 1—

A: Mean (±SE) change in fructosamine from day 1 to day 28 of treatment with 0.08 μg/kg AC2993 or placebo, summarized for each treatment group separately (intent-to-treat population). B: Mean (±SE) change in HbA1c from baseline (day –1) to day 28 of treatment with 0.08 μg/kg AC2993 or placebo, summarized for each treatment group separately (intent-to-treat population). C: Mean (±SE) β-cell index calculated by HOMA at baseline (day −1), day 1, day 14, and day 28 of treatment with 0.08 μg/kg AC2993 or placebo, summarized for each treatment group separately (intent-to-treat population). b, breakfast; d, dinner; s, bedtime.

Close modal
Figure 2—

Mean (±SE) postprandial plasma glucose concentration profile at baseline day −1 (A), day 1 (B), and day 28 (C) of treatment with 0.08 μg/kg AC2993 (all arms combined) or placebo (intent-to-treat population).

Figure 2—

Mean (±SE) postprandial plasma glucose concentration profile at baseline day −1 (A), day 1 (B), and day 28 (C) of treatment with 0.08 μg/kg AC2993 (all arms combined) or placebo (intent-to-treat population).

Close modal
Table 1—

Demographic and baseline characteristics of patients treated with AC2993 or placebo

CharacteristicAC2993 BID (bd)AC2993 BID (bs)AC2993 TID (bds)Placebo
n 26 27 28 28 
Sex     
 Men 62 (16) 63 (17) 57 (16) 75 (21) 
 Women 38 (10) 37 (10) 43 (12) 25 (7) 
Race     
 White 54 (14) 44 (12) 54 (15) 68 (19) 
 Hispanic 27 (7) 30 (8) 21 (6) 25 (7) 
 Black 15 (15) 22 (6) 18 (5) 7 (2) 
 Other 4 (1) 4 (1) 7 (2) 0 (0) 
Age (years) 54 ± 8 51 ± 9 50 ± 9 53 ± 9 
Weight (kg) 97.0 ± 16.7 98.2 ± 16.0 96.8 ± 17.5 97.6 ± 16.7 
BMI (kg/m232.8 ± 3.8 33.5 ± 4.7 33.2 ± 4.8 32.8 ± 4.1 
Fasting plasma glucose (mmol/l)     
 Day −1 11.2 ± 2.8 11.6 ± 3.3 10.9 ± 2.4 12.3 ± 3.6 
 Day 1 11.1 ± 3.4 11.7 ± 3.6 10.7 ± 2.5 12.0 ± 3.4 
Serum fructosamine (μmol/l)* 344 ± 74 340 ± 71 331 ± 59 346 ± 56 
HbA1c (%) 9.1 ± 1.2 9.3 ± 1.0 9.2 ± 1.1 9.4 ± 1.3 
 (Range) (7.7–12.0) (7.6–12.6) (7.7–11.7) (7.7–11.60 
Oral agent use     
 Sulfonylurea alone 12 (3) 15 (4) 36 (10) 18 (5) 
 Metformin alone 23 (6) 22 (6) 21 (6) 21 (6) 
 Sulfonylurea plus metformin 65 (17) 63 (17) 43 (12) 61 (17) 
Antihypertensive agent 42 (11) 63 (17) 54 (15) 46 (13) 
CharacteristicAC2993 BID (bd)AC2993 BID (bs)AC2993 TID (bds)Placebo
n 26 27 28 28 
Sex     
 Men 62 (16) 63 (17) 57 (16) 75 (21) 
 Women 38 (10) 37 (10) 43 (12) 25 (7) 
Race     
 White 54 (14) 44 (12) 54 (15) 68 (19) 
 Hispanic 27 (7) 30 (8) 21 (6) 25 (7) 
 Black 15 (15) 22 (6) 18 (5) 7 (2) 
 Other 4 (1) 4 (1) 7 (2) 0 (0) 
Age (years) 54 ± 8 51 ± 9 50 ± 9 53 ± 9 
Weight (kg) 97.0 ± 16.7 98.2 ± 16.0 96.8 ± 17.5 97.6 ± 16.7 
BMI (kg/m232.8 ± 3.8 33.5 ± 4.7 33.2 ± 4.8 32.8 ± 4.1 
Fasting plasma glucose (mmol/l)     
 Day −1 11.2 ± 2.8 11.6 ± 3.3 10.9 ± 2.4 12.3 ± 3.6 
 Day 1 11.1 ± 3.4 11.7 ± 3.6 10.7 ± 2.5 12.0 ± 3.4 
Serum fructosamine (μmol/l)* 344 ± 74 340 ± 71 331 ± 59 346 ± 56 
HbA1c (%) 9.1 ± 1.2 9.3 ± 1.0 9.2 ± 1.1 9.4 ± 1.3 
 (Range) (7.7–12.0) (7.6–12.6) (7.7–11.7) (7.7–11.60 
Oral agent use     
 Sulfonylurea alone 12 (3) 15 (4) 36 (10) 18 (5) 
 Metformin alone 23 (6) 22 (6) 21 (6) 21 (6) 
 Sulfonylurea plus metformin 65 (17) 63 (17) 43 (12) 61 (17) 
Antihypertensive agent 42 (11) 63 (17) 54 (15) 46 (13) 

Data are % (n) or means ± SD.

*

The upper limit of the normal range for serum fructosamine is 285 μmol/l. The upper limit of the normal range for HbA1c is 6.0%.

All of the authors hold stock in Amylin Pharmaceuticals.

A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.

This study was funded by Amylin Pharmaceuticals, Inc. We thank Dr. Loretta Nielsen for manuscript revisions.

1
American Diabetes Association: Postprandial blood glucose.
Diabetes Care
24
:
775
–778,
2001
2
Weyer C, Bogardus C, Mott DM, Pratley RE: The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus.
J Clin Invest
104
:
787
–794,
1999
3
UK Prospective Diabetes Study Group: UK Prospective Diabetes Study 16: overview of 6 years’ therapy of type II diabetes: a progressive disease.
Diabetes
44
:
1249
–1258,
1995
4
UK Prospective Diabetes Study Group: Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in subjects with type 2 diabetes (UKPDS 33).
Lancet
352
:
837
–853,
1998
5
Mudaliar S, Edelman SV: Insulin therapy in type 2 diabetes.
Endocrinol Metabol Clinics N Am
30
:
935
–982,
2001
6
Wright A, Burden ACF, Paisey RB, Cull CA, Holman RR, UKPDS Study Group: Sulfonylurea inadequacy: efficacy of addition of insulin over 6 years in patients with type 2 diabetes in the U.K. Prospective Diabetes Study (UKPDS 57). 
25
:
330
–336,
2002
7
Harris MI, Eastman RC, Cowie CC, Flegal KM, Eberhardt MS: Racial and ethnic differences in glycemic control of adults with type 2 diabetes.
Diabetes Care
22
:
403
–408,
1999
8
Inzucchi SE: Oral antihyperglycemic therapy for type 2 diabetes.
JAMA
287
:
360
–372,
2002
9
Young AA, Gedulin BR, Bhavsar S, Bodkin N, Jodka C, Hansen B, Denaro M: Glucose-lowering and insulin-sensitizing actions of exendin-4: studies in obese diabetic (ob/ob, db/db) mice, diabetic fatty Zucker rats, and diabetic rhesus monkeys (Macaca mulatta).
Diabetes
48
:
1026
–1034,
1999
10
Parkes DG, Pittner R, Jodka C, Smith P, Young A: Insulinotropic actions of exendin-4 and glucagon-like peptide-1 in vivo and in vitro.
Metabolism
50
:
583
–589,
2001
11
Egan JM, Clocquet AR, Elahi D: The insulinotropic effect of acute exendin-4 administered to humans: comparison of nondiabetic state to type 2 diabetes.
J Clin Endocrinol Metab
87
:
1282
–1290,
2002
12
Gedulin B, Jodka C, Hoyt J: Exendin-4 (AC2993) decreases glucagon secretion during hyperglycemic clamps in diabetic fatty Zucker rats (Abstract).
Diabetes
48(Suppl. 1)
:
A199
,
1999
13
Jodka C, Gedulin B, Young A: Exendin-4 potently regulates gastric emptying in rats (Abstract).
Diabetes
47(Suppl. 1)
:
403A
,
1998
14
Rayner CK, Samsom M, Jones KL, Horowitz M: Relationships of upper gastrointestinal motor and sensory function with glycemic control.
Diabetes Care
24
:
371
–381,
2001
15
Goke R, Fehmann HC, Linn T, Schmidt H, Krause M, Eng J, Goke B: Exendin-4 is a high potency agonist and truncated exendin-(9-39)-amide an antagonist at the glucagon-like peptide 1-(7-36)-amide receptor of insulin-secreting beta-cells.
J Biol Chem
268
:
19650
–19655,
1993
16
Xu G, Stoffers DA, Habener JF, Bonner-Weir S: Exendin-4 stimulates both β-cell replication and neogenesis, resulting in increased β-cell mass and improved glucose tolerance in diabetic rats.
Diabetes
48
:
2270
–2276,
1999
17
Tourrel C, Bailbé D, Meile M-J, Kergoat M, Portha B: Glucagon-like peptide-1 and exendin-4 stimulate β-cell neogenesis in streptozotocin-treated newborn rats resulting in persistently improved glucose homeostasis at adult age.
Diabetes
50
:
1562
–1570,
2001
18
Tourrel C, Bailbé D, Lacorne M, Meile MJ, Kergoat M, Portha B: Persistent improvement of type 2 diabetes in the Goto-Kakizaki rat model by expansion of the beta-cell mass during the prediabetic period with glucagon-like peptide-1 or exendin-4.
Diabetes
51
:
1443
–1452,
2002
19
Bhavsar S, Watkins J, Young A: Comparison of central and peripheral effects of exendin-4 and GLP-1 on food intake in rats (Abstract). In
Program and Abstracts: 80th Annual Meeting of the Endocrine Society
. Bethesda, MD, The Endocrine Society,
1998
, p.
433
20
Szayna M, Doyle ME, Betkey JA, Holloway HW, Spencer RGS, Greig NH, Egan JM: Exendin-4 decelerates food intake, weight gain, and fat deposition in Zucker rats.
Endocrinol
141
:
1936
–1941,
2000
21
Johnson RN, Metcalf PA, Baker JR: Fructosamine: A new approach to the estimation of serum glycosylprotein: an index of diabetic control.
Clin Chim Acta
127
:
87
–95,
1983
22
The DCCT Research Group: Feasibility of centralized measurements of glycated hemoglobin in the Diabetes Control and Complications Trial: a multicenter study.
Clin Chem
33
:
2267
–2271,
1987
23
Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RL: Homeostasis model assessment: insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man.
Diabetologia
28
:
412
–419,
1985
24
Haffner SM, Miettinen H, Stern MP: The homeostasis model in the San Antonio heart study.
Diabetes Care
20
:
1087
–1092,
1997
25
Young A, Blase E, Petrella E, Seward M: Exendin-4 is a circulating meal-related peptide in the Gila monster (Heloderma suspectum) (Abstract).
Diabetes
48(Suppl. 1)
:
A425
,
1999
26
Chen YE, Drucker D: Tissue-specific expression of unique mRNAs that encode pro-glucagon-derived peptides or exendin-4 in the lizard.
J Biol Chem
272
:
4108
–4115,
1997
27
Drucker DJ: Glucagon-like peptides.
Diabetes
47
:
159
–169,
1998
28
Greig NH, Holloway HW, De Ore KA, Jani D, Wang Y, Zhou J, Garant MJ, Egan JM: Once daily injection of exendin-4 to diabetic mice achieves long-term beneficial effects on blood glucose concentrations.
Diabetologia
42
:
45
–50,
1999
29
Gedulin B, Lawler R, Jodka C, Young A: Amylin inhibits pentagastrin-stimulated gastric acid secretion: comparison with glucagon-like peptide-1 and exendin-4 (Abstract).
Diabetes
46(Suppl. 1)
:
188A
,
1997
30
Nishizawa M, Nakabayashi H, Kawai K, Ito T, Kawakami S, Nakagawa A, Niijima A, Uchida K: The hepatic vagal reception of intraportal GLP-1 is via receptor different from the pancreatic GLP-1 receptor.
J Auton Nerv Syst
80
:
14
–21,
2000
31
American Diabetes Association: Standards of medical care for patients with diabetes mellitus (Position Statement).
Diabetes Care
25(Suppl. 1)
:
S33
–S49,
2002
32
Zander M, Madsbad S, Madsen JL, Holst JJ: Effect of 6-week course of glucagon-like peptide 1 on glycaemic control, insulin sensitivity, and β-cell function in type 2 diabetes: a parallel-group study.
Lancet
359
:
824
–830,
2002
33
Turner RC, Cull CA, Frighi V, Holman RR, UKPDS Group: Glycemic control with diet, sulfonylurea, metformin, or insulin in patients with type 2 diabetes mellitus. Progressive requirement for multiple therapies (UKPDS 49).
JAMA
281
:
2005
–2012,
1999
34
Purnell JQ, Hokanson JE, Marcovina SM, Steffes MW, Cleary PA, Brunzell JD: Effect of excessive weight gain with intensive therapy of type 1 diabetes on lipid levels and blood pressure: results from the DCCT.
JAMA
280
:
140
–146,
1998
35
Edwards CM, Stanley SA, Davis R, Brynes AE, Frost GS, Seal LJ, Ghatei MA, Bloom SR: Exendin-4: reduces fasting and postprandial glucose and decreases energy intake in healthy volunteers: Am J Physiol Endocrinol Metab
281
:
E155
–E161,
2001
36
Kolterman O, Gottlieb A, Gaines E, Fineman M: Five-day dosing of synthetic exendin-4 (AC2993) in people with type 2 diabetes reduces postprandial glucose, glucagon and triglyceride concentrations (Abstract).
Diabetologia
43(Suppl. 1)
:
A189
,
2000
37
Ryan AS, Egan JM, Habener JF, Elahi D: Insulinotropic hormone glucagon-like peptide-1-(7-37) appears not to augment insulin-mediated glucose uptake in young men during euglycemia.
J Clin Endocrinol Metab
83
:
2399
–2404,
1998