Original Article Philip A. Levin, MD1; James H. Mersey, MD1; Steven Zhou, PhD2; Lee A. Bromberger, BS1 ABSTRACT Objective: To examine the long-term effects of combination insulin glargine/exenatide treatment on glycemic control. Methods: We conducted a 24-month retrospective US chart review of patients with inadequately controlled type 2 diabetes (T2DM) and hemoglobin A1c (A1C) levels >7.0% for whom glargine and exenatide were coprescribed in differing order (glargine added after exenatide [exenatide/ glargine]; exenatide added after glargine [glargine/exenatide]). Treatment order groups were combined to form a pooled treatment group. Changes from baseline in A1C, patients with A1C ≤7.0%, body weight, glargine/exenatide daily dose, oral antidiabetic drug (OAD) use, and hypoglycemia were evaluated. Results: Treatment groups were similar at baseline; however, patients in the glargine/exenatide group (n = 121) (vs exenatide/glargine group [n = 44]) had longer disease duration (11.8 vs 8.0 years) and took fewer OADs (1.7 vs 2.3). Overall, baseline A1C was 8.8 ± 1.3% and weight was 109.5 ± 25.3 kg. Significant A1C reductions emerged at month 6 and persisted throughout 24 months (vs baseline) in both treatment groups (pooled: –0.7 ± 1.6; P<.001), and 33.0% of patients achieved an A1C level ≤7.0%. After Submitted for publication April 6, 2011 Accepted for publication May 20, 2011 From 1MODEL Clinical Research, Baltimore, Maryland, and 2sanofi-aventis U.S., Bridgewater, New Jersey. Address correspondence to Dr. Philip A. Levin, MODEL Clinical Research, Greater Baltimore Medical Center, 6535 N Charles St, Ste 400 N, Baltimore, MD 21204. E-mail: firstname.lastname@example.org. Published as a Rapid Electronic Article in Press at http://www. endocrine practice.org on July 8, 2011. DOI:10.4158/EP11097.OR Copyright © 2012 AACE. 24 months of exenatide/glargine, body weight remained unchanged (0.7 ± 8.3 kg; P = .640). With glargine/exenatide, body weight decreased (–2.5 ± 6.7 kg; P = .001). At month 24, daily glargine dose was 0.40 ± 0.23 units/kg for the exenatide/glargine group and 0.47 ± 0.30 units/kg for the glargine/exenatide group. Hypoglycemia frequency was similar in both treatment groups. Conclusions: Regardless of treatment order, longterm combined therapy with glargine and exenatide for up to 24 months in patients with inadequately controlled T2DM suggests reduction of A1C without significant weight gain or increased hypoglycemia risk. (Endocr Pract. 2012;18:17-25) Abbreviations: A1C = hemoglobin A1c; OADs = oral antidiabetic drugs; T2DM = type 2 diabetes mellitus INTRODUCTION Over time and largely because of declining b-cell function in type 2 diabetes mellitus (T2DM), lifestyle modifications and oral antidiabetic drugs (OADs) become less effective in achieving glycemic control, at which point insulin therapy is typically considered (1,2). Current guidelines suggest adding a long-acting basal formulation to an existing OAD regimen when initiating insulin therapy (3). Because of its longer duration of action and lower risk of hypoglycemia, insulin glargine may be preferred over other insulin options (4-7). Glucagonlike peptide 1 analogues, such as exenatide, also are effective for reducing hyperglycemia in patients with T2DM (3,8). Although insulin therapy, including insulin glargine, reliably reduces hemoglobin A1c (A1C) and fasting plasma glucose, it may cause weight gain (3). Conversely, exenatide can significantly reduce weight, although a significantly lower glycemic efficacy was found in comparison with insulin glargine (9,10). These complimentary clinical effects provide a rationale for combining insulin glargine and exenatide in 17 18 patients with inadequate glycemic control despite OAD use. Randomized and naturalistic, observational studies generally have described favorable efficacy and safety with this approach in patients with T2DM (11-15). In a 4-week, randomized, open-label, proof-of-concept study in patients with T2DM and inadequate glycemic control, Arnolds et al reported significant reductions in postprandial blood glucose excursions (P<.01) and A1C levels (P<.0001) after the addition of exenatide or sitagliptin to glargine + metformin therapy (vs glargine + metformin only) (11). Significantly greater weight loss was also seen with exenatide + glargine + metformin vs glargine + metformin (–0.9 vs 0.4 kg; P<.05) (11). While encouraging, these results prompt the question of whether the changes in A1C and weight from combining exenatide with glargine could be maintained for longer periods in a real-world setting. Moreover, the clinical importance of the chronologic order in which insulin and exenatide are combined is unknown; investigators have examined the impact of adding exenatide to an established insulin regimen, but the impact of adding glargine to an existing regimen of exenatide is unknown. The objective of this study was to examine the realworld use and outcomes after the combined use and different ordering of glargine and exenatide for up to 24 months—a longer duration than most previous observational studies—in patients with T2DM and inadequate glycemic control. This is the first study to evaluate the order in which insulin and exenatide are combined during treatment. The naturalistic study design allows for the observation of therapeutic outcomes in a wider range of clinically typical patients with T2DM. RESEARCH DESIGN AND METHODS Study Design This was a naturalistic, retrospective, multicenter study designed to evaluate and compare the effectiveness of glargine and exenatide, each given as monotherapy or added to an existing antidiabetic regimen that already included the other. Clinical medical records of patients with T2DM from 26 large endocrinology, internal medicine, and family and primary care practices throughout the continental United States were reviewed from January 2005 through June 2010. Deidentified, extracted data included assessment of glycemic control, body weight, medication discontinuation, use of concomitant medications, and incidence of hypoglycemia for up to 24 months from baseline. Treatment Groups Four treatment groups were identified: 2 using glargine or exenatide monotherapy and 2 using a combination of glargine and exenatide therapy. This report focuses on data from the 2 combination therapy groups. The 2 groups were distinguished based on the order of prescribed treatment and consisted of patients who initiated glargine before adding exenatide (glargine/exenatide) and those who initiated exenatide before adding glargine (exenatide/glargine). Once the patient first started either insulin glargine or exenatide, the second agent could be added (glargine or exenatide) to the patient’s regimen at any time the treating physician thought it was clinically necessary without any time restriction. During follow-up, there were no restrictions on the medications that treating physicians could prescribe once glargine or exenatide was added. Inclusion Criteria Selected patients with T2DM were aged 18 years or older and had inadequate glycemic control (A1C >7.0%) with oral therapy. Medical records were required to meet the following criteria: available for 4 months before and 16 or more months after glargine or exenatide initiation; A1C and body weight data available at baseline or from 4 weeks before baseline or 10 or more days after baseline; and at least 2 A1C values and 1 body weight measure recorded during the 24-month follow-up period. For combination treatment patients, the initial drug prescribed was required to be continued at the time the second drug was added (eg, baseline), and they could not have used the second drug within the past 12 months. Patients with A1C levels of 7.0% or less, a terminal illness, or HIV/AIDS and pregnant women were excluded. Baseline Dates and Data Extracted Baseline was established as the day glargine was first added to the existing exenatide treatment or vice versa. Data were extracted from the period 4 months before baseline to 16 or fewer months following baseline, until the last date that data were available, regardless of whether the treatment regimen was discontinued. Data obtained included patient age, sex, body weight, duration of diabetes, A1C, duration of glargine and/or exenatide therapy, and frequency of hypoglycemic episodes. Medical records included information obtained from office visits, over-thephone medication dosage titrations, prescription refills, and clinical laboratory reports. Study Endpoints Diabetes-related endpoints were examined at months 6, 12, 18, and 24 after the baseline date. Primary efficacy outcomes included change from baseline in A1C concentration and percentage of patients achieving an A1C level of 7.0% or lower. Secondary outcomes included weight and blood pressure change from baseline, use of concomitant OAD medications, therapy discontinuation, and total daily glargine and/or exenatide dosing units. Also of interest was the duration of A1C control (≤7.0%) during the follow-up period. Hypoglycemic events were captured from patient self-report and from glucose meter downloads, when available, or from in-office finger sticks. Severe hypoglycemia 19 was denoted by a call to 911 (emergency) and defined as an episode during which the patient had a blood glucose reading less than 56 mg/dL, required assistance from others, and recovered after administration of countermeasures. Statistical Analyses Analyses were performed on available data from all patients treated with study medication as per protocol at baseline, and who had at least 2 postbaseline A1C values. For analysis, the 2 treatment-order groups were combined to form a third, pooled combination-treatment group. For all 3 groups, descriptive statistics (mean, standard deviation for continuous variables, and number and percentage for category variables) were determined for baseline and outcome variables. For comparisons among treatment groups at baseline, F tests were performed on continuous variables while chi-square testing was performed on categorical variables. For comparisons of outcome variables between the follow-up periods and baseline, the paired t test was applied to the continuous variables while chisquare testing was applied to the categorical variables. RESULTS Demographic and Clinical Characteristics Overall, 165 medical charts were identified in which combination treatment was administered (exenatide/ glargine, n = 44; glargine/exenatide, n = 121). Table 1 summarizes the baseline demographic and clinical characteristics of each group. Patients in the exenatide/glargine and glargine/exenatide treatment-order groups were similar with regard to baseline A1C, body weight, age, and sex (race was not captured in the available medical records). Similar proportions of patients in each treatment-order group received concomitant metformin; the proportions of patients receiving thiazolidinedione and/or sulfonylurea therapy was more variable, but not significant. At baseline, however, patients in the glargine/exenatide group had significantly longer diabetes duration than patients in the exenatide/glargine group and were taking fewer OADs. A1C Reductions Regardless of the order in which glargine and exenatide were prescribed, combining these agents was associated with significant reductions in A1C from baseline (exenatide/glargine, –1.0 ± 1.2%; glargine/exenatide, −0.7 ± 1.7%; P<.001 for both). Fig. 1A illustrates that the unadjusted reductions from baseline in A1C concentrations emerged at month 6 post baseline and were maintained throughout the 24-month observation period. For the pooled combination-treatment group, reduction in A1C was –0.7 ± 1.6% after 24 months. Beginning at month 6 after baseline and continuing throughout 24 months of follow-up, the proportion of patients achieving an A1C level of 7.0% or less ranged between 29.4% and 42.5% in the exenatide/glargine group (24 months: 30.7%; P<.001) and between 31.7% and 36.8% in the glargine/exenatide group (24 months: 33.7%; P<.001) group (Fig. 1B). For the pooled combination treatment group, proportions of patients reaching an A1C level of 7.0% or less ranged from 31.7% to 37.1% throughout follow-up (24 month: 33.0%; P<.001). Body Weight Overall, changes in body weight were observed within the first 6 months of therapy and were maintained through the 24-month endpoint (Table 2). By the 24-month followup, body weight was significantly lower than baseline in the glargine/exenatide group (–2.5 ± 6.7 kg; P = .001). Body weight remained statistically unchanged from baseline in the exenatide/glargine group after 24 months of continuous therapy (0.7 ± 8.3 kg; P = .640) (Fig. 2, Table 2). Blood Pressure In the exenatide/glargine group, systolic and diastolic blood pressures were largely unchanged from baseline (ie, P = not significant vs baseline) throughout the 24-month follow-up period. Similarly, for the glargine/exenatide group, clinically nonsignificant decreases from baseline in blood pressure were observed throughout follow-up. For the pooled combination-treatment group, systolic blood pressure and diastolic blood pressure also were largely unchanged from baseline throughout the 24-month observation period (Table 2). Total Daily Exenatide Dose, Glargine Dose, and Total Insulin Dose In the exenatide/glargine group, a significant increase from baseline in daily dose of exenatide (0.16 ± 0.05 vs 0.19 ± 0.05 mg/kg; P = .019) was observed at month 24. Also at the 24-month follow-up, the daily glargine dose was 0.40 ± 0.23 units/kg. At baseline, 1 patient was using prandial insulin (daily dose: 0.07 ± 0 units/kg); by the 24-month follow-up, 8 patients used prandial insulin with a total insulin dose (ie, glargine + prandial insulin) of 0.48 ± 0.37 units/kg (Table 2). In the glargine/exenatide group, a significant increase from baseline in daily glargine dose (0.37 ± 0.21 units/kg vs 0.47 ± 0.30 units/kg; P = .003) was observed at month 24. At month 24, the daily exenatide dose was 0.19 ± 0.05 mg/kg. At baseline, 24 patients were using prandial insulin (0.32 ± 0.21 units/kg); by the 24-month follow-up, 34 patients were using prandial insulin, with a total insulin dose of 0.60 ± 0.45 units/kg (P = .002 vs baseline) (Table 2). Concomitant OAD Use During the 24-month follow-up, the number of OADs used gradually decreased in the exenatide/glargine group from 2.1 ± 1.0 to 1.7 ± 1.0 OADs (vs baseline; P<.001). By 20 Table 1 Patient Demographics and Baseline Clinical Characteristics Characteristics Age, y Mean (±SD) Range Female, No. (%) T2DM duration, y Mean (±SD) Range Hemoglobin A1c, % Mean (±SD) Range Body mass index, mean (±SD), kg/m2 Weight, kg Mean (±SD) Range Systolic blood pressure, mm Hg Mean (±SD) Range Diastolic blood pressure, mm Hg Mean (±SD) Range Insulin use, No. (%) Insulin glargine daily dose/body weight, units/kg Mean (±SD) Range Prandial daily dose/body weight, units/kg Mean (±SD) Range Premix daily dose/body weight, units/kg Mean (±SD) Range Total insulin daily dose /body weight, units/kg Mean (±SD) Range Exenatide daily dose/body weight, μg/kg Mean (±SD) Range Metformin, No. (%) Thiazolidinedione, No. (%) Sulfonylureas, No. (%) Number of OADs Mean (±SD) Range 2 or more OADs, No. (%) Combination group (n = 165) Exenatide/glargine (n = 44) Glargine/exenatide (n = 121) 59 (±9.6) 36-81 21 (47.7) 61 (±9.5) 36-83 69 (57.0) .268 .289 60 (±9.5) 36-83 90 (54.5) 8.0 (±5.6) 0-21 11.8 (±7.1) 1-34 .014 NAb 8.9 (±1.3) 7.3-12.4 36.2 (±6.7) 8.7 (±1.3) 7.1-14.0 38.1 (±6.6) .460 8.8 (±1.3) 7.1-14.0 37.6 (±6.7) 112.2 (±35.7) 69.9-283.5 108.4 (±20.4) 67.1-163.3 .395 109.5 (±25.3) 67.1-283.5 130.0 (±17.2) 94-180 129.7 (±16.1) 88-180 .898 129.8 (±16.4) 88-180 76.1 (±10.1) 60-100 3 (6.8) 76.0 (±12.2) 52-160 121 (100) .948 76.0 (±11.7) 52-160 NAb NAb 0.37 (±0.21) 0.10-1.34 0.07 (±0.00) 0.07-0.07 0.32 (±0.21) 0.02-0.83 NAb 0.62 (±0.54) 0.24-1.00 0.07 (±0.00) 0.07-0.07 0.45 (±0.32) 0.10-1.97 0.16 (±0.05) 0.07-0.29 24 (54.5) 21 (47.7) 27 (61.3) NAb 2.3 (±1.0) 0-5 35 (79.5) P valuea <.001 0.37 (±0.21) 0.10-1.34 .245 NAb 0.62 (±0.54) 0.24-1.00 .244 NAb 67 (55.3) 44 (36.3) 58 (47.9) .925 .186 .127 0.16 (±0.05) 0.07-0.29 91 (55.1) 65 (39.3) 85 (51.5) 1.7 (±1.0) 0-4 73 (60.3) .002 NAb .022 NAb Abbreviations: NA, not applicable; OADs, oral antidiabetic drugs; T2DM, type 2 diabetes mellitus. a P values based on comparisons of the combination groups with one another, using an F test for continuous variables and chi-square test for categorical variables. b Not applicable; pooled-combination group values for the indicated measures are not applicable, either because of baseline group definition (eg, not using insulin glargine or exenatide at baseline), or because a significant difference was detected between the treatment arms. 21 A B Fig. 1. Hemoglobin A1c (A1C) outcomes during 24 months of combination therapy. Change in A1C (%) from baseline (Panel A) and proportion of patients reaching A1C ≤7.0% at 6, 12, 18, and 24 months post baseline (Panel B). a P<.001 vs baseline; b P<.001 vs baseline. contrast, the number of OADs used in the glargine/exenatide group remained unchanged throughout the 24-month follow-up vs baseline (1.7 ± 1.0 vs 1.7 ± 0.9 OADs; Table 2). Hypoglycemia Hypoglycemic events occurred in similar proportions in the exenatide/glargine group (n = 5/44; 11.4%) and the glargine/exenatide group (n = 15/121; 12.3%). The treatment-order groups were also similar in terms of number of hypoglycemic events (4.0 ± 4.1 vs 3.1 ± 4.2), time to the first hypoglycemic episode (203.4 ± 190.8 days vs 228.9 ± 202.8 days), and hypoglycemic blood glucose values (57.2 ± 7.5 mg/dL vs 52.0 ± 12.1 mg/dL). For the pooled combination-treatment group, the mean number of days to a first hypoglycemic event was 222.5 ± 195.2 and the hypoglycemic blood glucose value was 53.3 ± 11.2 mg/dL. Reported cases of severe hypoglycemia were rare over the 24-month period; in the exenatide/glargine group, 1 patient called the physician, and in the glargine/exenatide group, 3 patients called 911 for emergency medical assistance. Treatment Discontinuation at Month 24 Regardless of treatment order, most patients continued with the combined treatment for up to 12 months. In both groups, most discontinuations of 1 or the other medication occurred between follow-up months 18 and 24. At month 24, 16 of 44 patients in the exenatide/glargine group had discontinued 1 or the other injectable medication (36.4%; n = 4 glargine discontinuation, n = 12 exenatide discontinuation) and 51 of 121 patients in the glargine/exenatide group had discontinued 1 or the other medication (42.1%; n = 11 glargine discontinuation, n = 40 exenatide discontinuation). Overall, 15 (9.1%) and 52 (31.5%) patients discontinued glargine and exenatide therapy, respectively, by the month 24 follow-up. Although the reasons associated 22 Fig. 2. Body weight (kg), change from baseline. White columns: exenatide therapy was started first followed by the addition of insulin glargine. Gray columns: insulin glargine therapy was started first followed by the addition of exenatide. a P<.001 vs baseline; b P = .001 vs baseline. with discontinuation of either medication were to be extracted from the medical records, they frequently were not available. DISCUSSION This 24-month naturalistic study demonstrated that in patients with inadequately controlled T2DM taking OADs, combining glargine and exenatide, regardless of the chronologic order in which they were prescribed, significantly reduced A1C, with 31% to 34% of patients achieving a A1C level of 7.0% or less. Moreover, the clinical benefits of combined glargine and exenatide therapy were maintained during the 24-month treatment. Body weight and blood pressure values at month 24 were either unchanged from baseline (with exenatide/ glargine) or slightly decreased (with glargine/exenatide). Because patients who had glargine added to exenatide were already slightly heavier at baseline than those who added exenatide to glargine (by ~4 kg), the lack of significant weight gain with the addition of glargine is clinically significant because it suggests that patients on this therapeutic combination may achieve A1C goals without incurring additional risks of significant weight gain over a 24-month period and in terms of patient adherence to the combined therapy. Therefore, the weight increase sometimes seen with the addition of glargine may be balanced by the weight-reducing effects of exenatide, making the combination therapy weight neutral. Patients in the glargine/exenatide group experienced an average loss of –2.5 kg with up to 24 months of combination therapy. This suggests that exenatide may be useful for patients on an established regimen of glargine who may benefit from weight loss and who have suboptimal glycemic control. This weight loss is slightly less than that reported previously with longerterm treatment (~5-6 kg) (12-14), but may be due to the current patient sample having a lower baseline body mass index (overall body mass index at baseline: 37.6 ± 6.7 kg/ m2) than in 2 of the 3 previous studies where the mean baseline body mass index was 39.0 and 43.6 kg/m2 (12,14). Overall, these findings suggest that combining exenatide and glargine can positively affect glycemic control without significantly affecting weight gain during a 24-month period. Reported episodes of hypoglycemia were of similar frequency and severity (based on reported hypoglycemic blood glucose values) in both treatment-order groups, with 11% to 12% of patients in either group experiencing at least 1 hypoglycemic episode during the 24-month follow-up. This frequency was similar to the 10% previously reported in a retrospective chart review (13,16). In the present analysis, the 2 treatment-order groups exhibited similar, significant, and durable improvements in glycemic control on the basis of A1C and the proportion of patients achieving a target A1C of less than 7.0%. In the pooled combination-therapy group at the 24-month follow-up, A1C levels were significantly reduced from baseline (–0.7 ± 1.6%) and 33% of patients achieved a target A1C level less than 7.0%. These results are consistent with a 30-week, randomized, controlled study by Buse et al (17) who reported that A1C decreased by 1.74% (baseline: 8.35%) and weight decreased by 1.8 kg (baseline: 95.4 kg; baseline body mass index: 33.8 kg/m2) after the addition of exenatide to insulin glargine therapy (17). The findings from the present analysis are also similar to those of previous chart reviews of insulin/exenatide combination therapy (12-14). With length of follow-up varying from 26 weeks (12) to 27 months (14), endpoint A1C reductions persisted with longer-term treatment, ranging between 0.54% (14) and 0.87% (13). Moreover, the proportion of patients achieving a target A1C level less than 7.0% in the analysis by Sheffield et al (37%) (13) was 23 Table 2 Summary of Key Outcomes Over 24 Months of Follow-up Outcome Exenatide/glargine Glycemic control A1C change, mean (±SD), % A1C ≤7.0%, No. (%) Weight change, mean (±SD), kgb Blood pressure, mean (±SD), mm Hg Systolic Diastolic Total daily dose, per kg, mean (±SD)b Insulin glargine, units Prandial insulin, units Total insulin, units Exenatide, mg No. of OADs used, follow-up, mean (±SD)b Metformin, No. (%) Thiazolidinedione, No. (%) Sulfonylureas, No. (%) Glargine/exenatide Glycemic control A1C change, mean (±SD), % A1C ≤7.0%, No. (%) Weight change, mean (±SD), kgb Blood pressure, mm Hg Systolic Diastolic Total daily dose, per kg, mean (±SD)b Insulin glargine, units Prandial insulin, units Total insulin, units Exenatide, mg No. of OADs used, mean (±SD)b Metformin, No. (%) Thiazolidinedione, No. (%) Sulfonylureas, No. (%) All treatment-order patients, pooled Glycemic control A1C change, mean (±SD), % A1C ≤7.0%, No. (%) Blood pressure, mean (±SD), mm Hg Systolic Diastolic 6 Months 12 Months 18 Months 24 Months n = 44 n = 43 n = 40 n = 30 –1.0 (±1.3)a 10 (29.4)a 0.6 (±3.6) –0.9 (±1.2)a 17 (42.5)a 0.6 (±5.0) –0.9 (±1.3)a 14 (37.8)a 1.4 (±5.8) –1.0 (±1.2)a 8 (30.7)a 0.7 (±8.3) 129.8 (±15.1) 76.0 (±9.6) 129.4 (±19.8) 77.9 (±8.1) 128.5 (±13.7) 75.7 (±9.5) 128.9 (±14.3) 75.8 (±8.4) 0.22 (±0.11) 0.12 (±0.09) 0.26 (±0.18) 0.16 (±0.05) 0.33 (±0.19) 0.32 (±0.36) 0.40 (±0.38) 0.17 (±0.05) 0.36 (±0.22) 0.24 (±0.29) 0.44 (±0.34) 0.16 (±0.05) 0.40 (±0.23) 0.27 (±0.28) 0.48 (±0.37) 0.19 (±0.05)c 2.1 (±1.0) 29 (65.9) 17 (38.6) 27 (61.3) 2.0 (±1.0) 29 (67.4) 13 (30.2) 22 (51.1) 2.0 (±1.0) 26 (65.0) 12 (30.0) 21 (52.5) 1.7 (±1.0)d 20 (66.6) 9 (30.0) 13 (43.3) n = 121 n = 121 n = 112 n = 106 –0.8 (±1.5)a 31 (32.6)a –1.8 (±3.6)a –0.8 (±1.3)a 33 (31.7)a –2.3 (±5.3)a –0.7 (±1.3)a 35 (36.8)a –2.5 (±5.7)a –0.7 (±1.7)a 26 (33.7)a –2.5 (±6.7)d 128.9 (±17.2) 73.1 (±10.8) 128.5 (±17.4) 74.6 (±11.2) 127.9 (±17.6) 73.3 (±10.1) 125.4 (±14.5) 72.7 (±7.8) 0.40 (±0.22)c 0.31 (±0.28) 0.52 (±0.38)d 0.15 (±0.04) 1.7 (±1.0) 69 (57.0) 44 (36.3) 57 (47.1) 0.42 (±0.25)c 0.27 (±0.29) 0.53 (±0.37)d 0.18 (±0.05) 1.6 (±1.0) 67 (59.8) 36 (32.1) 44 (39.2) 0.46 (±0.26)a 0.28 (±0.26) 0.59 (±0.42)a 0.18 (±0.05) 1.6 (±0.9) 67 (63.2) 32 (30.1) 40 (37.7) 0.47 (±0.30)d 0.30 (±0.27) 0.60 (±0.45)d 0.19 (±0.05) 1.7 (±0.9) 47 (57.3) 26 (31.7) 29 (35.3) n = 165 n = 165 n = 155 n = 146 –0.8 (±1.3)a 50 (34.7)a –0.8 (±1.3)a 49 (37.1)a –0.7 (±1.6)a 34 (33.0)a 128.7 (±18.0) 75.5 (±10.5) 128.0 (±16.6) 73.9 (±10.0) 126.2 (±14.5) 73.4 (±8.0) –0.9 (±1.4)a 41 (31.7)a 129.1 (±16.6) 73.9 (±10.5) Abbreviations: A1C, hemoglobin A1c; OADs, oral antidiabetic drugs; SD, standard deviation. a P<.001 vs baseline. b Because of the differences between the treatment-order groups at baseline and during follow-up in patterns of weight change, insulin glargine and exenatide dosing, and the number of OADs used, data for these variables are not provided for the pooled combination-treatment group. c P≤.02 vs baseline. d P≤.003 vs baseline. 24 similar to that observed in the present analysis at endpoint (33%). In a prospective, uncontrolled, naturalistic medical chart review of insulin/exenatide combination therapy, Nayak et al (15) showed that adding exenatide to insulin therapy significantly decreased A1C at 3 months’ followup (8.4 ± 1.8; P = .001), but the decreases were similar to baseline values after 1 year (9.1 ± 2.0; P = .74) (15). Body weight, daily insulin dose, and blood pressure were significantly reduced at the 1-year follow-up (P<.05). However, the specific type(s) of insulin taken was not noted and 65% of patients were taking large doses of insulin (>100 units daily) at baseline (15). These observational studies and the current results suggest that combination glargine/ exenatide treatment leads to significant and durable glycemic improvements over time. Findings from the present investigation, which was the first to examine the order in which glargine and exenatide are added to an antidiabetic regimen using a naturalistic study design, demonstrate that A1C improvements occur regardless of the order in which these medications are prescribed and administered. Exenatide dosing throughout the follow-up period was largely constant, considering initial upward titration, with the average daily exenatide dose at 0.19 mg/kg at month 24 for both treatment-order groups. Of note, there was more than a 3-fold greater rate of discontinuation among patients who had received exenatide vs the discontinuation rate of those who had received glargine. It was expected that patients in the glargine/exenatide group (eg, patients with an established insulin regimen at baseline) may require lower total daily insulin doses after adding exenatide on the basis of previous reports by Sheffield (13) and Yoon (14), who found reduced prandial insulin requirements with such treatment. In the present investigation, however, the prandial insulin dose remained constant, and the total daily insulin dose and total daily glargine dose showed clinically small but significant increases from baseline in this group. Nevertheless, these patients had an 11.8-year history of T2DM at baseline, and progressing disease severity may have made such dosing increases necessary. Moreover, patients in the glargine/exenatide group exhibited less sulfonylurea and thiazolidinedione use, with a slightly higher proportion taking metformin at the 24-month endpoint, although the average number of OADs used did not change during follow-up. This relative lack of change in OAD use may reflect the potentially more advanced state of illness in these patients. Similar total daily insulin doses (0.48 units/kg) and changes in OAD use were seen at month 24 among exenatide/glargine patients, marked by relatively larger decreases in sulfonylurea and thiazolidinedione use, and by increases in the use of metformin. In this group, the number of OADs used was significantly decreased during follow-up (2.3 vs 1.7), which may have contributed to increased requirements for basal and prandial insulin. Relative to the glargine/exenatide group, it is reasonable to hypothesize that these patients may have exhibited less advanced disease at baseline (8.0- vs 11.8-year diabetes history) and may have responded more favorably to combination therapy. Limitations The present study has several limitations. This was a retrospective chart review that lacked a comparison group, such as patients who continued their baseline antidiabetic treatment regimen. Thus, it is possible that treatment patterns and subsequent outcomes may have been influenced by individual physician preferences for particular medication combinations/doses. Findings based on the overall pooled combination-therapy group should also be interpreted with caution, because the groups who pooled together differed from one another in certain parameters at baseline (eg, diabetes duration, number of OADs used) and during followup. Baseline differences in the number of patients in each group and in illness duration and treatment pattern between the groups may account for some of the differences in outcomes. The patients in the glargine/exenatide group may have been sicker, given their nearly mean 12 years of diabetes duration vs 8 years for the patients in the exenatide/ glargine group. Information regarding patients’ race was not available for evaluation. Furthermore, adverse events (other than hypoglycemia) and reasons for discontinuation associated with combination therapy were not systematically determined, and hypoglycemia was based on patient report, possibly resulting in underestimation. Nevertheless, the naturalistic, observational nature of the investigation provides a snapshot of how a heterogeneous population of patients with poorly controlled T2DM may respond to combined glargine and exenatide therapy in real-world clinical settings, thus providing a more realistic observation of therapeutic responses over a larger sample of the treatment population. CONCLUSION The combination of glargine and exenatide, regardless of the chronologic order of their addition to the therapeutic regimen, significantly improved A1C and glycemic goal attainment (A1C ≤7.0%) without additional risk of weight gain. This evaluation also suggested no increased risk of hypoglycemia. Combination glargine/exenatide therapy also may help stabilize insulin dose and blood pressure while reducing OAD therapy. Thus, it represents an effective, persistent, and well-tolerated strategy for achieving measurable clinical benefits in patients with poorly controlled T2DM already receiving glargine or exenatide monotherapy. The results of this retrospective naturalistic study suggest that combining glargine and a glucagonlike peptide 1 analogue, such as exenatide, has clinical benefits and that further examination in randomized clinical trials is warranted. 25 ACKNOWLEDGMENT Study funding was provided by sanofi-aventis U.S. Editorial support was provided by Richard Fay, PhD, and Eleanor O’Rangers, PharmD, of Embryon, LLC, and was funded by sanofi-aventis U.S. The authors, however, were fully responsible for all content and editorial decisions and received no financial support or other form of compensation related to the development of this article. Preliminary results from this study were presented in abstract and poster form at the annual meeting of the American Diabetes Association, June 25-29, 2010, in Orlando, Florida. DISCLOSURE Dr. Levin is a consultant to sanofi-aventis U.S. Dr. Zhou is an employee of sanofi-aventis U.S. Dr. Mersey and Mr. Bromberger have no multiplicity of interest to disclose. REFERENCES 1. Östgren CJ, Lindblad U, Ranstam J, Melander A, Råstam L; Skaraborg hypertension and Diabetees Project. Glycaemic control, disease duration and betacell function in patients with type 2 diabetes in a Swedish community. Skaraborg Hypertension and Diabetes Project. Diabet Med. 2002;19:125-129. 2. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33) [Erratum in Lancet. 1999;354:602]. Lancet. 1998;352: 837-853. 3. Nathan DM, Buse JB, Davidson MB, et al; American Diabetes Association; European Association for Study of Diabetes. Medical management of hyperglycemia in type 2 diabetes: a consensus algorithm for the initiation and adjustment of therapy: a consensus statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care. 2009;32:193-203. 4. Home PD, Fritsche A, Schinzel S, Massi-Benedetti M. Meta-analysis of individual patient data to assess the risk of hypoglycaemia in people with type 2 diabetes using NPH insulin or insulin glargine. Diabetes Obes Metab. 2010;12: 772-779. 5. Riddle MC, Rosenstock J, Gerich J; Insulin Glargine 4002 Study Investigators. The Treat-to-Target Trial: randomized addition of glargine or human NPH insulin to oral therapy of type 2 diabetic patients. Diabetes Care. 2003; 26:3080-3086. 6. Rosenstock J, Dailey G, Massi-Benedetti M, Fritsche A, Lin Z, Salzman A. Reduced hypoglycemia risk with insulin glargine: a meta-analysis comparing insulin glargine with human NPH insulin in type 2 diabetes. Diabetes Care. 2005;28:950-955. 7. Forst T, Larbig M, Hohberg C, et al. Adding insulin glargine vs. NPH insulin to metformin results in a more efficient postprandial beta-cell protection in individuals with type 2 diabetes. Diabetes Obes Metab. 2010;12:437-441. 8. Rodbard HW, Blonde L, Braithwaite SS, et al. American Association of Clinical Endocrinologists medical guidelines for clinical practice for the management of diabetes mellitus. Endocr Pract. 2007;13(Suppl 1):1-68. 9. Heine RJ, Van Gaal LF, Johns D, Mihm MJ, Widel MH, Brodows RG; GWAA Study Group. Exenatide versus insulin glargine in patients with suboptimally controlled type 2 diabetes: a randomized trial. Ann Intern Med. 2005; 143:559-569. 10. Barnett AH, Burger J, Johns D, et al. Tolerability and efficacy of exenatide and titrated insulin glargine in adult patients with type 2 diabetes previously uncontrolled with metformin or a sulfonylurea: a multinational, randomized, open-label, two-period, crossover noninferiority trial. Clin Ther. 2007;29:2333-2348. 11. Arnolds S, Dellweg S, Clair J, et al. Further improvement in postprandial glucose control with addition of exenatide or sitagliptin to combination therapy with insulin glargine and metformin: a proof-of-concept study. Diabetes Care. 2010;33:1509-1515. 12. Viswanathan P, Chaudhuri A, Bhatia R, Al-Atrash F, Mohanty P, Dandona P. Exenatide therapy in obese patients with type 2 diabetes mellitus treated with insulin. Endocr Pract. 2007;13:444-450. 13. Sheffield CA, Kane MP, Busch RS, Bakst G, Abelseth JM, Hamilton RA. Safety and efficacy of exenatide in combination with insulin in patients with type 2 diabetes mellitus. Endocr Pract. 2008;14:285-292. 14. Yoon NM, Cavaghan MK, Brunelle RL, Roach P. Exenatide added to insulin therapy: a retrospective review of clinical practice over two years in an academic endocrinology outpatient setting. Clin Ther. 2009;31:1511-1523. 15. Nayak UA, Govindan J, Baskar V, Kalupahana D, Singh BM. Exenatide therapy in insulin-treated type 2 diabetes and obesity. QJM. 2010;103:687-694. 16. Byetta [package insert]. San Diego, CA: Amylin Pharmaceuticals, Inc; 2007. 17. Buse JB, Bergenstal RM, Glass LC, et al. Use of twicedaily exenatide in basal insulin-treated patients with type 2 diabetes: a randomized, controlled trial. Ann Intern Med. 2011;154:103-112.