Efficacy of Exenatide Administered Twice Daily in Body Mass Index Reduction in Patients with Type 2 Diabetes Mellitus

Background Exenatide is a glucagon-like peptide-1 receptor agonist that can reduce body weight. This study aimed to determine the efficacy of exenatide on body mass index (BMI) reduction in patients with type 2 diabetes mellitus (T2DM) with differing baseline body weight, blood glucose, and atherosclerotic status and to determine if there is a correlation between BMI reduction and cardiometabolic indices in these patients. Methods This retrospective cohort study used data from our randomized controlled trial. A total of 27 T2DM patients treated with combination therapy of exenatide twice daily and metformin for 52 weeks were included. The primary endpoint was a change in the BMI from the baseline to week 52. The secondary endpoint was a correlation between BMI reduction and cardiometabolic indices. Findings. The BMIs of overweight and obesity patients and those with glycated hemoglobin (HbA1c) ≥ 9% significantly decreased −1.42 ± 1.48 kg/m2(P=0.015) and −0.87 ± 0.93 kg/m2(P=0.003), respectively, at the baseline after 52 weeks of treatment. There was no reduction in BMI in patients with normal weight, HbA1c <9%, the nonatherosclerosis group, and the atherosclerosis group. The decrease in BMI was positively correlated with changes in blood glucose, high-sensitivity C-reactive protein (hsCRP), and systolic blood pressure (SBP). Conclusion BMI scores improved after exenatide treatment for 52 weeks in T2DM patients. Weight loss was affected by baseline body weight and blood glucose level. In addition, BMI reduction from the baseline to 52 weeks was positively correlated with baseline HbA1c, hsCRP, and SBP. Trial Registration. Chinese Clinical Trial Registry (ChiCTR-1800015658).


Introduction
Obesity is characterized by excessive accumulation of visceral and subcutaneous fat and is the common concomitant condition of type 2 diabetes mellitus (T2DM) [1]. Approximately 60% of T2DM patients in China are overweight and obese [2]. Weight gain is a major risk factor for T2DM as it induces chronic infammation, endoplasmic reticulum stress, mitochondrial dysfunction, and insulin resistance [1,3]. Abdominal obesity is also an important factor for arteriosclerosis [4]. Obesity promotes atherosclerosis by activating migration of endothelial cells, increasing smooth muscle cell proliferation, and promoting vascular calcifcation [5] In addition, fbrinogen, platelet count, blood lipids, and plasma viscosity are signifcantly higher in obese individuals than in normal-weight people [5] Deformation of red blood cells also decreases with obesity, resulting in abnormal hemorheology, circulatory dysfunction, tissue blood perfusion disorders, and increased risk of thrombosis and arteriovenous atherosclerosis formation [6]. Sustained moderate weight loss of more than 5% of initial body weight provides signifcant benefts for blood glucose, blood pressure, blood lipid, and other metabolic indices in patients with T2DM [7][8][9]. Terefore, weight reduction has been considered a key element of patient-centered approaches for the management of T2DM [10,11].
A previous study also suggested that the baseline BMI was closely related to weight change and glycemic control after exenatide monotherapy [24]. However, the efect of GLP-1RAs on body weight reduction in patients with T2DM with difering baseline BMI measurements has not been extensively studied [24]. Furthermore, little is known about the infuence of baseline blood glucose and the presence of atherosclerosis on weight loss efcacy after GLP-1RA treatment in T2DM patients. Te purpose of this retrospective cohort study was to compare weight loss efcacy of exenatide in T2DM patients with difering baseline BMI, blood glucose, and atherosclerotic levels and to determine if there is a correlation between BMI reduction and cardiometabolic indices in order to identify patient characteristics that could result in good response to exenatide.

Patients.
Tis was a retrospective cohort study using data from our randomized clinical trial that investigated the efects of exenatide, administered twice daily, combined with metformin when compared to insulin on carotid intimamedia thickness (cIMT) in patients with T2DM [23]. Te original study was performed in accordance with the Declaration of Helsinki, and the study protocol was reviewed and approved by the Research Ethics Committee of Beijing Hospital (No. 2013 BJYYEC-017A-03). Te study was registered in the Chinese Clinical Trial Registry (ChiCTR-1800015658). After all participants were informed of the study details, they signed the corresponding consent forms.
Exenatide was administered 5 μg twice daily subcutaneously, 60 min before breakfast and dinner. Te exenatide dose was titrated to 10 μg twice daily after 4 weeks. Metformin 500 mg was administered three times daily according to the blood glucose level.

Outcomes.
Te primary endpoint was a change in the BMI from the baseline to week 52. Te secondary endpoints were correlations between BMI reduction and cardiometabolic indices: HbA1c, high-sensitivity C-reactive protein (hsCRP), systolic blood pressure (SBP), and lowdensity lipoprotein cholesterol (LDL-C).

Statistical
Analysis. Data were analyzed using SPSS statistical software version 22 for Windows (IBM Corp., Armonk, NY, USA). Data are expressed as means and standard deviations (SD). Te normality of the data was analyzed using the Kolmogorov-Smirnov test. Te paired ttest or the Wilcoxon signed-rank test was used for withingroup comparisons. Multiple linear regression was used to analyze correlations between weight, hsCRP, SBP, and LDL-C. A P value <0.05 indicated statistical signifcance.

Discussion
Tis retrospective cohort study showed that exenatide administered twice daily was clinically sufcient to signifcantly reduce body weight in T2DM patients with varying baseline levels of the BMI, suggesting a correlation between weight loss efcacy of exenatide and obesity.

2
International Journal of Clinical Practice Te BMI of patients in all subgroups decreased at the baseline after 52 weeks of exenatide treatment, and this result was afected by obesity (baseline BMI). Te statistically signifcant decrease in the BMI was found only in patients with a BMI ≥24 kg/m 2 but not those with <24 kg/m 2 . A previous study found that exenatide reduced body weight in T2DM patients with diferent levels of obesity [25]. Another study observed the benefts of exenatide in the second week of treatment [26,27]. A meta-analysis [28] of Asians also demonstrated that GLP-1RAs exhibited more benefts in T2DM patients with an overweight/obese BMI (n � 14; −0.44 kg; 95% confdence interval (CI) −0.62, −0.26; I 2 � 85.3%) than in people with a normal BMI (n � 6; −0.03 kg; 95% CI −0.38, 0.32; I 2 � 87.6%). However, body weight loss (P � 0.572) was similar between the two groups. Te benefcial efects of exenatide twice daily on body weight improvement were similar to those of exenatide once weekly [29] and liraglutide [30,31].
We also found that weight loss efcacy of exenatide was infuenced by hyperglycemic stages (baseline HbA1c). Te decrease in the BMI after exenatide treatment tended to be slightly higher in patients with high baseline HbA1c values (HbA1c ≥ 9%) than those with low values (HbA1c < 9%); however, there was no signifcant diference between the   groups. Tis fnding suggests that exenatide could exhibit weight-lowering benefts for patients regardless of the baseline HbA1c level. Te correlation analysis also revealed that BMI reduction after 52 weeks was positively correlated with HbA1c levels (r � 0.419, (P � 0.030)). However, a prior study did not observe a correlation between weight loss and improvement in glycemic control [32]. Although 63.3% of the eligible participants who were treated with GLP-1RAs obtained dual benefts, including HbA1c reduction and weight loss, 3.8% of the patients did not respond to exenatide. Interestingly, 22% of the patients lost weight but had no improvement in glycemic control. In contrast, 11% of the patients gained weight but had blood glucose reduction. Tis is consistent with the result of another study of Korean obese T2DM patients that reported that the weight reduction efect of exenatide could not predict a decrease in HbA1c [33]. Te magnitude of BMI reduction in patients with a BMI <24 kg/m 2 and HbA1c < 9% was clinically signifcant, although there was lack of statistical signifcance. Tis might be related to the respective lifestyles of patients. Patients with normal weight tended to have healthier lifestyles and less weight changes than obese persons. Moreover, the average BMI value in the BMI <24 kg/m 2 group was in a normal range at the baseline, which may explain the lack of a statistically signifcant efect of exenatide on the normal BMI group. Patients with better blood glucose control tended to have more strict dietary control, resulting in a weakening efect of GLP-1. Tus, the weight improvement was not statistically signifcant in these patients. Our multivariate linear regression model showed that weight loss from the baseline to 12 months was positively correlated with a patient's baseline weight and LDL-C and triglycerides levels and negatively correlated with prior insulin treatment [31]. Moreover, it has been documented that exenatide treatment twice daily was efective at lowering body weight in T2DM patients regardless of age, sex, race, and duration of diabetes [34,35].
Obesity is a leading cause of cardiometabolic diseases [36]. Abnormal lipid metabolism, insulin resistance, infammation, endothelial dysfunction, imbalance of adipokines, and activation of infammatory bodies are considered the basis of atherosclerosis development in obese populations. Evidence showed that exenatide improved arteriosclerosis by reducing body weight, lowering LDL-C, and improving infammatory responses [23]. In this study, there was no signifcant diference in weight reduction among T2DM patients who had difering baseline atherosclerotic statuses, suggesting that the presence of atherosclerosis did not afect the weight loss beneft of exenatide. Correlation analysis further revealed that BMI reduction after 52 weeks was positively correlated with hsCRP levels (r � 0.440, P � 0.022), indicating that exenatide has potential for improving metabolic indices among T2DM patients. In our previous study, exenatide also signifcantly improved infammation indices (CRP, fbrinogen, 8-hydroxydeoxyguanosine, irisin, and brain natriuretic peptide) from baseline levels [23].
Tere are some limitations in our study that should be addressed. Patients who received metformin alone were not included in the control arm of the analysis. Te efects of exenatide treatment on clinical outcomes, especially BMI reduction, should be further analyzed in comparison with placebo as well as active comparators in future studies.

Conclusion
We found that exenatide has a signifcant efect on weight management in T2DM patients. Weight loss can be afected by baseline body weight and blood glucose level, and the benefts can be obtained regardless of arteriosclerosis status at the baseline.
Our study indicates that overweight or obese T2DM patients could respond better to exenatide than normalweight patients without an increased risk of excess weight loss. Patients with poor blood glucose control can also beneft more from the weight loss efect of exenatide than patients with good blood glucose control. Patients with severe atherosclerosis can achieve the same weight loss efect as patients with mild atherosclerosis. In addition, the BMI reduction from the baseline to 52 weeks was positively correlated with baseline HbA1c, hsCRP, and SBP.

Data Availability
Te datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

Additional Points
What is already known about this subject?: (i) Obesity is a common concomitant condition of T2DM. (ii) Sustained moderate weight loss of more than 5% of initial body weight provides signifcant benefts for blood glucose, blood pressure, blood lipid, and other metabolic indices in patients with T2DM. (iii) Glucagon-like peptide-1 receptor agonists (GLP-1RAs) have been considered for individuals with T2DM who require weight loss regardless of HbA1c levels. (iv) Te baseline BMI afects weight change and glycemic control after exenatide monotherapy. What are the new fndings?: (i) Exenatide signifcantly decreased BMI measurements in overweight and obese patients and in those with glycated hemoglobin (HbA1c) ≥ 9% after 52 weeks of treatment. (ii) Exenatide treatment did not reduce BMI measurements in healthy patients. (iii) BMI reduction after exenatide treatment for 52 weeks was positively correlated with baseline HbA1c, hsCRP, and systolic blood pressure. How might your results change the direction of research or the focus of clinical practice?: (i) Tere is a correlation between weight loss efcacy of exenatide and obesity. (ii) Overweight or obese T2DM patients can respond better to exenatide without an increased risk of excess weight loss. (iii) Weight loss efcacy of exenatide is also infuenced by hyperglycemia. Compared to those with normal blood glucose levels, patients with poor blood glucose control can beneft from the weight loss efect of exenatide. (iv) Te presence of atherosclerosis did not afect the weight loss beneft of exenatide.

Ethical Approval
Te Research Ethics Committee of Beijing Hospital reviewed and approved the study protocol before the enrollment of patients (No. 2013 BJYYEC-017A-03).

Conflicts of Interest
Te authors declare that they have no conficts of interest.

Authors' Contributions
TZX, LXG, and YT conceived and designed research; JZ, XZW, CL, MXW, WHW, FM, XBZ, and XXW collected data and conducted research; JZ analyzed and interpreted data; JZ wrote the initial paper; JZ revised the paper; JZ, TZX, and LXG had primary responsibility for fnal content. All authors read and approved the fnal manuscript. JZ, TZX, and YT contributed equally to the work.

Supplementary Materials
Supplemental Figure 1. Comparison of BMI change in the normal weight group (BMI <24 kg/m 2 ) and the overweight and obesity group (BMI ≥24 kg/m 2 ) after exenatide treatment for 52 weeks. Supplemental Figure 2. Comparison of BMI change in the HbA1c <9% group and the HbA1c ≥ 9% group after exenatide treatment for 52 weeks. Supplemental Figure 3. Comparison of BMI change in the mild atherosclerotic group (cIMT <1 mm) and severe atherosclerotic group (cIMT ≥1 mm) after exenatide treatment for 52 weeks.