Accepted Manuscript The effects of exenatide twice daily compared to insulin lispro added to basal insulin in Latin American patients with type 2 diabetes: A retrospective analysis of the 4B trial Silvia Beatriz Gorban de Lapertosa, Gustavo Frechtel, Elise Hardy, Leobardo Sauque-Reyna PII: S0168-8227(16)30722-7 DOI: http://dx.doi.org/10.1016/j.diabres.2016.10.001 Reference: DIAB 6765 To appear in: Diabetes Research and Clinical Practice Received Date: 15 April 2016 Revised Date: 21 September 2016 Accepted Date: 1 October 2016 Please cite this article as: S.B.G. de Lapertosa, G. Frechtel, E. Hardy, L. Sauque-Reyna, The effects of exenatide twice daily compared to insulin lispro added to basal insulin in Latin American patients with type 2 diabetes: A retrospective analysis of the 4B trial, Diabetes Research and Clinical Practice (2016), doi: http://dx.doi.org/10.1016/ j.diabres.2016.10.001 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. The effects of exenatide twice daily compared to insulin lispro added to basal insulin in Latin American patients with type 2 diabetes: A retrospective analysis of the 4B trial Silvia Beatriz Gorban de Lapertosaa, Gustavo Frechtelb, Elise Hardyc, Leobardo Sauque-Reynad aNational University of the Northeast, School of Medicine, Corrientes, Argentina; bHospital de Clinicas, Universidad de Buenos Aires, Buenos Aires, Argentina; cAstraZeneca, Gaithersburg, MD, USA; dInstituto de Diabetes Obesidad y Nutrición, Sociedad Civil, Cuernavaca, México Corresponding Author: Gustavo Frechtel, MD, PhD Hospital de Clínicas, Universidad de Buenos Aires Córdoba 2351 C1120AAR CABA, Buenos Aires, Argentina 1419 Tel: 0541145043658 Fax: 0541145043259 Email: [email protected] Running Head: Exenatide twice daily in Latin American patients 1 Abstract Aims: Socioeconomic changes in Latin American countries have led to an increased prevalence of type 2 diabetes (T2D). We examined the effects of exenatide twice daily (BID) or insulin lispro, each added to insulin glargine, in Latin American patients with T2D. Methods: This was a subgroup analysis of patients from Argentina and Mexico in the 4B study (N=114). Patients with HbA1c 7.0–10.0% (53–86 mmol/mol) after 12 weeks of intensive basal insulin optimization were randomized to exenatide BID or thrice-daily insulin lispro added to insulin glargine and metformin. Results: After 30 weeks, addition of exenatide BID or insulin lispro resulted in significant (P<0.0001) reductions in HbA1c (exenatide BID: −0.9% [−10 mmol/mol]; insulin lispro: −1.2% [−13 mmol/mol]). Weight was stable in the exenatide BID group (–0.1 kg) and increased significantly (+3.4 kg, P<0.0001) with insulin lispro. Major and minor hypoglycemia occurred less frequently (40 vs. 253 events) with exenatide BID compared with insulin lispro. Gastrointestinal adverse events of nausea, diarrhea, and vomiting occurred more frequently with exenatide BID than with insulin lispro. Conclusions: Both exenatide BID and prandial insulin lispro, each added to basal insulin glargine, were effective at reducing HbA1c in Latin American patients. Treatment with exenatide BID resulted in stable weight but more gastrointestinal adverse events. Treatment with insulin lispro resulted in weight gain and an increased risk of hypoglycemia. These findings support the addition of exenatide BID to insulin glargine as an option for Latin American patients unable to achieve glycemic control on basal insulin alone. Key Words: exenatide; insulin lispro; type 2 diabetes; Latin America 1 1. Introduction The population of patients with type 2 diabetes is heterogeneous, with patients differing in key aspects such as the rate of disease progression and presence of obesity [1]. Since separate phenotypes of type 2 diabetes have not been fully identified, subgroups are often defined based on geographic location. Populations from specific regions may share unique genetic and/or dietary characteristics that affect glycemic control and may therefore impact the response to therapy. Characterizing responses to treatment within patient subgroups from early to late stages of diabetes may inform individualized treatment decisions. In Latin American countries, socioeconomic changes have led to lifestyle changes and subsequent increases in the prevalence of obesity, type 2 diabetes, and cardiovascular disease [2]. The average prevalence of type 2 diabetes in South and Central American countries was reported as 8.0% in 2013 and is expected to reach 9.8% by 2035 [3], while in Mexico, the prevalence of type 2 diabetes was reported as 14.4% in 2006 [4]. In light of these trends, evaluation of antidiabetic treatment choices in Latin American patients is of interest, particularly for therapeutic options with novel mechanisms of action. In a multi-national, multi-center study of patients who were unable to achieve glycemic control after 12 weeks of basal insulin optimization (BIO) and metformin, the Basal Insulin Glargine +Exenatide BID vs. Basal Insulin Glargine + Bolus Insulin Lispro (4B) study group found that the addition of the glucagon-like peptide-1 (GLP-1) receptor agonist exenatide twice daily (BID) produced a comparable reduction in glycated hemoglobin (HbA1c) to that of prandial insulin lispro [5]. Treatment with exenatide BID was also associated with weight loss, a reduction in systolic blood pressure, and greater treatment satisfaction, despite an increased 2 prevalence of gastrointestinal adverse events. Treatment with insulin lispro was associated with weight gain and a higher rate of hypoglycemia. Differences in the effects of these treatments are likely due to their differing mechanisms of action; GLP-1 receptor agonists increase glucose- dependent insulin release and decrease glucose-dependent glucagon release, delay gastric emptying, and reduce food intake [6]. The glucose-dependent nature of GLP-1 receptor agonists improves glycemic control at elevated blood glucose levels, without increasing the risk of hypoglycemia. Insulin stimulates glucose reuptake in skeletal muscle and glycogenesis in the liver and inhibits glucagon secretion [7]. When administered exogenously, insulin reaches high concentrations in peripheral tissue, which increases the risk for hypoglycemia and weight gain. Here, we conducted a subgroup analysis of Latin American patients enrolled in the primary 4B study to evaluate the safety and efficacy of exenatide BID compared with insulin lispro, added-on to basal insulin glargine and metformin, in Latin American patients who failed to achieve glycemic control on titrated insulin glargine and metformin alone. 2. Materials and methods 2.1 Study design This was a post-hoc subgroup analysis of Latin American patients who participated in the open- label, randomized, controlled, noninferiority 4B study (NCT00960661), the methods of which have previously been reported [5]. Briefly, male and female patients, aged ≥18 years, with type 2 diabetes currently treated with insulin glargine and metformin (with or without sulfonylurea), and with an HbA1c of 7.0% to 10.0% (53 to 86 mmol/mol) and a BMI of 25.0 kg/m2 to 45.0 kg/m2, were eligible for enrollment. Sulfonylurea use was discontinued upon study entry. Type 2 diabetes was diagnosed by World Health Organization criteria [8]. Islet cell antibodies, C- 3 peptide, or insulin levels were not tested to further classify the type of diabetes. Patients with medical conditions that had the potential to impact patient safety, such as clinically significant cardiovascular disease (New York Heart Association Class III or IV for congestive heart failure or judged by the investigator to be exclusionary), ≥1 episode of major hypoglycemia in the 6 months prior to enrollment, fasting triglycerides ≥5.7 mmol/L, or a history of confirmed pancreatitis, were excluded from the study. For this subgroup analysis of Latin American patients, safety and efficacy data were pooled from the study centers in Mexico and Argentina and not from self-identified race. The study consisted of a 12-week BIO phase, followed by randomization and a 30-week intervention phase [5]. During the BIO phase, insulin glargine was titrated to maintain a self- monitored fasting glucose (FG) of 5.6 mmol/L or lower without hypoglycemia (ie, glucose <3.0 mmol/L). At the end of the BIO phase, patients who failed to achieve an HbA1c of 7.0% (53 mmol/mol) or less were eligible to be randomized and to continue in the study. To minimize hypoglycemia associated with medication changes, patients randomized to the exenatide BID group and with an HbA1c of ≤8.0% (≤64 mmol/mol) reduced their insulin glargine dose by ≥10%. Exenatide BID was added at a dose of 5 µg BID for the first four weeks and 10 µg BID thereafter, taken with the two largest meals (breakfast or lunch and dinner) at least 6 hours apart. Similarly, patients randomized to insulin lispro with an HbA1c of ≤8.0% (≤64 mmol/mol) reduced their insulin glargine dose by one-half to one-third, and added the equivalent dose (one- half or one-third of their total glargine dose) in insulin lispro, so that the total insulin dose remained the same. Insulin lispro was dosed three times daily prior to each meal (breakfast, lunch, and dinner). Throughout the intervention phase, insulin glargine was titrated as in the BIO phase. A 4-point self-monitoring of blood glucose profile, recorded with a Roche plasma- 4 equivalent (Roche Diagnostics, Indianapolis, IN, USA) or Roche International Federation of Clinical Chemistry and Laboratory Medicine plasma-equivalent referent meter (provided to all patients), was collected twice weekly. Measurements were taken pre-breakfast, pre-lunch, pre- dinner, and before bed. Insulin lispro was titrated based on self-monitored pre-meal glucose to achieve a target of <6.1 mmol/L without hypoglycemia.