Exenatide Therapy in Obese Patients With Type 2 Diabetes Mellitus Treated With Insulin

Viswanathan P, Chaudhuri A, Bhatia R, Al-Atrash F, Mohanty P, Dandona P. Endocr Pract. 2007;13:444-450.

The incretin mimetic exenatide has recently been approved for the treatment of patients with type 2 diabetes mellitus (T2DM) not adequately controlled with oral agents. Exenatide has been shown to be associated with reductions in both fasting and postprandial glucose concentrations, as well as concomitant weight loss. Reductions in body weight are a benefit that is unique to exenatide since all other treatments for diabetes, other than metformin, are associated with weight gain. This study was designed to assess the efficacy of exenatide among patients who experienced weight gain while treated with insulin and oral hypoglycemic agents, during which time their hyperglycemia was not adequately controlled.

This retrospective study included 52 patients with T2DM who were obese, had uncontrolled hyperglycemia, and continued to gain weight with combinations of oral hypoglycemic agents and insulin. Their mean A1C level was 7.9%, body weight (± standard error of the mean) was 116.8 ± 2.9 kg, and body mass index (BMI) was 43.4 ± 1.3 kg/m². Mean systolic blood pressure (SBP) and diastolic BP (DBP) were 127.8 ± 2.7 mm Hg and 76.04 ± 1.5 mm Hg, respectively. Patients treated with insulin whose A1C levels were > 7% were offered initiation of exenatide treatment. Among these patients, 38 were able to take exenatide regularly without problems related to medical insurance coverage (Group A). Patients in Group B (n = 14), who served as the control group, had problems either obtaining prescriptions or remaining compliant with them for insurance-related, personal, or economic reasons, and their maximum duration of treatment with exenatide was < 2 weeks.

Patients were prescribed exenatide, 5 mcg twice daily, before breakfast and dinner after receiving training on how and when to use the pen device for administration. Their previous antihypertensive and cholesterol-lowering drug regimens were not altered, but the dosages of rapid-acting and mixed insulins were reduced by 10% for those with A1C levels < 7.5%. If preprandial or bedtime glucose levels were > 130 mg/dL, insulin dosages were increased, but they were reduced if levels were < 110 mg/dL. Patients were examined 1 week after initiation of exenatide therapy and every 2 to 6 weeks thereafter, depending on information obtained during clinical evaluations. Blood glucose (BG) concentrations and A1C levels were measured at all clinic visits, and measurements of lipid levels and high-sensitivity C-reactive protein (CRP) concentrations during fasting were performed in a laboratory. The mean follow-up period among all patients was 26 weeks.

Patients in Group A had a mean reduction in body weight from 116.4 ± 3.5 kg to 110 ± 3.2 kg (P < .001), and patients in Group B had an increase from 118 ± 5.7 kg to 120.4 ± 6.1 kg (P < .001). Mean BMI decreased from 43.3 ± 1.5 kg/m² to 40.7 ± 1.4 kg/m² in Group A (P < .001) but increased from 43.6 ± 2.6 kg/m² to 44.5 ± 2.8 kg/m² in Group B (P = .006). There was a significant correlation between decline in body weight and the baseline weight in Group A (r = 0.41[lap1] , P = .01), in contrast to a strong inverse correlation in Group B (r = –0.75, P = .01). There was also a trend toward a relationship between weight loss and treatment duration in Group A (r = 0.25, P = .1) that was not evident in Group B (r = 0.20, P = .50). A 4.8-kg [lap2] reduction in mean body weight was observed in Group A patients treated with thiazolidinediones (TZDs) (n = 20; from 113.81 ± 404 kg to 109 ± 4 kg; P < .001), compared to an 8.4-kg reduction among patients in that group who were not treated with TZDs (n = 17; from 119.86 ± 5.8 kg to 111.51 ± 5.6 kg; P < .001). Group B patients receiving TZDs had an increase in body weight from 119.8 ± 7.5 kg to 122.38 ± 7.8 kg (P = .002), and those who were not had smaller increases, from 116.21 ± 12 kg to 118.41 ± 13.5 kg (P = .29).

A1C levels decreased from 7.7 ± 0.2% to 7.1 ± 0.2% (P = .007) in Group A, but remained unchanged from 8.4 ± 0.5% at baseline in Group B. The decreases in A1C were correlated with baseline A1C levels in Group A (r = 0.68, P < .0001), but there was no such relationship seen in Group B (r = 0.46, P = .27). The decline in A1C level was also correlated with duration of treatment in Group A (r = 0.40, P = .01), but changes in A1C level were not related to changes in body weight. Nine patients from Group A who were obese but had well-controlled hyperglycemia received exenatide and benefited from significant decreases in body weight, from 124 ± 9.2 kg to 115.9 ± 7.5 kg (P = .004), and a decline in systolic BP, from 136 ± 6.9 mm Hg to 122.9 ± 3.3 mm Hg (NS), but had no change in A1C levels (6.1 ± 0.1% to 6 ± 0.2%). These 9 patients also required significantly lower dosages of insulin than patients with A1C levels > 6.5%. Furthermore, after they were excluded[lap3]  from Group A, they had significant reductions in A1C levels (from 8.2 ± 0.2% to 7.4 ± 0.2%; P = .007) and mean body weight (from 114 ± 3.5 kg to 108.2 ­± 3.5 kg; P < .001).

Dosages of short-acting insulin (insulin aspart, insulin lispro) required by Group A decreased from 50.4 ± 6.7 U to 36.6 ± 5.1 U (28% reduction, P = .002). Dosages of mixed insulins, which included 75% insulin lispro protamine suspension/25% insulin lispro injection and 70% insulin aspart protamine suspension/30% insulin aspart injection, decreased from 72.9 ± 15.6 U to 28.3 ± 14.8 U in Group A (60% reduction, P = .02). However, dosages of basal insulins, neutral protamine Hagedorn[lap4] , and insulin glargine did not change significantly.

Total cholesterol concentration decreased from 163.9 ± 8.2 mg/dL to 149.8 ± 5.9 mg/dL (P = .03) in Group A patients, and from 168.1 ± 16.3 mg/dL to 144.33 ± 10.39 mg/dL (P = .08) in Group B patients. The mean low-density lipoprotein [lap5] and high-density lipoprotein cholesterol concentration did not change significantly in either group. The concentrations of plasma triglycerides (TG) decreased from 202.5 ± 28.8 mg/dL to 149.9 ± 17.3 mg/dL (P = .01) in Group A, and from 182.7 ± 23.9 mg/dL to 171.1 ± 39.2 mg/dL in Group B (NS). Mean SBP in Group A decreased from 129.9 ± 3.1 mm Hg to 120.7 ± 1.7 mm Hg (P = .02), but DBP did not change significantly in either group (from 76.6 ± 1.4 mm Hg to 73.9 ± 1.6 mm Hg in Group A, and from 76.3 ± 5.6 mm Hg to 81.4 ± 1.8 mm Hg in Group B). The declines in SBP were not related to decreases in BMI among either group. There was a significant decrease in high-sensitivity CRP concentrations in Group A, from 11.2 ± 2.7 mg/L to 7.4 ± 1.9 mg/L (P = .05), as well as a significant correlation between that reduction and baseline CRP concentrations (r = 0.68, P = .001). In addition, the relationship between reductions in CRP and A1C was significant in Group A (r = 0.66, P = .0006), and there was a trend toward such a correlation between reductions in TG and CRP concentrations (r = 0.42, P = .10). There was no correlation between decreases in CRP concentration and changes in body weight among patients in Group A. Within the first week of therapy, 4 individuals from Group B halted treatment after suffering severe nausea and/or vomiting, but none from Group A withdrew for such reasons. In contrast, hypoglycemia, defined as an ambient BG level < 60 mg/dL, was rare, did not result in any withdrawals, and was avoided by reducing fast-acting or mixed insulin dosages by 10% among patients with initial A1C levels < 7.5%.

This study demonstrated that exenatide can provide improvements in glucose homeostasis and body weight, and can also help to reduce the doses of insulin required by obese patients with T2DM. Such declines in required insulin dosage, especially among patients treated with rapid-acting or mixed insulins, can help to reduce their risk of hypoglycemia. Hyperglycemia can be difficult to control among individuals with T2DM who are obese, even if they are already receiving large doses of oral agents and insulin. The potential benefits of adding exenatide to the treatment regimen of such patients appear to deserve further investigation, including assessment of its potential to provide anti-inflammatory effects even among patients already treated with TZDs, metformin, and insulin. The decreases in TG and CRP concentrations observed in this study also justify use of exenatide. The results obtained during this trial emphasize the need for prospective, controlled studies designed to further investigate the potential benefits of treatment with exenatide in conjunction with insulin among obese patients with T2DM.