tion therapy with metformin and a thiazolidinedione, the addition of glimepiride improved glycemic control compared with placebo with an acceptable tolerabil-.
inical Therapeutics/Volume 27, Number 10, 2005
Triple Therapy with Glimepiride in Patients with Type 2 Diabetes MeUitus Inadequately Controlled by Metformin and a Thiazolidinedione: Results of a 30-Week, Randomized, Double-Blind, Placebo-Controlled, Parallel-Group Study Victor Lawrence Roberts, MD, MBA, CCTI1;John Stewart, MSc2; Maher Issa, Barbara Lake, MSN, ARNP, CDE, CCRC1; and Robert Melis, MD3
MSc2;
1private practice, Winterpark, Florida; 2sanofi-aventis, Lava/, Quebec, Canada; and 3sanofi-aventis, Bridgewater, NewJersey ABSTRACT Objective: This study evaluated the efficacy and tolerability of glimepiride in patients with type 2 diabetes mellitus that was inadequately controlled with a combination of immediate- or extended-release metformin and a thiazolidinedione. Methods: This was a multicenter, randomized, doubleblind, placebo-controlled, parallel-group, 2-arm study consisting of a 4-week stabilization and eligibility period and a 26-week treatment period. Patients with a diagnosis of type 2 diabetes for a minimum of 1 year received glimepiride (titrated sequentially from 2 to 4 to 8 mg/d over 6 weeks, followed by 20 weeks of maintenance therapy) or placebo in combination with an established regimen of immediate- or extendedrelease metformin and rosiglitazone or pioglitazone. The primary efficacy outcome was the change in glycosylated hemoglobin (HbAlc) from baseline. The safety analysis was based on the incidence of hypoglycemia, adverse events, and laboratory abnormalities. Changes in lipid levels (high-density lipoprotein cholesterol, total cholesterol, low-density lipoprotein cholesterol, very low density lipoprotein cholesterol, and triglycerides) were evaluated, and health-related quality of life was assessed based on scores on the Diabetes Care Profile (DCP) and Health Utilities Index Mark 3 (HUI3). Results: Of 170 randomized patients, 159 were included in the efficacy analysis and 168 were included in the safety analysis. Demographic variables were similar at baseline between the glimepiride and placebo groups (mean age, 56.5 and 56.4 years, respectively; percent men/women, 61.0%/39.0% and 62.3%/ 37.7%; weight, 100.9 and 96.3 kg). HbAlc was significantly improved at end point with glimepiride combi-
nation therapy compared with placebo (mean [SE], -1.31% [0.08] vs -0.33% [0.08], respectively; P < 0.001). The majority of patients (62.2%) who received glimepiride achieved an HbA k value of ~7%, compared with 26.0% of patients receiving placebo (P < 0.001 between groups). At end point, the adjusted mean differences between treatments significantly favored the glimepiride combination in terms of fasting plasma glucose (-37.4 [4.0] mg/dL; P < 0.001), fasting insulin (4.06 [1.69] pIU/mL; P < 0.03), and C-peptide (124.5 [35.9] pmol/L; P < 0.001). The adjusted mean changes in body mass index from baseline to end point were 1.26 (0.16) kg/m2 with glimepiride and 0.17 (0.16) kg/m2 with placebo (P < 0.001). Similarly, the mean change in weight was greater with glimepiride than with placebo (3.76 [0.54] vs 0.45 [0.52] kg; P < 0.001). There were no significant differences in lipid levels between groups. Clinically significant adverse events, laboratory abnormalities, and rates of severe hypoglycemia were similar between treatment groups. The overall incidence of hypoglycemia, however, was 51.2% in the glimepiride group and 8.3% in the placebo group (P < 0.001). In general, there was no significant difference between treatment groups with respect to scores on the DCP or HUI3 over the study period. Conclusions: In these patients with type 2 diabetes that was not adequately controlled by dual combination therapy with metformin and a thiazolidinedione, the addition of glimepiride improved glycemic control compared with placebo with an acceptable tolerabilAccepted fiorpublicationJu~ 29, 2005. doi:l 0.1016/j.clinthera.2005.10.017 0149-2918/05/$19.00 Printed in the USA. Reproduction in whole or part is not permitted. Copyright © 2005 Excerpta Medica, Inc.
i
::~
i
ity profile. Although there were significantly more episodes of hypoglycemia with triple therapy than with dual therapy and placebo, the risk for severe hypoglycemia was low. (Clin Tber. 2005;27:1535-1547) Copyright © 2005 Excerpta Medica Inc. Key words: diabetes, HbAlc , sulfonylurea, thiazolidinedione, biguanide.
INTRODUCTION Type 2 diabetes mellitus is a chronic, progressive metabolic disease characterized by insulin resistance, reduced insulin secretion, decreased glucose uptake, and increased hepatic glucose output. Oral antidiabetic drugs (OADs) targeting each of these defects are currently used for the safe and effective management of type 2 diabetes and to reduce the risk of diabetesrelated complications. 1 The United Kingdom Prospective Diabetes Study showed that intensive glucose control with a sulfonylurea or insulin maintained lower long-term glycosylated hemoglobin (HbAlc) values (median, 7.0%) than did conventional therapy (median, 7.9%), which was defined as dietary modification, with the addition of drugs only if hyperglycemic symptoms occurred or the fasting plasma glucose (FPG) concentration was >270 mg/dL. 2 Only 50% of patients taking a single oral agent will have target HbAlc levels after 3 years of therapy. 3 Combination therapy with oral agents is initiated when single agents are no longer effective.4 In addition to improving glycemic control, combination therapy may reduce the risk of adverse effects because it is administered at lower doses than monotherapy, s-7 The availability of newer classes of OADs improves options for the treatment of type 2 diabetes. 4 Starting OAD therapy with a sulfonylurea is an efficacious and cost-effective approach. 2,8-1° Sulfonylureas are insulin secretagogues that promote the release of insulin from functional 13-cells to increase the metabolism of glucose; this is currently the most widely used class of OADs. 11 Other OADs, used alone or in combination with sulfonylureas, target insulin resistance. Metformin is a biguanide that acts primarily by reducing hepatic gluconeogenesis. The thiazolidinediones rosiglitazone and pioglitazone enhance insulin activity mainly by promoting glucose utilization in peripheral tissues and reducing the incidence of hyperglycemia without causing hyperinsulinemia or hypertriglyceridemia.12,73 Combining metformin with a thiazolidinedione focus1536
'==:=: =:===:=:
: :: =:= : 7. =,:i,= = ;
':7:: : :
::::::::::::: : : : :
.... :,,::, :::::=;i:~ , :
~
"
7
es the effects of 2 different drug classes on a single pathophysiologic target--insulin resistance. This approach retains the option for adding a sulfonylurea, which enhances the insulin secretory pathway. Some sulfonylureas may be associated with hyperinsulinemia, hypoglycemia, and weight gain11,14; when these conditions occur, a 3-drug regimen may be difficult to tolerate. Glimepiride, on the other hand, is a second-generation sulfonylurea that provides good metabolic control with once-daily dosing, an acceptable safety profile with low risk of hypoglycemia, and lipid and weight neutrality, ls,16 Glimepiride has been shown to normalize glucose levels, measured by postprandial glucose, HbAlc , and FPG. 1r-z° Using a sulfonylurea such as glimepiride or glyburide as part of triple therapy in combination with oral agents having different mechanisms of action, such as metformin and a thiazolidinedione, is a well-tolerated and efficacious treatment option. 21-23 The objective of the present study was to assess the efficacy and safety profile of glimepiride in patients with type 2 diabetes inadequately controlled with the combination of immediate- or extended-release metformin and a thiazolidinedione (rosiglitazone or pioglitazone). PATIENTS Inclusion
AND and
METHODS
Exclusion
Criteria
M e n and w o m e n aged 18 to 80 years w i t h a m i n i m u m 1-year h i s t o r y of type 2 diabetes w h o had been
performing self-monitored blood glucose (SMBG) testing at home were eligible for participation. Patients had to be receiving a stable regimen of immediaterelease metformin (1.0-2.5 g/d) or extended-release metformin (up to 2 g/d) and a half-maximum to maximum dose of a thiazolidinedione (up to 8 mg/d for rosiglitazone, 45 mg/d for pioglitazone) for at least 3 months before study entry. At screening, HbAlc values had to be between 7.5% and 9.5%, and body mass index (BMI) had to be between 26 and 42 kg/m 2. In addition, patients were required to have insulin secretory capacity (defined as a fasting C-peptide concentration ->0.27 nmol/L) during the stabilization phase and an FPG level between 130 and 235 mg/dL within 42 to 72 hours of randomization. Concomitant medications--excluding sulfonylureas, insulin, acarbose, repaglinide, or continuous chronic systemic steroids--were permitted for coexisting conditions. Patients who required insulin therapy, were taking other sulfonylureas, or had a history of sulfonylurea :
:. . . . .
:::: :
= Volume
, : ,
27 Number10
hypersensitivity were excluded. Additional exclusions were a history of severe hypoglycemia while taking OAD therapy, acute metabolic complications, an increase in the thiazolidinedione dose during the previous 2 months, an increase in the metformin dose during the previous month, or clinically significant abnormalities in laboratory values at baseline.
Study Design This was a multicenter, randomized, double-blind, placebo-controlled, parallel-group study. The study protocol, informed-consent form, and other studyrelated documents were reviewed and approved by an independent ethics committee/institutional review board before initiation of the study. Two protocol amendments relating to the conduct of the study (ie, inclusion/exclusion criteria, screening procedures, duration of the study, number of study visits, study discontinuations) were also reviewed and approved. All patients gave written informed consent. The study included an initial 4-week stabilization and eligibility period during which patients continued to take fixed doses of their current OAD therapy and a 26-week treatment period that consisted of a 6-week titration phase and a 20-week maintenance phase. Patients were randomized in blocks of 2 to undergo forced titration of either glimepiride 2 mg or placebo as add-on therapy. At biweekly visits, the dose of glimepiride or placebo was titrated in 2 steps (to 4 and then 8 mg/d) as needed until the therapeutic goal of serum glucose 1 patient. None of the serious AEs were considered related to study medication and none resulted in death. Seven patients in each treatment group discontinued study medication as the result of an AE. Six of the AEs leading to discontinuation in the glimepiride group and 3 of the AEs leading to discontinuation in
::
Volume 27 N.mbor 10
V.L. Roberts e t al.
Table II. Changes in glycemic parameters from baseline to end point (week 26). Adjusted Mean (SE)Value Parameter
Difference Between Treatments
Glimepiride
Placebo
Mean (95% CI)
P*
Glycosylated hemoglobin, % Baseline Change from baseline
8.14 (0.08) -1.31 (0.08)
8.16 (0.08) -0.33 (0.08)
-0.98 (-1.20 to -0.76)
6 patients in either treatment group) adverse events (AEs). Glimepiride Placebo (n = 84) (n = 84) All AEs
67 (79.8)
60 (71.4)
Specific AEs Hypoglycemia Gastrointestinal disorders Upper respiratory infection Peripheral edema Headache Weight increase
43 (51.2)* 7 (8.3) 13 (15.5) 11 (13.1) 8 (9.5) 9 (10.7) 7 (8.3) 2 (2.4) 6 (7.1) 4 (4.8) 6 (7.1) 0 (0)
compared with placebo recipients. None of these changes led to serious AEs or to treatment discontinuation. There were no clinically noteworthy differences between treatment groups in any other laboratory parameter.
HRQoL DCP scale scores and HUI3 total utility scores were similar across assessments within treatment groups and were not significantly different between treatment groups. Mean scores on the DCP scales for support attitudes and importance of care indicated good HRQoL relative to the other DCP scales. Mean scores for individual attributes of the HUI3 indicated good HRQoL for both treatment groups at all assessments (Table V).
*P < 0.001.
DISCUSSION the placebo group were considered possibly related to study medication. Reasons for discontinuation in the glimepiride group were hyperglycemia (1), hypoglycemia (1), weight increase (1), hypoglycemia and weight increase (1), peripheral edema (1), fluid retention (1), and MI (1). Reasons for discontinuation in the placebo group were hyperglycemia (2), prostate cancer (1), MI (1), cerebral artery occlusion (1), and peptic ulcer (1). One patient in the placebo group discontinued due to multiple AEs, including diarrhea, headache, tremor, irritability, and dyspnea. There was an expected increase in the rate of hypoglycemia associated with the addition of the secretagogue glimepiride. The incidence of hypoglycemia was significantly greater in the glimepiride group than in the placebo group (43 [51.2%] vs 7 [8.3%], respectively; P < 0.001) (Table IV). Only 1 (1.2%) patient in the glimepiride group had a severe hypoglycemic episode, which was readily reversed by the administration of oral carbohydrates. The incidence of hypoglycemia did not vary by the thiazolidinedione used. There were no clinically significant changes in vital signs or hematologic parameters associated with glimepiride treatment and no changes in electrocardiographic findings. Small increases in levels of uric acid and creatine kinase from baseline to weeks 10, 18, and 26 were seen in the glimepiride group, compared with mean decreases at the same time points in the placebo group. A higher proportion of glimepiride recipients, however, had a weight increase of >5%
In this study, the addition of glimepiride to the combination of immediate- or extended-release metformin plus a thiazolidinedione resulted in improved glycemic control in patients in whom type 2 diabetes was not controlled with a 2-drug regimen. Significantly greater improvements were noted with glimepiride compared with placebo in HbA k values and FPG levels at all time points (all, P < 0.001) and in the proportion of patients achieving an HbAlc value
