a randomized, double-blind, placebo-controlled, clinical ...hera.ugr.es/doi/14977527.pdf ·...

A Randomized, Double-Blind, Placebo-Controlled, Clinical Trial of the Effects of Pioglitazone on Glycemic Control and Dyslipidemia in Oral Antihyperglycemic Medication-Naive Patients with Type 2 Diabetes Mellitus

Matthias Herz, MD, 1 Don Johns, PhD, ~ Jesus Reviriego, MD, PhD, 1 Loren D. Grossman, MD, 1 Chantal Godin, MD, 2 Santiago Duran, MD, PhD, 3 Federico Hawkins, MD, PhD, 4 Heather Lochnan, MD, 5 Fernando Escobar-Jim~nez, MD, PhD, 6 Philip A. Hardin, MD, 7 Christopher S. Konkoy, PhD, l and Meng H. Tan, MD, 1 for the GLAB Study Group* 1Lilly Research Laboratories, Eli Lilly and Company, Lilly Corporate Center, Indianapolis, Indiana, 2Centre de Recherche d'Endocrinologie, Sherbooke, Quebec, Canada, 3Hospital Universitario Nuestra Senora de Valme, Sevilla, 4Hospital 12 de Octubre, Madrid, Spain, 5Ottawa Civic Hospital, Ottawa, Ontario, Canada, 6Hospital Cl*nico Universitario San Cecilio, Granada, Spain, and 7West Edmonton Diabetes Center, Edmonton, Alberta, Canada

ABSTRACT

Objective: The goal of this study was to compare the effects of 2 doses of pioglitazone hydrochloride (a thiazolidinedione insulin sensitizer) with placebo on glycated hemoglobin (HbAlc), insulin sensitivity, and lipid profiles in patients with type 2 diabetes mellitus who had suboptimal glycemic control and mild dyslipidemia.

Methods: Patients with type 2 diabetes mellitus (HbAlc >6.5% and <9.8%) who had not previously received insulin or oral antihyperglycemic medications (OAMs) were randomized to treatment with placebo, piogfitazone 30 mg QD, or pioglitazone 45 mg QD in double-blind fashion for 16 weeks at 41 centers in Canada and Spain.

This work was presented in part at the 62nd Annual Meeting and Scientific Sessions of the American Diabetes Association, June 14-18, 2002, San Francisco, California (published as abstracts in Diabetes. 2002;51[Suppl 21: 414P, 1976PO, 2042PO); and at the Canadian Diabetes Association Meeting, October 3-5, 2002, Vancouver, Canada (published as an abstract in Can J Diabetes. 2002;26[Suppl 1]:A222). *Members of the GLAB Study Group are listed in the Acknowledgments.

Accepted Jor publication February 13, 2003. Printed in the USA. Reproduction in whole or part is not permitted. 0149-2918/03/$ L9.00

] 0 7 4 Copyright © 2003 Excerpta bledica, Inc

H. Her'z et al.

Results: A total of 297 patients were randomized (99 in each group). Overall, 286 (96.3%) were white. Mean (SD) age was 58.4 (10.9) years (range, 24-85 years), mean (SD) body mass index was 31.4 (4.8) kg/m 2, mean (SD) duration of type 2 diabetes mellitus was 20.0 (37.4) months, and 30.6% of patients were receiving medication for dyslipidemia. Treatment with pioglitazone 30 or 45 mg QD for 16 weeks reduced mean HbAlc by 0.8% and 0.9% from baseline, respectively (both P < 0.001 vs baseline and placebo). A reduction in HbAlc of 0.2% was observed in the placebo group (P = 0.025). In patients with medium (>7% to <8%) or high (>8% to <9.8%) baseline HbAlc, both doses of pioglitazone significantly reduced HbAlc (both P < 0.001 vs placebo). Pioglitazone 30 and 45 mg significantly reduced fasting serum insulin versus placebo (P = 0.008 and P = 0.006, respectively) and increased insulin sensitivity by Homeostasis Model Assessment versus placebo (P = 0.039 and P = 0.001, respectively). Relative to placebo, pioglitazone 30 and 45 mg significantly increased high-density lipoprotein cholesterol (HDL-C [P = 0.028 and P < 0.001, respectively]) and lowered the atherogenic index of plasma (P = 0.018 and P < 0,001, respectively). Pioglitazone 45 mg also significantly reduced serum triglycerides, apolipoprotein B, and total cholesterol:HDL-C ratio versus placebo (P = 0,007, P = 0.015, and P = 0.005, respectively). Pioglitazone 30 and 45 mg were associated with a significant reduction in serum alanine aminotransferase relative to placebo (P = 0.036 and P = 0.005, respectively). Pioglitazone appeared to be safe and was well tolerated.

Conclusions: In the present study, pioglitazone 30 and 45 mg produced significant improvements in HbAlc, insulin sensitivity, and lipid profile in OAM- naive patients with type 2 diabetes mellitus with suboptimal glycemic control and mild dyslipidemia. (Clin Ther. 2003;25:1074-1095) Copyright © 2003 Excerpta Medica, Inc.

Key words: pioglitazone, type 2 diabetes mellitus, hyperglycemia, lipoprotein, insulin resistance, monotherapy.

INTRODUCTION Type 2 diabetes mellitus carries risks for multiple complications, including mi- crovascu]ar (eg, retinopathy, nephropathy, neuropathy) and macrovascular disease (eg, coronary heart disease, stroke, peripheral vascular disease). 1-3 Maintaining adequate glycemic control is an important therapeutic goal for lowering the risk of microvascular complications in patients with type 2 diabetes mellitus.1 An increased risk of cardiovascular disease often predates the onset of type 2 diabetes mellitus, 4 suggesting that reduction of cardiovascular risk may require an early and aggressive treatment strategy to promote lipid homeostasis (ie, reducing

1075

CLINICAL THERAPEUTICS ~

triglycerides [TGs] and increasing high-density lipoprotein cholesterol [HDL-C]) and improve glycemic control.

Insulin resistance with compensatory hyperinsulinemia is a defining feature of type 2 diabetes mellitus and an independent risk factor for coronary heart disease. 5,6 Pharmacotherapy that increases insulin sensitivity has been shown to improve both glycemic and lipid profiles] Pioglitazone, a thiazolidinedione insulin sensitizer, has been shown in clinical studies to reduce glycated hemoglobin (HbAI,.), TGs, and free fatty acids while enhancing insulin sensitivity and elevat- ing HDL-C. 8-1° Moreover, a recent study 11 has shown that pioglitazone reduces the atherogenic index of plasma (AIP), 12 an indicator of risk for cardiovascular disease. Thiazolidinediones are thought to produce beneficial effects on glucose and lipid levels through activation of peroxisome proliferator-activated receptor gamma, which is involved in transcriptional regulation of genes related to glucose homeostasis and lipid metabolism. 7

The goal of the present study was to compare pioglitazone 30 and 45 mg with placebo in controlling hyperglycemia and dyslipidemia in patients with recently diagnosed type 2 diabetes mellitus who had not been treated with insulin or oral antihyperglycemic medications (OAMs). The patient population under study had suboptimal glycemic control and mild dyslipidemia at baseline. In the present study, we also analyzed response rates to American Diabetes Association (ADA) criteria for glycemic control (HbAlc) and lipid levels (HDL-C and TGs). 13 Our findings add to the results of previous clinical trials of pioglitazone monotherapy in which pioglitazone improved fasting plasma glucose (EPG), HbAk, insulin sensitivity, and lipid profiles in patients with type 2 diabetes mellitus of longer duration, less satisfactory glycemic control, and more severe dyslipidemia. >,15

PATIENTS A N D M E T H O D S Study Design

This was a randomized, double-blind, placebo-controlled, parallel-group study involving patients at 41 centers in Canada and Spain. The study consisted of 2 periods: a lead-in period in which all patients received placebo QD for 3 to 5 weeks, and a treatment period in which eligible patients were randomized to receive placebo, pioglitazone 30 mg QD, or pioglitazone 45 mg QD orally for 16 weeks. Patients were asked to maintain a consistent diet and exercise regimen throughout the study. Patients recorded their self-monitored blood glucose (SMBG) levels, hypoglycemic episodes, adverse events, and concomitant medications in a study diary. Each patient provided written informed consent for participation in the study. Eli Lilly and Company (Indianapolis, Indiana) sponsored the study and performed site monitoring and data collection and analysis.

Study conduct was approved by the institutional review board at each center, and the study was conducted in compliance with Good Clinical Practices and in

1076

M. Herz et al.

accordance with the principles of the Declaration of Helsinki as amended in 1996.

Inclusion a n d Exclusion Crt ter /a

There were 2 major inclusion criteria: (1) diagnosis of type 2 diabetes mellitus that was not controlled by diet and exercise and (2) no previous treatment with insulin or OAMs. At screening, HbAlc values for eligible patients were _>6.5% and <9.8%, generally indicative of mild to moderate hyperglycemia in type 2 diabetes mellitus.

Patients were excluded from the study for the following reasons: cardiac disease with marked limitation of functional capacity (New York Heart Association class III or IV clinical status); serum TGs >500 mg/dL (5.6 mmol/L) or total cholesterol (TC) >300 mg/dL (7.8 mmol/L); serum creatinine _>1.8 mg/dL (0.16 mmol/L); renal transplant or current renal dialysis; serum alanine aminotransferase (ALT) or aspartate aminotransferase (AST) >2.5 times the upper limit of normal for the central laboratory; clinical signs or symptoms of liver disease; hemoglobin or hematocrit below the lower limit of normal for the central laboratory; previous HIV infection; treatment with systemic glucocorticoids (excluding topical and inhaled preparations) within the previous 4 weeks; body mass index (BMI) <25 kg/m2; signs or symptoms of substance abuse; or life expectancy <3 years.

Patients were allowed to use concomitant medications as required, except for insulin, other OAMs, systemic glucocorticoids (excluding topical and inhaled preparations), or nicotinic acid. Patients receiving lipoprotein-altering agents be- fore enrollment were required to remain on the same dose for the duration of the study.

Laboratory and Safety Measurements A central laboratory (Covance Laboratories, Indianapolis, Indiana) was used for

analyses of all laboratory samples. FPG was measured using the hexokinase enzymatic method on Hitachi 747-200 chemistry analyzers (Roche Diagnostics, In- dianapolis). HbAlc levels were measured from whole-blood samples by ion- exchange high-performance liquid chromatography (Bio-Rad Variant analyzer, Bio-Rad Laboratories, Hercules, California). This system is certified by the Na- tional Glycohemoglobin Standardization Program to be traceable to the Diabetes Control and Complications Trial 16 reference method and values. The limits of de- tection were 3.6% to 17.8% (inclusive). The upper limit of the normal range was 6.1%. Intra-assay and interassay coefficients of correlation were 1.2% and 1.6%, respectively. Fasting serum insulin (FSI) levels were measured with the Mi- croparticle Enzyme Immunoassay technique (Abbott Laboratories, Abbott Park, Illinois). The lipoprotein panel was performed with serum samples in accordance

1077

CLINICAL THERAPEUTICS <"~

with the National Heart, Lung, and Blood Institute Lipid Standardi7ation Pro- gram. TC and TGs were measured photometrically using Hitachi analyzers. HDL-C was measured by dextran sulfate precipitation followed by assay of cholesterol on Hitachi analyzers. Low-density lipoprotein cholesterol (LDL-C) was measured directly by micellary solubilization followed by enzymatic assay on Hi- tachi analyzers. Assay of serum apolipoprotein B (apo B) was performed with the IMMAGE Immunochemistry System (Beckman Coulter, Inc., Palo Alto, Califor- nia). The Homeostasis Model Assessment of Insulin Sensitivity (HOMA-S) was also used. Values of HOMA-S were derived from fasting insulin and FPG using a computer program from the Diabetes Research Laboratories (Oxford, United Kingdom). lr AlP was calculated as logl0(TGs/HDL-C), as previously described.11 Urine creatinine was measured using a Hitachi 911 analyzer.

Safety assessments consisted of adverse-event reporting as follows: hypoglycemic episodes; vital sign measurements, including changes in body weight; and laboratory abnormalities, with particular attention to hemoglobin, hematocrit, and serum ALl and AST. Hypoglycemic episodes were defined by either of the following: (1) a sign or symptom of hypoglycemia recorded in the patient diary, or (2) an SMBG of <50 mg/dL, regardless of the presence of hypoglycemic signs or symptoms. Adverse events were coded using a modified MedDRA dictionary.

Statistical Analysis Descriptive statistics were used to summarize baseline characteristics for all pa-

tients. The primary time point for efficacy analyses was the study end point, defined as the last double-blind visit at which data were collected. For patients who completed the study, the end point took place after 16 weeks of treatment with study medication. Unless otherwise indicated, data are reported for the intent-to- treat (ITT) population--all randomized patients who received _>1 dose of study medication and had both a baseline measurement and _>1 measurement of the dependent variable during the treatment period. Data describing least squares mean (LSM) changes from baseline to end point for HbAlc and HOMA-S are expressed as percent absolute units. Continuous efficacy variables were assessed by a fixed- effects analysis of covariance (ANCOVA), performed as a last observation carried forward analysis in which missing values for postbaseline measurements were im- puted from the previous nonmissing postbaseline measurement for that variable. The ANCOVA model included treatment, investigator, and baseline value of the dependent variable as covariates. The overall test for treatment effect was performed with a significance level of P _< 0.05. Pairwise t tests between pioglitazone and placebo were calculated from LSM values derived from the model and were adjusted for multiplicity using Hochberg's method. 18 Differences among treatments for additional categorical variables were analyzed using chi-square tests.

1078

M. Herz et al.

We conducted 3 exploratory analyses in this study. First, treatment effects on HbAlc values were measured in patients stratified by baseline HbAlc as follows: low (_>6.5% to <7%); medium (_>7% to <8%); and high (_>8% to _<9.8%). Second, the proportion of patients achieving ADA target goals was assessed within these baseline strata. Third, we measured the proportion of patients meeting ADA target goals for TGs and HDL-C at baseline and end point. The overall level of significance of P _< 0.05 was maintained within each analysis using Hochberg's adjustment for multiplicity. 18

RESULTS Patient Characteristics

A total of 297 patients were randomized to receive placebo, pioglitazone 30 mg QD, or pioglitazone 45 mg QD (n = 99 in each group). No significant differences in baseline characteristics were found between the 3 treatment groups (Table I). The majority of patients were white, with a mean (SD) age of 58.4 (10.9) years (range, 24-85 years) and mean (SD) BMI of 31.4 (4.8) kg/m 2. The mean (SD)

Table I. Baseline characteristics of patients with type 2 diabetes with suboptimal glycemic control and mild dyslipidemia randomized to 16 weeks of therapy with placebo (n = 99), pioglitazone 30 mg (n = 99), or pioglitazone 45 mg (n = 99). No significant differences were found between groups for any variable at baseline.

Characteristic Placebo Pioglitazone 30 mg Pioglitazone 45 mg

Sex, no. (%) Men 49 (49.5) 59 (59.6) 52 (52.5) Women 50 (50.5) 40 (40.4) 47 (47.5)

Age, y Mean (SD) 58.0 (10.7) 59.0 (I 1.0) 58.1 (I 1.0) Range 33-85 24-79 24-84

Race, no. (%) White 96 (97.0) 97 (98.0) 93 (93.9) Asian 3 (3.0) I (I.0) 3 (3.0) Hispanic 0 (0.0) I (I.0) 3 (3.0)

Body weight, mean (SD), kg 86.3 (I 7.4) 86.6 (I 5.9) 84. I (I 6.8)

Body mass index, mean (SD), kg/m 2 31.7 (4.5) 31.7 (4.6) 30.8 (5. I)

Duration of type 2 diabetes mellitus, mean (SD), mo 17.4 (29.4) 22.5 (38.3) 20.1 (43.4)

Patients taking lipoprotein-altering drugs, no. (%) 31 (31.3) 29 (29.3) 31 (31.3)

1079

CLINICAL THERAPEUTICS ~!

duration of type 2 diabetes mellitus was 20.0 (37.4) months, and 30.6% of patients were receiving medication for dyslipidemia.

Of the 297 patients randomized to receive treatment, 272 (91.6%) completed the study. A similar proportion of patients in each treatment group completed the study (88, 92, and 92 patients in the placebo, pioglitazone 30-mg, and pioglitazone 45-mg groups, respectively). In the placebo, pioglitazone 30-mg, and pioglitazone 45-mg groups, respectively, 5, 1, and 0 patients discontinued therapy due to lack of efficacy (as perceived by the physician and/or the patient); 1, 2, and 2 discontinued due to adverse events; 0, 4, and 1 were lost to follow-up; 1, 0, and 2 discontinued due to patient decision; 1, 0, and 1 did not meet the protocol entry criteria; 2, 0, and 0 violated the protocol; 0, 0, and 1 were discontinued due to the sponsor's decision; and 1, 0, and 0 were discontinued due to the physician's decision.

Glycemic Control and Insulin Sensitivity As shown in Figure 1, baseline mean HbA]~ values were 7.5%, 7.5%, and 7.6%

for placebo, pioglitazone 30 mg, and pioglitazone 45 mg, respectively. HbAlc was

8.0

7.5

v 7.0- .O -i-

6.5-

6.0

T

[ ] Placebo (n = 96) [ ] Pioglitazone 30 mg (n = 95) [ ] Piogiitazone 45 mg (n = 96)

' I - - ] Baseline End Point

Figure I. Glycated hemoglobin (HbAic) at baseline and after 16 weeks of treatment with placebo, pioglitazone 30 mg, or pioglitazone 45 mg among patients with type 2 diabetes mellitus with suboptimal glycemic control and mild dyslipidemia. Data are expressed as mean (SE). *P = 0.025 versus baseline; tp < 0.001 versus baseline; *P < 0.001 versus placebo.

1080

M. Herz et al.

reduced in the placebo group by 0.2% from baseline (P = 0.025). Treatment with pioglitazone 30 and 45 mg for 16 weeks reduced mean HbAlc by 0.8% and 0.9% from baseline, respectively (both P < 0.001 vs baseline and placebo).

Subgroup analysis was conducted to examine the effect of pioglitazone in patients stratified by baseline HbAlc. Both doses of pioglitazone significantly reduced HbAlc levels at end point in patients with medium (>7% to <8%) or high (_>8% to _<9.8%) baseline HbAlc (both doses P < 0.001 vs placebo for both HbAlc categories; Figure 2). Mean reductions in HbA k for both doses were similar within these subgroups, ranging from 0.7% (medium, both doses) to 1.4% (high, 45-mg dose). In patients with low baseline HbAlc (>6.5% to <7%), pioglitazone 30 and 45 mg reduced end point values by 0.3% and 0.4%, respectively; HbAlc did not change in those patients who received placebo. However, in patients with low baseline HbAlc, changes in HbAlc did not differ significantly between the placebo group and the pioglitazone groups.

o.o

-0.2

-0.4

v -0.6 , j

I

< -0.8

I E "- -l.O-

c

-1.2 - f-

u

-IA-

-1.6 -

-I .8 -

Baseline HbAic

Low Medium High

• Placebo [ ] Pioglitazone 30 mg [ ] Pioglitazone 45 rng

~c

Figure 2. Reduction in glycated hemoglobin (HbAic) from baseline to the end of 16 weeks of treatment among patients with type 2 diabetes mellitus with subop-

timal glycemic control and mild dyslipidemia, stratified by baseline HbA~c: low, >6.5% to <7% (placebo, n = 29; pioglitazone 30 mg, n = 33; pioglitazone 40 mg, n = 30); medium, >7% to <8% (placebo, n = 37; pioglitazone 30 mg, n = 36; pioglitazone 45 mg, n = 36); or high, >8% to <9.8% (placebo, n = 25; pioglitazone 30 mg, n = 25; pioglitazone 45 mg, n = 30). Data are expressed as least squares mean (SE). *P < 0.001 versus placebo (Hochberg adjustment).

1081


Table II shows the proportion of patients achie~qng the ADA target glycemic goal (HbAI~. <7%) at study end. In the IIT population~ 70.5% (67/95) and 68.8% (66/96) of patients receiving pioglitazone 30 and 45 mg achieved the ADA glycemic goal, respectively, compared with 42.7% (41/96) of patients receiving placebo (P < 0.001 and P = 0.001 vs placebo, respectively). Among individuals with medium baseline HbA~c, 78.4% of patients who received pioglitazone 30 mg (29/37) and 72.2% of those who received pioglitazone 45 mg (26/36) reached the ADA glycemic goal, compared with 35.-/% of patients who received placebo (15/42) (P = 0.002 and P = 0.01 vs placebo, respectively); among those with high baseline HbAl~, 28.0% of those who received pioglitazone 30 mg (7/25) and 40.0% of those who received pioglitazone 45 mg (12/30) reached the ADA goal, compared with 4.0% of those who received placebo (1/25) (P = NS and P = 0.03 vs placebo, respectively). Among individuals with low baseline HbA~c, 93.9% (31/33) and 93.3% (28/30) of those treated with pioglitazone 30 and 45 mg, respectively, and 86.2% of those who received placebo (25/29) reached the target goal (both pioglitazone groups P = NS vs placebo).

HOMA-S was used to estimate insulin sensitivity from measurements of FPG and FS[. Sixteen-week treatment with pioglitazone 30 and 45 mg significantly reduced FPG by 1.4 mmol/L (baseline, 8.9 mmol/L) and 1.6 mmol/L (baseline, 8.6 mmol/L), respectively (all changes P < 0.001 vs baseline and placebo). Placebo had no significant effect on FPG (reduction, 0.1 mmol/L; baseline, 8.8 mmol/L). As shown in Table III, pioglitazone 30 and 45 mg produced similar reductions in FSI relative to

Table II. Proportion of patients with type 2 diabetes with suboptimal glycemic control and mild dyslipidemia who achieved American Diabetes Association target glycated hemoglobin (HbAic) values (<7%) after 16 weeks of treatment, according to baseline disease severity.

T r e a t m e n t G r o u p

Placebo, Pioglitazone 30 mg, Pioglitazone 45 mg, HbA k Category n/N (%) n/N (%) n/N (%)

{-iT population 41/96 (42.7) 67/95 (70.5)" 66/96 (68.8)~ Low (>6.5% to <7%) 25/29 (86.2) 31/33 (93.9) 28/30 (93.3) Medium (>7% to <8%) 15/42 (35.7) 29/37 (78.4)~ 26/36 (72.2)§ High (_>8% to <9.8%) 1/25 (4.0) 7/25 (28.0) 12/30 (40.0)11

{Ti- = intent to treat.

*P < 0,001 versus placebo. tp = 0.00l versus ptacebo.

*P = 0.002 versus placebo,

§P = 0.01 versus placebo.

lip = 0.03 versus placebo.

1082

N. Herz et al.

Table III. Effect of 16 weeks of treatment on insulin sensitivity in patients with type 2 diabetes with suboptimal glycemic control and mild dyslipidemia.

Treatment Group

Variable Placebo Pioglitazone 30 mg Pioglitazone 45 mg

Fasting serum insulin* Baseline, mean (SE), pmol Change at end point, LSM (SE), pmol P versus baseline P versus placebo

HOMA-St Baseline, mean (SE), % 67.8 (4. I) Change at end point, LSM (SE), % 7.3 (4.2) P versus baseline NS P versus placebo

99.3 (5.4) 116.7 (I 0.2) 106.6 (9.4) -13.0 (4.5) -31.4 (4.9) -31.8 (4.8)

0.004 <0.001 <0.00 I 0.008 0.006

62.2 (4.0) 70.3 (4.4) 21.2 (4.6) 27.3 (4.5) <0.001 <0.00 I

0.039 0.001

LSM = least squares mean; HOMA-S = Homeostasis Model Assessment of Insulin Sensitivity) 7 *Placebo, n = 85; pioglitazone 30 rng, n = 75; pioglitazone 45 mg, n = 83. tPlacebo, n = 83: pioglitazone 30 rag, n = 75; pioglitazone 45 mg, n = 82.

baseline (31.4 pmol/L and 31.8 pmol/L, respectively); these changes were statistically significant compared with baseline (both P < 0.001) and placebo (P = 0.008 and P = 0.006, respectively). A reduction of 13.0 pmol/L was observed in the placebo group. In addition, pioglitazone 30 and 45 mg significantly increased HOMA-S values compared with baseline (both P < 0.001) and with placebo (P = 0.039 and P = 0.001, respectively); the LSM changes (minus the placebo values) were 13.9% and 20.0% for pioglitazone 30 and 45 mg, respectively.

Lipid Profile Serum lipids were measured in all study patients at baseline and after 16 weeks

of treatment with either placebo or pioglitazone (Figure 3). Mean baseline levels of HDL-C were 1.20 mmol/L, 1.14 mmol/L, and 1.13 mmol/L for the placebo, pioglitazone 30-mg, and pioglitazone 45-mg groups, respectively. Pioglitazone 30 and 45 mg significantly increased levels of HDL-C from baseline (16% and 20%, respectively; both P < 0.001); this increase was significantly greater than that of the placebo group, which had an increase of 9% in HDL-C levels (P = 0.028 and P < 0.001, respectively). Mean baseline TGs were 1.72 mmol/L, 1.91 mmol/L, and 1.99 mmol/L for placebo, pioglitazone 30 mg, and pioglitazone 45 mg, respectively Although both pioglitazone 30 and 45 mg reduced TGs (5% and 16%, respectively), only the 45-mg dose produced a significant effect versus baseline and placebo (P < 0.001 and P = 0.007, respectively). Pioglitazone 45 mg pro-

1083

CLINICAL THERAPEUTICS ~'

t* t§ • Placebo (n = 96)

[ ] Pioglitazone 30 mg (n = 95) [ ] Pioglitazone 45 mg (n = 96)

r- t~ ¢- u

*¶#

Figure 3.

tll Total HDL-C** LDL-C t* Triglycerides** Apo B *¢

Cholesterol** Lipid Parameter

Effect of 16 weeks of treatment on lipid parameters among patients with type

2 diabetes mellitus with suboptimal glycemic control and mild dyslipidemia.

Data are expressed as least squares mean (SE). HDL-C = high-density l ipopro-

tein cholesterol; LDL-C = low-densi ty l ipoprotein cholesterol; apo B =

apolipoprotein B. *P < 0.05 versus baseline; tp < 0.001 versus baseline; *P =

0.028 versus placebo; §P < 0.001 versus placebo; lip = 0.007 versus placebo;

¶P = 0.025 versus pioglitazone 30 mg; #P = 0.015 versus placebo; **placebo, n = 97, pioglitazone 30 mg, n = 95, pioglitazone 45 mg, n = 96; ttplacebo, n = 96,

pioglitazone 30 mg, n = 95, pioglitazone 45 mg, n = 96; **placebo, n = 94,

pioglitazone 30 mg, n = 90, pioglitazone 45 mg, n = 94.

duced a small but significant reduction in apo B (5%; P = 0.020 vs baseline, P = 0.015 vs placebo, and P = 0.025 vs pioglitazone 30 mg), whereas pioglitazone 30 mg had no effect on this level. Pioglitazone 45 mg had no effect on TC and LDL-C. Pioglitazone 30 mg increased TC and LDL-C over baseline by 4% and 7%, respectively (P = 0.033 and P = 0.020). However, these changes were not significant compared with those of placebo.

Overall, the effects of pioglitazone were similar in the subset of patients cur- rently receiving lipoprotein-altering medications (data not shown).

As shown in Figure 4, pioglitazone 30 and 45 mg significantly reduced AIP (both P < 0.001 vs baseline; P = 0.018 and P < 0.001 vs placebo, respectively)

1084

Figure 4.

M. Herz et al.

A

o_ m

0.25 -

0.20 -

0.15 -

0.10-

0.05 -

0.00 -

-0.05

-O- F~acebo (n = 97) I~oglitazone 30 mg (n = 95)

. - ~ Pioglitazone 45 mg (n = 96)

i 1 I

Baseline End Poin t

B

5.4-

O °_

o£

u, ..J £3 .z. u F-

5 . 2 -

5.0-

4.8-

4.6-

4.4-

4.2-

4.0-

3.8 I I Baseline End Point

Effect of 16 weeks of t reatment on (A) atherogenic index of plasma (ALP) and

(B) rat io of total cholesterol (TC) to high-density l ipoprotein cholesterol

(HDL-C) among patients with type 2 diabetes mellitus with suboptimal glycemic

control and mild dyslipidemia.AIP was calculated as IOgl0(triglycerides/HDL-C ). Data are expressed as mean (SE). *P < 0.001 versus baseline; tp = 0.018 ver-

sus placebo; *P < 0.001 versus placebo; §P = 0.005 versus placebo.

1085

CLINICAL THERAPEUTICS:"

and -FC:HDL-C ratio (both P < 0.001 vs baseline). In the 45-rag group, the decrease in TC:HDL-C ratio was significant compared with placebo (P = 0.005).

Table IV shows the proportion of patients meeting ADA target goals for TGs and HDL-C at baseline and end point. In the ITT population, the proportion of patients reaching the target goal for TGs (<1.7 mmol/L) increased from 52.6% (50195) to 60.0% (57195) in the pioglitazone 30-mg group and from 52.1% (50196) to 70.8% (68196) in the pioglitazone 45-mg group, compared with a change from 55.7% (54/97) to 56.7% (55/97) in the placebo group. (Statistical significance was not assessed.) In a subanalysis of patients with elevated TGs at baseline (>1.7 mmol/L), 47.8% (22/46) achieved the ADA target for TG at end point after receiving pioglitazone 45 mg, compared with 27.9% (12/43) receiving placebo. The proportion of patients who met the ADA target for HDL-C (>1.15 and >1.4 mmol/L for males and females, respectively) increased from 36.1% (35197) to 45.4% (44/97) in the placebo group, from 23.2% (22/95) to 52.6% (50195) in the pioglitazone 30-mg group, and from 28.1% (27196) to 56.3% (54/96) in the pioglitazone 45-mg group. (Statistical significance was not assessed.)

Safety Profile Although the incidence of treatment-emergent adverse events was higher in pa-

tients treated with pioglitazone 45 mg compared with either pioglitazone 30 mg or placebo (79.8% [79/99], 63.6% [63/99], and 68.7% [68/99], respectively), nei-

Table IV. Proportion of patients with type 2 diabetes with suboptimal glycemic control and mild dyslipidemia (intent-to-treat population) who met American Diabetes Association goals 13 for triglycerides (TGs) (<1.7 mmol/L) and high-density lipoprotein cholesterol (HDL-C) (men, >1.15 mmol/L; women, > 1.40 mmol/L) at baseline and after 16 weeks of treatment.*

Variable

Treatment Group

Placebo, Pioglitazone 30 mg, Pioglitazone 45 mg, n/N (%) n/N (%) n/N (%)

HDL-C Baseline 35/97 (36. I) 22/95 (23.2) 27/96 (28. I) End point 44/97 (45.4) 50/95 (52.6) 54/96 (56.3)

TGs Baseline 54/97 (55.7) 50/95 (52.6) 50/96 (52.1) End point 55/97 (56.7) 57/95 (60.0) 68/96 (70.8)

"Statistical significance not assessed.

1086

M. Herz et al.

ther pioglitazone treatment group was statistically significantly different from placebo. The most frequently reported treatment-emergent adverse events (total incidence _>5%) were edema, nasopharyngitis, dizziness, back pain, headache, and arthralgia. Only the incidence of arthralgia was significantly different with pioglitazone 45 mg than with placebo (P : 0.017), at 2%, 3%, and 10% for placebo, pioglitazone 30 mg, and pioglitazone 45 mg, respectively. Serious adverse events (2% to 3% across groups) and discontinuations due to adverse events (1% to 2% across groups) were similar among the treatment groups. No statistically significant differences among the treatment groups were observed for incidences of hypoglycemic episodes--11%, 10%, and 11% for placebo, pioglitazone 30 rag, and pioglitazone 45 mg, respectively--or edema (coded as not otherwise specified, peripheral, and pitting edema)--16%, 14%, and 16% for placebo, pioglitazone 30 rag, and pioglitazone 45 mg, respectively. In general, adverse events were mild or moderate in severity.

Body weight increased slightly in patients receiving pioglitazone (mean increase of 0.35 and 0.82 kg in the 30- and 45-mg groups, respectively; both P < 0.001 vs placebo); body weight decreased 1.58 kg in patients receiving placebo. Patients receiving pioglitazone 30 and 45 mg had slightly greater changes in hemoglobin (-3.4% and -4.9%, respectively, vs -0.7% with placebo; P = 0.001 and P < 0.001 vs placebo, respectively) and hematocrit (-0.6% and-2 .1%, respectively, vs 1.2% with placebo; P = 0.004 and P < 0.001 vs placebo, respectively), but the investigators deemed these changes as not clinically meaningful. As shown in Table V, there were no significant differences in mean baseline serum ALT and AST values among treatment groups. Relative to baseline, serum ALT decreased in all groups and serum AST decreased in the pioglitazone-treated groups. Reduc- tions in serum ALT for pioglitazone 30 mg and 45 mg were statistically significant versus placebo (P = 0.036 and P = 0.005, respectively). No treatment effect on urinary albumin/creatinine was observed in this s tudy (data not shown).

DISCUSSION The major finding of this large-scale, randomized, double-blind study was that administration of pioglitazone 30 or 45 mg for 16 weeks provided effective treatment for patients with type 2 diabetes mellitus with suboptimal glycemic control and mild dyslipidemia; our results support those of previous clinical trials of pioglitazone monotherapy. 14,15,19 The exclusion criteria used in the present study limit extrapolation to the general population. For example, individuals with a BMI <25 kg/m 2 were excluded from participation in the study, thereby eliminat- ing lean patients whose level of insulin sensitivity and lipid profile may not match those of the overweight or obese patient with type 2 diabetes mellitus. However, examination of the characteristics of the patients included in this study suggests

1087

CLINICAL THERAPEUTICS <''

Table V. Change from baseline in serum aminotransferases among patients with type 2 diabetes mellitus with suboptimal glycemic control and mild dyslipidemia at the end of 16 weeks of treatment with placebo (n = 97), pioglitazone 30 mg (n = 95), or pioglitazone 45 mg (n = 96).

Variable Placebo

Treatment Group

Pioglitazone 30 mg Pioglitazone 45 mg

ALT Baseline, mean (SE), U/L 26.5 (I. I) 29.4 (I .3) 30.7 (2.6) Change at end point, LSM (SE), U/L -4.0 (I. I) 7.5 (I.2) -8.6 (I.2) P versus baseline <0.001 <0.001 <0.00 l P versus placebo - 0.036 0.005

AST Baseline, mean (SE), U/L 22.0 (0.7) 23.9 (I.0) 23.6 (0.8) Change at end point, LSM (SE), U/L q).5 (I. I) 3.4 (I.2) 3.8 (I. f) P versus baseline NS 0.004 0.001 P versus placebo - NS NS

ALT = alanine aminotransferase; LSM = least squares mean; AST= aspartate aminotransfeease.

that these results may be applicable to the average patient with type 2 diabetes mellitus with suboptimal glycemic control.

Pioglitazone reduced mean baseline HbAlc by 0.8% and 0.9% in the 30- and 45- mg groups, respectively (both P < 0.001 vs placebo). Both doses of pioglitazone reduced HbAlc by -1.4% in the subgroup with the highest baseline HbAlc (ranging from >8% to <9.8%) (both P < 0.001 vs placebo). These changes in patients with high baseline HbAlc were comparable to those observed previously 14 in patients who received piogfitazone monotherapy. In contrast, patients in the low-baseline subgroup (HbAlc >6.5% to <7%) showed only minimal (0.3% to 0.4%) reductions in HbAlc, suggesting that improvement in absolute HbAlc may depend on baseline levels. However, these reductions in HbAlc may prove clinically meaningful; long- term outcome studies indicate that even small reductions in HbAlc have been associated with improvements in the late stages of complications of diabetes.l,2

The results of the responder analysis were consistent with findings from the HbAlc analysis. In the ITT population, more patients in the pioglitazone 30- and 45-mg groups achieved the ADA target of HbAlc <7% than in the placebo group (P < 0.001 and P = 0.001 vs placebo, respectively). This benefit was most ap- parent in patients with high baseline HbAlc. Although 93.7% of pioglitazone- treated patients in the low-baseline subgroup of the ITT population (59/63) reached ADA target levels for HbAlc at end point, no definitive statement re-

1088

M. Herz et al.

garding the benefit of pioglitazone for maintaining or achieving glycemic goals in patients with low-baseline disease can be made without further study.

The hyperinsulinemic euglycemic clamp technique is the gold standard for as- sessing insulin sensitivity. 2° However, this method is laborious and invasive, mak- ing it impractical to use in large-scale studies. In the present study, we used FSI and HOMA-S as indices of insulin sensitivity. Estimates of insulin sensitivity using HOMA-S have been shown to correlate strongly with estimates obtained by euglycemic c lamp-measured glucose disposal. 21,22 The HOMA-S method is a simpler alternative to the euglycemic insulin clamp and provides an acceptable index of insulin sensitivity in both diabetic and nondiabetic individuals. Relative to baseline, pioglitazone 30 and 45 mg lowered the LSM (SE) ESI (by 31.4 [4.9] and 31.8 [4.8] pmol/L, respectively; both P < 0.001) and increased the LSM (SE) HOMA-S value (by 21.2% [4.6%] and 27.3% [4.5%], respectively; both P < 0.001), suggesting that both doses improved insulin sensitivity while reducing FPG. The 2 doses of pioglitazone were not significantly different from one another on measures of insulin sensitivity. Although this finding suggests that the 30-mg dose may be optimal for enhancing insulin sensitivity, Miyazaki et a119 have shown that pioglitazone dose-dependently improves whole-body and hepatic insulin sensitivity within a dose range of 7.5 to 45 mg.

The risk of coronary heart disease in patients with type 2 diabetes mellitus is 2- to 4-fold higher than that in those without diabetes. 5 Furthermore, cardiovascular complications account for 50% to 75% of deaths in patients with the disease, 23,24 underscoring the need for treatments that ameliorate cardiovascular risk in this population. Together with obesity and hypertension, dyslipidemia is a major cardiovascular risk factor in patients with type 2 diabetes mellitus and often precedes the onset of clinical disease. 25-27 In patients with type 2 diabetes mellitus, dyslipidemia often consists of elevated TGs, decreased HDL-C, and reduced LDL particle size, whereas TC and LDL may or may not be in the normal range. 5,28

In the present study, pioglitazone 30 and 45 mg increased HDL-C compared with placebo (P = 0.028 and P < 0.001, respectively) and had no treatment effect on TC or LDL-C.

These data confirm results from previous studies 14,15 of pioglitazone monotherapy in patients with type 2 diabetes mellitus. Although the previous studies detected significant reductions from baseline in serum TGs after treatment with pioglitazone 30 mg, the present data found no significant difference in TGs with the 30-mg dose, but significant differences from baseline and placebo suggest that the 45-rag dose may benefit patients with elevated TGs (P < 0.001 vs baseline and P = 0.007 vs placebo).

We detected a significant reduction in apo B after treatment with pioglitazone 45 mg (P = 0.02 vs baseline; P = 0.025 vs pioglitazone 30 rag; P = 0.015 vs placebo). Two large studies 29,3° in the general population have shown that ele-

1089

CLINICAL THERAPEUTICS ~J

vated apo B is a good predictor of ischemic heart disease and fatal myocardial in- farctions. Thus, the possible dose dependency of this finding warrants further in- vestigation in persons with type 2 diabetes mellitus. In addition, pioglitazone 30 and 45 mg significantly reduced AIP (both P < 0.001 vs baseline; P = 0.018 and P < 0.001 vs placebo, respectively), which has been shown to inversely correlate with LDL particle size.ll Small, dense LDL is associated with an increased risk of coronary artery disease. 3k32 The reduction in AlP observed in the present study implies a possible increase in LDL particle size, suggesting a reduction of the atherogenicity of LDL particles. Taken together, our data add to the results of previous studies >'~5 showing that monotherapy with pioglitazone improves lipid profiles associated with cardiovascular disease in patients with type 2 diabetes mellitus. Such effects may be specific to pioglitazone rather than the thiazolidinedione class as a whole; treatment with another thiazolidinedione, rosigfita- zone, has been shown to increase TC, TGs, and LDL-C. 33-35 In the present study, the proportion of patients achieving ADA target goals for HDL-C increased after treatment with pioglitazone 30 and 45 mg (statistical significance not assessed), and the proportion of those receiving the 45-mg dose achieving ADA target goals for TGs also increased (statistical significance not assessed), further supporting a role for pioglitazone in ameliorating cardiovascular risk factors.

In the present study, pioglitazone 30 and 45 mg provided comparable improvements in glycemic control and insulin sensitivity These findings contrast with those of Aronoff et al, 14 who observed that 26-week treatment of OAM-naive patients with pioglitazone 45 mg lowered HbAlc by 2.6% relative to placebo, whereas pioglitazone 30 mg lowered HbA k. by only 1.3% (P < 0.05). However, compared with the present study, baseline HbAI~. values were 9.3% and 10.0% in the pioglitazone 30- and 45-mg groups, respectively, suggesting poorer glycemic control at baseline in their study Our results suggest that pioglitazone 45 mg improves dyslipidemia more than the 30-mg dose. Thus, it is possible that higher doses of pioglitazone are required to achieve maximal effects on dyslipidemia than those required for optimal glycemic control.

Treatment with pioglitazone was well tolerated. Whereas placebo-treated patients experienced mild weight loss, pioglitazone-treated patients experienced mild weight gain. Although the cause of weight loss in the placebo group is not clear, it may have resulted from continued glycosuria as a consequence of poor glycemic control or from ongoing improvements in diet and lifestyle. Assessment of potential risks associated with the modest weight gain commonly observed during pioglitazone therapy awaits long-term morbidity and mortality studies. However, several studies 36-3s have demonstrated that treatment with thiazolidinediones leads to an increase in subcutaneous fat accompanied by a decrease in visceral fat. Recently, Miyazaki et aP 8 suggested that the redistribution of body fat during pioglitazone therapy is associated with improved insulin sensitivity.

1090

l . Herz et al.

Treatment with pioglitazone 30 and 45 mg was associated with nonclinically relevant decreases in hemoglobin and hematocrit, as well as with reductions in ALT (both P < 0.001 vs baseline; P = 0.036 and P = 0.005 vs placebo, respectively) and AST (P = 0.004 and P = 0.001 vs baseline, respectively; both P = NS vs placebo). Patients with type 2 diabetes mellitus have a high background preva- lence of liver disease, including nonalcoholic steatohepatitis and fatty liver, 39,4° which have been associated with insulin resistance. 41,42 Although the clinical relevance of this finding is unknown, decreased serum aminotransferases after pioglitazone therapy may be associated with a reduction in hepatic steatosis as a result of improved insulin sensitivity Further studies will be necessary to inves- tigate this possibility.

C O N C L U S I O N S In this randomized, double-blind, placebo-controlled, multicenter study, pioglitazone 30 and 45 mg was administered to OAM-naive patients with type 2 diabetes mellitus with suboptimal glycemic control and mild dyslipidemia. Both doses of pioglitazone significantly reduced HbAlc in the ITT population. Com- pared with patients receiving placebo, more patients in both pioglitazone groups with medium or high baseline HbA~c achieved the ADA glycemic goal (HbAlc <7%) at end point. Both doses of pioglitazone improved insulin sensitivity (as demonstrated by decreased FSI and increased HOMA-S) and had a beneficial effect on lipid profiles, resulting in an improvement in AIP. Maximal improvement in lipid profiles (increased HDL-C and decreased TGs and apo B) was observed with pioglitazone 45 mg. The beneficial effects of pioglitazone on glycemic control, insulin sensitivity, and dyslipidemia may reduce morbidity and mortality in patients with type 2 diabetes mellitus.

A C K N O W L E D G M E N T S This work was sponsored by Eli Lilly and Company (Indianapolis, Indiana) and is related to study protocol H6E-MC-GLAB. Dr. Lochnan has received honoraria from Eli Lilly and Company.

The authors wish to thank Belinda Schluchter, PhD, for her assistance in the preparation of this manuscript.

The following are the primary investigators in the GLAB Study Group, by coun- try: Canada: R. Aronson, MD, Trafalgar Professional Center, Oakville, Ontario; G.A. Bailey, MD, Bailey Clinic, Red Deer, Alberta; A. Belanger, MD, Centre de Recherche Clinique de Laval, Laval, Quebec; J. Garon, MD, Clinique d'Endo- crinologie de l'Outauais, Hull, Quebec; C. Godin, MD, Centre de Recherche d'Endocrinologie Godin et St-Pierre, Sherbooke, Quebec; Ronald Goldenberg, MD, Lifestyle Metabolism Center, Thornhill, Ontario; J.E Halle, MD, Hospital Maisonneuve-Rosemont, Montreal, Quebec; P.A. Hardin, MD, West Edmonton

1091

CLINICAL THERAPEUTICS cp~

Diabetes Center, Edmonton, Alberta; C. Joyce, MD, General Health Science Cen- ter, St John's, Newfoundland; H. Lochnan, MD, Ottawa Civic Hospital, Ottawa, Ontario; S. Mann, MD, Metabolism and Diabetes Education Centre, Regina, Saskatchewan; H.P. Marshall, MD, Apex Clinical Research, Victoria, British Co- lumbia; R.M. McManus, MD, London Health Sciences Center, London, Ontario; A. Nadeau, MD, Centre Hospitalier Universitaire de Quebec, Sainte-Foy, Quebec; P. Perron, MD, Complexe Hospitalier de la Sagamie, Chicoutimi, Quebec; E Rosen, MD, Scarborough, Ontario; J. Shaban, MD, Windsor, Ontario; Spain: A. Barrios Merino, MD, Hospital Infanta Elena, Huelva; A. Diez, MD, Centro de An- tecion Primaria Sant Andreu, Barcelona; S. Duran, MD, PhD, Hospital Universi- tario Nuestra Senora de Valme, Sevilla; L. Escobar, MD, PhD, Hospital Puerta del Mar; E Kscobar-Jimenez, MD, PhD, Hospital Clinico Universitario San Cecilio, Granada; R. Gomis de Barbara, MD, Hospital Clinic i Provincial, Barcelona; E Hawkins, MD, PhD, Hospital 12 de Octubre, Madrid; A.M. Hernandez, MD, PhD, Hospital Universitario Dr. Pesset, Valencia; A. Leiva, MD, PhD, Hospital Santa Creu i Saint Paul, Barcelona; J.E Marafies, PhD, Hospital Clinico de San Carlos, Madrid; J. Mesa Manteca, MD, PhD, Cuitat Sanitaria de ka Vail d'Hebron, Barcelona; V. Miralles Belda, MD, Centro de Antecion Primaria Sant Andreu, Barcelona; L.K Pallardo, MD, PhD, Hospital La Paz, Madrid; A. Pico, MD, PhD, Hospital Gral Universitario de Alicante, Alicante; M. Puig, MD, Hospital de Mataro, Mataro, Barcelona; E. Ruiz Perez, MD, Hospital General Yag~ie, Burgos; J. Sales, MD, Hospital Vega Baja, Orihuela, Alicante; M. Serrano-Rios, MD, PhD, Hospital Clinico de San Carlos, Madrid; E Soriguer, MD, Complejo Hospitalario Carlos Haya, Malaga; M.C. Troyano, MD, Centro de Atencion Primaria Ciutat Mediriana, Barcelona; J.A. Vazquez, MD, Hospital de Cruces, Barakaldo, Vizcaya.

REFERENCES 1. Intensive blood-glucose control with sulphonylureas or insulin compared with

conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group. Lancet. 1998;352:837- 853.

2. Ohkubo Y, Kishikawa H, Araki E, et al. Intensive insulin therapy prevents the progression of diabetic microvascular complications in Japanese patients with non- insulin-dependent diabetes mellitus: A randomized prospective 6-year study. Diabetes Res Clin Pratt. 1995;28:103-117.

3. Tan MH. Diabetes and coronary artery disease. Diabetes Spectrum. 1999;12:81-83. 4. Haffner SM. Management of dyslipidemia in adults with diabetes. Diabetes Care.

t999;22(Suppl):S56-S59. 5. Fontbonne A, Charles MA, Thibult N, et al. Hyperinsulinaemia as a predictor of coro-

nary heart disease mortality in a healthy population: The Paris Prospective Study, 15- year follow-up. Diabetologia. 1991 ;34:356-361.

1092

M. Herz et al.

6. Reaven GM. Insulin resistance and its consequences: Type 2 diabetes mellitus and coronary heart disease. In: LeRoith D, Taylor SI, Olefsky JM, eds. Diabetes Mellitus: A Fundamental and Clinical Text. Philadelphia, Pa: Lippincott, Williams and Wilkins; 2000:604-615.

7. Vosper H, Khoudoli GA, Graham TL, Palmer CN. Peroxisome proliferator-activated receptor agonists, hyperlipidaemia, and atherosclerosis. Pharmacol Ther. 2002;95:47-62.

8. Kipnes MS, Krosnick A, Rendell MS, et al. Pioglitazone hydrochloride in combination with sulfonylurea therapy improves glycemic control in patients with type 2 diabetes mellitus: A randomized, placebo-controlled study. Am J Med. 2001; 111:10-17.

9, Miyazaki Y, Mahankali A, Matsuda M, et al. Improved glycemic control and enhanced insulin sensitivity in type 2 diabetic subjects treated with pioglitazone. Diabetes Care. 2001;24:710-719.

10. Rosenstock J, Einhorn D, Hershon K, et al. Efficacy and safety of pioglitazone in type 2 diabetes: A randomised, placebo-controlled study in patients receiving stable insulin therapy. Int J Clin Pract. 2002;56:251-257.

11. Dobiasova M, Frohlich J. The plasma parameter log (TG/HDL-C) as an atherogenic index: Correlation with lipoprotein particle size and esterification rate in apoB- lipoprotein-depleted plasma (FER(HDL)). Clin Biochem. 2001;34:583-588.

12. Tan MH, Johns D, Glazer B, Pioglitazone Study 001 Group. Pioglitazone reduces atherogenic index of plasma, an indirect indicator of LDL particle size. Diabetes. 2001;50:A133. Abstract.

13. American Diabetes Association. Standards of medical care for patients with diabetes mellitus. Diabetes Care. 2002;25:213-229.

14. Aronoff S, Rosenblatt S, Braithwaite S, et al. Pioglitazone hydrochloride monotherapy improves glycemic control in the treatment of patients with type 2 diabetes: A 6- month randomized placebo-controlled dose-response study. The Pioglitazone 001 Study Group. Diabetes Care. 2000;23:1605-1611.

15. Rosenblatt S, Miskin B, Glazer NB, et al. The impact of pioglitazone on glycemic control and atherogenic dyslipidemia in patients with type 2 diabetes mellitus. Coron Artery Dis. 2001;12:413-423.

16. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. The Diabetes Con- trol and Complications Trial Research Group. N Engl J Med. 1993;329:977-986.

17. Levy JC, Matthews DR, Hermans MP Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care. 1998;21:2191-2192.

18. Hochberg Y. A sharper Bonferroni procedure for multiple significance testing. Bio- metrika. 1988;75:800-803.

19. Miyazaki Y, Matsuda M, DeFronzo RA. Dose-response effect of pioglitazone on insulin sensitivity and insulin secretion in type 2 diabetes. Diabetes Care. 2002;25:517-523.

20. Radziuk J. Insulin sensitivity and its measurement: Structural commonalities among the methods. J Clin Endocrinol Metab. 2000;85:4426-4433.

1093


21. Bonora E, Targher G, Alberiche M, et al. Homeostasis model assessment closely mir- rors the glucose damp technique in the assessment of insulin sensitivity: Studies in subjects with various degrees of glucose tolerance and insulin sensitivity. Diabetes

Cart'. 2000;23:57-63. 22. Matthews DR, Hosker JE Rudenski AS, et al. Homeostasis model assessment: Insulin

resistance and beta-cell function from fasting plasma glucose and insulin concentra- tions in man. Diabetologa. 1985;28:412-419.

23. Huang ES, Meigs JB, Singer DE. The effect of interventions to prevent cardiovascular disease in patients with type 2 diabetes mellitus. Am.l Med. 2001;111:633-642.

24. Gu K, Cowie CC, Harris MI. Mortality in adults with and without diabetes in a national cohort of the U.S. population, 1971-1993. Diabetes Care. 1998;21:1138-1145.

25. Kannel WB. Lipids, diabetes, and coronary heart disease: Insights from the Framing- ham Study. Am HeartJ. 1985;110:1100-1107.

26. UK Prospective Diabetes Study. IV Characteristics of newly presenting type 2 diabetic patients: Male preponderance and obesity at different ages. Multi-center Study. Diabet

Med. 1988;5:154-159. 27. Grundy SM, Benjamin IJ, Burke GL, et al. Diabetes and cardiovascular disease: A

statement for healthcare professionals from the American Heart Association. Circula- tion. 1999;100:1134-1146.

28. Turner RC, Millns H, Neff HA, et al. Risk factors for coronary artery disease in non- insulin dependent diabetes mellitus: United Kingdom Prospective Diabetes Study (UKPDS: 23). BMJ. 1998;316:823-828.

29. Walldius G, Jungner I, Holme I, et al. High apolipoprotein B, low apotipoprotein A-l, and improvement in the prediction of fatal myocardial infarction (AMORIS study): A prospective study Lancet. 2001;358:2026-2033.

30. Lamarche B, Moorjani S, Lupien PJ, et al. Apolipoprotein A-I and B levels and the risk of ischemic heart disease during a five-year follow-up of men in the Quebec cardiovascular study. Circulation. 1996;94:273-278.

31. Austin MA, King MC, Vranizan KM, Krauss RM. Atherogenic lipoprotein phenotype. A proposed genetic marker for coronary heart disease risk. Circulation. 1990;82: 495-506.

32. Coresh J, Kwiterovich PO Jr, Smith HH, Bachorik PS. Association of plasma triglyc- eride concentration and LDL particle diameter, density, and chemical composition with premature coronary artery disease in men and women. J Lipid Res. 1993;34: 1687-1697.

33. Khan MA, St Peter JV, Xue JL. A prospective, randomized comparison of the meta- bolic effects of pioglitazone or rosiglitazone in patients with type 2 diabetes who were previously treated with troglitazone. Diabetes Care. 2002;25:708-711.

34. Boyle PJ, King AB, Olansky L, et al. Effects of pioglitazone and rosiglitazone on blood lipid levels and glycemic control in patients with type 2 diabetes mellitus: A retro- spective review of randomly selected medical records. Clin The~: 2002;24:378-396.

1094

M. Herz et al.

35. Gegick CG, Altheimer MD. Comparison of effects of thiazolidinediones on cardiovascular risk factors: Observations from a clinical practice. Endocr Pract. 2001;7: 162-169.

36. Mori Y, Murakawa Y, Okada K, et al. Effect of troglitazone on body fat distribution in type 2 diabetic patients. Diabetes Care. 1999;22:908-912.

37. Nakamura T, Funahashi T, Yamashita S, et al. Thiazolidinedione derivative improves fat distribution and multiple risk factors in subjects with visceral fat accumulation-- double-blind placebo-controlled trial. Diabetes Res Clin Pract. 2001;54:181-190.

38. Miyazaki Y, Mahankali A, Matsuda M, et al. Effect of pioglitazone on abdominal fat distribution and insulin sensitivity in type 2 diabetic patients. J Clin Endocrinol Metab. 2002;87:2784-2791.

39. Tolman KG. Thiazolidinedione hepatotoxicity: A class effect? Int J Clin Pract Suppl. 2000;113:29-34.

40. Jick SS, Stender M, Myers MW. Frequency of liver disease in type 2 diabetic patients treated with oral antidiabetic agents. Diabetes Care. 1999;22:2067-2071.

41. Luyckx FH, Lefebvre PJ, Scheen AJ. Non-alcoholic steatohepatitis: Association with obesity and insulin resistance, and influence of weight loss. Diabetes Metab. 2000;26:98-106.

42. Yu AS, Keeffe EB. Nonalcoholic fatty liver disease. Rev Gastroenterol Disord. 2002;2: 11-19.

Address correspondence to: Matthias Herz, MD, Lilly Deutschland GmbH, Saalburgstrasse 153, 61350 Bad Homburg, Germany. E-mail: Herz_Matthias@ Lilly.corn

1095

a randomized, double-blind, placebo-controlled, clinical ...hera.ugr.es/doi/14977527.pdf ·...

Documents