Skip to content

Treatment Options for Type 2 Diabetes in Adolescents and Youth (TODAY)

Studies to Treat Or Prevent Pediatric Type 2 Diabetes (STOPP-T2D) Treatment Options for Type 2 Diabetes in Adolescents and Youth (TODAY) Clinical Trial

Status
Completed
Phases
Phase 3
Study type
Interventional
Source
ClinicalTrials.gov
Registry ID
NCT00081328
Acronym
TODAY
Enrollment
699
Registered
2004-04-12
Start date
2004-05-31
Completion date
2014-02-28
Last updated
2021-07-30

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Diabetes Mellitus, Type II

Brief summary

The National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) of the National Institutes of Health (NIH) has sponsored a consortium of investigators to conduct a clinical treatment trial, Treatment Options for type 2 Diabetes in Adolescents and Youth (TODAY). The primary objective of the TODAY trial is to compare the efficacy of three treatment arms on time to treatment failure based on glycemic control. The secondary aims are to: * compare and evaluate the safety of the three treatment arms; * compare the effects of the three treatments on the pathophysiology of type 2 diabetes (T2D) with regards to beta cell function and insulin resistance, body composition, nutrition, physical activity and aerobic fitness, cardiovascular risk factors, microvascular complications, quality of life, and psychological outcomes; * evaluate the influence of individual and family behaviors on treatment response; and * compare the relative cost effectiveness of the three treatment arms. The three treatment regimens are: (1) metformin alone, (2) metformin plus rosiglitazone, and (3) metformin plus an intensive lifestyle intervention called the TODAY Lifestyle Program (TLP). The study recruits patients over a three-year period and follows patients for a minimum of two years. Patients are randomized within two years of the diagnosis of T2D.

Detailed description

T2DM has dramatically increased throughout the world in many ethnic groups and among people with diverse social and economic backgrounds. Over the last decade, the increase in the number of children and youth with T2DM has been labeled an epidemic. Before the 1990s, it was rare for most pediatric centers to have patients with T2DM. By 1994, T2DM patients represented up to 16% of new cases of diabetes in children in urban areas, and by 1999, depending on geographic location, the range of percent of new cases due to T2DM was between 8-45% and disproportionately represented in minority populations. T2DM in children and youth, as in adults, is due to the combination of insulin resistance and relative β-cell failure. It appears that there are a host of genetic and environmental risk factors for insulin resistance and limited β-cell reserve. The epidemic of pediatric T2DM is coincident with the rise in the number of children who are overweight or at risk for overweight and with a decrease in the physical activity pattern of youth. There has been a strong association between T2DM and the onset of puberty, a positive family history of T2DM, and elements of the metabolic syndrome such as acanthosis nigricans and polycystic ovarian syndrome (PCOS). Preceding the development of frank diabetes, children and youth experience a period of prediabetes. Prediabetes is defined as either elevated fasting glucose or impaired glucose tolerance. Despite the dramatic increase in the number of cases of prediabetes and T2DM in pediatric populations, there have been no published large-scale studies investigating the pathophysiology, treatment, and complications of these disorders in children and youth. The long-term complications and costs associated with T2DM make such studies imperative. Between 1997 and 2002, the estimated cost of diabetes with regard to direct medical cost increased from $44 billion to $92 billion, and the total cost increased from $98 billion to $132 billion. The vast majority of monies are spent on the long-term complications of this disorder. Since the long-term microvascular and cardiovascular complications relate to duration of diabetes and to control of glycemia, it could be hypothesized that the increasing number of children and youth diagnosed with T2DM, if not effectively treated, could dramatically add to the economic burden of this disease over the ensuing decades. Except in American Indian youth, there are no population-based data available with regard to prevalence of T2DM. Instead, only clinic-based reports indicate that there has been a tremendous increase in the number of children and adolescents with T2DM. T2DM occurs almost exclusively in children and youth who are overweight or at risk for overweight (BMI \> 85th percentile for age). At the time of diagnosis, most pediatric patients are in the midst of Tanner Stage 2-4 puberty. Puberty contributes to insulin resistance due to augmentation of growth hormone secretion, and if these normal pubertal physiologic changes are not compensated for by increased insulin secretion, frank diabetes will develop. Half to three-quarters of patients have a parent and close to ninety percent have at least one first or second degree relative with T2DM. The clinical presentation of T2DM in youth ranges from mild asymptomatic hyperglycemia to severe ketoacidosis. In those who present with clinical symptoms due to hyperglycemia, glycosuria and weight loss are present in 20-40%, ketonuria is present in 33% and ketoacidosis is found in 5-10%. Patients without clinical symptoms are diagnosed as the result of routine blood or urine testing during a health care visit or by investigating a variety of complaints such as chronic infection, sleep apnea, hyperlipidemia, hypertension, and hirsutism or irregular periods associated with PCOS. It may be difficult to distinguish T1DM from T2DM at presentation. The absence of autoantibodies is a prerequisite for the diagnosis of T2DM. In addition, evidence of residual insulin secretion is suggestive of T2DM rather than T1DM. Patients with T2DM have dual abnormalities of insulin resistance and insulin deficiency. It is hypothesized that to achieve the level of glycemic control required to optimize long-term outcome and decrease or prevent microvascular complications, treatment regimens should theoretically be designed to improve insulin resistance and preserve residual β-cell function. The available anti-diabetic agents have not been adequately evaluated in pediatric patients. This is particularly relevant with regard to using combination therapy to improve glycemic control or lifestyle interventions aimed at obesity and sedentary behavior.

Interventions

DRUGMetformin

capsule, 1000 mg bid

DRUGRosiglitazone

capsule, 4 mg bid

a lifestyle change (LC) phase of weekly sessions for months 1-6, followed by a bi-weekly lifestyle maintenance (LM) phase through months 7-12, and a continued contact (CC) phase from months 13 through the end of the study. The CC phase sessions are scheduled monthly for the initial 12 months (study months 13-24) and then quarterly or 4 times a year to the end of the study

Sponsors

National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)
Lead SponsorNIH

Study design

Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
QUADRUPLE (Subject, Caregiver, Investigator, Outcomes Assessor)

Eligibility

Sex/Gender
ALL
Age
10 Years to 17 Years
Healthy volunteers
No

Inclusion criteria

(during Screening and Run-in period): * Diabetes by ADA criteria (laboratory determinations of fasting glucose ≥ 126 mg/dL, random glucose ≥ 200 mg/dL, or two-hour OGTT glucose ≥ 200 mg/dL) documented and confirmed in medical record. For patients diagnosed with diabetes during screening who have a normal fasting glucose but an elevated two-hour glucose during an OGTT, the HbA1c must be ≥ 6%. * Duration since diagnosis less than two years by date of randomization. * BMI ≥ 85th percentile documented at time of diagnosis or at screening. * Fasting C-peptide at screening (drawn at least one week after treatment for ketosis or acidosis, if applicable) \> 0.6 ng/mL. * Absence of pancreatic autoimmunity (both GAD and ICA512 negative). * Age 10-17, with randomization prior to 18th birthday. * Signed informed consent/assent forms for the pre-randomization period. * A family member or adult closely involved in the daily activities of the child agrees to participate in the child's treatment. * Fluency in English or Spanish for both child and family member. * Patient and family able to fully participate in trial protocol in the opinion of the investigator.

Exclusion criteria

(during Screening and Run-in period): * Participating in another interventional research study protocol in the past 30 days. * Genetic syndrome or disorder known to affect glucose tolerance other than diabetes. * Patient on inhaled steroids at dose above 1000 mcg daily Flovent equivalent. * Patient on a course of oral steroids within the last 60 days or on oral steroids more than 20 days during the past year. * Patient on medication(s) that are known to affect insulin sensitivity or secretion within the last 30 days. * Patient on medication(s) that are known to cause weight gain within the last 30 days. * Patient on any weight-loss medication(s) within the last 30 days. * Patient on medication(s) known to affect the metabolism of study drug. * Inability to comprehend the lowest grade level at which lifestyle intervention materials are prepared, for both child and participating family member. * Females who are pregnant, planning to become pregnant within two years of enrollment, or who admit sexual activity without appropriate contraception. * Calculated creatinine clearance \< 70 mL/min. * Any transaminase \> 2.5 ULN. If any transaminase 1.5-2.5 times ULN, then patient must be appropriately evaluated by PCP (minimum evaluation includes ceruloplasmin level, alpha-1 antitrypsin phenotype, ANA, anti-smooth muscle antibody, anti-LKM antibody, anti-HCV, and anti-HBc total antibody not IgM, iron, and TIBC) and is eligible if all other causes for elevation are ruled out and it is presumed due only to non-alcoholic fatty liver disease (NAFLD). * Diabetic ketoacidosis (DKA) at any time after diagnosis unless only a single episode of DKA related to a significant medical illness. * Physical limitations preventing patient from being randomized to the lifestyle intervention. * Patient plans to leave the geographic area within one calendar year. * Abnormal reticulocyte count or HbA1c chromatogram at time of screening. * Admitted use of anabolic steroids within the past 60 days. * Other significant organ system illness or condition (including psychiatric or developmental disorder) that would prevent participation in the opinion of the investigator. * Patient participates in a formal weight-loss program. Inclusion Criteria (post Run-in and Randomization): * Duration since diagnosis less than 2 years at randomization. * HbA1c \< 8% on metformin alone. * Age 10-17, with randomization before patient is 18 years old. * Signed consent/assent forms for randomization and the post-randomization phase. * A family member or adult closely involved in the daily activities of the child agrees to participate in the child's treatment. * Fluency in English or Spanish for both child and family member. * Patient and family able to fully participate in trial protocol in the opinion of the investigator.

Design outcomes

Primary

MeasureTime frameDescription
Treatment Failure (Loss of Glycemic Control)Study duration - 2 years to 6.5 years of follow up from randomizationDefined as A1c persistently \>=8% over a 6-month period or persistent metabolic decompensation (inability to wean insulin within 3 months of initiation or the occurrence of a second episode within three months of discontinuing insulin)

Secondary

MeasureTime frameDescription
Number of Serious Adverse EventsReported as occurred during study follow-up - 2 years to 6.5 years from randomization.Number of serious adverse events reported during the trial. Participant could have multiple episodes reported.
Insulin Secretion24 monthsInsulinogenic index determined from OGTT as difference in insulin at 30 minutes minus 0 minutes divided by difference in glucose at 30 minutes minus 0 minutes. The analysis sample includes only participants with 24 month data who had not experienced the primary outcome by that time.
Body Composition -- BMI24 monthsBody mass index (BMI) measured in kg per meters squared. The analysis sample includes only participants with 24 month data who had not experienced the primary outcome by that time.
Body Composition -- Waist Circumference24 monthsWaist circumference (cm) measured at the iliac crest at its outermost point with the measuring tape placed around the participant in a horizontal plane parallel to the floor at the mark and the measurement teken at the end of normal expiration without the tape compressing the skin. The analysis sample includes only participants with 24 month data who had not experienced the primary outcome by that time.
Insulin Sensitivity24 monthsAll participants were followed to 24 months. Insulin sensitivity is measured from OGTT as inverse of fasting insulin (mL/uU). The analysis sample includes only participants with 24 month data who had not experienced the primary outcome by that time.
Body Composition -- Fat Mass24 monthsDetermined by DXA whole body scan. The analysis sample includes only participants with 24 month data who had not experienced the primary outcome by that time. In addition, in about 1/3 of participants DXA scans could not be obtained on participants weighing more than 300 pounds (136 kg), the upper limit in size set by the machine manufacturers. Scans were considered invalid if a body part (e.g., arm, leg) was completely off or partially off the scanner, there was hand-hip overlap, or there was motion or movement during the scan.
Comorbidity -- HypertensionData collected at baseline and during follow-up - 2 years to 6.5 years from randomization.A diagnosis was made by an out-of-range value \>=95th percentile or systolic \>=130 or diastolic \>=80 sustained over 6 months or on an anti-hypertensive medication.
Comorbidity -- LDL DyslipidemiaData collected at baseline and during follow-up - 2 years to 6.5 years from randomization.A diagnosis was made from out-of-range value \>= 130 mg/dL sustained over 6 months or put on lipid lowering medication.
Comorbidity -- Triglycerides DyslipidemiaData collected at baseline and during follow-up - 2 years to 6.5 years from randomization.A diagnosis was made by an out-of-range value \>=150 mg/dL sustained over 6 months or on appropriate lipid lowering medication.
Body Composition -- Bone Density24 monthsMeasured by DXA, both whole body scan and AP-spine scan. The analysis sample includes only participants with 24 month data who had not experienced the primary outcome by that time. In addition, in about 1/3 of participants DXA scans could not be obtained on participants weighing more than 300 pounds (136 kg), the upper limit in size set by the machine manufacturers. Scans were considered invalid if a body part (e.g., arm, leg) was completely off or partially off the scanner, there was hand-hip overlap, or there was motion or movement during the scan.

Countries

United States

Participant flow

Recruitment details

Participants were recruited from July 2004 to February 2009. Participants were recruited from the patient populations of pediatric endocrine clinics at the participating study clinical centers, including satellite clinics. Posters announced the study. Study staff approached youth and their families during medical visits.

Pre-assignment details

Prior to randomization, eligible subjects entered a 2-6 month run-in period, with goals of weaning from non-study diabetes medications, tolerating metformin at 1000 mg bid but no less than 500 mg bid, attaining glycemic control on metformin alone, mastering standard diabetes education, and adhering to study medication and visit attendance.

Participants by arm

ArmCount
1 Metformin Alone
Metformin alone Metformin: capsule, 1000 mg bid
232
2 Metformin + Rosiglitazone
Metformin + Rosiglitazone Metformin: capsule, 1000 mg bid Rosiglitazone: capsule, 4 mg bid
233
3 Metformin + Lifestyle Program
Metformin + Lifestyle Program Metformin: capsule, 1000 mg bid Lifestyle Program: a lifestyle change (LC) phase of weekly sessions for months 1-6, followed by a bi-weekly lifestyle maintenance (LM) phase through months 7-12, and a continued contact (CC) phase from months 13 through the end of the study. The CC phase sessions are scheduled monthly for the initial 12 months (study months 13-24) and then quarterly or 4 times a year to the end of the study
234
Total699

Baseline characteristics

Characteristic1 Metformin Alone3 Metformin + Lifestyle ProgramTotal2 Metformin + Rosiglitazone
Age, Continuous14.1 years
STANDARD_DEVIATION 1.9
13.8 years
STANDARD_DEVIATION 2
14.0 years
STANDARD_DEVIATION 2
14.1 years
STANDARD_DEVIATION 2.1
Body mass index (BMI)35.8 kg per m squared
STANDARD_DEVIATION 8.1
34.1 kg per m squared
STANDARD_DEVIATION 7.1
34.9 kg per m squared
STANDARD_DEVIATION 7.6
35.0 kg per m squared
STANDARD_DEVIATION 7.7
Body mass index z-score2.27 z-score
STANDARD_DEVIATION 0.45
2.18 z-score
STANDARD_DEVIATION 0.46
2.23 z-score
STANDARD_DEVIATION 0.47
2.22 z-score
STANDARD_DEVIATION 0.49
Bone density from DXA1.09 g per cm squared
STANDARD_DEVIATION 0.12
1.08 g per cm squared
STANDARD_DEVIATION 0.12
1.09 g per cm squared
STANDARD_DEVIATION 0.13
1.10 g per cm squared
STANDARD_DEVIATION 0.13
Duration of diabetes7.8 months
STANDARD_DEVIATION 6
7.6 months
STANDARD_DEVIATION 5.8
7.8 months
STANDARD_DEVIATION 5.8
8.0 months
STANDARD_DEVIATION 5.7
Fat mass from DXA33.7 kg
STANDARD_DEVIATION 9.9
32.6 kg
STANDARD_DEVIATION 9.8
33.3 kg
STANDARD_DEVIATION 10
33.5 kg
STANDARD_DEVIATION 10.2
Highest household education level
College no degree
77 participants63 participants218 participants78 participants
Highest household education level
Graduate degree
35 participants38 participants114 participants41 participants
Highest household education level
High school, GED, business or technical
57 participants66 participants172 participants49 participants
Highest household education level
Less than high school
60 participants62 participants182 participants60 participants
Highest household education level
Unknown
3 participants5 participants13 participants5 participants
Insulinogenic index from OGTT1.02 uU/mL divided by mg/dL.87 uU/mL divided by mg/dL.93 uU/mL divided by mg/dL.92 uU/mL divided by mg/dL
Insulin sensitivity (inverse of fasting insulin from OGTT).036 mL/uU.040 mL/uU.039 mL/uU.040 mL/uU
Nuclear family + grandparents history of diabetes
No
17 participants29 participants72 participants26 participants
Nuclear family + grandparents history of diabetes
Unknown
4 participants5 participants16 participants7 participants
Nuclear family + grandparents history of diabetes
Yes
211 participants200 participants611 participants200 participants
Nuclear family history of diabetes
No
97 participants86 participants276 participants93 participants
Nuclear family history of diabetes
Unknown
4 participants5 participants16 participants7 participants
Nuclear family history of diabetes
Yes
131 participants143 participants407 participants133 participants
Percent overweight82.1 100% X (BMI-50th %ile)/50th %ile
STANDARD_DEVIATION 38.3
75.6 100% X (BMI-50th %ile)/50th %ile
STANDARD_DEVIATION 35.3
78.9 100% X (BMI-50th %ile)/50th %ile
STANDARD_DEVIATION 37.3
79.1 100% X (BMI-50th %ile)/50th %ile
STANDARD_DEVIATION 38.1
Race/Ethnicity, Customized
American Indian
12 participants13 participants41 participants16 participants
Race/Ethnicity, Customized
Asian Non-Hispanic
3 participants3 participants11 participants5 participants
Race/Ethnicity, Customized
Black Non-Hispanic
77 participants86 participants227 participants64 participants
Race/Ethnicity, Customized
Hispanic
91 participants86 participants278 participants101 participants
Race/Ethnicity, Customized
White Non-Hispanic
49 participants46 participants142 participants47 participants
Region of Enrollment
United States
232 participants234 participants699 participants233 participants
Sex: Female, Male
Female
146 Participants154 Participants452 Participants152 Participants
Sex: Female, Male
Male
86 Participants80 Participants247 Participants81 Participants
Tanner stage by physical examination
1, 2, or 3
23 participants30 participants79 participants26 participants
Tanner stage by physical examination
4 or 5
209 participants204 participants620 participants207 participants
Total annual household income
< $25,000
81 participants92 participants259 participants86 participants
Total annual household income
$25,000-49,999
83 participants66 participants210 participants61 participants
Total annual household income
>=$50,000
44 participants53 participants155 participants58 participants
Total annual household income
unknown
24 participants23 participants75 participants28 participants
Waist circumference110.4 cm
STANDARD_DEVIATION 16.6
106.6 cm
STANDARD_DEVIATION 16.2
108.6 cm
STANDARD_DEVIATION 16.7
109.0 cm
STANDARD_DEVIATION 17

Adverse events

Event typeEG000
affected / at risk
EG001
affected / at risk
EG002
affected / at risk
deaths
Total, all-cause mortality
— / —— / —— / —
other
Total, other adverse events
215 / 232196 / 233213 / 234
serious
Total, serious adverse events
42 / 23234 / 23358 / 234

Outcome results

Primary

Treatment Failure (Loss of Glycemic Control)

Defined as A1c persistently \>=8% over a 6-month period or persistent metabolic decompensation (inability to wean insulin within 3 months of initiation or the occurrence of a second episode within three months of discontinuing insulin)

Time frame: Study duration - 2 years to 6.5 years of follow up from randomization

Population: The entire cohort of 699 participants was included in the analysis.

ArmMeasureGroupValue (NUMBER)
1 Metformin AloneTreatment Failure (Loss of Glycemic Control)Treatment failure120 participants
1 Metformin AloneTreatment Failure (Loss of Glycemic Control)Did not fail treatment during trial112 participants
2 Metformin + RosliglitazoneTreatment Failure (Loss of Glycemic Control)Treatment failure90 participants
2 Metformin + RosliglitazoneTreatment Failure (Loss of Glycemic Control)Did not fail treatment during trial143 participants
3 Metformin + Lifestyle ProgramTreatment Failure (Loss of Glycemic Control)Treatment failure109 participants
3 Metformin + Lifestyle ProgramTreatment Failure (Loss of Glycemic Control)Did not fail treatment during trial125 participants
Secondary

Body Composition -- BMI

Body mass index (BMI) measured in kg per meters squared. The analysis sample includes only participants with 24 month data who had not experienced the primary outcome by that time.

Time frame: 24 months

Population: Cohort measured at 24 months and had not experienced treatment failure.

ArmMeasureValue (MEAN)Dispersion
1 Metformin AloneBody Composition -- BMI36.7 kg per meters squaredStandard Deviation 9.1
2 Metformin + RosliglitazoneBody Composition -- BMI38.2 kg per meters squaredStandard Deviation 8.1
3 Metformin + Lifestyle ProgramBody Composition -- BMI35.3 kg per meters squaredStandard Deviation 8.4
Secondary

Body Composition -- Bone Density

Measured by DXA, both whole body scan and AP-spine scan. The analysis sample includes only participants with 24 month data who had not experienced the primary outcome by that time. In addition, in about 1/3 of participants DXA scans could not be obtained on participants weighing more than 300 pounds (136 kg), the upper limit in size set by the machine manufacturers. Scans were considered invalid if a body part (e.g., arm, leg) was completely off or partially off the scanner, there was hand-hip overlap, or there was motion or movement during the scan.

Time frame: 24 months

Population: Members of the cohort measured at 24 months who did not experience treatment failure.

ArmMeasureValue (MEAN)Dispersion
1 Metformin AloneBody Composition -- Bone Density1.15 g/cm squaredStandard Deviation 0.1
2 Metformin + RosliglitazoneBody Composition -- Bone Density1.15 g/cm squaredStandard Deviation 0.11
3 Metformin + Lifestyle ProgramBody Composition -- Bone Density1.15 g/cm squaredStandard Deviation 0.12
Secondary

Body Composition -- Fat Mass

Determined by DXA whole body scan. The analysis sample includes only participants with 24 month data who had not experienced the primary outcome by that time. In addition, in about 1/3 of participants DXA scans could not be obtained on participants weighing more than 300 pounds (136 kg), the upper limit in size set by the machine manufacturers. Scans were considered invalid if a body part (e.g., arm, leg) was completely off or partially off the scanner, there was hand-hip overlap, or there was motion or movement during the scan.

Time frame: 24 months

Population: Members of the cohort measured at 24 months who did not experience treatment failure.

ArmMeasureValue (MEAN)Dispersion
1 Metformin AloneBody Composition -- Fat Mass36.1 kgStandard Deviation 12
2 Metformin + RosliglitazoneBody Composition -- Fat Mass39.7 kgStandard Deviation 10.5
3 Metformin + Lifestyle ProgramBody Composition -- Fat Mass32.2 kgStandard Deviation 9.4
Secondary

Body Composition -- Waist Circumference

Waist circumference (cm) measured at the iliac crest at its outermost point with the measuring tape placed around the participant in a horizontal plane parallel to the floor at the mark and the measurement teken at the end of normal expiration without the tape compressing the skin. The analysis sample includes only participants with 24 month data who had not experienced the primary outcome by that time.

Time frame: 24 months

Population: Members of cohort measured at 24 months who had not experience treatment failure.

ArmMeasureValue (MEAN)Dispersion
1 Metformin AloneBody Composition -- Waist Circumference110.8 cmStandard Deviation 17.5
2 Metformin + RosliglitazoneBody Composition -- Waist Circumference114.0 cmStandard Deviation 16.3
3 Metformin + Lifestyle ProgramBody Composition -- Waist Circumference108.6 cmStandard Deviation 16.7
Secondary

Comorbidity -- Hypertension

A diagnosis was made by an out-of-range value \>=95th percentile or systolic \>=130 or diastolic \>=80 sustained over 6 months or on an anti-hypertensive medication.

Time frame: Data collected at baseline and during follow-up - 2 years to 6.5 years from randomization.

Population: Entire cohort.

ArmMeasureValue (NUMBER)
1 Metformin AloneComorbidity -- Hypertension57 participants
2 Metformin + RosliglitazoneComorbidity -- Hypertension53 participants
3 Metformin + Lifestyle ProgramComorbidity -- Hypertension45 participants
Secondary

Comorbidity -- LDL Dyslipidemia

A diagnosis was made from out-of-range value \>= 130 mg/dL sustained over 6 months or put on lipid lowering medication.

Time frame: Data collected at baseline and during follow-up - 2 years to 6.5 years from randomization.

Population: Entire cohort.

ArmMeasureValue (NUMBER)
1 Metformin AloneComorbidity -- LDL Dyslipidemia18 participants
2 Metformin + RosliglitazoneComorbidity -- LDL Dyslipidemia16 participants
3 Metformin + Lifestyle ProgramComorbidity -- LDL Dyslipidemia15 participants
Secondary

Comorbidity -- Triglycerides Dyslipidemia

A diagnosis was made by an out-of-range value \>=150 mg/dL sustained over 6 months or on appropriate lipid lowering medication.

Time frame: Data collected at baseline and during follow-up - 2 years to 6.5 years from randomization.

Population: Entire cohort.

ArmMeasureValue (NUMBER)
1 Metformin AloneComorbidity -- Triglycerides Dyslipidemia20 participants
2 Metformin + RosliglitazoneComorbidity -- Triglycerides Dyslipidemia28 participants
3 Metformin + Lifestyle ProgramComorbidity -- Triglycerides Dyslipidemia22 participants
Secondary

Insulin Secretion

Insulinogenic index determined from OGTT as difference in insulin at 30 minutes minus 0 minutes divided by difference in glucose at 30 minutes minus 0 minutes. The analysis sample includes only participants with 24 month data who had not experienced the primary outcome by that time.

Time frame: 24 months

Population: Participants who were measured at 24 months and had not experience treatment failure.

ArmMeasureValue (MEDIAN)
1 Metformin AloneInsulin Secretion.75 uU/mL divided by mg/dL
2 Metformin + RosliglitazoneInsulin Secretion.83 uU/mL divided by mg/dL
3 Metformin + Lifestyle ProgramInsulin Secretion.71 uU/mL divided by mg/dL
Secondary

Insulin Sensitivity

All participants were followed to 24 months. Insulin sensitivity is measured from OGTT as inverse of fasting insulin (mL/uU). The analysis sample includes only participants with 24 month data who had not experienced the primary outcome by that time.

Time frame: 24 months

Population: Participants who were measured at 24 months and had not experienced treatment failure.

ArmMeasureValue (MEDIAN)
1 Metformin AloneInsulin Sensitivity0.037 mL/uU
2 Metformin + RosliglitazoneInsulin Sensitivity0.049 mL/uU
3 Metformin + Lifestyle ProgramInsulin Sensitivity0.039 mL/uU
Secondary

Number of Serious Adverse Events

Number of serious adverse events reported during the trial. Participant could have multiple episodes reported.

Time frame: Reported as occurred during study follow-up - 2 years to 6.5 years from randomization.

Population: Entire cohort.

ArmMeasureValue (NUMBER)
1 Metformin AloneNumber of Serious Adverse Events42 episodes of serious adverse event
2 Metformin + RosliglitazoneNumber of Serious Adverse Events34 episodes of serious adverse event
3 Metformin + Lifestyle ProgramNumber of Serious Adverse Events58 episodes of serious adverse event

Source: ClinicalTrials.gov · Data processed: Mar 27, 2026