Insulin Resistance, Overweight and Obesity, PreDiabetes
Conditions
Brief summary
The purpose of this research study is to determine whether continuous glucose monitors (CGM) are a safe, effective, and acceptable way to evaluate type 2 diabetes risk in youth as compared to the standard 2-hour oral glucose tolerance test (OGTT). This study will involve wearing a CGM, wearing a physical activity tracker, responding to surveys, and completing at-home glucose and mixed food challenge while wearing the CGM. Subjects will also be asked to complete an interview by phone or videoconference after wearing the CGM.
Detailed description
Youth-onset type 2 diabetes (T2D), an increasingly common and aggressive disease that disproportionately impacts racial and ethnic minorities, presents several diagnostic challenges. The rapid loss of glycemic control soon after T2D diagnosis makes early identification an important goal in pediatrics, and risk-based screening has been recommended for children for over two decades. However, the screening tests currently used in practice, including hemoglobin A1c (HbA1c), fasting glucose, and 2-hour oral glucose tolerance test (OGTT), have significant drawbacks. Real-time continuous glucose monitors (CGM) are promising new tools for risk stratification in individuals with dysglycemia. These patient-worn medical devices measure interstitial glucose every 5 minutes and offer glimpses into an individual's everyday life, which may include consumption of carbohydrates in amounts larger than that assessed in an OGTT, enhancing test sensitivity. Whether use of CGMs to evaluate type 2 diabetes risk in free-living youth is feasible and acceptable to patients and families is unknown. Assessment of CGM with standardized, but practical, home-based glucose and mixed-meal challenges is needed to determine facilitators and barriers to use of this technology-driven approach in clinical practice. Without advancement in the approach to screening for youth-onset T2D, there is ongoing risk of failure to understand which individuals are at highest risk for complications and could benefit the most from intervention early in the disease process. In this single-arm prospective study, participants will attend one in-person study visit to complete 2-hour oral glucose tolerance test after overnight fast. Participants will wear a continuous glucose monitor and a wrist-worn actigraphy device for physical activity and sleep measurement for 10 days beginning on the day of the study visit, then return the monitors by mail. During the 10 day wear period, participants will respond to text message-based surveys to document food intake, physical activity, and sleep times and will complete two at-home challenges: 1) glucose challenge using study-provided glucose beverage that is identical to that they consume during clinical oral glucose tolerance test; and 2) food challenge with mixed carbohydrate/fat/protein-containing food of their choice. Glucose trends during these at-home challenges will be compared with oral glucose tolerance test obtained during the study visit. Individual interviews will be conducted upon completion of CGM wear to assess acceptability of the use of at-home glucose challenge and CGM wear as a strategy to evaluate diabetes risk in youth.
Interventions
DEVICECGM
Participants will wear blinded Dexcom G6 Pro continuously for 10 days in their home environment. Two challenges will be completed at home: 1) glucose, using 1.75 g/kg (max 75g) glucose beverage provided by study team, and 2) mixed food, containing 50g carb as well as protein and fat content of participant's choice.
Sponsors
Mary Ellen Vajravelu, MD
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
DIAGNOSTIC
Masking
NONE
Intervention model description
Subjects will wear a continuous glucose monitor (CGM) to determine the feasibility and acceptability of the device to evaluate risk of youth developing Type 2 Diabetes (T2D).
Eligibility
Sex/Gender
ALL
Age
8 Years to 18 Years
Healthy volunteers
Yes
Inclusion criteria
* Any gender, age 8-18 years * Tanner 2 or higher pubertal development * Overweight or obese (BMI ≥85th percentile for age/sex, or ≥ 25 kg/m2 for participants =18 years) * A) No previously documented abnormal HbA1c or glucose but at higher risk for T2D based on race/ethnicity (Black, Hispanic, Asian, Native American, Pacific Islander), diagnosis with dyslipidemia, hypertension, polycystic ovary syndrome, presence of acanthosis nigricans, or first degree relative with T2D; or B) previously documented (within 6 months) HbA1c 5.7-7.0%, fasting glucose ≥100 mg/dL, or 2-hour plasma glucose on OGTT of ≥140 mg/dL * Consent (adult subjects), parental/guardian permission (if applicable), assent (if applicable) * Willingness to wear ActiGraph watch and CGM continuously for 10 day study duration
Exclusion criteria
* Current or recent (within 1 month) use of diabetes-related medication * Current use of hydroxyurea (due to interference with CGM) * Known type 1, cystic fibrosis-related, or medication-induced diabetes * Potential subject unable to speak or read in English * Severe cognitive impairment * Current or previous pregnancy
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Feasibility of CGM Use to Determine Type 2 Diabetes Risk | 10 days | Completeness of CGM data (≥ 80% of days with data) |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Positive Predictive Value of At-home, CGM-measured Glucose Challenge | 2 Hour Point of Home-OGTT | Positive predictive value of CGM-glucose 2 hours after at-home glucose beverage ingestion to predict OGTT-2h≥140 mg/dL |
| Specificity of At-home CGM-measured Fasting Glucose | 5 Minute Point of Home-OGTT | Specificity of baseline CGM-glucose at the time of at-home glucose beverage ingestion to predict fasting OGTT ≥100 mg/dL |
| Positive Predictive Value of At-home CGM-measured Fasting Glucose | 5 Minute Point of Home-OGTT | Positive predictive value of baseline CGM-glucose at the time of at-home glucose beverage ingestion to predict fasting OGTT ≥100 mg/dL |
| Negative Predictive Value of At-home CGM-measured Fasting Glucose | 5 Minute Point of Home-OGTT | Negative predictive value of baseline CGM-glucose at the time of at-home glucose beverage ingestion to predict fasting OGTT ≥100 mg/dL |
| Acceptability of CGM Use for At-home T2D Risk Evaluation in Youth | 30 minutes | Interview-based assessment. Participants rated their experience with CGM-OGTT using a Likert-like scale from 1-5 (1= terrible, 2=bad, 3=neutral, 4=good, 5=excellent). Acceptability was defined as a pre-specified threshold of ≥80% of participants providing a neutral or higher rating. |
| Sensitivity of At-home CGM-measured Glucose Challenge | 2 Hour Point of Home-OGTT | Sensitivity of CGM-glucose 2 hours after at-home glucose beverage ingestion to predict OGTT-2h≥140 mg/dL |
| Specificity of At-home CGM-measured Glucose Challenge | 2 Hour Point of Home-OGTT | Specificity of CGM-glucose 2 hours after at-home glucose beverage ingestion to predict OGTT-2h≥140 mg/dL |
| Negative Predictive Value of At-home, CGM-measured Glucose Challenge | 2 Hour Point of Home-OGTT | Negative predictive value of CGM-glucose 2 hours after at-home glucose beverage ingestion to predict OGTT-2h≥140 mg/dL |
| Sensitivity of At-home CGM-measured Fasting Glucose | 5 Minute Point of Home-OGTT | Sensitivity of baseline CGM-glucose at the time of at-home glucose beverage ingestion to predict fasting OGTT ≥100 mg/dL |
Other
| Measure | Time frame | Description |
|---|---|---|
| Laboratory-measured 2-hour Glucose | 2 Hour Point of Lab-OGTT | 2-hour OGTT glucose will be measured in the laboratory at baseline to investigate associations with CGM indices obtained during at-home glucose challenge. |
| Laboratory-measured Fasting Glucose | Start of Lab-OGTT | Fasting glucose will be measured in the laboratory at baseline to investigate associations with CGM indices obtained during at-home glucose challenge. |
| Laboratory-measured HbA1c | Measured at Research Visit | HbA1c will be measured in the laboratory at baseline to investigate associations with CGM indices obtained during at-home glucose challenge. |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| CGM Use for T2D Risk Evaluation All participants will complete a standard 2-hour oral glucose tolerance test (OGTT), wear a CGM for 10 days, complete at-home glucose challenge, and provide qualitative feedback about their experiences with both OGTT and CGM use for risk evaluation.
CGM: Participants will wear blinded Dexcom G6 Pro continuously for 10 days in their home environment. Two challenges will be completed at home: 1) glucose, using 1.75 g/kg (max 75g) glucose beverage provided by study team, and 2) mixed food, containing 50g carb as well as protein and fat content of participant's choice. | 39 |
| Total | 39 |
Withdrawals & dropouts
| Period | Reason | FG000 |
|---|---|---|
| Completed 75-gram, 2-hour CGM-OGTT | CGM fell off early | 3 |
| Completed 75-gram, 2-hour CGM-OGTT | Incomplete due to illness | 1 |
| Completed Study Visit | Withdrawal by Subject | 1 |
| Completed, Valid Lab-OGTT and CGM-OGTT | Incomplete glucose beverage for lab-OGTT | 1 |
| Valid 75-gram, 2-hour CGM-OGTT | No date/time recorded | 4 |
| Valid 75-gram, 2-hour CGM-OGTT | Vomited after drinking glucose beverage at home | 1 |
Baseline characteristics
| Characteristic | CGM Use for T2D Risk Evaluation |
|---|---|
| Age, Continuous | 14.6 years |
| BMI | 35.9 kg/m^2 |
| Race/Ethnicity, Customized Race/Ethnicity Hispanic/Latinx | 3 Participants |
| Race/Ethnicity, Customized Race/Ethnicity multi-racial | 5 Participants |
| Race/Ethnicity, Customized Race/Ethnicity non-Hispanic Black | 13 Participants |
| Race/Ethnicity, Customized Race/Ethnicity non-Hispanic White | 18 Participants |
| Region of Enrollment United States | 39 participants |
| Sex: Female, Male Female | 21 Participants |
| Sex: Female, Male Male | 18 Participants |
Adverse events
| Event type | EG000 affected / at risk |
|---|---|
| deaths Total, all-cause mortality | 0 / 39 |
| other Total, other adverse events | 3 / 39 |
| serious Total, serious adverse events | 0 / 39 |
Outcome results
Primary
Feasibility of CGM Use to Determine Type 2 Diabetes Risk
Completeness of CGM data (≥ 80% of days with data)
Time frame: 10 days
Population: Of the 39 participants who completed the study visit, 38 returned their CGM for analysis.
| Arm | Measure | Category | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| CGM Use for T2D Risk Evaluation | Feasibility of CGM Use to Determine Type 2 Diabetes Risk | CGM Wear ≥ 8 Days | 33 Participants |
| CGM Use for T2D Risk Evaluation | Feasibility of CGM Use to Determine Type 2 Diabetes Risk | CGM Wear < 8 Days | 5 Participants |
Secondary
Acceptability of CGM Use for At-home T2D Risk Evaluation in Youth
Interview-based assessment. Participants rated their experience with CGM-OGTT using a Likert-like scale from 1-5 (1= terrible, 2=bad, 3=neutral, 4=good, 5=excellent). Acceptability was defined as a pre-specified threshold of ≥80% of participants providing a neutral or higher rating.
Time frame: 30 minutes
Population: Of the 39 participants enrolled in the study, only 36 completed the exit interview.
| Arm | Measure | Category | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| CGM Use for T2D Risk Evaluation | Acceptability of CGM Use for At-home T2D Risk Evaluation in Youth | Rated as excellent/good | 33 Participants |
| CGM Use for T2D Risk Evaluation | Acceptability of CGM Use for At-home T2D Risk Evaluation in Youth | Rated as neutral | 3 Participants |
Secondary
Negative Predictive Value of At-home CGM-measured Fasting Glucose
Negative predictive value of baseline CGM-glucose at the time of at-home glucose beverage ingestion to predict fasting OGTT ≥100 mg/dL
Time frame: 5 Minute Point of Home-OGTT
Population: Of the 35 participants completing the 2-hr CGM-OGTT, 31 had valid data for baseline fasting glucose.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| CGM Use for T2D Risk Evaluation | Negative Predictive Value of At-home CGM-measured Fasting Glucose | 100 percent accuracy of negative tests |
Secondary
Negative Predictive Value of At-home, CGM-measured Glucose Challenge
Negative predictive value of CGM-glucose 2 hours after at-home glucose beverage ingestion to predict OGTT-2h≥140 mg/dL
Time frame: 2 Hour Point of Home-OGTT
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| CGM Use for T2D Risk Evaluation | Negative Predictive Value of At-home, CGM-measured Glucose Challenge | 85.7 percent accuracy of negative tests |
95% CI: [0.18, 0.94]
Secondary
Positive Predictive Value of At-home CGM-measured Fasting Glucose
Positive predictive value of baseline CGM-glucose at the time of at-home glucose beverage ingestion to predict fasting OGTT ≥100 mg/dL
Time frame: 5 Minute Point of Home-OGTT
Population: Of the 35 participants completing the 2-hr CGM-OGTT, 31 had valid data for baseline fasting glucose.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| CGM Use for T2D Risk Evaluation | Positive Predictive Value of At-home CGM-measured Fasting Glucose | 3.6 percent accuracy of positive tests |
Secondary
Positive Predictive Value of At-home, CGM-measured Glucose Challenge
Positive predictive value of CGM-glucose 2 hours after at-home glucose beverage ingestion to predict OGTT-2h≥140 mg/dL
Time frame: 2 Hour Point of Home-OGTT
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| CGM Use for T2D Risk Evaluation | Positive Predictive Value of At-home, CGM-measured Glucose Challenge | 18.2 percent accuracy of positive tests |
95% CI: [0.18, 0.94]
Secondary
Sensitivity of At-home CGM-measured Fasting Glucose
Sensitivity of baseline CGM-glucose at the time of at-home glucose beverage ingestion to predict fasting OGTT ≥100 mg/dL
Time frame: 5 Minute Point of Home-OGTT
Population: Of the 35 participants completing the 2-hr CGM-OGTT, 31 had valid data for baseline fasting glucose.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| CGM Use for T2D Risk Evaluation | Sensitivity of At-home CGM-measured Fasting Glucose | 100 percentage of true positives |
Secondary
Sensitivity of At-home CGM-measured Glucose Challenge
Sensitivity of CGM-glucose 2 hours after at-home glucose beverage ingestion to predict OGTT-2h≥140 mg/dL
Time frame: 2 Hour Point of Home-OGTT
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| CGM Use for T2D Risk Evaluation | Sensitivity of At-home CGM-measured Glucose Challenge | 80 percentage of true positives |
95% CI: [0.18, 0.94]
Secondary
Specificity of At-home CGM-measured Fasting Glucose
Specificity of baseline CGM-glucose at the time of at-home glucose beverage ingestion to predict fasting OGTT ≥100 mg/dL
Time frame: 5 Minute Point of Home-OGTT
Population: Of the 35 participants completing the 2-hr CGM-OGTT, 31 had valid data for baseline fasting glucose.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| CGM Use for T2D Risk Evaluation | Specificity of At-home CGM-measured Fasting Glucose | 10 percentage of true negatives |
Secondary
Specificity of At-home CGM-measured Glucose Challenge
Specificity of CGM-glucose 2 hours after at-home glucose beverage ingestion to predict OGTT-2h≥140 mg/dL
Time frame: 2 Hour Point of Home-OGTT
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| CGM Use for T2D Risk Evaluation | Specificity of At-home CGM-measured Glucose Challenge | 25 percentage of true negatives |
95% CI: [0.18, 0.94]
Other Pre-specified
Laboratory-measured 2-hour Glucose
2-hour OGTT glucose will be measured in the laboratory at baseline to investigate associations with CGM indices obtained during at-home glucose challenge.
Time frame: 2 Hour Point of Lab-OGTT
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| CGM Use for T2D Risk Evaluation | Laboratory-measured 2-hour Glucose | 112 mg/dL |
Comparison: Correlation analyses were performed to assess the relationship between home-OGTT fasting glucose and lab-OGTT 2-hour glucose.p-value: 0.6Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between home-OGTT 2-hour glucose and lab-OGTT 2-hour glucose.p-value: 0.1Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between delta (difference between fasting and postprandial) home-OGTT glucose and lab-OGTT 2-hour glucose.p-value: 0.2Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between home-OGTT peak glucose and lab-OGTT 2-hour glucose.p-value: 0.2Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between time of home-OGTT peak glucose and lab-OGTT 2-hour glucose.p-value: 0.2Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between CGM mean glucose and lab-OGTT 2-hour glucose.p-value: 0.08Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between CGM coefficient of variability (CV) and lab-OGTT 2-hour glucose.p-value: 0.4Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between CGM standard deviation (SD) and lab-OGTT 2-hour glucose.p-value: 0.1Spearman correlation
Other Pre-specified
Laboratory-measured Fasting Glucose
Fasting glucose will be measured in the laboratory at baseline to investigate associations with CGM indices obtained during at-home glucose challenge.
Time frame: Start of Lab-OGTT
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| CGM Use for T2D Risk Evaluation | Laboratory-measured Fasting Glucose | 84 mg/dL |
Comparison: Correlation analyses were performed to assess the relationship between home-OGTT fasting glucose and laboratory-OGTT fasting glucose.p-value: 0.2Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between home-OGTT 2-hour glucose and lab-OGTT fasting glucose.p-value: 0.3Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between delta (difference between fasting and postprandial) home-OGTT and lab-OGTT fasting glucose.p-value: 0.6Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between home-OGTT peak glucose and lab-OGTT fasting glucose.p-value: 0.3Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between time of home-OGTT peak glucose and lab-OGTT fasting glucose.p-value: 0.6Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between CGM mean glucose and lab-OGTT fasting glucose.p-value: 0.05Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between CGM coefficient of variability (CV) and lab-OGTT fasting glucose.p-value: 0.3Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between CGM standard deviation (SD) and lab-OGTT fasting glucose.p-value: 0.7Spearman correlation
Other Pre-specified
Laboratory-measured HbA1c
HbA1c will be measured in the laboratory at baseline to investigate associations with CGM indices obtained during at-home glucose challenge.
Time frame: Measured at Research Visit
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| CGM Use for T2D Risk Evaluation | Laboratory-measured HbA1c | 5.7 percent |
Comparison: Correlation analyses were performed to assess the relationship between home-OGTT fasting glucose and HbA1c.p-value: 0.3Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between home-OGTT 2-hour glucose and HbA1c.p-value: 0.005Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between delta (difference between fasting and postprandial) home-OGTT glucose and HbA1c.p-value: 0.01Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between home-OGTT peak glucose and HbA1c.p-value: 0.2Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between time of home-OGTT peak glucose and HbA1c.p-value: 0.5Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between CGM mean glucose (mg/dL) and HbA1c.p-value: 0.1Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between CGM correlation of variability (CV) and HbA1c.p-value: 0.4Spearman correlation
Comparison: Correlation analyses were performed to assess the relationship between CGM standard deviation and HbA1c.p-value: 0.7Spearman correlation