|
|
| General |
| Study Status |
Completed |
Application Number /
Requirement Number |
P120005 S041/ PAS001 |
| Date Original Protocol Accepted |
12/20/2016
|
| Date Current Protocol Accepted |
01/09/2024
|
| Study Name |
Dexcom G5 Mobile Cont Glucose Monitoring Sys -PAS
|
| Device Name |
DEXCOM G5 MOBILE CONTINUOUS GLUCOSE MONITORING SYSTEM
|
| General Study Protocol Parameters |
| Study Design |
Prospective Cohort Study
|
| Data Source |
New Data Collection
|
| Comparison Group |
Device Subjects Serve as Own Control
|
| Analysis Type |
Analytical
|
| Study Population |
Child: 2 yrs < 12 yrs,
Adolescent: 13-18 yrs,
Transit.Adolescent B(as adults): 18 yrs < 22 yrs,
Adult: At least 22 yrs
|
| Detailed Study Protocol Parameters |
| Study Objectives |
This is a prospective, observational, non-inferiority, adaptive design study intended to evaluate the safety of non-adjunctive CGM use in CGM-naïve adult and pediatric subjects, by assessing the number of hypoglycemic and diabetic ketoacidosis (DKA) events with use of CGM. The sponsor has defined hypoglycemia as (1) severe hypoglycemia resulting in a seizure or loss of consciousness or (2) moderate hypoglycemia defined as the inability to self-treat due to weakness or confusion to take carbohydrates, requiring assistance by a third party. The sponsor defined DKA as requiring an emergency room (ER) visit or hospitalization.
The study has two cohorts:
(1) Cohort A, enrolling adults and pediatric subjects, includes a baseline period of 6 months using a blood glucose meter (BGM) as a comparator. Subjects will perform self-monitored blood glucose (SMBG) testing using their personal BGM. Each subject will serve as their own control defined as the subject’s first 6 months in the study with diabetes management using SMBG. The second six months comprise the intervention period during which subjects will use CGM to manage their diabetes. Cohort A subjects will be in the study period for approximately 12 months and will have approximately 3 clinic visits and 10 monthly telephone calls.
(2) Cohort B, enrolling pediatric subjects, includes only an intervention period of 6 months of study participation, using a CGM. Cohort B participants cannot serve as their own controls because they will not participate in a BGM period. These subjects will be included in all analyses that do not require a control. In order to ensure the safety results are the same for pediatric subjects in Cohort A and those in Cohort B subjects, additional sensitivity analyses will be performed to compare the safety outcomes during the CGM phase for Cohort A and Cohort B to ensure they are similar. Cohort B subjects will be in the study period for approximately 6 months and will have approximately 2 clinic visits and 5 monthly telephone calls.
|
| Study Population |
Adult and pediatric subjects, less than or equal to 2 years of age, with type 1 diabetes (T1D) or insulin-requiring type 2 diabetes (T2D), naïve to CGM (either naïve to real-time CGM or may have used professional CGM). Subpopulation enrollment is targeted for 1/3 pediatric (2/3 in Cohort A and 1/3 in Cohort B) and 2/3 adult subjects (all in Cohort A), with approximately 10% of adult participants greater than or equal to 65 years of age.
|
| Sample Size |
1110 participants are required to demonstrate non-inferiority of CGM to SMBG. Target enrollment specified up to 1,388 subjects (or more depending on the sample size re-estimation procedure) to allow for an assumed 20% lost to follow-up.
|
| Key Study Endpoints |
Primary Endpoint: The primary endpoint for this clinical study is to assess change in the average number of hypoglycemic events per participant after 6-months of CGM use compared to 6-months of BGM use in Cohort A.
Secondary endpoints: The secondary endpoints for this clinical study will be to assess patient reported outcomes, clinical measurements, and CGM metrics after 6-months of CGM use compared with 6-months of baseline data with BGM use in Cohort A. For Cohort B, secondary analyses will assess outcomes after 6-months of CGM use compared with the baseline status collected at the initial visit.
Additional analyses include, but are not limited to the following:
Clinical Measurements
1. A1C change, overall and across the A1C range
2. Change in percentage of participants with at least one (1) hypoglycemia event
3. Change in the percentage of participants with at least one (1) diabetic ketoacidosis event
4. Percentage of participants with increased and decreased number of hypoglycemia events
5. Change in the number & proportion of hypoglycemia events stratified by day versus night
Patient Reported Outcomes
1. Glucose Monitoring Satisfaction Survey (GMSS)
2. CGM Satisfaction
3. Diabetes Distress Scale (DDS)
4. Hypoglycemia Fear Survey-Worry subscale (HFS-W)
5. Hypoglycemia Confidence
CGM Metrics-comparing Month 7 (after CGM initiated) with Month 12:
1. Average of CGM Glucose
2. Standard Deviation (SD) of CGM glucose
3. Percentage of CGM glucose within range: 70-180 mg/dL
4. Percent Time-in-Hypoglycemia (<70 mg/dL)
5. Percentage of CGM glucose in Hypoglycemia: (<55 mg/dL)
6. Percent Time-in-Hyperglycemia (>250 mg/dL)
7. Percentage of CGM glucose in Hyperglycemia (>300 mg/dL)
Other Metrics
1. CGM adherence: frequency of CGM use at month 12 compared to month 7 of CGM use
2. Change in SMBG frequency after initiation of CGM compared to baseline period, defined aa time point when BGM or CGM is established (months 3-6 for BGM and months 9-12 for CGM)
3. Change in the numbers & proportions of hypoglycemic events stratified by sensor location (used in approved wear locations or off-label)
4. A subpopulation analyses will be conducted at the end of the study, including but not limited to: Type 1 diabetes, Type 2 diabetes, adult, pediatric, and elderly.
In addition, the safety outcomes for the additional outcomes listed above will be repeated to compare the outcomes for pediatric subjects in Cohort A and those in Cohort B.
|
| Follow-up Visits and Length of Follow-up |
Follow-up Visits and Length of Follow-up Subjects in Cohort A will be in the study period for approximately 12 months and will have approximately 3 clinic visits and monthly telephone calls (total of 10). Subjects in Cohort B will be in the study period for approximately 6 months and will have approximately 2 clinic visits and monthly telephone calls (total of 5).
|
| Interim or Final Data Summary |
| Actual Number of Patients Enrolled |
1,092 subjects were enrolled, of which 1,088 met all inclusion/exclusion criteria and were included in the Intent-to-Treat (ITT) population.
|
| Actual Number of Sites Enrolled |
27
|
| Patient Follow-up Rate |
A total of 1,092 subjects were enrolled across the two cohorts of the study. The sponsor had an overall enrollment goal of 1,388 participants (allowing an assumed 20% loss to follow-up), targeted for 1/3 pediatrics (2/3 in Study Cohort A and 1/3 in Study Cohort B) and 2/3 adults (all in Study Cohort A), with approximately 10% of participants targeted to be 65 years of age or older. The study's overall follow-up rate was 83.2%. Across all cohorts, there were 184 subject withdrawals, the most common reasons being lost to follow-up (n=42, 3.8%), non-compliance (n=40, 3.7%), and other (n=68, 6.2%). Follow-up is summarized by study cohort below.
Cohort A: A total of 911 subjects enrolled in the study and met all inclusion criteria and none of the exclusion criteria. Of these subjects, a total of 743 completed the 12-month final follow-up visit (519 adults and 224 pediatric participants). The follow-up rate was 83.8% for adults and 78.0% for pediatric participants.
Cohort B: A total of 181 subjects were enrolled in the study and met all inclusion criteria and none of the exclusion criteria. Of these subjects, a total of 161 completed the 6-month final follow-up visit. The follow-up rate was 89.0%.
|
| Final Safety Findings |
Adults: There was a total of 975 adverse events (AEs) in the ITT adult population (446 in the SMBG period and 529 in the CGM period), of which 17 were device-related. For adults, the majority of AEs (399/446 = 89.5% in the SMBG period and 489/529 = 92.4% in the CGM period) were disease-related hypoglycemia. The majority of AEs were mild in the SMBG period (360/446 = 80.7%) and in the CGM period (488/529 = 92.2%). The percentage of participants with hypoglycemia (of any severity) was relatively stable across the two periods (26.8% in the SMBG period and 28.6% in the CGM period). The percentage of participants with DKA was very small in each period (0.7% in the SMBG period and 0.2% in the CGM period). Across the two periods, there were 5 DKA events (4 in the SMBG period and 1 in the CGM period) and 888 hypoglycemia events (399 in the SMBG period and 489 in the CGM period). Of the 17 device related AEs in the CGM period, most were insertion site AEs (13/17=76.5%). There were 2 deaths, both of which were determined not to be device-related.
Pediatrics (COHORT A): There were 511 AEs in the ITT pediatric Cohort A population (262 in the SMBG period and 249 in the CGM period), of which 8 were device-related. The majority of AEs were mild in the SMBG period (204/262 = 77.9%) and in the CGM period (205/249 = 82.3%). The percentage of patients with hypoglycemia (of any severity) was 23.5% in the SMBG period and 30.5% in the CGM period. The percentage of patients with DKA was 5.7% in the SMBG period and 6.6% in the CGM period. Across the two periods, there were 43 DKA events (22 in the SMBG period and 21 in the CGM period) and 406 hypoglycemia events (210 in the SMBG period and 196 in the CGM period). Of the 8 device related AEs in the CGM period, 4 were insertion site AEs, 3 were hypoglycemia, and 1 was a sensor/catheter or introducer needle fracture and retained in situ (characterized as other).
Pediatrics (COHORT B): There were 110 AEs in the ITT pediatric Cohort B population. There was a similar amount of mild and moderate AEs (53 mild, 52 moderate, and 5 severe). There were 4 DKA events and 3 hypoglycemia events reported. There were 15 device-related AEs, the majority of which were insertion site AEs, with the remaining categorized as "other".
There were no reports of serious unanticipated adverse device effects in any cohort and no reports of deaths in any pediatric cohort.
|
| Final Effect Findings |
The primary endpoint was based on Cohort A and met. A modified statistical methodology from the approved protocol was used for this evaluation. All secondary endpoint analyses were exploratory and not adjusted for multiple hypothesis tests.
Cohort A Adults: The Hypo event rate over a 6-month period decreased by 62% (95% CI: 22%, 81%) from 0.081 in the SMBG period to 0.031 in the CGM period. CGM was shown to be noninferior to SMBG (p-value = 0.0018). The percentage of adult participants with at least one Hypo event decreased from 5.6% in the SMBG period to 2.3% in the CGM period, and the percentage of participants with at least one DKA event decreased from 0.4% in the SMBG period to 0.0% in the CGM period.
Cohort A Pediatrics: The Hypo event rate decreased by 53% (95% CI: 12%, 75%) from 0.076 in the 6-month SMBG period to 0.036 in the 6 month CGM period. CGM was shown to be non-inferior to SMBG (p value = 0.004). The percentage of Cohort A pediatric participants with at least one Hypo event decreased from 5.4% in the SMBG period to 3.1% in the CGM period, though the percentage of participants with at least one DKA event increased from 4.5% in the SMBG period to 5.4% in the CGM period.
Cohort B Pediatrics: There was no 6-month SMBG wear period, so the comparison of the CGM period to the Initial visit may not be representative of the difference between SMBG and CGM use. The change in rates (CGM period minus Initial visit) is reported. The Hypo event rates were similar in the two periods (Initial visit = 0.000 and CGM = 0.006), with just one additional Hypo event in the CGM period. The percentages of participants with at least one Hypo event based on recall and in the CGM period were similar as well (Initial visit = 0.0% and CGM = 0.6%). However, the DKA rate decreased by 95% from 0.342 at the Initial visit to 0.019 with CGM. Further, the percentage of participants with at least one DKA event decreased from 34.2% to 1.9%.
|
| Study Strengths & Weaknesses |
Strengths: This post-approval study to evaluate the non-adjunctive use of Dexcom G5, G6, or G7 continuous glucose monitoring systems (CGM) included CGM-naïve pediatric and adult users and had a control phase (Cohort A with SMBG for diabetes management). The average number of hypoglycemia events per participant over a 6-month period was decreased in the CGM period compared to the preceding SMBG control period, and noninferiority of CGM to SMBG was demonstrated in adults and pediatric subjects in Cohort A. The post-approval study confirms that the Dexcom G5 CGM is safe for non-adjunctive use in CGM-naïve users.
Weaknesses: The study did not meet its sample size requirement for the primary endpoint evaluation despite multiple efforts from the sponsor to increase enrollment. This was due to, at least in part, the protocol changes to remove the SMBG phase in Cohort B to address the pediatric enrollment challenges. However, available data from this post-approval study support safety of the Dexcom G5 CGM when used as intended.
|
| Recommendations for Labeling Changes |
Results of the post-approval study do not change the benefits and risks conclusions of the original PMA approval. Additionally, the PMA device (Dexcom G5 CGM) is no longer being marketed. Therefore, labeling changes are not recommended.
|
