MiniMed 670G System

P160017S031 · Medtronic Minimed, Inc. · OZP · Jun 21, 2018 · Clinical Chemistry

Device Facts

• Record ID: P160017S031
• Device Name: MiniMed 670G System
• Applicant: Medtronic Minimed, Inc.
• Product Code: OZP · Clinical Chemistry
• Decision Date: Jun 21, 2018
• Decision: APPR
• Device Class: Class 3
• Attributes: Therapeutic, Real-World Evidence, Pediatric

Real-World Evidence

Submission	Device	Sponsor	RWD Sources	RWE Use Summary	Key Tags
P160017S031 · Jun 21, 2018	MiniMed 670G System	Medtronic Minimed, Inc.	Medical records/source documentation for clinical diagnosis; Post-market adverse event reports for similar CGM devices	Medical records were used to verify clinical diagnosis of Type 1 diabetes for study enrollment. Post-market experience with similar devices was used to support the safety profile and expected adverse event rates for the CGM component.	Medical record verification; Post-market surveillance; Safety profile support

Clinical Evidence

Study Design	Population	Comparator	Key Endpoints
Safety Evaluation of the Hybrid Closed Loop (HCL) System in Pediatric Subjects with Type 1 Diabetes (G150247); Multi-center, single-arm, home and hotel clinical study; Follow-up/Duration: 3.5 months; Study Period: April 18, 2016 to October 09, 2017	Pediatric subjects ages 7-13 years with Type 1 diabetes; Sample Size: 105; Number of Sites: 9	Not applicable for this study	Safety (SAE, SADE, UADE, severe hypoglycemia, DKA) and descriptive effectiveness (HbA1c, TDD, time in range)
A Performance Evaluation of the Enlite® and Enlite 3 Glucose Sensor to Support Use in Children; Phase 2 (Enlite 3) (G120262); Prospective, single arm, multi-center, in-clinic study; Follow-up/Duration: 7-day training period followed by 7-day study period; Study Period: May 12, 2015 to December 13, 2016	Pediatric subjects ages 7-13 years; Sample Size: 50; Number of Sites: 11	Yellow Springs Instrument 2300 Stat Plus Glucose Lactate Analyzer	Sensor accuracy and precision

Indications for Use

The Medtronic MiniMed 670G system is intended for continuous delivery of basal insulin (at user selectable rates) and administration of insulin boluses (in user selectable amounts) for the management of Type 1 diabetes mellitus in persons, seven years of age and older, requiring insulin as well as for the continuous monitoring and trending of glucose levels in the fluid under the skin. The MiniMed 670G System includes SmartGuard technology, which can be programmed to automatically adjust delivery of basal insulin based on Continuous Glucose Monitor sensor glucose values, and can suspend delivery of insulin when the sensor glucose value falls below or is predicted to fall below predefined threshold values. The Medtronic MiniMed 670G System consists of the following devices: MiniMed 670G insulin pump, the Guardian Link (3) Transmitter, the Guardian Sensor (3), One-Press Serter, and the Contour NEXT Link 2.4 Glucose Meter. The system requires a prescription. The Guardian Sensor (3) glucose values are not intended to be used directly for making therapy adjustments, but rather to provide an indication of when a finger stick may be required. All therapy adjustments should be based on measurements obtained using a home glucose monitor and not on values provided by the Guardian Sensor (3).

Device Story

MiniMed 670G System is a hybrid closed-loop (HCL) insulin delivery system; manages Type 1 diabetes in patients ≥7 years. System components: MMT-1780 insulin pump; Guardian Sensor (3) (interstitial glucose monitor); Guardian Link (3) transmitter; Contour NEXT Link 2.4 blood glucose meter. Input: interstitial glucose signals (every 5 min) via RF telemetry; manual blood glucose meter inputs for calibration/bolus calculations. Operation: SmartGuard technology adjusts basal insulin delivery automatically (Auto Mode) or suspends insulin based on low/predicted-low glucose thresholds (Manual Mode). Output: automated basal insulin delivery; real-time glucose trends/alerts on pump display. Used in home settings; operated by patients/caregivers. Healthcare providers use downloaded CareLink data to track patterns/adjust therapy. Benefits: improved glycemic control; reduced hypoglycemia/hyperglycemia; automated insulin adjustments reduce burden of manual dosing.

Clinical Evidence

Evidence includes a pivotal multi-center, single-arm, home/hotel study (G150247) of 105 pediatric subjects (7-13 years) and a sensor performance study (G120262) of 50 subjects. Primary endpoints were descriptive safety and system performance. Results showed HbA1c decreased from 7.9% to 7.5% (p-value not reported). Time in range (71-180 mg/dL) increased from 56.16% to 65.01%. No severe hypoglycemia or DKA events were reported in the pivotal study. Sensor accuracy (MARD) was 11.73% (abdomen) and 9.14% (buttock) with 12-hour calibration.

Technological Characteristics

System includes an ambulatory micro-infusion pump (MMT-1780), glucose transmitter (MMT-7811), and sterile single-use glucose sensor (MMT-7020). Sensing principle: electrochemical glucose oxidase. Energy source: battery-operated pump and transmitter. Connectivity: RF telemetry between sensor/transmitter and pump; USB/CareLink for data download. Software: proprietary SmartGuard algorithm (hybrid closed-loop). Sterilization: sterile sensor components.

Indications for Use

Indicated for management of Type 1 diabetes mellitus in persons 7 years of age and older requiring insulin. Contraindicated in patients under 7 years old, patients requiring <8 units of insulin per day, patients unable to perform ≥4 blood glucose tests daily, patients unable to maintain contact with healthcare professionals, or patients unable to recognize pump alarms.

Regulatory Classification

Identification

An insulin-only automated insulin dosing system that achieves a target glucose level by automatically increasing or reducing insulin infusion from an insulin pump. Insulin adjustments are based upon specified thresholds of measured glucose levels.

Predicate Devices

• MiniMed 530G System (P120010)
• MiniMed 630G System (P150001)

Reference Devices

• Enlite Sensor (P120010)
• Contour Next Link 2.4 Wireless Blood Glucose Monitoring System (K110894)

Submission Summary (Full Text)

{0}

# SUMMARY OF SAFETY AND EFFECTIVENESS DATA (SSED)

## I. GENERAL INFORMATION

Device Generic Name: Automated Insulin Dosing Device System, Single Hormone

System Device Trade Name: MiniMed 670G System

Device Procode: OZP

Applicant's Name and Address: Medtronic MiniMed, Inc.
18000 Devonshire Street
Northridge, CA 91325

Date(s) of Panel Recommendation: Not applicable

Premarket Approval Application (PMA) Number: P160017/S031

Date of FDA Notice of Approval: June 21, 2018

Breakthrough Device: Granted breakthrough device status (formerly known as the Expedited Access Pathway, or EAP) on June 21, 2018 because the Medtronic MiniMed 670 System will provide more effective management of type 1 diabetes, an irreversibly debilitating disease.

The 670G System was previously approved for use in ages 14 years and up and was approved on September 28, 2016. The SSED for the original approval (P160017) can be found on the CDRH website. The current supplement was to expand the indications for the device to include a user population of 7 to 13 years of age.

## II. INDICATIONS FOR USE

## I. GENERAL INFORMATION

The Medtronic MiniMed 670G system is intended for continuous delivery of basal insulin (at user selectable rates) and administration of insulin boluses (in user selectable amounts) for the management of Type 1 diabetes mellitus in persons, seven years of age and older, requiring insulin as well as for the continuous monitoring and trending of glucose levels in the fluid under the skin. The MiniMed 670G System includes SmartGuard technology, which can be programmed to automatically adjust delivery of basal insulin based on Continuous Glucose Monitor sensor glucose values, and can suspend delivery of insulin when the sensor glucose value falls below or is predicted to fall below predefined threshold values.

The Medtronic MiniMed 670G System consists of the following devices:

MiniMed 670G insulin pump, the Guardian Link (3) Transmitter, the Guardian Sensor (3), One-Press Serter, and the Contour NEXT Link 2.4 Glucose Meter. The system requires a prescription.

The Guardian Sensor (3) glucose values are not intended to be used directly for making therapy adjustments, but rather to provide an indication of when a finger stick may be required. All therapy adjustments should be based on measurements obtained using a home glucose monitor

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 1

{1}

and not on values provided by the Guardian Sensor (3).

#### Guardian Sensor (3)

The Guardian Sensor (3) is intended for use with the MiniMed™ 670G system to continuously monitor glucose levels in persons with diabetes. It is intended to be used for detecting trends and tracking patterns, and to be used by the MiniMed 670G system to automatically adjust basal insulin levels. It is indicated for use as an adjunctive device to complement, not replace, information obtained from standard blood glucose monitoring devices. The sensor is intended for single use and requires a prescription. The Guardian Sensor (3) is indicated for 7 days of continuous use.

#### One-press Serter

The One-press Serter is used as an aid for inserting the sensor. It is indicated for single-patient use and it is not intended for multiple-patient use.

#### Guardian Link (3) Transmitter

The Guardian Link (3) Transmitter is intended for use with the MiniMed 670G System. The Guardian Link (3) Transmitter powers the glucose sensor, collects and calculates sensor data, and wirelessly sends the data to the MiniMed 670G insulin pump. The Transmitter is intended for single-patient multi-use.

#### Contour NEXT Link 2.4 Glucose Meter

The Contour Next Link 2.4 Wireless Blood Glucose Monitoring System is an over the counter (OTC) device utilized by persons with diabetes in home settings for the measurement of glucose in whole blood, and is for single patient use only and should not be shared. The Contour Next Link 2.4 wireless blood glucose monitoring system is indicated for use with fresh capillary whole blood samples drawn from the fingertip and palm only.

The Contour NEXT Test Strips are intended for self-testing by persons with diabetes for the quantitative measurement of glucose in whole blood samples from 20 to 600 mg/dL. The Contour Next Link 2.4 wireless blood glucose monitoring system is intended to be used to transmit glucose values to the MiniMed 670G pump and facilitate transfer of information to Medtronic CareLink Software through the use of radio frequency communication. The Contour Next Link 2.4 Wireless Blood Glucose Monitoring System is not intended for the diagnosis of, or screening for, diabetes mellitus. It is not intended for use on neonates.

### III. CONTRAINDICATIONS

A prominent boxed warning is included in the labeling regarding use of the device in subjects under the age of 7 years as follows:

“Medtronic performed an evaluation of the 670G closed loop system and determined that it may not be safe for use in children under the age of 7 because of the way that the system is designed and the daily insulin requirements. Therefore this device should not be used in anyone under the age of 7 years old. This device should also not be used in patients who require less than a total daily insulin dose of 8 units per day because the device requires a minimum of 8 units per day to operate safely.”

The following contraindications for this device are also described in the labeling:

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 2

{2}

- Insulin pump therapy is not recommended for people who are unwilling or unable to perform a minimum of four blood glucose tests per day. As insulin pumps use rapid acting insulin only, blood glucose testing is required to help identify rapid glycemic deterioration due to insulin infusion occlusion, infusion site problems, insulin stability issues, user error, or a combination of these.
- Pump therapy is not recommended for people who are unwilling or unable to maintain contact with their healthcare professional.
- Pump therapy is not recommended for people whose vision or hearing does not allow recognition of pump signals and alarms.
- Do not use serter on products other than the Enlite Sensor (P120010) or Guardian Sensor (3). Medtronic cannot guarantee the safety or efficacy of this product if used with other products.
- The reservoir is contraindicated for the infusion of blood or blood products.
- Infusion sets are indicated for subcutaneous use only and not for intravenous (IV) infusion or the infusion of blood or blood products.

# IV. WARNINGS AND PRECAUTIONS

The warnings and precautions can be found in the MiniMed 670G System User Guide labeling.

# V. DEVICE DESCRIPTION

There is no physical change to the MiniMed 670G System, previously approved under P160017, as a result of this panel track supplement. This current panel track supplement supports the expansion of the indicated population to include 7-13 year old people with type 1 diabetes.

The MiniMed 670G system is comprised of the following devices:

# MiniMed 670G Pump (MMT-1780)

The MiniMed 670G pump (model MMT-1780) is an ambulatory, battery operated, rate programmable micro infusion pump designed to deliver insulin from a reservoir. The reservoir is driven by a motor to deliver determined basal rate profiles and user selected bolus amounts of insulin into the subcutaneous tissue through an infusion set.

The MiniMed 670G pump is offered in one model (MMT-1780). The pump houses electronics, a pumping mechanism, a user interface, and a medication reservoir within the same physical device. The reservoir is attached to a tube that connects to the user's infusion site on their body. It is intended to deliver insulin through a diffusion mechanism. Model MMT-1780 is compatible with a 3.0 mL reservoir. The pump only displays blood glucose level units in mg/dL and cannot be reconfigured by the user.

In addition to its delivery of insulin, the MiniMed 670G pump is designed to receive and display real-time interstitial fluid glucose values via the Guardian Link (3) Transmitter. When used in combination with Guardian Sensor (3), the transmitter sends sensor signals to the MiniMed 670G pump via radiofrequency (RF) telemetry. The 670G Pump has the following features and capabilities:

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 3

{3}

- Provides predictive sensor glucose alerts when sensor glucose values are high or low (please see 'Manual Mode' section below for details).
- Can receive blood glucose values from the Contour NEXT Link 2.4 Meter to use for sensor calibration.
- The pump can display Guardian Sensor (3) glucose values in real-time, and also store those values (and blood-glucose values from the meter) into its pump memory.
- "SmartGuard" Technology: There are two levels of this technology; the first is available in Manual Mode and the second in Auto Mode.

o The first level of SmartGuard technology is available in Manual Mode:

- This technology automatically suspends insulin when the sensor reaches a preset low limit (referred to as 'Suspend on Low')
- This technology automatically suspends insulin when the sensor glucose value is predicted, using a proprietary predictive suspend algorithm, to reach a pre-set low limit, i.e., suspends before the low limit is reached (referred to as 'Suspend before Low').
- When a Suspend event occurs, the user can choose to continue to keep insulin suspended, or the user can choose to resume insulin delivery.
- Following a Suspend event, insulin delivery will automatically resume when the sensor glucose levels rise above the pre-set suspend threshold.
- The 'Suspend on low' and 'Suspend before low' features are optional features available when the system is in Manual Mode.
- This technology (in Manual Mode) provides a bolus calculator called the 'Bolus Wizard' that uses your settings to calculate an estimated bolus amount based on the meter blood glucose values and carbohydrates that the user enters. These settings should be set up with the help of a Health Care Practitioner before using the Bolus Wizard. Those settings include:

- Carbohydrate Ratio
- Insulin Sensitivity Factor
- Blood Glucose Target
- Active Insulin Time

o The second level of SmartGuard technology is available in Auto Mode:

- This technology automatically adjusts basal insulin delivery using continuous glucose monitor data, referred to as Auto Mode.
- The Auto Mode feature can automatically increase or decrease the amount of insulin delivered based on sensor glucose values.

During Auto Mode operation, the user must manually deliver meal boluses that they calculate using the estimated amount of carbohydrates for meals at the time they are eaten. The user must also use the Auto Mode bolus feature to calculate boluses while in Auto Mode.

The MiniMed 670G Insulin Pump is designed to receive and display real-time glucose values received from the Guardian Link (3) transmitter. Guardian sensor (3) signals are

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 4

{4}

transmitted from the transmitter to the MiniMed 670G Insulin Pump via RF telemetry and converted into glucose concentrations based on calibration values from the Contour Next Link 2.4 meter. Signals are updated and transmitted to the pump every five minutes.

The real time sensor glucose values, displayed by the MiniMed 670G Insulin Pump, are not intended to be used directly for making therapy adjustments. The user can use the tracking and trending of sensor glucose values to help determine if an unplanned finger stick measurement may be needed. In addition, sensor glucose values should not be used to modify insulin therapy. All manual insulin therapy adjustments should be based on measurements obtained using a blood glucose meter and not based on the sensor glucose value displayed by the MiniMed 670G Insulin Pump.

#### Manual Mode

The user can set the pump to temporarily suspend insulin delivery automatically when the sensor glucose level is equal to or less than a selected threshold. The user has the capability to select a 'Suspend on Low' threshold within the 50 mg/dL to 90 mg/dL range. When the 'Suspend on Low' tool is set to 'ON', the system compares the sensor glucose value and the programmed Suspend threshold whenever the sensor glucose value is updated (every five minutes). If the sensor glucose value falls below the set threshold, insulin delivery will be suspended. Once the sensor glucose rises above that set threshold, insulin delivery will resume. The use of the 'Suspend on Low' tool is optional and the user can turn the tool 'ON' and 'OFF'.

The pump also includes the new Predictive Low Management tool that allows the user to set a glucose value threshold where the pump will suspend the insulin delivery if the blood glucose value is predicted to reach the selected threshold in the near future ('Suspend before Low'). The user can set their low sensor glucose threshold value from 50 mg/dL to 90 mg/dL; insulin delivery will suspend when the sensor glucose value is predicted to reach or fall below a level that is 20 mg/dL above the set low limit threshold value within approximately 30 minutes. The user has the ability to program the length of time in which suspension will occur pending the blood glucose threshold 30 minutes before the threshold is reached. Once the sensor glucose rises above that set threshold, insulin delivery will resume. The use of the 'Suspend on Low' tool is optional and the user can turn the tool 'ON' and 'OFF'.

When the sensor glucose value is below or predicted to be below the set threshold, an alarm and siren occurs and the pump suspends, and the user may elect to continue the suspend, or cancel the temporary pump suspension of insulin delivery at any time.

If the user does not respond to the 'suspend on low' or 'suspend before low' alarm or siren, the pump will automatically suspend for up to two hours.

If the user cancels the suspension of insulin delivery, the system will continue to deliver insulin at the programmed basal rate until the next time the sensor glucose value is below the set threshold value. The alarm and siren will then re-sound, and the pump will suspend (unless canceled by the user).

If the user responds to the alarm or siren by electing to accept the insulin suspension, the pump will suspend for at least 30 minutes and up to a maximum of 2 hours; the pump may resume insulin delivery if the sensor glucose value rises above the set low threshold after 30 minutes have passed and before or up to the 2 hours maximum suspension time. The user can manually

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 5

{5}

resume insulin at any time. At the end of the two-hour maximum, the pump will resume insulin delivery until the next sensor glucose value is below the set threshold suspend value.

The user can cancel the temporary pump suspension at any time during the two-hour period regardless if the suspension occurred because the user was not able to respond to the initial alarm or he/she accepted the suspension. The pump may also resume insulin delivery on its own based on sensor glucose values or predicted sensor glucose values (if the sensor glucose value indicates that the user has recovered from the set threshold low glucose value). The table below summarizes the suspend features.

*Table 1: Suspend on Low and Suspend before Low Feature and Related Resumption Options*

| What Happens | Suspend Features | |
| --- | --- | --- |
| | Suspend on Low | Suspend before Low |
| The pump suspends insulin delivery | The predetermined sensor blood glucose threshold is reached. | The predetermined sensor blood glucose threshold is predicted to be reached within 5 to 30 minutes (time period predetermined by |
| The user accepts the pump suspension | The pump will suspend for at least 30 minutes and up to a maximum of 2 hours. | The pump will suspend for at least 30 minutes and up to a maximum of 2 hrs. |
| The user responds to the insulin suspend and the system suspends insulin for two hours (maximum suspension time) | If insulin was suspended for two hours, then at the end of the two-hour maximum, the pump will resume insulin delivery. The pump will not suspend insulin again until after the refractory period* and the next sensor glucose value is below the set low glucose suspend value. | If insulin was suspended for two hours, then at the end of the two-hour maximum, the pump will resume insulin delivery. The pump will not suspend insulin again until after the refractory period* and the next sensor glucose value is predicted to go below the set low glucose suspend value. |
| The user cancels suspension | The system will resume insulin delivery at the programmed basal rate until the next time the sensor glucose value is below the set threshold value. | The system will resume insulin delivery at the programmed basal rate until the next time the sensor glucose value is below the set threshold value. |
| The user does not respond to the suspend alert | The pump will automatically suspend for up to two hours if the sensor glucose does not detect that the user's glucose values are or are predicted to go above the low glucose level. | The pump will automatically suspend for up to two hours if the sensor glucose does not detect that the user's glucose values are or are predicted to go above the low glucose level. |

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 6

{6}

| What Happens | Suspend Features | |
| --- | --- | --- |
| | Suspend on Low | Suspend before Low |
| The sensor glucose a glucose value above low glucose pre-set level | Pump may resume insulin delivery on its own after 30 minutes based on sensor glucose values glucose values. | Pump may resume insulin delivery on its own after 30 minutes based on sensor glucose values or predicted sensor glucose values. |

* After a Suspend event occurs, there is a period of time when the suspend functionality is unavailable (refractory period). This time will vary depending on whether or not the user responds to the Suspend event. Please see the labeling for the MiniMed 670G System regarding how the suspend functions work. The user can manually suspend insulin delivery at any time.

The MiniMed 670G Insulin Pump is capable of storing 90 days of pump history and glucose sensor data. The pump has a graphical display that the user can use to view the glucose history for the past 3, 6, 12 and 24 hours, high/low glucose alarms and display of retrospective glucose trend information.

Stored pump history and glucose data can be downloaded to a personal computer for review and analysis, to track patterns and improve diabetes management. Data is downloaded from the pump to CareLink therapy management software.

The MiniMed 670G pump is compatible with commercially available Medtronic Paradigm infusion sets and their 3 mL volume reservoirs. It was not necessary to develop new infusion sets or reservoirs for use with this pump.

#### Auto Mode

The MiniMed 670G Insulin Pump contains an Auto Mode feature; this new tool uses an algorithm to automatically adjust basal insulin delivery using continuous glucose monitor data. When in Auto Mode, the pump responds to fluctuations in interstitial fluid glucose levels measured by the continuous glucose monitor; the Auto Mode feature can automatically increase or decrease the amount of basal insulin delivered based on sensor glucose values.

Auto Mode does not administer meal boluses. During Auto Mode operation, users must deliver meal boluses by entering the amount of insulin they want to deliver based on the estimated amount of carbohydrates they are eating. Failure to deliver meal boluses in association with meals during Auto mode operation can result in significant post meal hyperglycemia.

The Auto Mode algorithm is designed to adjust the user's basal insulin rates to try to keep them at a target blood glucose level. The standard target glucose setting in Auto Mode is 120 mg/dL, and the target can also be set temporarily to 150 mg/dL for exercise and other events. In addition, blood glucose readings above 150 mg/dL will prompt the Auto Mode feature to calculate if a correction bolus is needed; if needed, a correction bolus will be recommended to the users, who can choose whether they want to deliver that correction

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 7

{7}

bolus. Users should check their blood glucose levels using a blood glucose meter before administering a correction bolus.

When first using the device, Auto Mode cannot be activated until the system completes a 48-hour warm-up period while the user uses the pump in manual mode to deliver insulin. In addition, before activation, the user must cancel any temporary basal rates, ensure delivery is not suspended, set a carbohydrate ratio, set high and low glucose settings, and enter a blood glucose value obtained by using a blood glucose meter if one has not been entered in the last 12 minutes. Carbohydrate ratios and high and low glucose settings should be discussed and established with the user's healthcare practitioner before use of Auto Mode begins.

There is an additional feature, called the Safe Basal feature, which will be activated in Auto Mode when the system encounters specific issues that the user should address; the user cannot manually enable Safe Basal as it is a mandatory automatic feature that activates when the pump encounters these types of issues. This feature is designed to be a safety net when in Auto Mode. The pump will transition into Safe Basal if the system detects the user is getting too little or too much basal insulin, if the pump detects an issue with the sensor or a discrepancy between blood glucose meter and sensor glucose readings, or if the pump has not received sensor glucose values in over 5 minutes. When the pump encounters an issue, it transitions to Safe Basal; in Safe Basal, the pump supplements the user's basal insulin needs by delivering a steady state basal rate, which allows the user time to perform the additional actions prompted by the pump that are required to ensure Auto Mode stays active. Safe Basal does not automatically adjust basal insulin and instead delivers a steady basal rate until the user has addressed the issue that triggered the pump to activate Safe Basal. After 90 minutes in Safe Basal, if the condition(s) that caused the pump to transition into Safe Basal have not been resolved, the pump will automatically exit Auto Mode and enter Manual Mode. If Auto Mode is active, then the 'Suspend on Low' and 'Suspend before Low' features from the 'Manual Mode' are unavailable and inactive. However, while in Auto Mode, basal insulin is adjusted by decreasing or suspending automatically based on continuous glucose monitor sensor values if the user is experiencing low glucose. The table below describes the available features and accessibility within each mode (Auto Mode and Manual Mode). Note that, although Safe Basal is not mentioned in the table below, it is only available in Auto Mode and cannot be activated by the user, as it is a mandatory automatic feature that activates when the pump encounters an issue that the user should address (see description above).

*Table 2: Modes and Related Accessibility*

| Mode | Description | When is it Active? | Will I receive Alerts? |
| --- | --- | --- | --- |
| Manual Mode: Sensor Augmented Pump | This mode is when the device is functioning as a sensor and pump, but the device is not in Auto Mode and the insulin suspend features are not turned on. | This is the default mode and the user does not have to specifically turn this mode on. | There is a mandatory severe low alarm at 50 mg/dL; The user can also set optional high and low alerts to sound on or before set sensor glucose levels. |

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 8

{8}

| Mode | Description | When is it Active? | Will I receive Alerts? |
| --- | --- | --- | --- |
| Manual Mode: Suspend On Low | When this feature is active the device detects that your sensor glucose level has reached a pre-set sensor glucose value and it automatically suspends basal insulin delivery when that value is reached. | The user has to turn this feature on. It is not available when Auto Mode is turned on, and it cannot be turned on if Suspend before Low is turned on. | There is a mandatory severe low alarm at 50 mg/dL and at the pre-set low level. The user can also set optional high alerts to sound on or before set sensor glucose levels, and an optional alert before low alert. |
| Manual Mode: Suspend Before Low | When this feature is active the device detects when your sensor glucose is predicted to reach a pre-set value and it automatically suspends basal insulin delivery before that value is reached. | The user has to turn this feature on. It is not available when Auto Mode is turned on, and it cannot be turned on if Suspend before Low is turned on. | There is a mandatory severe low alarm at 50 mg/dL and at the pre-set low level. The user can also set optional high alerts to sound on or before set sensor glucose levels, and an optional alarm before low alert. |
| Auto Mode | When this mode is active, the device can automatically adjust basal insulin by increasing, decreasing, or turning off basal insulin delivery based on sensor glucose levels. | The user has to turn this mode on and certain pre-defined conditions have to be met. | There is a mandatory severe low alarm at 50 mg/dL; The user can also set optional high and low alerts to sound on or before set sensor glucose levels. |

### **Guardian Link (3) Transmitter System (MMT-7811)**

The Guardian Link (3) Transmitter System consists of the Guardian Link (3) Transmitter (MMT-7811), the Charger (model MMT-7715), and the Tester (model MMT-7736).

The Guardian Link (3) Transmitter interfaces directly with the glucose-sensor assembly. The Guardian Link (3) Transmitter provides power to the glucose sensor, and measures the sensor signal current from the glucose sensor.

The sensor signal current is an electrical current level that is proportional to the glucose level in the user's subcutaneous interstitial fluid. The sensor signal current is converted to a digital signal, which is filtered to reduce noise artifacts. This digital signal is sent to the MiniMed 670G pump every 5 minutes, using radio frequency (RF).

### **Guardian Sensor (3) (MMT-7020)**

The Guardian Sensor (3) is a sterile, single-use, single patient glucose sensing component for continuous monitoring of glucose levels in the user's interstitial fluid, when inserted in the user's abdomen (or in the abdomen or the buttock for users 7-13 years old) for up to seven days. The Sensor is inserted into the subcutaneous tissue using the One-Press Serter and is taped to the user's skin. It connects to the Guardian Link (3) Transmitter, which in turn communicates with the MiniMed 670G Pump.

When making treatment decisions, such as determining insulin dose for meals, the MiniMed

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 9

{9}

670G continuous glucose monitor (CGM) values should not be used, as they are not intended to be used to make such treatment decisions. The MiniMed 670G continuous glucose monitor does not replace a blood glucose meter. Users should always use the values from a blood glucose meter for treatment decisions. Blood glucose values may differ from sensor glucose values. Using the sensor glucose readings for treatment decisions could lead to unwanted high or low blood glucose.

Users should calibrate the Guardian Sensor (3) at least every 12 hours using meter blood glucose values. Calibration is necessary for sensor function, and more frequent calibration can help to increase the accuracy of the sensor. The system requires a minimum of two calibrations per day, and four calibrations per day are recommended. The system is contraindicated for patients unwilling or unable to do frequent blood glucose meter measurements.

If the user obtains blood glucose values using the Contour Next Link 2.4 Meter, the user may transmit blood glucose values via radiofrequency to the MiniMed 670G pump to be used for sensor calibrations. If the user uses a different FDA cleared blood glucose meter to calibrate the Guardian Sensor (3), the user must manually input the blood glucose values into the pump to be used for sensor calibration. Additionally, users who use the Contour Next Link 2.4 should calibrate with values obtained using fingersticks; users should not use readings obtained from blood from alternative sites (e.g. palm).

### **One-Press Serter**

The One-Press serter is a sensor insertion device which aids the user in inserting the Guardian Sensor. The serter was also previously reviewed and approved under P120010/S070. The user must use the One-Press Serter in order to insert the Guardian Sensor (3).

### **Contour Next Link 2.4 Meter (MMT-1352 and MMT-1152) and Test Strips**

The Contour Next Link 2.4 Meter can be used with the MiniMed 670G system; the meter wirelessly sends blood glucose values to the insulin pump for sensor calibration via radiofrequency. The meter was also previously cleared under K110894. Specifications and performance requirements were established for the meter and evaluated as part of the class III MiniMed 670G System. The sponsor verified and validated the specifications and performance requirements of the meter for the MiniMed 670G System. The sponsor provided blood glucose meter specifications, rationale for requirements for the meter, and impact of error on the sensor, predictive low alerts, threshold glucose suspend, and the predictive low glucose management and hybrid closed loop features in the original PMA submission. The sponsor carried out error impact analysis in order to determine the lot release criteria for the meter test strips. Based on the information provided in the original PMA submission, P160017, the meter specifications meet the clinical needs of the MiniMed 670G system.

### **Additional System Devices**

The following additional accessory devices are compatible with the MiniMed 670G Insulin Pump:

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 10

{10}

*Table 3: Additional System Devices*

| Device | Model |
| --- | --- |
| **Reservoirs and Infusion Sets** | **Model Numbers** |
| MiniMed Quick Set Infusion Set | MMT-386, MMT-387, MMT-394, MMT-396, MMT-397, MMT-398, MMT-399 |
| MiniMed Silhouette Infusion Set | MMT-368, MMT-369, MMT-370, MMT-377, MMT-378, MMT-381, MMT-382, MMT-383, MMT-384 |
| MiniMed Mio Infusion Set | MMT-921, MMT-923, MMT-925, MMT-941, MMT-943, MMT-945, MMT-965, MMT-975 |
| MiniMed Sure-T Infusion Set | MMT-862, MMT-864, MMT-866, MMT-874, MMT-876, MMT-886 |
| Paradigm Reservoir | MMT-332A |
| **Optional Devices** | **Model Numbers** |
| CareLink USB 2.4 | MMT-7306 |
| CareLink Online (Personal) | MMT-7333 |
| CareLink Pro | MMT-7335 |

## **VI. ALTERNATIVE PRACTICES AND PROCEDURES**

Control of diabetes can be achieved through a combination of various behaviors and methods.

Self-behaviors include healthy eating, taking the clinically indicated medications, and being active. Persons with diabetes may also administer insulin by injection or using other insulin infusion pumps as prescribed by their physician. An insulin pump is an alternative to multiple daily insulin injections (via insulin syringe or an insulin pen). Periodic self-glucose monitoring using home use blood glucose meters provides information regarding variations in glucose levels.

Methods of monitoring glycemic control include periodic measurement of Hemoglobin A1c (HbA1c) which reflects blood glucose control over a three-month period. Self-monitoring of blood glucose using glucose meters and test strips provides quantitative measurements of blood glucose at a single point in time for users and their healthcare providers. This helps to monitor the effectiveness of glycemic control, as well as make more immediate treatment modifications.

Currently, cleared or approved insulin infusion pumps may be used for continuous subcutaneous insulin infusion. Additionally, commercially available sensor-augmented insulin infusion pumps or continuous glucose monitoring systems may be used to record continuous interstitial glucose information and provide real-time hypoglycemia and hyperglycemia alerts.

Each alternative method for monitoring glycemic control has its own advantages and disadvantages. A user should fully discuss these alternatives with his/her physician to select the method that best meets expectations and lifestyle.

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 11

{11}

## **VII. MARKETING HISTORY**

The MiniMed 670G System was approved for marketing in the United States in September 2016. The device has not been withdrawn from the market for any reason related to its safety or effectiveness.

The insulin reservoirs and infusion sets used with the 670G System are also the same as those currently used with the MiniMed 530G System (P120010) and the MiniMed 630G System (P150001). These devices have not been withdrawn from commercial distribution for any reason, related to either safety or effectiveness.

## **VIII. PROBABLE ADVERSE EFFECTS OF THE DEVICE ON HEALTH**

Potential device-related serious adverse events include

- Diabetic ketoacidosis (DKA) resulting from high blood glucose due to suspension of insulin delivery or inadequate insulin delivery (which may result from catheter occlusion, hardware or software malfunction, erroneous CGM readings in Auto Mode or suspend mode, or inadequate insulin dosing).
- Severe hypoglycemia resulting from over-delivery of insulin (which can result from hardware or software malfunction, erroneous CGM readings in Auto Mode, or erroneous insulin dosing), which may lead to seizure, unconsciousness, and rarely death.

Potential device related non-serious events include:

- Skin irritation or redness
- Infection
- Pain or discomfort
- Bruising
- Edema
- Rash
- Bleeding
- Induration of skin
- Allergic reaction to adhesives

Sensor breakage with fragments retained under the skin is a potential adverse event related to use of the CGM component of the 670G system, but this was not observed during these studies. Based on post-market experience with similar devices and the results observed in these clinical studies, the occurrence and severity of these events are low.

Infection at the insulin pump infusion set insertion site and sensor insertion site is a potential complication related to insertion of the CGM or the insulin pump infusion set. Based on post-market experience with similar devices, and the results observed in these clinical studies, the occurrence and severity of these events are not expected to be different from other approved infusion sets and CGM devices.

Use of insulin pumps are known to carry an increased risk of DKA. However, FDA has received information indicating some patients are willing to accept an increased risk of DKA

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 12

{12}

or ketosis and hyperglycemia (severe hyperglycemia) because of the benefits of pump use (see also Section XII below).

Like other insulin pumps, there is an inherent risk that users of the device who do not use the 670G device as intended could harm themselves. Therefore, the device is for prescription use only and contraindicated for people unwilling or unable to perform a minimum of four fingerstick blood glucose meter tests per day and for people unwilling or unable to maintain contact with their healthcare professional.

As demonstrated under P120010/S046 for the MiniMed 530G System (which has the same 'suspend on low' feature, where the insulin delivery will suspend for two hours after the low glucose threshold has been reached), two hour suspension of insulin delivery is unlikely to lead to clinically significant ketosis or ketoacidosis even if the pump inappropriately suspends when the blood sugar is normal or elevated, and should respond to insulin therapy and hydration within a few hours.

There is a theoretical risk of insulin over-delivery due to device malfunction which has a risk of leading to severe hypoglycemia due to malfunction of the 670G System. This event did not occur during the pivotal study or the continuation phase of the pivotal study. If insulin over-delivery were to occur, there are several mechanisms in place designed to help detect and mitigate the risk of impending and/or current hypoglycemia, including the presence of alarms/alerts and the suspension/reduction of insulin delivery.

There is a theoretical risk of insulin under-delivery (due to a hardware or software malfunction) which may lead to severe hyperglycemia or DKA due to malfunction of the 670G system. This event did not occur during the pivotal study or the continuation phase of the pivotal study. If insulin under-delivery were to occur, there are mechanisms in place to help detect impending and/or current hyperglycemia, including the presence of alerts and alarms.

The consequences of falsely high glucose reading on the continuous glucose monitor would be potential over-delivery of insulin via automated insulin delivery and missed low glucose suspensions and alerts/alarms, which have the potential to lead to severe hypoglycemia. The consequences of falsely low glucose reading on the continuous glucose monitor would be potential under-delivery of insulin and missed high glucose alerts, which have the potential to lead to severe hyperglycemia or DKA.

## IX. SUMMARY OF NONCLINICAL STUDIES

### A. Laboratory Studies

The MiniMed 670G System remains unchanged as a result of this panel track supplement, though the indicated population has been expanded to include 7 to 13 year old people with type 1 diabetes. However, a human factors study of use of the MiniMed 670G System by a pediatric population ages 7-13 was conducted. A summary of this human factors study is provided below. Please refer to the Summary of Safety and Effectiveness Data (SSED) for the original PMA, P160017, for other pre-clinical/laboratory studies that supported the approval of the MiniMed 670G System and its components.

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 13

{13}

# **Human Factors Testing**

Medtronic conducted a usability validation study to evaluate the MiniMed 670G System, with continuous glucose monitoring and auto mode (HCL) function, to provide evidence that younger children (7 to 13 years of age) diagnosed with T1DM can, under the supervision of a caregiver, safely and effectively use the MiniMed 670G System.

Thirty users representing both pediatric novice and pediatric experienced patients and their caregivers participated in the usability validation study. These representative users performed critical tasks associated with using the MiniMed 670G system in both Manual and Auto Modes with the continuous glucose monitoring system. Task Analysis was used to determine critical tasks. Two representative user groups were studied (15 per group):

- Pediatric Novice Insulin Pump Users (ages 7-13 years)
- Users in this group were not currently external pump users, had less than 6 months of experience with a Medtronic pump, or were currently using a competitor pump.
- Pediatric Experienced Insulin Pump Users (ages 7-13 years)
- Users in this group currently were using a Medtronic external insulin pump for more than 6 months.

All use difficulties and close calls observed during completion of critical tasks were analyzed, and the root causes and impacts were assessed. For any close calls, a residual risk analysis was performed to determine whether design changes would further reduce the risks and to assess the residual risks related to the benefits to the patient.

The Human Factors usability validation study demonstrated that the MiniMed 670G System, with continuous glucose monitoring and auto mode (hybrid closed loop (HCL)) function, is safe and effective for use by patients with T1DM 7 to 13 years of age with assistance from their adult caregivers.

# **B. Animal Studies**

None.

# **C. Additional Studies**

None

# **X. SUMMARY OF PRIMARY CLINICAL STUDIES**

The original approval of P160017 was based on three pivotal studies to establish a reasonable assurance of safety and effectiveness of the MiniMed 670G System for its intended use, including the accuracy performance of the Guardian Sensor (3) in people 14 and older. Please refer to the Summary of Safety and Effectiveness Data (SSED) for the original PMA, P160017, for details on those studies.

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 14

{14}

Medtronic performed clinical studies to establish a reasonable assurance of safety and effectiveness of the 670G System in pediatric subjects ages 7-13 years. A summary of the clinical studies is presented below.

*Table 4: Summary of P160017/S031 Clinical Studies*

| Clinical Study | Patient Population | Study Design/Objective |
| --- | --- | --- |
| Safety Evaluation of the Hybrid Closed Loop (HCL) System in Pediatric Subjects with Type 1 Diabetes (described in Sections A-D below) | 7-13 years* | Multi-center, single-arm, home and hotel clinical study. The study evaluated the safety of the 670G System and its algorithm with the Guardian Sensor in subjects 7 to 13 years. |
| A Performance Evaluation of the Enlite® and Enlite 3 Glucose Sensor to Support Use in Children; Phase 2 (Enlite 3) (Described in Sections E-H, below) | 7-13 years* | Multi-center, prospective, single-sample correlational design without controls. The study demonstrated the measurement performance of the Guardian Sensor 7 days in subjects 2 to 18 years. |

*Note: These studies were designed for broader pediatric patient populations (2-13 years for G150247 and 2-18 years for G120262). However, only data for the stipulated patient population, ages 7-13 years, from G150247 was provided in support of this PMA supplement.

# **Pivotal study: Safety Evaluation of the Hybrid Closed Loop (HCL) System in Pediatric Subjects with Type 1 Diabetes (G150247):**

# **A. Study Design**

Subjects were treated between April 18, 2016 and October 09, 2017 and included 105 patients. There were 9 investigational sites.

The pivotal study was a multi-center, single-arm home and hotel clinical evaluation in subjects with type 1 diabetes on insulin pump therapy. The sponsor enrolled 111 subjects (ages 7-13 years) at 9 investigational centers (see subject accountability below).

Of the 111, 106 entered the run-in period. One subject withdrew prior to the start of the study period. Therefore, 105 subjects entered the study period. The 105 study subjects wore the MiniMed 670G pump with the Guardian Link (3) Transmitter, the Guardian Sensor (3) and infusion sets for approximately 3.5 months and participated in all three study phases: a two-week run-in period, a three-month at-home use period, and a 6 days/5 nights hotel study, which occurred during month 1, month 2, or month 3 of the study. Subjects were instructed to use the device in auto mode for the duration of the 3-month at home study.

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 15

{15}

*Run-in period*

During the two-week run-in period, subjects used the study pump (670G) with only the sensor augmented pump function activated (all automated features were off).

Prior to wearing study devices, all subjects and their companions were trained on the devices as well as diabetes management principles, such as the treatment of hyperglycemia and hypoglycemia.

At the end of the Run-In period at Run-In Visit 4, subjects 7-13 years of age were asked to undergo a 12 hour (maximum) Frequent Sample Testing procedure to assess the Suspend before Low feature.

*At-Home Study Period*

Following the two-week run-in period using the Study Pump (670G), a total of 105 subjects participated in a 3-month at-home study period. Prior to entry into Auto Mode, subjects used the pump in Manual Mode during the first 6 days of the study period in order to collect data on insulin utilization and sensor glucose levels. After this 6-day period, the subjects were allowed to enter Auto Mode.

*Hotel Study Period*

All 105 subjects participated in the Hotel portion of the study (6 days, 5 nights). There was an average of 4-6 subjects participating in each hotel stay. The hotel stays were to occur during the first, second or third month of study, with a minimum of 20 subjects completing the hotel stay each month. The purpose of the hotel portion of the study was to stress the user with sustained daily exercise and unrestricted eating. Subjects participated in daily sustained exercise/activity regimens for a minimum of 4 hours spread throughout the day. With respect to meals, subjects were allowed to eat what they chose.

1. Clinical Inclusion and Exclusion Criteria

Subjects were considered for enrollment in the study if they met all of the following criteria:

*General Inclusion Criteria*

1. Subject is age 7-13 years at time of screening
2. Subject age 7 -13 years has a clinical diagnosis of type 1 diabetes for 1 year or more as determined via medical record or source documentation by an individual qualified to make a medical diagnosis

*Study-Specific Inclusion Criteria*

3. Subject must have a minimum daily insulin requirement (Total Daily Dose) of greater than or equal to 8 units.
4. Subjects 7-13: Subjects and their parent(s)/guardian(s) are willing to participate in an overnight visit at the end of the Run-In period.
5. Subject 7-13 years of age and their parent(s)/guardian(s) are willing to participate in a Hotel study for the specified duration of Hotel period.
6. Subject must have companion 18 years or older who will sleep in the same dwelling place every night during the Study period. This requirement may be verified by subject report at screening visit.
7. Subject is willing to perform ≥ 4 finger stick blood glucose measurements daily

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 16

{16}

8. Subject is willing to perform required sensor calibrations
9. Subject is willing to wear the system continuously throughout the study
10. Subject has a Glycosylated hemoglobin (HbA1c) value less than 10.0% (as processed by Central Lab) at time of screening visit
11. Subject has TSH in the normal range OR if the TSH is out of normal reference range the Free T3 is below or within the lab's reference range and Free T4 is within the normal reference range.
12. Subject 7 -13 years of age has had pump therapy for greater than 6 months prior to screening (with or without CGM experience)
13. Subjects and their parent(s)/guardian(s) are willing to upload data from the study pump; must have Internet access and a computer system that meets the requirements for uploading the study pump
14. If subject has celiac disease, it has been adequately treated as determined by the investigator
15. Subjects and their parent(s)/guardian(s) are willing to take one of the following insulins and can financially support the use of either of the 2 insulin preparations throughout the course of the study (i.e. co-payments for insulin with insurance or able to pay full amount)
o Humalog® (insulin lispro injection)
o NovoLog® (insulin aspart)
16. Subjects and their parent(s)/guardian(s)/companions must be able to speak and be literate in English as verified by the investigator

Subjects were not permitted to enroll in the pivotal study if they met any of the following exclusion criteria:

Exclusion Criteria:

1. Subject has a history of 2 or more episodes of severe hypoglycemia, which resulted in any the following during the 6 months prior to screening:
a.Medical assistance (i.e. Paramedics, Emergency Room (ER) or Hospitalization)
b.Coma
c.Seizures
2. Subject is unable to tolerate tape adhesive in the area of sensor placement
3. Subject has any unresolved adverse skin condition in the area of sensor placement (e.g., psoriasis, dermatitis herpetiformis, rash, Staphylococcus infection)
4. Females who are sexually active and able to conceive will be excluded if they are not using an effective method of contraception and do not agree to continue using an effective method of contraception for the duration of the study as determined by investigator.
5. Subject has a cardiovascular condition which the investigator determines should exclude the subject, i.e. ventricular rhythm disturbance, hypertrophic cardiomyopathy
6. Subject is being treated for hyperthyroidism at time of screening
7. Subject has diagnosis of adrenal insufficiency
8. Subject 7-13 years of age has had DKA in the 6 months prior to screening visit.
9. Subject has taken any oral, injectable, or intravenous (IV) glucocorticoids within 8 weeks from time of screening visit, or plans to take any oral, injectable, or IV glucocorticoids during the course of the study

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 17

{17}

10. Subject is actively participating in an investigational study (drug or device) wherein he/she has received treatment from an investigational study drug or investigational study device in the last 2 weeks

11. Subject 7-13 years of age has been hospitalized or has visited the ER in the 6 months prior to screening resulting in a primary diagnosis of uncontrolled diabetes

12. Subject is currently abusing illicit drugs

13. Subject is currently abusing marijuana.

14. Subject is currently abusing prescription drugs

15. Subject is currently abusing alcohol

16. Subject is using pramlintide (Symlin), DPP-4 inhibitor, liraglutide (Victoza or other GLP-1 agonists), metformin, canagliflozin (Invokana or other SGLT2 inhibitors) at time of screening

17. Subject has a history of visual impairment which would not allow subject to participate in the study and perform all study procedures safely, as determined by the investigator

18. Subject has elective surgery planned that requires general anesthesia during the course of the study

19. Subject has a sickle cell disease, hemoglobinopathy; or has received red blood cell transfusion or erythropoietin within 3 months prior to time of screening

20. Subject plans to receive red blood cell transfusion or erythropoietin over the course of study participation

21. Subject diagnosed with current eating disorder such as anorexia or bulimia

22. Subject has been diagnosed with chronic kidney disease that results in chronic anemia

23. Subject has a hematocrit that is below the normal reference range of lab used.

24. Subject is on dialysis

25. Subject has serum creatinine of >2 mg/ dL.

# 2. Follow-up Schedule

Throughout the study period there were a number of scheduled telephone calls. These calls were meant to make sure that the subject was healthy and to remind them about adherence to study requirements, for example uploading the study pump data to CareLink. During the final study visit, subjects were asked to complete some questionnaires about their experience and also had blood collected for an HbA1c test.

# 3. Clinical Endpoints

There were no statistically powered endpoints in the pivotal study, nor was there any hypothesis testing. This was a descriptive study to evaluate the safe use of the Auto Mode. The analyses used for the various study phases are as follows:

# During the Entire Study

# Descriptive Endpoints

- The mean change in HbA1c is presented from baseline to end of study
- Change of Total Daily Dose (TDD) of insulin from baseline to end of

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 18

{18}

study

- Change of weight from baseline to end of study
- Time spent in Auto Mode versus time spent in Manual Mode
- Time in different range (% of sensor glucose): sensor glucose ≤ 50 mg/dL, ≤ 54 mg/dL, ≤ 60 mg/dL, ≤ 70 mg/dL, 71 mg/dL - ≤ 180 mg/dL, sensor glucose > 180 mg/dL, > 250 mg/dL, > 350 mg/dL
- Number of Events, area under the curve (AUC) and Time in the high sensor glucose range: sensor glucose (SG) > 180 mg/dL, > 250 mg/dL, > 350 mg/dL
- Number of Events, AUC and Time in the low sensor glucose range: sensor glucose ≤ 50 mg/dL, ≤ 54 mg/dL, ≤ 60 mg/dL, and ≤ 70 mg/dL

# Safety Data Summarized

- Serious Adverse Events (SAE), Serious Adverse Device Effects (SADE)
- Unanticipated Adverse Device Effects (UADE)
- Incidence of Severe Hypoglycemia
- Incidence of DKA

# B. Accountability of Study Cohort

A total of 106 subjects entered the run-in period, 1 subject withdrew during the run-in period and 105 subjects entered the study period. No subject was withdrawn during the study period, therefore 105 subjects completed the study phase.

# C. Study Population Demographics and Baseline Parameters

The demographics of the study population are typical for studies performed in the Type 1 diabetes population performed in the US.

Table 5: Study Population Demographics and Baseline Parameters

| Characteristic | Number of Subjects =105 |
| --- | --- |
| **Age (Years)** | |
| n | 105 |
| Mean (SD) | 10.8 (1.8) |
| Median | 11.0 |
| Min, Max | 7.0, 13.0 |
| **Gender N(%)** | |
| Female | 49 (46.7%) |
| Male | 56 (53.3%) |
| **Race N(%)** | |

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 19

{19}

| Characteristic | Number of Subjects =105 |
| --- | --- |
| Asian | 2 (1.9%) |
| Black/African American | 1 (1.0%) |
| Other | 8 (7.6%) |
| White | 94 (89.5%) |
| **Ethnicity N(%)** | |
| Hispanic/Latino | 8 (7.6%) |
| Non-Hispanic/Non-Latino | 97 (92.4%) |
| **Diabetes History(Years)** | |
| n | 105 |
| Mean (SD) | 5.6 (2.9) |
| Median | 4.9 |
| Min, Max | 1.1, 13.0 |
| **Height(cm)** | |
| n | 104* |
| Mean (SD) | 148.6 (13.8) |
| Median | 148.2 |
| Min, Max | 118.3, 184.6 |
| **Weight (kg)** | |
| n | 104* |
| Mean (SD) | 42.8 (13.0) |
| Median | 40.1 |
| Min, Max | 23.4, 83.0 |
| **BMI (kg/m²)** | |
| n | 104* |
| Mean (SD) | 19.1 (4.3) |
| Median | 18.0 |
| Min, Max | 14.0, 41.0 |
| **Baseline HbA1c (%)** | |
| n | 105 |
| Mean (SD) | 7.9 (0.8) |
| Median | 7.9 |
| Min, Max | 5.7, 9.6 |

* One subject's height and weight were not measured at enrollment

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 20

{20}

# **D. Safety and Effectiveness Results**

# 1. Safety Results

The safety of the 670G System was assessed by evaluation of the incidence of all adverse device events (ADEs), serious adverse device events (SADEs), and unanticipated adverse device events (UADEs) experienced by study subjects. AEs were listed in terms of severity and relationship to the device.

There was one report of a serious adverse event. This event was an episode of DKA that occurred prior to the subject beginning use of the 670G system, and therefore the event was not related to the 670G system or the other components of the 670G system.

There were no reports of unanticipated serious adverse device effects.

There were no reports of severe hypoglycemia events.

There was one report of a non-SADE. This event involved anxiety associated with sensor insertion that progressed throughout the subject's participation in the study.

There were 5 procedure-related events:

- neurocardiogenic syncope
- headache
- angioedema
- hyperglycemia
- skin irritation

There were 104 severe hyperglycemia events reported:

Severe hyperglycemia was defined in the protocol as a blood glucose concentration >300 mg/dL with blood ketones >0.6mmol/L or accompanied by symptoms of nausea, vomiting or abdominal pain.

Of the 104 reported severe hyperglycemia events, 62 were thought to be device-related. The majority of the device-related severe hyperglycemia events were believed to be due to infusion set issues (occlusion, bent cannula or cannula pull out). Other infrequently (2 events or less) reported causes of the device related severe hyperglycemia events include events associated with both automated and manual functionalities of the 670G System (safe basal, suspend before low suspension, automatic and manual suspensions). (See Table 6 below).

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 21

{21}

*Table 6. Study Period Severe Hyperglycemia*

| Cause | Total |
| --- | --- |
| Infusion set change | 28 |
| Occlusion alarm | 12 |
| Infusion set fell out | 7 |
| Bent /Kinked Cannula | 5 |
| Infusion set change/safe basal | 3 |
| Safe basal | 2 |
| Suspend before low suspension | 1 |
| Automatic & manual suspensions | 1 |
| Unclipped infusion set | 1 |
| Internal Battery Connector Resistance | 1 |
| Manual suspension and safe basal | 1 |
| **Total** | **62** |

Of these 62 reported device related severe hyperglycemia events, all events were reported to have resolved without sequelae. There were no reports of emergency room visits or hospitalizations relating to these 62 reported device related severe hyperglycemia events.

Of the 42 non-device related severe hyperglycemia events reported, all events were reported to have recovered without sequelae. The majority of reported causes of these non-device related severe hyperglycemia events were reported to be due to illness, stress, and incorrect meal boluses. Of these 42 non-device related severe hyperglycemia events, one event resulted in a visit to the emergency room with the subject being discharged home on the same day. This event was reported to have resolved without sequelae.

## 2. Effectiveness Results

### *System Performance: Pivotal Study (G150247)*

As stated above, the study performed was an observational study and only designed to evaluate system safety. The data below describe how the device performed during the pivotal study. The study was not designed to determine the effectiveness of the device compared to alternative treatments such as manual daily injections or non-automated insulin pump therapy.

The table below provides an overall summary of the run-in phase and study phase (home and hotel) for all subjects in the study. The data presented in this table includes information about subjects' glucose levels, insulin delivered and weight during run-in versus study phases.

*Table 7: Percent of Time within Glucose Ranges, Mean Insulin Delivery, and Mean Weight*

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 22

{22}

of Subjects during Run-in and Study Phases

| Parameter | Run-In | Study |
| --- | --- | --- |
| Sensor glucose, mean ± SD (median) mg/dL | 168.6±21.6(168.0, 154.5-183.6) | 161.7±12.4(161.9, 153.5-169.0) |
| Sensor Glucose Range (mg/dL) | Percent of Time with Sensor Glucose Level in Range Mean ± SD (95% CI), % | |
| ≤ 50 mg/dL | 0.85±1.17(0.62, 1.07) | 0.55±0.51(0.45, 0.64) |
| ≤ 54 mg/dL | 1.29±1.55(0.99, 1.59) | 0.81±0.68(0.68, 0.95) |
| ≤ 60 mg/dL | 2.24±2.27(1.80, 2.68) | 1.41±0.98(1.22, 1.60) |
| ≤ 70 mg/dL | 4.74±3.85(4.00, 5.49) | 2.99±1.64(2.68, 3.31) |
| 71 -≤ 180 mg/dL | 56.16±11.35(53.96, 58.35) | 65.01±7.67(63.53, 66.50) |
| > 180 mg/dL | 39.10±12.80(36.63, 41.58) | 31.99±7.67(30.51, 33.48) |
| > 250 mg/dL | 13.26±7.74(11.76, 14.76) | 10.31±5.15(9.31, 11.30) |
| > 300 mg/dL | 4.73±3.80(3.99, 5.46) | 3.73±2.74(3.20, 4.26) |
| > 350 mg/dL | 1.60±1.73(1.26, 1.93) | 1.21±1.19(0.98, 1.44) |
| Within-day SD of sensor glucose -mean ± SD (median, interquartile range) mg/dL | 57.7±8.3(58.4, 52.2-65.0) | 54.7±7.5(55.0, 49.7-59.1) |
| Within-day coefficient of variation of sensor glucose (%) -mean ± SD (median, interquartile range) mg/dL | 34.8±4.3(34.4, 32.0-37.8) | 33.7±3.1(33.7, 31.4-35.8) |
| Glycated hemoglobin -mean ± SD (median, interquartile range), % | 7.9±0.8(7.9, 7.2-8.4) | 7.5±0.6(7.5, 7.1-7.8) |
| Total daily dose of insulin -mean ± SD (median, interquartile range), U | 35.7±14.6(32.1, 24.4-45.0) | 38.5±15.5(34.3, 27.0-50.1) |
| Weight -mean ± SD (median, interquartile range), kg | 42.8±13.0(40.1, 32.4-51.6) | 44.9±13.4(42.3, 33.8-53.3) |

The baseline HbA1c value was collected at the first office visit for subjects during the study phase. The end of study HbA1c was collected at the last visit of the study phase. A summary of HbA1c data is provided in the table below.

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 23

{23}

*Table 8: Percent Change in HbA1c from Baseline to End of Study*

| | Baseline % (SD) | End of Study % (SD) |
| --- | --- | --- |
| Number of Subjects | 105 | 105 |
| HbA1c, %, Mean(SD) | 7.9 (0.8) | 7.5 (0.6) |
| HbA1c %, Median | 7.9 | 7.5 |
| 95% Confidence Interval | (7.7, 8.0) | (7.4, 7.6) |
| HbA1c, %, Min, Max | 5.7, 9.6 | 6.4, 9.3 |

The table below provides data regarding the subject baseline HbA1c collected at the beginning of the study and the number of subjects who with decreased, unchanged, or increased HbA1c values at the end of the study.

*Table 9: Number of subjects with change in HbA1C at different baselines*

| HbA1c Range Baseline HbA1c (%) | Number of Subject (%) with Change in HbA1c | | | | |
| --- | --- | --- | --- | --- | --- |
| | Decrease > 1% | Decrease > 0 to 1% | No Change | Increase > 0 to 1% | Increase > 1% |
| 5% ≤ HbA1c <6% | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 1 (1.0%) | 0 (0.0%) |
| 6% ≤ HbA1c <7% | 0 (0.0%) | 1 (1.0%) | 0 (0.0%) | 9 (8.6%) | 0 (0.0%) |
| 7% ≤ HbA1c <8% | 3 (2.9%) | 22 (21.0%) | 6 (5.7%) | 16 (15.2%) | 0 (0.0%) |
| 8% ≤ HbA1c <9% | 9 (8.6%) | 21 (20.0%) | 3 (2.9%) | 4 (3.8%) | 0 (0.0%) |
| 9% ≤ HbA1c <10% | 4 (3.8%) | 6 (5.7%) | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) |
| Overall | 16 (15.2%) | 50 (47.6%) | 9 (8.6%) | 30 (28.6%) | 0 (0.0%) |

The table below summarizes time spent in Auto Mode and summary of sensor wear, from start of Auto-Mode (Day 7 of the Study Phase) until the end of the study period.

*Table 10: Summary of Sensor Wear and Time Spent in Auto Mode, From Start of Auto-Mode to End of the Study Period*

| Category | Percentage of Time |
| --- | --- |
| Time spent wearing sensor | 95.1% |
| Time spent not wearing sensor | 4.9% |
| Auto Mode (core algorithm) | 71.9% |
| Auto Mode (safe basal) | 6.6% |
| Time spent in Manual Mode | 21.5% |

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 24

{24}

# **Guardian Sensor Performance Study (G120262)- A Performance Evaluation of the Enlite® and Enlite 3 Glucose Sensor to Support Use in Children; Phase 2 (Enlite 3):**

# **E. Study Design**

This study was performed to assess the analytical performance of the Guardian sensor. It ran between May 12, 2015 and December 13, 2016 and included 61 subjects (different from subjects who participated in the pivotal study above). There were 11 investigational sites.

This study was a prospective, single arm, multi-center, in-clinic study. All subjects wore one receiver, one transmitter, one transmitter used as a recorder, and two sensors.

Subjects wore the Guardian Sensor (3) in the abdomen and buttock insertion sites for a 7-day training period (that included a minimum 6 days of sensor wear), followed by a 7-day study period. During the study period, each subject participated in one in-clinic, frequent sample testing (FST) evaluation. FST occurred at the beginning (Day 1), middle (Day 3) or end (Day 7) of the Guardian sensor system use. During these FST sessions, intravenous (IV) blood samples were drawn every 5 to 15 minutes and analyzed for plasma blood glucose levels using the comparator method (CM). The CM is this study was the Yellow Springs Instrument 2300 Stat Plus Glucose Lactate Analyzer. FST with the CM lasted approximately 6 hours during the in-clinic visit.

Subjects were randomly assigned to one of three sensor insertion combinations; abdomen/abdomen (17 subjects), buttock/buttock (10 subjects), or abdomen/buttock (23 subjects). Subjects were randomized to one of 10 groups that determined when they participated in the in-clinic frequent sample testing.

Subjects continued with their current diabetes regimen independent of the study devices. Subjects were instructed by the investigational center that they were not to use the investigational devices for the management of their diabetes.

There was no control group as this study was an observational study to observe the accuracy and precision of the Guardian sensor. Accuracy was assessed by comparing the sensor values to the CM, and precision of the sensor system was assessed by comparing sensor values to each other in subjects wearing two sensors.

# **1. Clinical Inclusion and Exclusion Criteria**

Enrollment in the Guardian sensor study was limited to subjects who met the following inclusion criteria:

1. Subject is 7-13 years of age at time of screening (which was a subset of the entire study population)
2. Subject has been diagnosed with insulin requiring diabetes mellitus for at least one year.
3. Subject is willing to perform greater than or equal to 4 finger stick blood glucose measurements daily

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 25

{25}

4. Subject is willing to perform required sensor calibrations
5. Subject is willing to wear the system (Guardian Mobile application, pumps, sensors, meter) continuously throughout the study
6. Adequate venous access for subjects requiring IV for their FST, as assessed by investigator or appropriate staff

Subjects were not permitted to enroll in the Guardian sensor study if they met any of the following exclusion criteria:

1. Subject is unable to tolerate tape adhesive in the area of sensor placement
2. Subject has any unresolved adverse skin condition in the area of sensor or device placement (e.g., psoriasis, rash, Staphylococcus infection)
3. Subject is actively participating in an investigational study (drug or device) wherein they have received treatment from an investigational study (drug or device) in the last 2 weeks
4. Females of child-bearing potential who have a positive pregnancy test at screening or plans to become pregnant during the course of the study
5. Subjects with hematocrit lower than the normal age specific reference range per central or local lab testing

2. Follow-up Schedule

At the end of the study, subjects removed all study devices. Upon removal, all the Sensor insertion sites were examined and evaluated by the study staff. Sensors were visually inspected at the site. Study investigators documented any Adverse Device Effects (including skin irritations) and evaluated safety issues related to system use during the study. No long-term follow up was included in this study protocol.

3. Clinical Endpoints

Because this was an observational study, it did not include traditional analysis of clinical endpoints. The data were presented using multiple analyses as described in the Study Results section below.

Safety of the sensor was determined by skin and insertion site reactions.

F. Accountability of Study Cohort

Of the 61 subjects that entered the study, 3 completed Phase 1 (Enlite phase) of the study without participating in Phase 2 (Enlite 3 Phase). An additional eight (8) subjects withdrew before entering Phase 2 (Enlite 3 phase) for the following reasons:

- One subject decided not to be in the study any longer.
- Two subjects withdrew due to school schedule
- One subject withdrew due to work and school schedule
- One subject withdrew because subject did not feel comfortable inserting sensors at home.
- One subject withdrew because subject did not like first sensor insertion (after training)
- One subject withdrew because of major psychosocial issues and stress in family

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 26

{26}

• One subject withdrew because subject found wearing sensor too difficult – too much stimulus and information for patient to track

Among the 50 randomized subjects, 5 subjects did not complete the study for the following reasons:

• Four subjects withdrew due to repeated sensor failures
• One subject withdrew due to school schedule

A total of 44 subjects underwent frequent sample testing and completed the study. Twenty-one (21) subjects completed the first frequent sample testing on day 1, 13 subjects completed frequent sample testing on day 3, and 10 subjects completed frequent sample testing on day 7. One (1) subject completed the study, but did not complete frequent sample testing.

# G. Study Population Demographics and Baseline Parameters

Table 11: Study Population Demographics and Baseline Parameters

| Characteristic | All Subjects N=50 |
| --- | --- |
| **Age (Years)** | |
| Number of Subjects | 50 |
| Mean (SD) | 11.4 (1.53) |
| Median | 12 |
| Min, Max | 8.0 , 13.0 |
| **Gender, Number(%)** | |
| Female not of child bearing potential | 17 ( 34.0%) |
| Female of child bearing potential | 6 ( 12.0%) |
| Male | 27 ( 54.0%) |
| **Race, Number(%)** | |
| Asian | 1 ( 2.0%) |
| Other | 6 ( 12.0%) |
| White | 43 ( 86.0%) |
| **Ethnicity, Number(%)** | |
| Hispanic/Latino | 6 ( 12.0%) |
| Non-Hispanic/Non-Latino | 43 ( 86.0%) |
| Subject refused | 1 ( 2.0%) |
| **Height (cm)** | |
| Number of Subjects | 50 |
| Mean (SD) | 153.5 (13.24) |
| Median | 154 |
| Min, Max | 125.0 , 180.3 |
| **Weight (kg)** | |
| Number of Subjects | 50 |
| Mean (SD) | 49.8 (16.86) |

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 27

{27}

| Characteristic | All Subjects N=50 |
| --- | --- |
| Median | 47 |
| Min, Max | 23.0 , 104.5 |
| **Body Mass Index (kg/m2)** | |
| Number of Subjects | 50 |
| Mean (SD) | 20.7 (5.08) |
| Median | 19.3 |
| Min, Max | 14.7 , 36.9 |
| **A1C (%)** | |
| Number of Subjects | 50 |
| Mean (SD) | 8.0 (0.99) |
| Median | 8 |
| Min, Max | 6.0 , 11.0 |
| **Hematocrit (%)** | |
| Number of Subjects | 50 |
| Mean (SD) | 41.9 (2.49) |
| Median | 41.9 |
| Min, Max | 37.2 , 49.9 |

## H. Safety and Effectiveness Results

### 1. Safety Results

The safety of the Guardian Sensor (3) was assessed by evaluation of the incidence of all adverse events, ADEs, SADEs, and UADEs experienced by study subjects. AEs were listed in terms of severity and relationship to device. Sensor insertion site and adhesive area were examined for erythema, edema and infection. The local skin reactions from the insertion site or the adhesive were also evaluated.

There were 10 adverse events reported during the study as summarized below. None of these events were believed to be device-related All adverse events were resolved and subjects recovered completely without residual sequelae:

- There was one report of influenza virus B.
- There were two reports of gastroenteritis.
- There were two reports of upper respiratory infection.
- There was one report of a sore throat.
- There was one report of a vasovagal event.
- There was one report of abrasions due to falling off a scooter.
- There was one report of severe hyperglycemia secondary to influenza B
- There was one report of otitis media.

There were no reports of subject death.

There were no reports of device-related SAEs. There were no reports of non-

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 28

{28}

device-related SAEs.

There were no reports of DKA.

There were no reports of severe hyperglycemia.

There were no reports of severe hypoglycemia.

There were no reports of device-related adverse events.

The incidence of adverse events directly related to the CGM in the intended use population is not expected to differ significantly from the event rate observed during the Sensor accuracy study or those observed for other approved CGM devices. Based on (FDA-analyzed) post-market adverse event reports for similar CGM devices, no additional concerns regarding adverse events were raised for CGMs.

# 2. Effectiveness Results

Study results from the Guardian sensor study are presented in Tables 12 to 46. Results are presented below by abdominal insertion site followed by buttock insertion site.

The sensor must be calibrated at least twice per day (every 12 hours). However, the sponsor recommends users calibrate more often for best results (3 to 4 times per day). Most tables below represent data for minimal calibration (every 12 hours). Please see table headers to understand calibration frequency for each analysis.

# Abdomen Insertion Site

Tables 12 and 13 below provide the Guardian Sensor (3) values and the percent of data points that fell within 15, 20, 30, 40, and >40 mg/dL, or percent, of a specific CM glucose range when the sensor worn in the abdomen was calibrated every 12 hours after: Days 1, 3, and 7 (table 12); Day 1 only (table 13), respectively.

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 29

{29}

Table 12. Agreement (%) of Sensor-CM Paired Points (15/15%- greater than 40/40%) Stratified by Different CM Glucose Ranges, Calibrated every 12 hours, Abdominal Insertion Site, Data From Subjects 7-13 Years of Age, Days 1, 3, and 7 combined

| CM Glucose Ranges (mg/dL) | Number of Subjects | Number of CGM-CM Pairs | Percent of CGM Within 15/15% of CM | Percent of CGM Within 20/20% of CM | Percent of CGM Within 30/30% of CM | Percent of CGM Within 40/40% of CM | Percent of CGM Within >40/40% of CM |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Overall | 30 | 733 | 78.9 | 87.7 | 95.9 | 98.9 | 1.1 |
| ≥40-60* | 2 | 4 | 50 | 50 | 75 | 100 | 0 |
| >60-80* | 5 | 20 | 70 | 80 | 90 | 95 | 5 |
| >80-180 | 27 | 378 | 74.1 | 83.1 | 92.9 | 98.1 | 1.9 |
| >180-300 | 26 | 290 | 83.1 | 93.1 | 100 | 100 | 0 |
| >300-350 | 6 | 32 | 100 | 100 | 100 | 100 | 0 |
| >350-400 | 3 | 9 | 100 | 100 | 100 | 100 | 0 |

*For glucose ranges ≤ 80 mg/dL, agreement was based on 15/20/30/40 mg/dL. Note: Sensor glucose readings are within 40-400 mg/dL

Table 13. Agreement (%) of Sensor-CM Paired Points (15/15%- greater than 40/40%) Stratified by Different CM Glucose Ranges, Calibrated every 12 hours, Abdominal Insertion Site, Data From Subjects 7-13 Years of Age, Day 1

| CM Glucose Ranges (mg/dL) | Number of Subjects | Number of CGM-CM Pairs | Percent of CGM Within 15/15% of CM | Percent of CGM Within 20/20% of CM | Percent of CGM Within 30/30% of CM | Percent of CGM Within 40/40% of CM | Percent of CGM Within >40/40% of CM |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Overall | 16 | 403 | 81.9 | 90.6 | 96.5 | 99 | 1 |
| ≥40-60* | 1 | 2 | 100 | 100 | 100 | 100 | 0 |
| >60-80* | 3 | 11 | 63.6 | 72.7 | 90.9 | 100 | 0 |
| >80-180 | 14 | 196 | 75.5 | 84.2 | 93.4 | 98 | 2 |
| >180-300 | 15 | 160 | 86.9 | 97.5 | 100 | 100 | 0 |
| >300-350 | 3 | 27 | 100 | 100 | 100 | 100 | 0 |
| >350-400 | 2 | 7 | 100 | 100 | 100 | 100 | 0 |

*For glucose ranges ≤ 80 mg/dL, agreement was based on 15/20/30/40 mg/dL. Note: Sensor glucose readings are within 40-400 mg/dL

P160017/S031: FDA Summary of Safety and Effectiveness Data

Page 30

{30}

The following tables (14 to 17) show the percentage of concurring CGM readings compared to CM values for sensors worn in the abdomen. With ideal performance, the CGM readings would match the CM values. For example, with perfect concurrence, the shaded boxes in the tables below would be 100 percent.

Tables 14 and 15 show the concurrence of the CGM values compared to CM values when calibrating every 12 hours, after Days 1, 3, and 7.

Tables 16 and 17 show the concurrence of the CGM values compared to CM values when calibrating three to four times per day, after Days 1, 3, and 7.

Page 31

{31}

Table 14. Concurrence of CM values and CGM readings using CM glucose ranges; Calibrating every 12 hours, Abdominal Insertion Site, Data From Subjects 7-13 Years of Age, Overall

| CM Glucose Ranges (mg/dL) | Percent of Matched Pairs-in Each CGM Glucose Range for Each CM Glucose Range | | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | CGM (mg/dL) | | | | | | | | | | | |
| | # of Paired CGM-CM | <40 | ≥40-60 | >60-80 | >80-120 | >120-160 | >160-200 | >200-250 | >250-300 | >300-350 | >350-400 | >400 |
| >=40-60 | 6 | 33.3% (2/6) | 33.3% (2/6) | 0.0% (0/0) | 33.3% (2/6) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) |
| >60-80 | 20 | 0.0% (0/0) | 10.0% (2/20) | 55.0% (11/20) | 35.0% (7/20) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) |
| >80-120 | 124 | 0.0% (0/0) | 4.8% (6/124) | 13.7% (17/124) | 66.1% (82/124) | 15.3% (19/124) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) |
| >120-160 | 169 | 0.0% (0/0) | 0.0% (0/0) | 0.6% (1/169) | 21.3% (36/169) | 62.1% (105/169) | 15.4% (26/169) | 0.6% (1/169) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) |
| >160-200 | 160 | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 1.9% (3/160) | 25.0% (40/160) | 64.4% (103/160) | 8.8% (14/160) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) |
| >200-250 | 151 | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 1.3% (2/151) | 40.4% (61/151) | 56.3% (85/151) | 2.0% (3/151) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) |
| >250-300 | 64 | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 32.8% (21/64) | 64.1% (41/64) | 3.1% (2/64) | 0.0% (0/0) | 0.0% (0/0) |
| >300-350 | 32 | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 40.6% (13/32) | 59.4% (19/32) | 0.0% (0/0) | 0.0% (0/0) |
| >350-400 | 9 | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 88.9% (8/9) | 11.1% (1/9) | 0.0% (0/0) |

Page 32

{32}

Table 15. Concurrence of CM values and CGM readings using CM glucose ranges; Calibrating three to four times per day, Abdominal Insertion Site, Data From Subjects 7-13 Years of Age, Day Overall

| CM Glucose Ranges (mg/dL) | Number of Paired CGM-CM | Percent of Matched Pairs-in Each CGM Glucose Range for Each CM Glucose Range | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | | CGM (mg/dL) | | | | | | | | | | |
| | | <40 | ≥40-60 | >60-80 | >80-120 | >120-160 | >160-200 | >200-250 | >250-300 | >300-350 | >350-400 | >400 |
| ≥40-60 | 2 | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 100.0% (2/2) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) |
| >60-80 | 18 | 0.0% (0/0) | 0.0% (0/0) | 61.1% (11/18) | 38.9% (7/18) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) |
| >80-120 | 120 | 0.0% (0/0) | 3.3% (4/120) | 15.8% (19/120) | 67.5% (81/120) | 13.3% (16/120) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) |
| >120-160 | 162 | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 17.9% (29/162) | 64.8% (105/162) | 16.7% (27/162) | 0.6% (1/162) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) |
| >160-200 | 161 | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 1.2% (2/161) | 25.5% (41/161) | 65.2% (105/161) | 8.1% (13/161) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) |
| >200-250 | 145 | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 1.4% (2/145) | 42.8% (62/145) | 53.8% (78/145) | 2.1% (3/145) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) |
| >250-300 | 61 | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 32.8% (20/61) | 65.6% (40/61) | 1.6% (1/61) | 0.0% (0/0) | 0.0% (0/0) |
| >300-350 | 32 | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 0.0% (0/0) | 37.5…