Zone-Model Predictive Control on the Diabetes Assistant (DiAs) Platform Efficacy

Diabetes Mellitus, Type 1 Clinical Trial

Official title:

Use of Zone-Model Predictive Control (MPC) Controller on the Diabetes Assistant (DiAs) Platform With Fault Detection in Adolescents With Type 1 Diabetes in the Outpatient Setting (UCSB Study)

Clinical Trial Details — Status: Not yet recruiting

Administrative data

• NCT number: NCT02514785
• Other study ID #: IRB34487
• Status: Not yet recruiting
• Phase: Phase 1
• First received: July 11, 2015
• Last updated: July 31, 2015
• Start date: July 2015

Study information

• Verified date: July 2015
• Source: Stanford University
• Contact: Trang Ly, MBBS PhD
• Phone: 650.215.0732
• Email: trangly[@]stanford.edu
• Is FDA regulated: No
• Health authority: United States: Food and Drug Administration
• Study type: Interventional

Clinical Trial Summary

This clinical trial is designed to test the efficacy of a portable closed-loop artificial pancreas (AP) system using the Zone-Model Predictive Control (MPC) algorithm on the Diabetes Assistant (DiAs) platform in adolescents with type 1 diabetes. The system will also have additional fault detection algorithms designed to detect sensor failure as well as infusion set failure. The fault detection algorithms will be implemented via the remote monitoring software and clinical staff will be alerted to potential failures to assist with clinical care.

Description:

This clinical trial is designed to test the efficacy of a portable closed-loop artificial pancreas (AP) system using the Zone-Model Predictive Control (MPC) algorithm on the Diabetes Assistant (DiAs) platform. The system will also have additional fault detection algorithms designed to detect sensor failure as well as infusion set failure. The fault detection algorithms will be implemented via the remote monitoring software and clinical staff will be alerted to potential failures to assist with clinical care. This information will not be directly communicated to subjects to influence diabetes management in real-time.

The study is designed to test efficacy of the AP device in adolescents with type 1 diabetes under outpatient conditions. For this phase, we will recruit up to 20 subjects, 10 at Stanford and 10 at Denver. At each site, 5 subjects will be randomized to closed-loop control and 5 subjects will be randomized to sensor-augmented pump therapy. Subjects in both groups will be studied for 5 days/5 nights. Subjects will reside in a hotel/camp setting where they will sleep overnight and engage in mild to moderate intensity exercise during the day. Subjects will be fully supervised with clinical staff in attendance at all times. There will be at least twice daily exercise. Subjects will choose their own meals and perform their own carbohydrate counting for pre-meal insulin delivery. This second phase will generate 50 days of closed-loop data to compare with 50 days of sensor-augmented pump therapy.

The primary outcome for the second phase of the study will be to determine the efficacy of the system, as assessed by the percentage of time spent in range for sensor glucose values between 70-180 mg/dL, for closed-loop control compared to sensor-augmented pump therapy.

Recruitment information / eligibility

• Status: Not yet recruiting
• Enrollment: 20
• Est. primary completion date: July 2016
• Accepts healthy volunteers: No
• Gender: Both
• Age group: 10 Years to 19 Years

Eligibility

Inclusion Criteria:

1. Clinical diagnosis of type 1 diabetes

2. Daily insulin therapy for at least 12 months

3. Aged between 10.0 to 19.0 years

4. Insulin pump use for at least 3 months

5. Subject is using a downloadable pump

6. Female participants who are sexually active must be on acceptable method of contraception e.g. oral contraceptive pill, diaphragm, IUD (Intrauterine Device)

7. Female participants who have entered menarche must have a negative urine pregnancy test

8. Informed consent form signed by the subject and/or parent and assent signed by the subject if < 18 years

9. Total daily insulin requirement > 0.4 units/kg/day over the preceding two weeks

10. Subject is able to understand and speak English

Exclusion Criteria:

1. Subject has a medical disorder that in the judgment of the investigator will affect the wearing of the devices or the completion of any aspect of the protocol

2. Diabetic ketoacidosis in the past month

3. History of seizure or loss of consciousness in the last 6 months

4. Subject has a respiratory condition such as asthma, treated with systemic or inhaled corticosteroids in the previous 6 months or cystic fibrosis

5. Subject has a history of any cardiac or vascular disorder such as myocardial infarction, unstable angina, coronary artery bypass surgery, coronary artery stenting, transient ischemic attack, cerebrovascular accident, angina, congestive heart failure, arrhythmia or thromboembolic disease

6. Subject has a history of liver or kidney disease (other than microalbuminuria)

7. Systolic blood pressure > 140 mmHg on screening visit

8. Diastolic blood pressure > 90 mmHg on screening visit

9. Subject has active Graves disease

10. Subjects with inadequately treated thyroid disease or celiac disease

Study Design

Allocation: Randomized, Endpoint Classification: Efficacy Study, Intervention Model: Parallel Assignment, Masking: Open Label, Primary Purpose: Treatment

Related Conditions & MeSH terms

• Diabetes Mellitus, Type 1

Intervention

Device:
• Artificial Pancreas using the Zone-Model Predictive Control
Artificial pancreas (AP) system using the Zone-Model Predictive Control (MPC) algorithm on the Diabetes Assistant (DiAs) platform

Locations

Country	Name	City	State
n/a

Sponsors (5)

Lead Sponsor	Collaborator
Stanford University	Rensselaer Polytechnic Institute, University of California, San Diego, University of California, Santa Barbara, University of Colorado, Denver

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Efficacy, percentage of time spent in range	to test efficacy of the AP system using the Zone-Model Predictive Control (MPC) algorithm on the Diabetes Assistant (DiAs) platform in adolescents with type 1 diabetes. This will be assessed by the percentage of time spent in range for sensor glucose values between 70-180 mg/dL, for closed-loop control compared to sensor-augmented pump therapy.	change in 5 days/ 5 nights	No
Secondary	Percent time in hypoglycemic range	Secondary outcomes will include percentage of time spent in the hypoglycemic range of < 70 mg/dL, < 60 mg/dL, < 50 mg/dL compared to the sensor augmented control group in the same time frame	change in 5 days/ 5 nights	No
Secondary	number of treatments for potential hypoglycemia	To evaluate system usability we will record the number of treatments for potential hypoglycemia, to include patient initiated treatments without insulin boluses, and recommended treatments by the Health Monitoring System.	change in 5 days/ 5 nights	No
Secondary	Percent time in hyperglycemic range	Secondary outcomes will include percentage of time spent in the hyperglycemic range of > 150 mg/dL, > 180 mg/dL, >250 mg/dL compared to the sensor augmented control group in the same time frame	change in 5 days/ 5 nights	No