NCT03577158: SAFE-AP3

SAFE-AP: Automatic Control of Blood Glucose Under Announced and Unannounced Exercise

Completed NA Last updated 7 August 2019

What this trial tests

NA trial testing Closed-loop controller with exercise mitigation module in Diabetes Mellitus, Type 1 in 10 participants. Completed in 31 July 2019.

Timeline

1 June 2018

Primary endpoint
30 April 2019

31 July 2019

Quick facts

Lead sponsor	Hospital Clinic of Barcelona
Phase	NA
Status	Completed
Study type	INTERVENTIONAL
Allocation	randomized
Design	crossover
Masking	single
Primary purpose	treatment
Enrollment	10
Start date	1 June 2018
Primary completion	30 April 2019
Estimated completion	31 July 2019
Sites	1 location across Spain

Drugs / interventions tested

• Closed-loop controller with exercise mitigation module
• Closed-loop controller without exercise mitigation module
• Open-loop insulin infusion system

Conditions studied

• Diabetes Mellitus, Type 1 — all drugs for Diabetes Mellitus, Type 1 →

Sponsor

Hospital Clinic of Barcelona

Who can join

Adults 18 to 65, any sex, with Diabetes Mellitus, Type 1. Patients with the condition only — healthy volunteers not accepted.

Sponsor's own description

Achieving near-normoglycemia has been established as the main objective for most patients with diabetes. Despite growing evidence about the health benefits of regular exercise in diabetes, exercise-associated glycemic imbalance remains a challenge in subjects with type 1 diabetes (T1D) due to a higher risk of hypoglycemia.

Automatic glucose control, the so-called artificial pancreas (AP) or closed-loop system, may represent the ideal solution for reaching the therapeutic goals in diabetic patients. Intuitively, closed-loop insulin delivery may be superior to open-loop insulin delivery due to a better compensation of the variability of subcutaneous insulin absorption and the intra-subject insulin sensitivity. However, several challenges exist to effectively realize an optimal closed-loop control of blood glucose during exercise. Indeed, the physical exercise induces one of the major glucose perturbations that need to be controlled by an artificial pancreas and is currently one of the main challenges found in clinical validations of the few existing prototypes of an artificial pancreas. In particular, experiments carried out with the currently used algorithms for glucose control (the so called PID (proportional-integrative-derivative) and MPC) showed that closed-loop insulin delivery is deteriorated by exercise, increasing the risk of hypoglycemia even with pre-exercise carbohydrate ingestion when non-announcement strategies are used.

In this project, a rigorous clinical testing of a novel closed-loop controller ('artificial pancreas') will be carried out in T1D patients treated with continuous subcutaneous insulin infusion (CSII). The innovative element of the controller is a safety auxiliary feedback based on sliding mode reference conditioning (SMRC), which has been demonstrated to limit over-insulinization and the resulting hypoglycemia, reducing glycaemic variability.

Announced and unannounced exercise will be performed in T1D subjects treated with CSII, comparing the number of hypoglycemic episodes with a controller-driven insulin delivery (closed-loop study) based on continuous subcutaneous glucose monitoring (CGM).

If any glucose value \<70 mg/dL 15 gr of glucose will be provided. Moreover 15gr of glucose will be provided in AP studies when AP system recommends it.

The hypothesis is that closed loop control will provide better glycemic control, with less hypoglycemia episodes.

Publications & conference data

1 peer-reviewed publication reference this trial (live from Europe PMC):

1. Artificial Pancreas With Carbohydrate Suggestion Performance for Unannounced and Announced Exercise in Type 1 Diabetes.
   Viñals C, Beneyto A, Martín-SanJosé JF, Furió-Novejarque C, et al · · 2021 · cited 12× · PMID 32852548 · DOI 10.1210/clinem/dgaa562