Achieving near-normoglycemia has been established as the main objective for most patients with Type 1 Diabetes (T1D). Automated insulin delivery (AID) systems, the so-called artificial pancreas (AP) or closed-loop systems, may represent the ideal solution, especially for patients not reaching the therapeutic goals with multiples doses of insulin or open-loop delivery systems. Despite the advances in recent years that have proven the efficacy and safety of these devices in clinical trials and clinical practice settings, such evidence cannot be extrapolated to highly unstable patients, as problematic hypoglycemia remains an exclusion criterion in most of the trials.

The SAFE-AP system is a single-hormone hybrid closed-loop controller based on a proportional derivative with an insulin feedback controller that integrates a safety layer with insulin-on-board constraints and sliding mode reference conditioning. The hybrid closed-loop system includes a second safety feedback loop with a controller that triggers carbohydrate recommendations to the patient. Both control loops are coordinated to ensure that the counter-regulatory effect of rescue carbohydrates is not counteracted with insulin. Such system has been previously proven effective in unannounced exercise, one of the main challenges in AID systems development. Additionally, the algorithm has been recently tailored to achieve a better control in the subgroup of T1D patients prone to hypoglycemia.

In this project, a rigorous clinical testing of the SAFE-AP system will be carried out in 12 patients with T1D and problematic hypoglycemia, despite treatment with continuous subcutaneous insulin infusion. The safety and performance of the system will be evaluated in a 32-hour pilot study, including 4 meals, one overnight period and 2 unannounced aerobic exercise sessions. The study will be performed in a hospital setting with the on-site supervision of a specialized nurse and a diabetologist, as well as an engineer in remote control.

Inclusion Criteria:

• Subjects aged 18 or more with Type 1 Diabetes for at least 5 years.

• Treated with continuous subcutaneous insulin infusion (CSII) for at least six months.

• Trained to carbohydrate counting.

• Subjects prone to hypoglycemia, despite optimal diabetes management and hypoglycemia-specific education programs, as defined by >4 hypoglycemic episodes per week and at least one of the following:

  1. Occurrence of at least 2 severe hypoglycemic episodes during the last 2 years (need for third party).
  2. Occurrence of at least 1 severe hypoglycemic episode during the last 2 years and high glycemic variability (coefficient of variation >36%).
  3. Impaired awareness of hypoglycemia (Clarke test ≥4).

• Physical examination, laboratory data and EKG without alterations. Abnormalities considered clinically irrelevant by the investigator will not be exclusion criteria.

• Negative SARS-CoV-2 PCR test performed at arrival.

• Subject willing to wear the SAFE-AP system continuously throughout the study.

Exclusion Criteria:

• Pregnancy and breastfeeding.
• History of drug or alcohol abuse.
• Progressive fatal disease.
• Clinically relevant microvascular complications (macroalbuminuria, pre-proliferative and proliferative retinopathy), cardiovascular, hepatic, neurological, endocrine or other systematic, apart from T1D, which may hamper the implementation of the clinical study protocol or interpretation of results study.
• Scheduled surgery during the study period.
• Mental conditions that prevent the subject to understand the nature, purpose and possible consequences of the study.
• Subjects who are unlikely to meet the clinical study protocol, eg. uncooperative attitude, inability to return for follow-up visits, or poor probability of completing the study.
• Using an experimental drug or device during the past 30 days.