Closed-loop Response to Unannounced Mixed and Carbohydrates-rich Breakfasts (CRUMB)

The CRUMB Study: Closed-loop Response to Unannounced Mixed and Carbohydrates-rich Breakfasts. A Randomized Controlled Crossover Pilot-study in a Cohort of Adolescents With Type 1 Diabetes.

The development of advanced hybrid closed-loop (a-HCL) systems represents a significant step toward in improving glucose control and reducing user-dependent variability, especially In pediatric patients. Systems can automatically deliver correction boluses and modulate insulin delivery based on CGM feedback, thereby compensating for some of the consequences of human error. Current evidence suggests that a-HCL systems can tolerate unannounced carbohydrate loads up to approximately 20 g without compromising time in range (TIR) or safety. However, the metabolic response to larger or compositionally complex meals remains variable and highly dependent on the specific algorithm governing insulin delivery.

Currently a variety of AID are commercially available: all of them present similarities and differences. The Medtronic MiniMed™ 780G uses a proportional-integral-derivative (PID) algorithm, a mathematical model that adjusts in real time insulin delivery rate based on 3 elements obtained from CGM reading values: the difference between the actual value and the chosen glucose target (proportional action), past values (integral action) and the glucose's rate of change (derivative action), In contrast, the Tandem t:slim X2™ with Control-IQ employs a model predictive control (MPC) algorithm, which aims, through a complex mathematical model, to predict glucose trends up to half-an-hour in the future, takin, also, in consideration actual and past glucose values. Despite sharing the same objective, said algorithms have different approaches, the former one being "reactive" and the latter "predictive". Therefore, their difference could result in different performances while facing mixed-nutrient meal or unannounced meals, defined as the consumptions of a meal with any prior insulin administration.

Pediatric patients represent certainly a unique subgroup in which therapeutic adherence is a relevant issue, due to cognitive, developmental and behavioral factors. Understanding how different AID algorithms respond to unannounced meals in this age group is therefore crucial for optimizing safety and personalization of diabetes management.

This study was designed to evaluate the strengths and limitations of two a-HCL systems, the Medtronic 780G (PID algorithm) and the Tandem t:slim X2 (MPC algorithm), in managing unannounced meals with different macronutrient compositions in children and adolescents with T1D. We also aim to better understand physiological and technological unannounced meal implications as to provide additional insight useful for the development of new fully closed loop algorithms, capable of minimizing glucose excursions and patient's burden.

Study Overview

• Status: Completed
• Conditions: Type 1 Diabetes Mellitus
• Intervention / Treatment: Other: Unannounced meal

• Study Type: Interventional
• Enrollment (Actual): 20
• Phase: Not Applicable

Contacts and Locations

Study Locations

• Italy
  • Sicily
    • Catania, Sicily, Italy, 95123
      • Azienda Ospedaliero Universitaria Policlinico "G.Rodolico - San Marco" - Catania

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Age between 11 and 18 years;
• Type 1 diabetes diagnosed for at least 12 months;
• Use of an A-HCL device for at least 3 months;
• Proficiency in carbohydrate counting.

Exclusion Criteria:

• Presence of malabsorption disorders;
• Presence of chronic diabetes complications.

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Treatment
• Allocation: Randomized
• Interventional Model: Crossover Assignment
• Masking: None (Open Label)

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
No Intervention: Announced meal (AM); Participants followed their usual therapy while consuming both the "CHO meal" and the "mixed meal" three times each. In this phase, the carbohydrate content of each meal was announced to the insulin pump, and pre-meal boluses were delivered as per standard practice.
Experimental: Unannounced meal (UM); Participants consumed the same meals, an equal number of times, without announcing their carbohydrate intake to the device. Consequently, no user-initiated boluses were administered during this period, and all insulin delivery adjustments were determined exclusively by the algorithm.	Other: Unannounced meal; Participants consumed both the "CHO meal" and the "mixed meal" three times each, without announcing their carbohydrate intake to the device. Consequently, no user-initiated boluses were administered during this period, and all insulin delivery adjustments were determined exclusively by the algorithm.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Change in blood glucose concentration from pre-meal to two hours post-meal (2h-ΔBG), as measured by CGM in mg/dl	Change in blood glucose concentration from pre-meal to two hours post-meal (2h-ΔBG), as measured by CGM	from pre-meal to two hours post-meal

Secondary Outcome Measures

Outcome Measure	Time Frame
Pre- and post-meal four-hour glucose difference (4h-ΔBG), as measured by CGM in mg/dl	from pre-meal to four hour postprandial
Peak CGM value in mg/dl	from pre-meal to four hour postprandial
Time to peak CGM value	from pre-meal to four hour postprandial
Percentage of spent time in range (TIR, 70-180 mg/dL), as measured by CGM	from pre-meal to four hour postprandial
Percentage of spent time in tight range (TITR, 70-140 mg/dL), as measured by CGM	from pre-meal to four hour postprandial
Percentage of spent time below range (TBR, < 70 mg/dL), as measured by CGM	from pre-meal to four hour postprandial
Percentage of spent time above range (TAR, > 180 mg/dL), as measured by CGM	from pre-meal to four hour postprandial
Percentage of spent time in second-level TAR (> 250 mg/dL), as measured by CGM	from pre-meal to four hour postprandial

Collaborators and Investigators

• Sponsor: University of Catania

Study record dates

Study Major Dates

• Study Start (Actual): April 20, 2025
• Primary Completion (Actual): May 20, 2025
• Study Completion (Actual): May 20, 2025

Study Registration Dates

• First Submitted: February 23, 2026
• First Submitted That Met QC Criteria: March 2, 2026
• First Posted (Actual): March 6, 2026

Study Record Updates

• Last Update Posted (Actual): March 6, 2026
• Last Update Submitted That Met QC Criteria: March 2, 2026
• Last Verified: March 1, 2026

More Information

Terms related to this study

• Keywords: type 1 diabetes mellitus; automated insulin delivery; unannounced meals; proportional integral derivative control; model predictive control
• Additional Relevant MeSH Terms: Endocrine System Diseases; Metabolic Diseases; Autoimmune Diseases; Immune System Diseases; Glucose Metabolism Disorders; Diabetes Mellitus; Nutritional and Metabolic Diseases; Diabetes Mellitus, Type 1
• Other Study ID Numbers: CRUMB Study

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No