980-P: Automated Insulin Delivery Clamp Technology—Proof of Concept Study

Introduction & Objective: Interventions designed to improve endogenous insulin production and/or reduce exogenous insulin requirements are currently assessed using C-peptide measurement & traditional clamp techniques. We propose a novel adjunctive protocol for determining insulin requirements using Automated Insulin Delivery (AID) systems, known as AID-Clamp.

Methods: With AID-Clamp (see video https://bit.ly/AIDClamp), an individual’s insulin requirement over a defined period is quantified using analysis of the AID event log. An insulin quantification script retrieves basal and bolus insulin data over the time period and provides an output of total insulin requirement.

A 55-year-old man with type 1 diabetes already using AID (open-source iAPS) opted to commence off-label tirzepatide as adjunct therapy. He reported a profound reduction in insulin requirement. We applied AID-Clamp analytics retrospectively to identical meals eaten regularly by this patient, to demonstrate its application.

Results: In the two weeks prior to tirzepatide commencement, mean insulin total daily dose was 194 units. The AID-Clamp 6-hour insulin requirement post standard-meal of 88g carbohydrate was 30.5 units. In the third week of tirzepatide use (dose 5 mg weekly), the mean insulin total daily dose was 113.4 units. The 6-hour insulin requirement post standard 88 g meal was 11.8 units.

Conclusion: AID-Clamp is a novel application for AID systems. When used in appropriate clinical or research settings, it is capable of detailed quantification of insulin requirements. The changes in insulin requirement in response to pharmacologic, lifestyle or other interventions, serve as a proxy for the effect of the intervention. AID-Clamp is non-invasive, carries no additional cost and is much quicker and less labor-intensive than traditional methods of assessing the efficacy of such interventions.