Pregnancy in patients with type 1 diabetes (T1D) is associated with high maternal and fetal risk. Tight glycemic control is critical to reduce pregnancy complications. However, achieving the recommended glycemic targets is very challenging for most pregnant women with T1D. Contemporary closed-loop control or artificial pancreas (AP) systems can offer optimal solutions to this unmet clinical need. However, diurnal physiological, hormonal and glucose changes throughout the ~40 week gestation, the changing glycemic risks to fetus and mother and the stringent regulations about research in this population have limited interventional studies in this cohort to date. As a result, AP technologies have not been studied in pregnant women with T1D in the US. The consortium of Harvard/Mayo Clinic/Mt. Sinai and Sansum Diabetes Research Institute aim to develop an AP system that is customized to the individual needs of pregnant women with T1D and is able to adapt to the physiologic changes during pregnancy. A series of short-term supervised clinical studies will provide feedback for the system performance and will indicate required improvements. Through an iterative process, the AP algorithm will be revised and finalized. The successful completion of the initial studies will lead to a 4-week single arm pregnancy specific AP study with the option of extension of AP for the duration of the pregnancy. The current studies will enroll subjects after organogenesis is complete and risk is relatively lower in the 14-28 week pregnancy period. The overall goal of this project is therefore to create adaptive pregnancy-informed AP technologies, for testing with subsequent phase 2 and phase 3 randomized clinical trials, in 3 Specific Aims as follows:
Specific Aim 1 : Develop a state of the art automated glucose control system for prolonged use during all stages of pregnancy in T1D. Leverage clinical information obtained by the analysis of a multicenter database of glycemic variability, insulin management and outcomes of pregnant women with T1D to design a cloud based adaptive decision support system (DSS) layer for the AP.
Specific Aim 2 : Evaluate the safety and feasibility of automated glucose control during the second trimester in a supervised setting for T1D. The control algorithm will be updated to enable real-time glucose management for pregnant women with time-varying insulin sensitivity profiles. Three clinical studies will be conducted, two supervised 48 hour and one 1-week home-based study to evaluate the safety and efficiency of the proposed control strategy and to finalize the algorithm details that will establish a safe AP system tailored for pregnant women.
Specific Aim 3 : Extended evaluation of automated glucose control during second and third trimesters of pregnancy in T1D. A prolonged multicenter single arm outpatient AP study (4 weeks and optional extension phase) will be conducted to evaluate the safety and effectiveness of the developed AP system.
Artificial pancreas (AP) as an effective treatment for type 1 diabetes (T1D) is in clinical practice at this time and rapidly being improved in clinical research studies. Despite suffering from higher rates of medical complications with their usual care, higher risk populations such as pregnant women with T1D have not been enrolled in the US in AP research and remain understudied. The consortium of the experienced engineering team at Harvard University and clinical research centers of Mayo Clinic, Mt. Sinai and Sansum Diabetes Research Institute will develop and then evaluate a pregnancy specific AP in a sequence of in-clinic and transitional environment trials to facilitate progression to a safe and effective at-home clinical trial with an extension phase to the end of pregnancy.
