MKC-TI-192 (NCT05243628) was an open-label, three-arm, proof-of-concept study of 90-day treatment with TI in adult participants with T1D using an AID system for diabetes management. Here, we describe findings from a post hoc analysis based on the initial meal test conducted to determine the starting study dose of TI. The full study results for the 90-day treatment period are reported in a forthcoming manuscript by Kaiserman et al.

The MKC-TI-192 study was conducted at two investigational sites between March 31, 2022 and October 17, 2022, an academic institution and a private clinic. All participants provided written informed consent prior to initiation of study procedures. The participants were randomized into three arms: TI + insulin degludec (TI for meals and corrections and SC insulin degludec for basal insulin), TI + AID (TI for meals and AID for correction and basal insulin), and AID control (AID using an RAA for meals and corrections and basal insulin coverage) for 90 days. The AID control group was not controlled for type of SC insulin. The study consisted of five clinic (screening visit, three treatment visits, and an end-of-treatment visit) and nine telephone visits. Participants continued their prestudy antihyperglycemic medications, and no changes to medications were allowed between visits 1 and 2. The initial meal challenge reported here was conducted at visit 2 and no other meal tests were conducted in this study thereafter. For this post hoc analysis, the TI + AID and TI + insulin degludec groups were combined and are referred to as the TI group hereafter because their meal challenge was identical.

Participants aged ≥ 18 years with a clinical diagnosis of T1D currently using a continuous subcutaneous insulin infusion (CSII) pump with an AID algorithm (Control IQ™ by Tandem Diabetes Care, San Diego, California, or MiniMed 670G™/770G™ by Medtronic, Northridge, California) with an A1C of 7% to 11% were eligible to participate in the study. Exclusion criteria included recent history/treatment of asthma, chronic obstructive pulmonary disease, or any other clinically important pulmonary disease (e.g., cystic fibrosis, bronchopulmonary dysplasia); significant congenital or acquired cardiopulmonary disease; history of smoking in the 6 months before screening; respiratory tract infection within 14 days before screening; use of antiadrenergic drugs; female participants who were pregnant, breastfeeding, intended to become pregnant or of child-bearing potential and not using adequate contraceptive methods; and a severe hypoglycemic event or episode of diabetic ketoacidosis within 180 days prior to screening.

For administration of basal insulin, participants receiving TI + insulin degludec received daily SC insulin degludec for basal coverage. Participants receiving TI + AID or AID controls continued use of their pre-study AID system and infused rapid-acting insulin with an individualized dose regimen of rapid-acting insulin aspart (Novolog®, Novo Nordisk, Plainsboro, New Jersey, or Fiasp®, Novo Nordisk, Plainsboro, New Jersey) or insulin lispro (Humalog®, Eli Lilly and Company, Indianapolis, Indiana; Admelog®, Sanofi-Aventis U.S. LLC, Bridgewater, New Jersey; or Lyumjev®, Eli Lilly and Company, Indianapolis, Indiana) for basal coverage and for prandial or correction dosing in the AID control group only. The study did not control for type of insulin in AID systems. Prandial insulin was delivered using TI in the TI + AID and TI + insulin degludec groups, and correction insulin was also delivered by TI in the TI + insulin degludec group.

The study protocol and informed consent form were approved by a central institutional review board (Advarra Institutional Review Board) before the start of the study (FDA/Office for Human Research Protections Institutional Review Board Registration Number IRB00000971). The study was conducted in compliance with applicable FDA regulations, local regulations (where applicable), the International Conference on Harmonisation Good Clinical Practice guidelines, and the Declaration of Helsinki. All participants provided written informed consent prior to initiation of study procedures. Manuscript development conformed to the International Committee of Medical Journal Editors Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals.

All participants consumed an in-clinic standardized meal (Boost®, Nestlé, Leeds, England) consisting of 37 g of carbohydrates. Participants fasted for 6 h prior to the meal and avoided manual bolus insulin 4 h prior to the meal; the meal challenge started 15 min after a glucose value ≥ 70 mg/dL was confirmed. Participants in both arms remained on their AID system during the meal, calculated an injectable RAA dose based on their standard of care carbohydrate ratio, and converted it to TI dose using the higher modified dose conversion, where the TI dose is calculated by multiplying the usual SC RAA dose by 2 and rounding it down to the nearest available TI dose (Table 1). No premeal correction dose was added. Participants administered an initial dose of ≤ 24 U of TI to cover the carbohydrate intake. Participants in the TI group dosed TI at the start of the meal and did not enter carbohydrate intake and/or meal announcements into their AID system. Participants in the AID control group dosed using their AID system 5 to 15 min prior to starting the meal. Self-monitored blood glucose (SMBG) values were taken at 0, 15, 30, 45, 60, 90, and 120 min, relative to the start of the meal using CONTOUR® meters (Ascensia Diabetes Care, Parsippany, NJ). In-clinic SMBG values were recorded by the site staff in the study case report form.

The primary endpoint for this post hoc analysis of the in-clinic meal study was the assessment of PPG and PPG excursion (PPGE) in the TI group compared with the AID control group. Safety endpoints for this post hoc analysis were the incidence of level 1 (< 70 mg/dL) and level 2 (< 54 mg/dL) hypoglycemia.

Summary tables (descriptive statistics and/or frequency tables) are provided for baseline and end-of-study variables assessing efficacy and safety by descriptive statistics (n, mean, standard deviation, minimum, and maximum). Frequency counts and percentage of participants within each category are provided for categorical data. No imputation method was applied for any missing postbaseline values. The Shapiro–Wilk test was used to assess the normality of data. The study was not powered for statistical significance, and P values from the two-sample t test should be considered hypothesis-generating only and interpreted with caution. Analyses were performed with Microsoft Excel for Microsoft 365, MATLAB R2022a, or Minitab 20.2, and P values < 0.05 were considered statistically significant.