Concurrent glucagon-like peptide 1 (GLP-1) receptor agonist use with automated insulin delivery (AID) systems has not been well studied in type 2 diabetes. GLP-1 receptor agonists are increasingly used before insulin therapy and are often added to insulin regimens in individuals with type 2 diabetes. Therefore, many individuals with type 2 diabetes may be on these medications before initiating AID therapy. Even so, it is not clear how GLP-1 receptor agonist therapy may affect outcomes and insulin doses with AID use.
We performed a subgroup analysis of GLP-1 receptor agonist use, as well as other adjuvant medications, from the clinical trial Beneficial Effects of Control-IQ Automated Insulin Delivery in Basal-Bolus and Basal-Only Insulin Users With Type 2 Diabetes (1). In that trial, 30 adults with type 2 diabetes (age 54 ± 12 years, 60% female, A1C 8.6 ± 1.2%, BMI 31.6 ± 6 kg/m2, diabetes duration 14 years [range 10–22 years]) using either basal insulin only or multiple daily injection (MDI) insulin therapy initiated Control-IQ technology use. Mean time in range (TIR; 70–180 mg/dL) for both prior MDI and basal-only insulin users improved 15% with 6 weeks’ use of the Tandem t:slim X2 insulin pump with Control-IQ technology, representing an increase of 3.6 hours/day, and mean glucose decreased by 22 mg/dL. Mean total daily insulin (TDI) was 0.62 units/kg/day at baseline and 0.67 units/kg/day with Control-IQ technology use. There were no severe hypoglycemia or diabetic ketoacidosis events.
In the trial, participants were required to be on a stable dose of any glucose-lowering or weight loss medications for at least 3 months before enrollment. Participants were not allowed to start a GLP-1 receptor agonist, sodium–glucose cotransporter 2 (SGLT2) inhibitor, or other glucose-lowering or weight loss medication or to adjust doses of these medications during the trial. Of the participants already taking a GLP-1 receptor agonist at a stable dose, seven were taking semaglutide with doses ranging from 0.5 to 1 mg/week, seven were taking dulaglutide with doses ranging from 0.75 to 4.5 mg/week, and one was taking liraglutide 1.8 mg/day.
We found that those not using a GLP-1 receptor agonist increased their TDI from 0.54 to 0.74 units/kg/day with Control-IQ technology use. Similar findings were seen for those using an SLGT2 inhibitor alone or using an SGLT2 inhibitor with a GLP-1 receptor agonist, with both groups showing an increase in TDI. However, those solely taking a GLP-1 receptor agonist showed a decrease in TDI from 0.72 units/kg/day at baseline to 0.63 units/kg/day with Control-IQ technology use. At the same time, the highest TIR achieved (79%) and the greatest improvement from baseline (+27% TIR) was seen in those solely taking a GLP-1 receptor agonist with Control-IQ technology use (Table 1).
Mean TDI, TIR, and Time in Hypoglycemia According to Use of SGLT2 Inhibitor and GLP-1 Receptor Agonist Drugs in Addition to Control-IQ Technology
| . | Not on SGLT2 Inhibitor or GLP-1 Receptor Agonist* . | On SGLT2 Inhibitor Only† . | On GLP-1 Receptor Agonist Only‡ . | On SGLT2 Inhibitor and GLP-1 Receptor Agonist§ . |
|---|---|---|---|---|
| TDI Overall n Pre-study, units/kg Open-loop TDI, units/kg Closed-loop TDI, units/kg Pre-study MDI/pump n Pre-study TDI, units/kg Open-loop TDI, units/kg Closed-loop TDI, units/kg Pre-study basal-only n Pre-study TDI, units/kg Open-loop TDI, units/kg Closed-loop TDI, units/kg | 7 0.54 0.50 0.74 5 0.64 0.59 0.90 2 0.30 0.29 0.34 | 7 0.64 0.56 0.71 3 0.80 0.68 1.00 4 0.52 0.46 0.49 | 12 0.72 0.58 0.63 8 0.83 0.64 0.69 4 0.49 0.45 0.52 | 3 0.54 0.55 0.62 1 0.67 0.82 0.83 2 0.47 0.42 0.52 |
| TIR (70–180 mg/dL) n CGM run-in, % Open-loop, % Closed-loop, % | 7 43 51 56 | 7 73 73 70 | 12 52 67 79 | 3 69 60 75 |
| Time <70 mg/dL n CGM run-in, % Open-loop, % Closed-loop, % | 7 0.34 0.15 0.25 | 7 0.30 0.41 0.20 | 12 0.60 0.46 0.22 | 3 0.08 0.24 0.58 |
| Time <54 mg/dL n CGM run-in, % Open-loop, % Closed-loop, % | 7 0.04 0.02 0.05 | 7 0.05 0.05 0.05 | 12 0.25 0.08 0.03 | 3 0.00 0.00 0.18 |
| . | Not on SGLT2 Inhibitor or GLP-1 Receptor Agonist* . | On SGLT2 Inhibitor Only† . | On GLP-1 Receptor Agonist Only‡ . | On SGLT2 Inhibitor and GLP-1 Receptor Agonist§ . |
|---|---|---|---|---|
| TDI Overall n Pre-study, units/kg Open-loop TDI, units/kg Closed-loop TDI, units/kg Pre-study MDI/pump n Pre-study TDI, units/kg Open-loop TDI, units/kg Closed-loop TDI, units/kg Pre-study basal-only n Pre-study TDI, units/kg Open-loop TDI, units/kg Closed-loop TDI, units/kg | 7 0.54 0.50 0.74 5 0.64 0.59 0.90 2 0.30 0.29 0.34 | 7 0.64 0.56 0.71 3 0.80 0.68 1.00 4 0.52 0.46 0.49 | 12 0.72 0.58 0.63 8 0.83 0.64 0.69 4 0.49 0.45 0.52 | 3 0.54 0.55 0.62 1 0.67 0.82 0.83 2 0.47 0.42 0.52 |
| TIR (70–180 mg/dL) n CGM run-in, % Open-loop, % Closed-loop, % | 7 43 51 56 | 7 73 73 70 | 12 52 67 79 | 3 69 60 75 |
| Time <70 mg/dL n CGM run-in, % Open-loop, % Closed-loop, % | 7 0.34 0.15 0.25 | 7 0.30 0.41 0.20 | 12 0.60 0.46 0.22 | 3 0.08 0.24 0.58 |
| Time <54 mg/dL n CGM run-in, % Open-loop, % Closed-loop, % | 7 0.04 0.02 0.05 | 7 0.05 0.05 0.05 | 12 0.25 0.08 0.03 | 3 0.00 0.00 0.18 |
Of the seven participants using neither a GLP-1 receptor agonist nor an SGLT2 inhibitor, six were using metformin. One participant using metformin plus a dipeptidyl peptidase 4 inhibitor was not included.
Of the seven participants using an SGLT2 inhibitor only, six were using metformin.
Of the 12 using a GLP-1 receptor agonist only, six were using metformin.
Of the three participants using both a GLP-1 receptor agonist and an SGLT2 inhibitor, one was using metformin, and one was using metformin and pioglitazone. CGM, continuous glucose monitoring.
Prior studies of Control-IQ technology have not identified a subgroup of users who showed such strong improvements in glycemic outcomes while taking less insulin with AID use (2). In fact, we would expect that, with an average A1C of 8.6%, participants were under-insulinized at baseline and would have received more insulin with AID therapy when achieving such significant improvements in glycemic outcomes. The cause of this reduced insulin need with GLP-1 receptor agonist use with AID may be multifactorial and could be related to reduced food intake, changes in insulin secretion, weight loss, and/or changes in insulin resistance. However, it appears the automated adjustments of insulin delivery were most effective in those solely using a GLP-1 receptor agonist. The same changes were not seen in those taking an SGLT2 inhibitor alone or those taking an SGLT2 inhibitor and a GLP-1 receptor agonist, but those participants already had a very high TIR at baseline (73 and 69%, respectively), and only a small number of participants were taking both medications.
Although this analysis is from a small sample size and the study was only of 6 weeks’ duration, results from participants using a GLP-1 receptor agonist in the ongoing Randomized Controlled Trial of Control-IQ Technology in Type 2 Diabetes (ClinicalTrials.gov identifier NCT05785832), which is a 13-week trial of Control-IQ technology use in 300 individuals with type 2 diabetes, will offer further insight into the synergy between these two therapies.
Duality of Interest
R.W.B.’s institution has received funding and study supplies from Tandem Diabetes Care and Dexcom. C.J.L. has been a paid consultant to Dexcom, and her institution has received research support from Abbott Diabetes, Dexcom, Insulet, and Tandem Diabetes Care. Y.C.K.’s institution has received research product support from Dexcom and Tandem Diabetes Care and payment for consultancy from Novo Nordisk. K.P. has received research support from Altimmune, Dexcom, Eli Lilly, Medtronic, Novo Nordisk, Oramed, and Tandem Diabetes Care. T.B. has received research support from Abbott Diabetes Care, Dexcom, Eli Lilly, MannKind, Medtronic, Novo Nordisk, Tandem, and Viacyte and speaker fees from Astra Zeneca, Boehringer Ingelheim, Dexcom, Eli Lilly, and Novo Nordisk. D.R.’s institution has received funding and study supplies from Dexcom and Tandem Diabetes Care. J.E.P. is an employee of Tandem Diabetes Care.
Author Contributions
R.W.B., D.R., and J.E.P. analyzed the data and wrote the manuscript. C.J.L., Y.C.K., K.P., and T.B. researched data. All authors edited and reviewed the manuscript and approved the final version. J.E.P. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
