GLP-1RAs showed no advantage for basal insulin discontinuation
1. In this target trial emulation, initiating a glucagon-like peptide-1 receptor agonist (GLP-1RA) was not associated with higher rates of basal insulin discontinuation than initiating a sodium-glucose cotransporter-2 inhibitor (SGLT-2i) or dipeptidyl peptidase-4 inhibitor (DPP4i) in US veterans with type 2 diabetes (T2D).
2. Although younger patients and those receiving lower baseline insulin doses were more likely to discontinue insulin overall, no subgroup demonstrated a comparative advantage for GLP-1RAs.
Evidence Rating Level: 2 (Good)
Study Rundown: Studies have shown that adding glucagon-like peptide-1 receptor agonists (GLP-1RAs) to insulin regimens can substantially reduce total daily insulin requirements. Their increasing use has raised interest in whether they may also facilitate safe discontinuation of insulin therapy in people with type 2 diabetes (T2D), although previous research has primarily focused on discontinuing bolus rather than basal insulin. This study evaluated whether initiating a GLP-1RA, compared with a sodium-glucose cotransporter-2 inhibitor (SGLT-2i) or dipeptidyl peptidase-4 inhibitor (DPP4i), increased rates of basal insulin discontinuation over 3 years in adult veterans with T2D. Basal insulin discontinuation occurred in 16.7% of GLP-1RA users, similar to SGLT-2i (17.9%) and DPP4i (17.1%) users, indicating that GLP-1RAs did not significantly increase the likelihood of insulin discontinuation. Younger patients (<65 years) and those using lower baseline insulin doses (<50 units/day) were more likely to discontinue insulin regardless of treatment. No subgroup demonstrated a comparative advantage for GLP-1RAs, while younger participants and those with a BMI ≥30 kg/m² were more likely to discontinue insulin with SGLT-2is or DPP4is. Findings were consistent using a less conservative definition of insulin discontinuation. Although limited by potential residual confounding, misclassification, and a predominantly older male veteran population, the study suggests that GLP-1RAs, despite their established cardiovascular, renal, and weight benefits, do not promote basal insulin discontinuation more than other glucose-lowering therapies.
Click to read this study in AIM
In-Depth [retrospective cohort]: This target trial emulation evaluated whether initiating a GLP-1RA, compared with a SGLT-2i or DPP4i, increased rates of basal insulin discontinuation among US veterans with T2D. Data were obtained from the Veterans Health Administration’s Corporate Data Warehouse. Eligible participants were adults (≥18 years) with T2D receiving basal insulin who had no contraindications to GLP-1RAs, SGLT-2is, or DPP4is. Patients were excluded if they had used any of these drug classes within the previous 2 years, had contraindications (including type 1 diabetes, pancreatitis, gastroparesis, thyroid cancer, MEN-2, diabetic ketoacidosis, or diabetes during pregnancy), or had missing baseline data for Care Assessment Needs score, BMI, hemoglobin A1c, or insulin dose. The primary outcome was basal insulin discontinuation within 3 years, defined as a gap of at least 12 months without an insulin prescription. The study included 9828 GLP-1RA initiators, 25,746 SGLT-2i initiators, and 9632 DPP4i initiators. Exact and propensity score matching generated 8869 matched sets. Over 3 years, insulin discontinuation occurred in 1480 GLP-1RA users (16.7%; 95% confidence interval [CI], 15.9%-17.5%), 1585 SGLT-2i users (17.9%; 95% CI, 16.7%-19.1%), and 1517 DPP4i users (17.1%; 95% CI, 15.4%-18.9%). No significant differences were observed between groups (GLP-1RA vs SGLT-2i: risk ratio [RR], 0.93 [95% CI, 0.86-1.01]; GLP-1RA vs DPP4i: RR, 0.98 [95% CI, 0.87-1.09]). Patients younger than 65 years and those receiving <50 units/day of insulin were more likely to discontinue insulin across all treatment groups. However, no subgroup demonstrated a comparative advantage for GLP-1RAs. A sensitivity analysis using a less conservative definition of insulin discontinuation (≥6-month gap over 2 years) produced similar findings, with no significant differences between treatment groups. Overall, GLP-1RA initiation did not significantly increase basal insulin discontinuation compared with SGLT-2is or DPP4is.
Image: PD
©2026 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.
