Evaluation of outcomes after initiating triple antidiabetic therapy with a GLP-1 RA in an integrated health care system

Abstract

BACKGROUND: Type 2 diabetes (T2D) is characterized by chronic hyperglycemia and can lead to life-threatening complications if not treated. A stepwise and patient-centered approach is recommended when managing patients with T2D. Metformin is the preferred first-line agent, while sulfonylureas (SU) are often chosen as second-line agents. If a patient’s hemoglobin A1c (A1c) goal is not achieved despite 3 months of treatment with dual therapy, then triple therapy is recommended. However, due to the lack of head-to-head trials for different triple antidiabetic regimens, the recommendations are unclear for selection of an optimal third-line agent. OBJECTIVE: To evaluate the comparative effectiveness of a glucagon-like peptide-1 receptor agonist (GLP-1 RA) compared with a thiazolidinedione (TZD) or insulin as a third-line add-on therapy in patients who have not achieved A1c goals while receiving metformin and SU dual therapy in the real-world setting within an integrated health care system. METHODS: This is a retrospective cohort study of adult patients with T2D who were not at goal A1c while on dual therapy with metformin and an SU and initiated triple antidiabetic therapy. The primary outcome was the proportion of patients who achieved goal A1c within 3-7 months after starting triple therapy with a GLP-1 RA compared with a TZD or insulin. Goal A1c was defined as an A1c of <7% for patients aged less than 65 years and A1c of <8% for patients aged 65 years or older. Secondary outcomes included mean change in A1c, mean change in weight, and the proportion of patients with an emergent health care encounter due to a hypoglycemic event. Propensity score matching was used to select comparison groups from the insulin and TZD groups with similar baseline characteristics to the GLP-1 RA group in a 4:1 ratio. RESULTS: 274 patients initiated a GLP-1 RA in addition to dual therapy with metformin and an SU. A propensity matched group of 1,096 patients who initiated insulin and 1,096 patients who initiated a TZD were selected as the control groups. Addition of a GLP-1 RA resulted in a significantly lower proportion of patients achieving goal A1c (23.0%) compared with the addition of a TZD (30.8%, P=0.011). There was no significant difference with the addition of a GLP-1 RA when compared with insulin (24.1%, P=0.704). CONCLUSIONS: This study reflects data from real-world practice in a large integrated health care system. Significantly less patients achieved goal A1c with the addition of a GLP-1 RA as a third-line add-on option to dual therapy with metformin and an SU compared with the addition of a TZD. Providers and patients should carefully weigh the risks and benefits of different antidiabetic agents when choosing triple therapy regimens.