MON-LB028 Comparing Cardiovascular Effects of GLP-1 Agonists in a Real World Setting

There is limited research using real-world data to study differences in the cardiovascular (CV) effects of glucagon-like peptide-1 (GLP-1) agonists. We conducted a retrospective active comparator Type 2 Diabetes (DM) cohort study using 2011-2015 administrative claims data to evaluate the cardiovascular (CV) effectiveness of GLP-1: exenatide extended-release (E-ER), exenatide immediate-release (E-IR), liraglutide, albiglutide, and dulaglutide. We included those with: 1) 1st pharmacy dispensing event (index date); 2) ≥ 2 fills; and 3) no evidence of antidiabetic medication (ADM) fills (besides metformin). Medical diagnoses associated with claims were coded by the International Classification of Diseases, 9(th) Revision (ICD-9). The primary study outcome was time to 1st major adverse CV event after starting index ADM and composite primary outcome of hospitalization including: ischemic heart disease, stroke, congestive heart failure, or peripheral arterial disease. Chi-square tests were used to examine associations between baseline covariates and the index GLP-1 medication. Cox proportional hazards regression was used to model the association between index GLP-1 and CV events (composite and separate models), adjusting for baseline patient, prescriber, and plan characteristics. Stratified analyses to measure effects on patient subgroups, and a time-varying Cox model, accounting for changes over time, were completed. Because too few patients received albiglutide (n=360) and dulaglutide (n=99), evaluation of CV outcomes was restricted to liraglutide (n=7531), E-ER (n=1451), or E-IR (n=1910). Prescription patterns changed over time; in 2011, 26.6% of patients were prescribed E-IR and dropped to 13.8% by 2015, with more patients prescribed E-ER. Using liraglutide as reference, neither formulation of exenatide was significantly associated with differences in the composite CV outcome [hazard ratios (HRs) for exenatide ER and IR are 1.33 [95% C.I. 0.73-2.39] and 1.30 [0.81-2.09]] respectively. Adjusting for time-varying covariates (i.e. DM and CVD Rx use during the follow up period) produced similar results. For individual CVD categories, there were no significant differences between GLP-1; for ischemic outcomes, the HR for E-IR was 1.85 [95% C.I. 0.97-3.53] relative to liraglutide. Stratified analyses resulted in similar non-significant results. The direction of the HR’s for exenatide differed for patients taking versus those not taking metformin, but were not statistically significant: E-ER with metformin: HR 1.72 [95% C.I. 0.82-3.62]; E-IR with metformin: HR 1.48 [95% C.I. 0.77-2.84]; E-ER without metformin: HR 0.92 [95% C.I. 0.31-2.74]; E-IR without metformin: HR 0.93 [0.43-2.03]. Initiating different GLP1, in comparison to liraglutide, early in the course of DM (as a 1st or 2(nd) medication to metformin), did not result in differences in CV effects in the real world. Unless otherwise noted, all abstracts presented at ENDO are embargoed until the date and time of presentation. For oral presentations, the abstracts are embargoed until the session begins. Abstracts presented at a news conference are embargoed until the date and time of the news conference. The Endocrine Society reserves the right to lift the embargo on specific abstracts that are selected for promotion prior to or during ENDO.