Tirzepatide Rivals Top GLP-1 Drugs on Heart Outcomes in Type 2 Diabetes

Over the past decade, GLP-1 receptor agonists (GLP-1RAs) have become pillars of cardiovascular risk reduction in type 2 diabetes. Tirzepatide, a novel dual glucose-dependent insulinotropic polypeptide (GIP) and GLP-1 receptor agonist, demonstrated non-inferiority to dulaglutide in the SURPASS-CVOT trial, but its broader standing relative to placebo and the full GLP-1RA class remained unclear—a gap this analysis was designed to fill.

Researchers conducted a frequentist network meta-analysis (NMA) using REML estimation with Hartung-Knapp-adjusted confidence intervals. They identified 11 eligible RCTs from MEDLINE and Embase (10 GLP-1RA trials and SURPASS-CVOT), all enrolling adults with T2D and established ASCVD or high cardiovascular risk, with ≥1,000 participants and ≥12 months follow-up. A hierarchical framework separated class-level (tirzepatide vs. pooled GLP-1RA class vs. placebo) from agent-level (tirzepatide vs. each individual GLP-1RA and placebo) analyses across five outcomes: MACE, CV mortality, all-cause mortality, non-fatal MI, and non-fatal stroke. All trials were rated low risk of bias via Cochrane RoB 2.

In the class-level analysis, tirzepatide produced numerically larger reductions than the pooled GLP-1RA class across every endpoint versus placebo. Reductions versus placebo were: MACE HR 0.79 (95% CI 0.69–0.91), CV mortality HR 0.77 (0.66–0.90), all-cause mortality HR 0.74 (0.65–0.83), non-fatal MI HR 0.77 (0.61–0.97), and non-fatal stroke HR 0.79 (0.64–0.97). The GLP-1RA class also significantly reduced all five outcomes versus placebo, with HRs ranging from 0.86 to 0.90. Heterogeneity was low to moderate (I² 2.8–31.4%). A formal statistical comparison between tirzepatide and the GLP-1RA class could not be made because the tirzepatide node was connected to the network solely through SURPASS-CVOT.

In the agent-level analysis, tirzepatide significantly reduced MACE versus placebo (HR 0.81, 95% CI 0.70–0.94) and versus lixisenatide (HR 0.79, 95% CI 0.65–0.97). Comparisons with albiglutide, dulaglutide, efpeglenatide, exenatide, liraglutide, and semaglutide showed no statistically significant differences, consistent with broadly comparable efficacy. Agent-level heterogeneity was negligible (I² = 0%). Subgroup analyses restricted to established ASCVD populations and sensitivity analyses excluding oral semaglutide or short-acting GLP-1RAs did not materially shift point estimates.

These results support tirzepatide as a cardiovascular risk-reducing agent with efficacy at least on par with the GLP-1RA class, and potentially superior to the weakest agents in that class. The dual GIP/GLP-1 mechanism may underlie the numerically favorable point estimates, though formal superiority cannot be claimed due to the single-trial constraint on the tirzepatide network node. The analysis strengthens the rationale for using tirzepatide in patients with T2D and high cardiovascular risk, particularly those who may also benefit from its well-established weight-reduction effects.