UNC Libraries
Estimating the True MACE Benefits From Tirzepatide in SURPASS-CVOT Using an Imputed Placebo Analysis of REWIND
By Steven E Kahn, Michael A Nauck, Imre Pavo and 18 others.
Abstract
Objective
In prespecified analyses, the treatment effect for MACE of tirzepatide compared with imputed placebo was estimated using SURPASS-CVOT and REWIND data.
Research design and methods
The indirect comparison with placebo was conducted for primary (MACE-3) and secondary outcomes of SURPASS-CVOT. The analysis included data from REWIND participants who would have been eligible for SURPASS-CVOT and all participants from SURPASS-CVOT. Propensity score estimation was used to adjust for differences in participant characteristics between studies. The indirect analysis of the treatment effect was derived by multiplying the hazard ratio (HR) for MACE-3 between tirzepatide and dulaglutide in SURPASS-CVOT by the HR for dulaglutide versus placebo in REWIND. Sensitivity analyses were performed using unadjusted analyses, including both the selected REWIND and the entire REWIND populations, and adjusted analysis in the entire REWIND population. Post hoc sensitivity analyses used data from a recent GLP-1RA meta-analysis that included REWIND.