Glucagon-Like Peptide-1 Receptor Agonists and Dual GIP/GLP-1 Receptor Co-Agonism in Heart Failure with Preserved Ejection Fraction: A Systematic Review of Randomized and Real-World Evidence

Background: This is a reconciliation (umbrella-style) systematic review of existing randomized and observational evidence; with the exception of one exploratory two-study pooled analysis reported in Section 3.6, it does not present new primary or patient-level data. Heart failure with preserved ejection fraction (HFpEF) and heart failure with mildly reduced ejection fraction (HFmrEF) are frequently associated with obesity and cardiometabolic dysfunction. Glucagon-like peptide-1 receptor agonists (GLP-1RAs) and the dual glucose-dependent insulinotropic polypeptide (GIP)/GLP-1 receptor co-agonist tirzepatide have emerged as candidate disease-modifying therapies for obesity-related HFpEF, but the consistency of benefit across trial designs, phenotypes, and diabetes status requires critical synthesis. Objective: To systematically review randomized and real-world evidence on GLP-1 receptor agonist and dual GIP/GLP-1 receptor co-agonist therapy in heart failure with preserved or mildly reduced ejection fraction, and to critically compare these findings against independently conducted meta-analyses identified during the search. Methods: We searched MEDLINE/PubMed, Embase, the Cochrane Central Register of Controlled Trials, Scopus, and Web of Science from inception to June 2026, supplemented by ClinicalTrials.gov and major cardiology/endocrinology congress proceedings, for randomized controlled trials (RCTs), prespecified and post hoc analyses, and pooled/participant-level analyses of GLP-1RA or GIP/GLP-1 co-agonist therapy in adults with HFpEF or HFmrEF. Risk of bias was assessed using the Cochrane RoB 2 tool. Given substantial cross-trial heterogeneity in ejection fraction thresholds, obligate comorbidities, and endpoint hierarchies, we conducted a structured narrative synthesis rather than a de novo quantitative meta-analysis, and we critically compared our findings against six independently conducted meta-analyses identified during the search, including one PROSPERO-registered meta-regression synthesis. Results: Six pivotal trial-level data sources met eligibility criteria: STEP-HFpEF, STEP-HFpEF DM, SUMMIT, the SELECT heart failure substudy, the FLOW heart failure substudy, and a four-trial pooled analysis. Semaglutide and tirzepatide produced consistent placebo-corrected improvements in Kansas City Cardiomyopathy Questionnaire Clinical Summary Score (7–14 points), 6-minute walk distance, and body weight (10–14% loss), with concordant reductions in C-reactive protein and loop diuretic requirement. In SUMMIT, tirzepatide reduced the composite of cardiovascular death or worsening heart failure events by 38% (hazard ratio [HR] 0.62, 95% confidence interval [CI] 0.41–0.95), driven by fewer heart failure hospitalizations/urgent visits (HR 0.54, 95% CI 0.34–0.85), without a significant mortality effect. An independently conducted, PROSPERO-registered meta-regression synthesis of six RCT/post hoc sources and five observational cohorts corroborated this pattern (pooled HR 0.68, 95% CI 0.52–0.88 for heart failure hospitalization/events; HR 0.76, 95% CI 0.65–0.87 for the composite endpoint) and found no significant effect modification by type 2 diabetes status for any outcome (all interaction P>0.05), while pooled cardiovascular mortality (HR 0.83, 95% CI 0.69–1.01) and all-cause mortality (HR 0.86, 95% CI 0.52–1.43) remained non-significant. Gastrointestinal adverse events were consistently more frequent with active therapy, with no HFpEF-specific safety signal identified. A post hoc fixed-effect pooling of the two largest independent trial-level estimates for this composite outcome (the semaglutide four-trial pooled analysis and the SUMMIT tirzepatide trial), conducted by the review authors from published summary statistics, yielded a combined hazard ratio of 0.67 (95% CI 0.54–0.84, P=0.0004) with no detectable heterogeneity (I²=0%), though this two-study estimate should be read as exploratory. Conclusion: GLP-1RAs and tirzepatide produce robust, reproducible improvements in symptoms, function, and inflammatory/congestive biomarkers in obesity-related HFpEF and HFmrEF, with an emerging and now externally corroborated signal for reduced heart failure hospitalization that is independent of diabetes status. A mortality benefit remains statistically unconfirmed across all currently available randomized and pooled evidence. These agents represent an evidence-based, phenotype-directed addition to HFpEF/HFmrEF management, consistent with their recent incorporation into at least one national cardiovascular society guideline, pending further regulatory label expansion and mortality-powered outcome data.