Tirzepatide reduced heart failure events and improved symptoms and renal function in HFpEF with obesity irrespective of CKD; creatinine- versus cystatin C–based eGFR can diverge and change differently with therapy.

Background

Obesity contributes to both HFpEF and CKD through hemodynamic, neurohormonal, and inflammatory pathways. Incretin-based therapies reduce adiposity and may directly affect renal hemodynamics and tubular function. Estimating GFR in obesity is challenging: muscle and fat mass skew creatinine and cystatin C synthesis, respectively, potentially biasing eGFR during weight-loss therapies.

Patients

• Adults ≥40 years with HFpEF (LVEF ≥50%) and BMI ≥30 kg/m².
• Functional limitation: 6-minute walk distance 100–425 m and KCCQ-CSS ≤80.
• Evidence of elevated filling pressures or left atrial enlargement (NT-proBNP elevation not required).
• Enrichment for CKD (eGFR <70 mL/min/1.73 m² allowed/encouraged); median age higher and HF severity greater in CKD subgroup.
• Baseline eGFR (mean): cystatin C 55.3 vs creatinine 64.4 mL/min/1.73 m² (≈9 mL/min/1.73 m² lower by cystatin C); CKD prevalence 61% (cystatin C) vs 46% (creatinine).

Intervention

Tirzepatide 2.5 mg subcutaneously weekly, titrated by 2.5 mg every 4 weeks to a maximum of 15 mg/week as tolerated; continued double-blind for the trial duration.

Control

Matching placebo, with standard-of-care therapies permitted in both groups.

Outcome

• Primary outcomes
  • Time-to-first cardiovascular death or worsening heart failure event (hospitalization, urgent IV therapy, or oral diuretic intensification).
  • Change in KCCQ-CSS at 52 weeks.
• Key secondary outcomes
  • Change in 6-minute walk distance at 52 weeks.
  • Percent change in body weight and hsCRP at 52 weeks.
• Additional prespecified assessments
  • EQ-5D-5L Health Index, PGIS, NYHA class.
  • Renal physiology: eGFR by creatinine (CKD-EPI) and cystatin C at 12/24/52 weeks; UACR at 24/52 weeks.

Study Design

SUMMIT was a multicenter, randomized, double-blind, placebo-controlled, parallel-group trial (1:1 allocation; n=731), stratified by recent HF decompensation, history of type 2 diabetes, and BMI ≥35 vs <35 kg/m². Intention-to-treat analyses used Cox models for time-to-event and ANCOVA/mixed models for continuous outcomes.

Level of Evidence

Level I (randomized, double-blind, placebo-controlled trial).

Follow up period

• Clinical events: median 104 weeks (treatment continued until last patient reached 1 year).
• Health status and physiological endpoints: primary 52-week assessments; eGFR at 12/24/52 weeks; UACR at 24/52 weeks.

Results

Primary outcomes

• Composite of CV death or worsening HF (overall): 9.9% with tirzepatide vs 15.3% with placebo; HR 0.62 (95% CI 0.41–0.95).
  • NNT ≈ 19 over ≈2 years.
• By CKD (cystatin C–defined)
  • CKD: 12.7% vs 18.8%; HR 0.67 (95% CI 0.41–1.08). NNT ≈ 16.
  • No CKD: 6.2% vs 9.8%; HR 0.56 (95% CI 0.24–1.32). NNT ≈ 28.
• KCCQ-CSS at 52 weeks (overall): +6.9 points (95% CI 3.3–10.6) vs placebo.
  • CKD: +7.0 (95% CI 2.8–11.2); No CKD: +6.2 (95% CI 0.9–11.5).

Key secondary outcomes

• 6-minute walk distance (52 weeks): improvement vs placebo; CKD +10 m (−4 to +24); no CKD +26 m (+11 to +42).
• Body weight: −11% to −12% greater reduction vs placebo across CKD strata.
• hsCRP: ~33%–37% greater reduction vs placebo across CKD strata.
• Quality of life and function: EQ-5D-5L, PGIS, and NYHA class improved similarly with and without CKD.

Renal outcomes and biomarker discrepancies

• Baseline eGFR discrepancy: eGFR-cystatin C ≈9 mL/min/1.73 m² lower than eGFR-creatinine across the range with wide individual variance; CKD identified more often by cystatin C (61% vs 46%).
• eGFR trajectory with tirzepatide:
  • Creatinine-based: early dip at 12 weeks (≈−3.0 mL/min/1.73 m²), followed by net improvement at 52 weeks (+1.9; 95% CI 0.2–3.7), most evident in baseline CKD (interaction trend).
  • Cystatin C–based: no significant early dip; clearer improvement at 52 weeks (+2.9; 95% CI 0.9–4.9), consistent across baseline eGFR.
• Albuminuria:
  • UACR reduction: −25% at 24 weeks; −15% at 52 weeks (borderline).
  • Macroalbuminuria at 52 weeks: 8.8% (tirzepatide) vs 15.0% (placebo). NNT ≈ 16 to prevent one case at 52 weeks.
• Interpretation: Long-term renal benefit is supported by both creatinine- and cystatin C–based eGFR, but each may be biased by changes in muscle (creatinine) and fat mass (cystatin C) during therapy; individual-level discordance was common.

Safety

• Gastrointestinal adverse events were more frequent with tirzepatide than placebo, irrespective of CKD status.

Limitations

• eGFR was estimated (creatinine, cystatin C); no gold-standard measured GFR (e.g., inulin/iohexol clearance).
• Body composition changes likely bias both creatinine- and cystatin C–based eGFR during incretin therapy.
• Trial underpowered for renal hard outcomes (e.g., ESKD); renal analyses were secondary/exploratory.
• Population enriched for CKD and obesity-related HFpEF; external generalizability to other HFpEF phenotypes may vary.

Funding

Sponsored by Eli Lilly and Company. Some authors are employees of Eli Lilly; others report research support and consultancies with multiple industry partners. Open access under CC BY-NC-ND 4.0.

Citation

Packer M, Zile MR, Kramer CM, Murakami M, Ou Y, Borlaug BA; the SUMMIT Trial Study Group. Interplay of Chronic Kidney Disease and the Effects of Tirzepatide in Patients With Heart Failure, Preserved Ejection Fraction, and Obesity: The SUMMIT Trial. Journal of the American College of Cardiology. 2025;-. doi:10.1016/j.jacc.2025.03.0092.