Abstract 15257: The “Free Choice” Diet-Induced Obese Nonalcoholic Steatohepatitis Hamster, a Clinically Relevant Model for Evaluating Drugs Targeting Heart Failure With Preserved Ejection Fraction

Intro: Heart failure with preserved ejection fraction (HFpEF) is frequently associated with multiple comorbidities including nonalcoholic steatohepatitis (NASH) and obesity. Drug development for HFpEF is partly hampered by the limitations of preclinical models, such as diet-induced obese mouse and rat.

Hypothesis: To overcome these limitations, we recently established a nutritional hamster model developing obesity, with human features for dyslipidemia NASH, and HFpEF. To confirm the relevance of our model, we tested whether the clinical benchmarks empagliflozin (EMPA, a SGLT2 inhibitor), vardenafil (VAR, a PDE5 inhibitor) and semaglutide (SEMA, a GLP-1r agonist) would improve HFpEF.

Methods: To induce obesity, NASH and HFpEF, hamsters were fed with a free choice diet, which presents hamsters with a choice between control chow or high fat/cholesterol diet, and normal or 10% fructose water for 20 weeks. Obese hamsters were then randomized based on echocardiography parameters, i.e., E/A ratio and ejection fraction (EF) and were treated QD for 5 weeks with vehicle, EMPA (30mg/kg), VAR (10mg/kg) or SEMA (0.06mg/kg). Chow fed hamsters were used as control.

Results: Compared with control, echocardiography in obese hamsters showed cardiac remodeling, preserved systolic function and normal EF. Free choice fed hamsters had impaired diastolic function with higher E/A (2.0±0.1 vs 1.41±0.03 in control, p<0.001), while the inverted E’/A’ (0.8± 0.1 vs 1.3±0.04, p<0.001), a sign of relaxation default, suggested restrictive filling pattern. Increased E/E’ suggested higher filling pressure (18-20 vs 12-14 in control, p<0.01). Plasma NT-proBNP levels were also increased (p<0.05 vs. control). Compared with vehicle, all drug treatments alleviated diastolic dysfunction: E/A returned to control values (E/A=1.4-1.5) with VAR (p<0.001 vs. vehicle), EMPA (p<0.01) and SEMA (p<0.05). As well, E’/A’ was normalized (p<0.01 for all), and E/E’ was also reduced (p<0.05 for all). EMPA and VAR reduced NT-proBNP levels (p<0.05 vs. vehicle).

Conclusion: Our data indicate that, VAR, EMPA and SEMA improved diastolic function in the obese NASH hamster. This novel model provides a unique alternative to mice and rats limitations for evaluating drugs targeting HFpEF.