Abstract Body (Do not enter title and authors here): Background: Mineralocorticoid receptor antagonists (MRAs) improve outcomes in patients (pts) with heart failure and reduced ejection fraction (HFrEF) but are underused due to hyperkalemia (HK) risk.
Hypothesis: We tested if sodium zirconium cyclosilicate (SZC) can enable optimization of the MRA spironolactone (SPL) in pts with HFrEF and HK.
Methods: REALIZE-K (NCT04676646) was a prospective, double-blind, placebo-controlled, randomized-withdrawal trial in pts with HFrEF (NYHA Class II–IV; left ventricular ejection fraction ≤40%), optimal HF therapy (except MRAs), and prevalent or incident MRA-induced HK. During the open-label run-in, pts underwent SPL titration (target: 50 mg/day); pts with prevalent HK or incident HK during SPL titration started SZC. Pts maintaining normokalemia (NK; serum K⁺ 3.5–5.0 mEq/L) on SZC and SPL ≥25 mg/day were randomized to continued SZC or placebo (PBO). The primary endpoint was optimal treatment response, defined as NK on SPL ≥25 mg/day without HK rescue therapy since prior visit (Months 1–6). The 5 key secondary endpoints were tested hierarchically.
Results: Overall, 203 pts were randomized. SZC led to a greater occurrence of response vs PBO (OR 4.45 [95% CI 2.89–6.86], P<0.001; estimated %: 71% vs 36%). SZC (vs PBO) improved the first 4 of 5 key secondary endpoints: NK on the SPL randomization dose without rescue therapy (OR 4.58 [2.78–7.55], P<0.001; estimated %: 58% vs 23%); being on SPL ≥25 mg/daily (OR 4.33 [2.50–7.52], P<0.001; estimated %: 81% vs 50%); time to first HK episode (HR 0.51 [0.37–0.71], P<0.001); and time to first decrease/discontinuation of SPL dose due to HK (HR 0.37 [0.17–0.73], P=0.006) (Figure). Treatment-emergent adverse events (AEs; 64% vs 63%) and serious AEs (23% vs 22%) were balanced for SZC and PBO, respectively. Numerically more pts had an adjudicated HF event with SZC (13%) vs PBO (4%). In post hoc analysis, imbalance in HF events was limited to pts with baseline NTproBNP >4000 pg/ml (10 vs 1, SZC vs PBO).
Conclusion: In pts with HFrEF and HK, SZC led to large improvements in the percentage of pts with NK while on optimal dose of SPL. Imbalance in HF events was observed in SZC vs PBO but was limited to those with very high natriuretic peptide levels at baseline.