A US Hospital-Perspective Decision-Analytic Model of Once-Weekly Rezafungin Versus Daily Echinocandin Therapy in Hospitalized Adults With Candidemia or Invasive Candidiasis: Potential Cost Savings From Earlier Discharge.

Background: Hospitalized adults with candidemia or invasive candidiasis (C/IC) often remain inpatient to complete daily intravenous echinocandin therapy. Rezafungin, a once-weekly echinocandin, may facilitate earlier discharge in clinically stable patients. We evaluated its potential economic impact from a US hospital perspective. Method: A decision-analytic probabilistic model compared 2 treatment-discharge strategies in 100 hypothetical hospitalized adults with C/IC who had reached clinical stability and were otherwise dischargeable: continued inpatient daily echinocandin therapy until observed discharge versus administration of a single 400 mg dose of rezafungin with earlier discharge. The rezafungin strategy incorporated earlier discharge distributions from a multicenter US study. Costs included antifungal acquisition, inpatient hospitalization during the final 4 hospital days, and peripherally inserted central catheter (PICC)-related costs (2026 US dollars). Deterministic, threshold, and probabilistic sensitivity analyses (10 000 Monte Carlo simulations) were performed, focusing on excess hospital costs. Results: In the base case, mean excess cost was lower with rezafungin than standard of care (SOC) (-$412 vs $708 per patient), corresponding to savings of $1120 per patient. Savings were greater when SOC required PICC placement (-$2269) versus no PICC placement (-$851). Rezafungin was associated with higher costs when no earlier discharge occurred (+$3471) or discharge occurred only 1 day earlier (+$1475), but became cost saving with ≥2 days earlier discharge (-$556 at 2 days, -$2903 at 3 days, and -$5313 at ≥4 days). Conclusions: Once-weekly rezafungin reduced hospital costs versus daily echinocandins when it enabled discharge ≥2 days earlier, with greater savings when PICC placement was avoided.