OBJECTIVES: In 2019, isavuconazole was approved in Brazil by ANVISA for the treatment of invasive aspergillosis (IA) and invasive mucormycosis (IM). The primary objective was to assess the cost-effectiveness of isavuconazole compared to reference voriconazole for suspected IA in the Brazilian Private Healthcare System (PHS). The secondary objective was to assess the price impact in an economic evaluation of generic voriconazole versus isavuconazole without discount on list price, and generic and reference voriconazole with a 60% discount.

METHODS: A 5-year decision-tree was developed from the PHS perspective. Patients with possible IA entered the model, with 5.75% of them assumed to actually have mucormycosis. It was also assumed that pathogen information would become available during the treatment course for only 61% of patients, with differential diagnosis unavailable for the remainder. Efficacy parameters were extracted from SECURE/VITAL trials. Costs included treatment acquisition, hospitalization, and adverse events. 3- and 10-year time horizon alternative scenarios and deterministic/probabilistic sensitivity analyses were conducted.

RESULTS: The base case analysis showed that isavuconazole was associated with a total cost saving of R$ 95,204 per patient, when compared to reference voriconazole. In this scenario versus generic voriconazole, isavuconazole was associated with a saving of R$ 66,685. Considering the discount simulation, isavuconazole compared to reference voriconazole resulted in a saving of R$ 46,333 and when compared to generic voriconazole resulted in a saving of R$ 34,926. All scenarios showed that isavuconazole dominated voriconazole. Results were robust in sensitivity analyses, 3- and 10-year ICURs were also dominant, relative to a willingness-to-pay threshold of R$ 40,688/QALY.

CONCLUSIONS: This study demonstrates that, in comparison to voriconazole, isavuconazole is the superior strategy for treatment of patients with suspected IA, regardless of it being the reference, voriconazole, or generic.