Abstract

Cryptococcal meningoencephalitis is an urgent global health problem. Induction regimens using 14 days of amphotericin B deoxycholate (dAmB) are considered the standard of care but may not be suitable for resource-poor settings. We investigated the efficacy of conventional and abbreviated regimens of dAmB for cryptococcal meningoencephalitis in both murine and rabbit models of cryptococcal meningoencephalitis. We examined the extent to which immunological effectors contribute to the antifungal effect. We bridged the results to humans as a first critical step to define regimens suitable for further study in clinical trials. There were significant differences in the murine plasma-versus-cerebrum dAmB concentration-time profiles. dAmB was detectable in the cerebrum throughout the experimental period, even following the administration of only three doses of 3 mg/kg. dAmB induced a fungistatic effect in the cerebrum with a 2- to 3-log10 CFU/g reduction compared with controls. The effect of 3 days of therapy was the same as that of daily therapy for 14 days. There was no evidence of increased numbers of CD3(+) CD4(+) or CD3(+) CD8(+) cells in treated mice to account for the persistent antifungal effect of an abbreviated regimen. The administration of dAmB at 1 mg/kg/day for 3 days was the same as daily therapy in rabbits. The bridging studies suggested that a human regimen of 0.7 mg/kg/day for 3 days resulted in nearly maximal antifungal activity in both the cerebrum and cerebrospinal fluid. An abbreviated regimen (3 days of therapy) of dAmB appears to be just as effective as conventional induction therapy for cryptococcal meningoencephalitis. IMPORTANCE: Cryptococcal meningitis is a significant and neglected infection that is associated with excessive morbidity and mortality. In well-resourced health care settings, induction therapy with at least 2 weeks of amphotericin B deoxycholate (dAmB) is advocated. Multiple clinical studies suggest that dAmB is the drug of choice for cryptococcal meningitis. In many parts of the world where the burden of cryptococcal meningitis is highest, it is infeasible to administer dAmB for prolonged periods. This paper provides the experimental basis for the efficacy of abbreviated regimens of dAmB for cryptococcal meningitis. The concept was explored in two well-validated laboratory animal models of cryptococcal meningitis, and the results were bridged to humans by using a range of mathematical modeling techniques. A 3-day regimen is as effective as the standard 14-day course. An abbreviated regimen is significantly more feasible and may enable better antifungal therapy to be administered to many patients with this frequently lethal disease.