Micafungin pharmacodynamics predict clinical outcomes in hospitalized patients with candidemia caused by certain Candida species.

Abstract

Echinocandins are guideline-preferred therapies for invasive candidiasis (IC). Fixed dosing of echinocandins is commonly used despite variations in body mass index and echinocandin susceptibility. The purpose of this study was to evaluate clinical outcomes of micafungin based on population-predicted pharmacokinetic/pharmacodynamic (PK/PD) factors and susceptibility. Candida isolate results were screened from bloodstream or intraabdominal cultures of hospitalized patients admitted to a quaternary-care teaching hospital. Patients with a first episode of IC who received micafungin for at least 48 h were included. Patients with mixed cultures or Candida species with no minimum inhibitory concentration (MIC) differences were excluded. Breakpoints for micafungin MIC and area under the curve (AUC)/MIC ratio were calculated using classification and regression tree (CART) analysis and related to clinical outcomes. Primary efficacy outcome was candida-contributable mortality, defined as mortality within 28 days of positive culture with concomitant micafungin treatment failure; secondary outcome was micafungin treatment failure within 28 days, MAIN RESULTS: Seventy-two patients were included of whom 15 (21%) had Candida-contributable mortality and 34 (47%) experienced micafungin treatment failure. C. albicans and C. tropicalis did not have differing MICs and these patients were excluded from the study. Mortality using a CART-derived MIC breakpoint of ≥1.0 mg/L was 38% compared to 9% in patients infected with lower MIC strains (p = 0.003). Patients with a CART-derived AUC/MIC value >138.5 had a mortality rate of 9% compared to 41% for patients with AUC/MIC values below the breakpoint (p = 0.0013). Results were similar for treatment failure rates, and both were confirmed using multivariable models. CART-derived micafungin MIC and AUC/MIC breakpoints predicted patient mortality and treatment failure for certain Candida species. These results support the need for further PK/PD studies to optimize echinocandin dosing and improve patient outcomes.