The objective of this study was to investigate the safety, tolerability, and pharmacokinetics of oral voriconazole in subjects at high risk of developingfungal infections. This was a multicenter, randomized, double-blind, double-dummy, parallel-group, dose escalation study with a fluconazole active control. Twenty-four subjects with hematological malignancies, solid tumors, or autologous bone marrow transplants were randomized to receive voriconazole 200 mg q 12 h (n = 9), voriconazole 300 mg q 12 h (n = 9), or fluconazole 400 mg OD (n = 6)for a period of 14 days. Blood samples were taken for the assessment of voriconazole pharmacokinetics in plasma on Days 1 and 14. Using a 200 mg q 12 h dosing regimen, geometric mean voriconazole peak plasma concentrations (Cmax) were 904 ng/ml on Day 1 and 2996 ng/ml on Day 14. Geometric mean voriconazole exposure, as measured by the area under the curve within a dosing interval (AUCtau), was 4044 and 20308 ng x h/ml on Days 1 and 14, respectively. On Day 1, geometric mean Cmax and AUC were 1.80- and 1.94-fold higher in subjects receiving voriconazole 300 mg q 12 h than in those receiving 200 mg q 12 h. Similarly, on Day 14, geometric mean Cmax and AUC were 1.56- and 1.80-fold greater in the high-dose group. Although the confidence intervals are large, this trend suggests nonlinearity in pharmacokinetics with respect to dose as seen in healthy volunteers. The absorption of orally administered voriconazole was relatively rapid, with t(max) achieved in 1.7 to 3.0 hours. There was a mean 5.4- and 5.0-fold accumulation of voriconazole over the 14-day study period in the 200 mg and 300 mg q 12 h dose groups, respectively. Voriconazole was generally safe and well tolerated. Mild, reversible visual disturbances were the most commonly reported adverse event but were not associated with treatment discontinuation. No patient developed a breakthrough fungal infection. It was concluded that in this group of patients at risk of fungal infection, voriconazole pharmacokinetics was consistent with that reported in healthy volunteers.