Abstract Body

The WHO recommends high dose 1200 mg fluconazole (FLU) if amphotericin B (AmB) is not available or not feasible to use in persons with AIDS associated cryptococcal meningitis (CM) in resource limited settings. We report the safety and efficacy of high dose (1200-2000mg/d) oral FLU for the initial treatment of AIDS-associated CM in an AmB-controlled 2-stage dose escalation study.

HIV-infected adults (18years and older) with acute CM were randomized 3:1(Stage 1) or 2:1 (Stage 2) to high dose oral FLU or AmB. Stage 1 was a dose-finding escalation study of FLU (1200, 1600 and 2000 mg/d for up to 10 weeks) compared with AmB (0.7 mg/kg daily for 2 wks followed by low-dose fluconazole). Stage 2 enrolled simultaneous cohorts of the safe and effective FLU (1600 and 2000 mg/d) vs. AMB for safety and efficacy. Quantitative CSF cultures were evaluated every 2 wks until negative or wk 10 (therapeutic efficacy). After wk 10, all participants took 200 mg of FLU through wk 24 (protocol completion). Antiretroviral therapy (ART) was deferred for 4 wks if ART naïve and in Stage 2, ART was also permitted if initiated before enrollment. FLU dose adjustments were for weight and concomitant rifampin. The 1200mg FLU dose showed inferior efficacy and was excluded from Stage 2. The participants from Stage 1 and Stage 2 were combined in the analysis.

A total of 48, 22, 50 and 48 patients were in the safety analysis in the AmB, 1200mg FLU, 1600mg FLU and the 2000mg FLU groups respectively. There were 46, 20, 45 and 43 in the efficacy analysis. The groups were matched for demographic, clinical and laboratory characteristics. The clearance of crypto (KM prop: 90% CI) was 81%:71-90; 45%:29-65; 56%:45-69; and 60%:49-73. The mortality (KM% over 24 weeks) was 24, 41, 30 and 36% respectively. Log10 baseline CSF crypto CFUs and baseline GCS (below 15) were significant independent determinants of efficacy: p<0.009 and p<0.0043 respectively. In the safety population baseline log10 HIV viral load, log10 baseline CFUs, and baseline GCS (less than 15) were significant independent determinants: p<0.020, p<0.003 and p<0.022 respectively.

These results suggest that the WHO recommendation of FLU 1200 mg may be too low. It appears that weight based1600mg oral FLU is the most effective dose and safer than 2000mg and may be preferable to use over 1200 mg in conditions where AmB is unavailable. Overall, higher dose FLU was found to safe and well tolerated but less effective than AMB across all doses.