Optimizing antifungal therapy for HIV-associated cryptococcal meningitis in Africa

We will test the activity of alternative doses and combinations of antifungal drugs in the treatment of cryptococcal meningitis in Africa. Cryptococcal meningitis is a very common infection and cause of death in patients with HIV and AIDS in Africa.
Current drug treatment takes a long time to control the infection. In addition, improvements in treatment have been slow because traditional trials of treatment need many patients and are therefore slow and expensive to undertake. However, we have now developed a new way to test the activity of new drug combinations and dosages in small numbers of patients, by measuring the rate of killing of the organism causing the infection.
Using this method, we will test alternative drug combinations and dosages in patients that laboratory testing and animal experiments have suggested will be more active and also safe. Parallel studies will be done: to optimize doses and combinations with the intravenous drug, amphotericin B, and with the oral drug, fluconazole.
The results of the studies will be used to determine how best to use the drugs that are currently available, and to select the best combinations for further trials to find out if the new combinations can reduce the number of deaths caused by this infection.

Cryptococcal meningitis remains a very common infection, and the cause of death in 13-44% of HIV-infected patients in Africa. Current antifungal drug regimens still frequently take more than two weeks to sterilise the cerebrospinal fluid (CSF), and improvements in treatment have been slow due to the large number of patients needed for clinical endpoint trials. However, we have now developed and validated a new and more powerful marker of treatment response, the rate of decrease in cryptococcal colony-forming units (CFU) in cerebrospinal fluid (CSF) over the first 2 weeks, or early fungicidal activity, based on serial quantitative cultures of CSF (Brouwer et al. Lancet 2004; 363:1764). Pre-clinical and limited clinical data suggest novel drug combinations and higher than standard doses of some antifungal drugs may be more effective in treatment of cryptococcosis, without a significant increase in toxicity. Therefore, using this new endpoint, the objective of the proposal is to optimise initial antifungal treatment for HIV-associated cryptococcal meningitis in Africa. In a centre where facilities make amphotericin B-based therapy possible, randomized studies will determine the optimal amphotericin B (AmB) dose(0.7 mg/kg/d versus 1 mg/kg/d), and the optimal second drug and dose (flucytosine 100 mg/kg/d versus fluconazole 800 mg/d versus fluconazole 1200 mg/d). A second, parallel, dose-ranging, cohort study, in a centre, like many in Africa, where safe amphotericin B-based therapy is currently not possible and fluconazole is the only option for treatment, will define the dose-activity relationship for fluconazole monotherapy as initial therapy. The power of the fungicidal activity endpoint means the aims can be achieved with relatively few patients, facilitating close hands-on supervision. Enrolment is underway in pilot phase studies at both sites, confirming the feasibility of the work at these sites and the estimates for recruitment. The results of the studies will be used to (a) determine best use of currently available drugs for treatment of cryptococcal meningitis in Africa (b) taking account of fungicidal activity results, safety, and accessability issues, select the best regimens for testing in a phase 3, clinical endpoint trial.