High Dose AMBISOME on a Fluconazole Backbone for Cryptococcal Meningitis Induction Therapy in sub-Saharan Africa: A Randomized Controlled Trial

Cryptococcal meningitis is a leading cause of death in HIV-infected individuals in Africa. The current recommended treatment is a drug called amphotericin B deoxycholate. Treatment with amphotericin B requires 14 days of intravenous infusions given in hospital, making it difficult and costly to administer. It also causes many side effects, including kidney failure and low blood count, making close laboratory monitoring essential. The combination of the costs associated with prolonged hospital admissions, the difficulties in administration and the need for laboratory monitoring make amphotericin B treatment difficult in much of Africa. The only alternative currently available treatment is called fluconazole. Treatment with fluconazole is inadequate, and is associated with death rates of approximately 60%.

A modified form of amphotericin B is available called liposomal amphotericin B (Ambisome). This is considerably less toxic than standard amphotericin B, and is known to be efficacious in treatment of cryptococcal meningitis. Its use has been limited by the high cost of therapy, but recent data suggest that much shorter courses of Ambisome may be effective in the treatment of cryptococcal meningitis. Due to its lower toxicity, higher doses of Ambisome can be given safely, and it also persists for a long time in the tissues, raising the possibility of delivering highly effective induction therapy with very few (1, 2, or 3) doses. A large reduction in the number of doses and duration of hospitalisation, together with reduced pricing of Ambisome, may result in cryptococcal meningitis treatment costs that are not more than those with 2 weeks of conventional amphotericin B, and provide a convenient, safe and efficacious alternative to conventional amphotericin B therapy.

This study aims to define the most effective and most cost-effective schedules for Ambisome use in the treatment of cryptococcal meningitis. A currently ongoing study is testing the safety and effect on rate of clearance of cryptococcal infection of one, two or three dose Ambisome treatment regimens compared to the standard 14-day course. The shortest of these Ambisome regimens that is found to be safe and effective will be utilized in this proposed large clinical trial to determine whether or not it is as effective as the standard 14-day amphotericin B deoxycholate treatment in terms of preventing deaths from cryptococcal meningitis. If short-course Ambisome treatment regimens were shown to be of comparable effectiveness in the treatment of HIV-associated cryptococcal meningitis, the results of this study would lead to changes in international treatment guidelines, and provide an effective and practical treatment option for HIV-associated cryptococcal meningitis with the potential to prevent many thousands of deaths.

Cryptococcal meningitis (CM) is associated with 10-20% of deaths in HIV-programmes in Africa. Ten-week mortality rates with current treatments range from 30-60%; novel treatments are urgently needed. Use of standard amphotericin B therapy is limited by side effects and the need for intensive nursing care and laboratory monitoring. Recent pharmacokinetic and clinical data show it may be possible to deliver safe and effective therapy for HIV-associated CM with very few large doses of Ambisome (liposomal amphotericin B, L-AmB). L-AmB has low rates of nephrotoxicity, long tissue half-life, and good brain penetration. The concept of single or intermittent dosing with very high doses is also established in prophylaxis in haematology patients and treatment of visceral leishmaniasis (VL). Single doses of 15mg/kg have been safely given, and 10mg/kg doses are routinely given for treatment of VL.

We propose to determine whether short-course high-dose L-AmB is as effective as 14-day amphotericin B therapy (standard of care) in averting all-cause mortality in HIV-associated CM patients in a phase 3 clinical-endpoint trial. 850 patients recruited at 6 east and southern African sites will be randomised to either short course high-dose L-AmB (the shortest-course L-AmB regimen found to be safe and effective in an ongoing phase-II dose finding study) or standard 14-day amphotericin B deoxycholate, both given with fluconazole 1200 mg/d for the first 2 weeks. The primary outcome will be all-cause mortality within the first 10 weeks of treatment. Secondary objectives are to determine the fungicidal activity and safety profile of short-course high-dose L-AmB, and determine the health service costs of the two treatment regimens.

If short-course treatment regimens were shown to be effective, the results of this study would lead to evidenced-based changes in treatment guidelines, and provide a practical treatment option with the potential to prevent many thousands of deaths.

We aim to demonstrate that it is possible to deliver highly effective induction therapy for HIV-associated cryptococcal meningitis with very few (1, 2, or 3) large doses of Ambisome (liposomal amphotericin B). Ambisome is less nephrotoxic than conventional amphotericin B, allowing much higher doses to be given safely, and has a long tissue half-life. The concept of effective therapy with just one or more doses of high dose Ambisome has recently been established in the treatment of visceral leishmaniasis, and the drug is now made available at a markedly reduced price for this indication. A similar large reduction in the number of doses and duration of hospitalisation required for the treatment of cryptococcal meningitis, together with a negotiated reduction in the pricing of Ambisome for this indication, will likely result in costs that are no more, and possibly less, than 2 weeks conventional amphotericin B therapy.

Use of a safe, sustainable regimen of high dose intermittent Ambisome in African centres still relying on fluconazole monotherapy would lead to reductions in 10-week mortality from approximately 60% to the 25-35% seen with amphotericin B-based combinations. It would also markedly facilitate effective treatment, reduce duration of hospital admissions, and lower the high burden of drug related side effects and other complications such as drip-site sepsis currently seen in centres using conventional amphotericin B based therapy. Overall, if short-course Ambisome treatment regimens were shown to be of comparable microbiological and clinical efficacy in the treatment of HIV-associated cryptococcal meningitis, the results of this study would lead to evidenced-based changes in regional and international treatment guidelines, and provide an effective and practical first-line treatment option with the potential to prevent many thousands of deaths.

In addition to direct benefits to HIV-infected patients with cryptococcal meningitis in low-resource settings, and the doctors and health services treating them, the proposed study will further scientific knowledge, providing pharmacokinetic and pharmacodynamics data applicable to a range of other diseases treated with Ambisome. The planned sub-studies and collaborations will advance understanding of the pathophysiology, immunology and genetics of cryptococcal infection in the context of HIV-infection. The study will also build strong research links between leading European and African research institutions, help to further develop clinical and research capacity at the trial sites, and facilitate the development of local investigators into future research leaders. All of the African partner sites have worked hard to develop clinical trials capability and expertise, and this phase 3 trial will ensure continued expansion and improvement in this research capacity. Ultimately it is planned that the study sites developed will form the basis of an African meningitis research network, providing a framework for future clinical trials investigating urgently needed novel treatments and diagnostics strategies aimed at reducing the morbidity and mortality due to meningitis in Africa.