Improved treatment for cryptococcal meningitis, a leading cause of death in people with HIV

Dr David Lawrence (bottom left) presenting to IAS 2021.
Dr David Lawrence (bottom left) presenting to IAS 2021.

A study conducted in sub-Saharan Africa has found that only one high-dose of liposomal amphotericin B is non-inferior to the standard treatment of HIV-associated cryptococcal meningitis, and much easier to administer. It has also revealed that the single high-dose of the drug is associated with significantly fewer drug-related adverse events than those found with the standard-of-care regimen.

The fungus Cryptococcus is the cause of a major opportunistic infection in people living with HIV and a CD4 count below 100. Infection often leads to meningitis, the dangerous swelling of the membrane surrounding the brain and spinal cord. Cryptococcal meningitis is the second leading cause of AIDS-related mortality after tuberculosis.

The World Health Organization (WHO) recommends the following as first-line treatment of the infection in adults living with HIV:

  • Induction: one week with daily intravenous amphotericin B (1mg/kg), in partnership with oral flucytosine 100mg/kg per day, followed by 7 extra days of daily oral fluconazole at high doses (1200mg).
  • Consolidation: 8 weeks of daily oral fluconazole 800mg.
  • Maintenance (or secondary prophylaxis, until the patient reaches a stable immune reconstitution): daily oral fluconazole 200mg.

However, amphotericin B is associated with significant adverse events such as anaemia, impaired renal function and abnormalities in electrolytes such as potassium and magnesium. Also, safe delivery of the treatment is not feasible or sustainable in many African healthcare centres, due to the need for two weeks hospitalisation, the complexities of providing the drug intravenously, the cost of extra intravenous fluids and electrolytes, and above all, the availability of rapid, reliable laboratory monitoring. All of this can be expensive for healthcare systems and patients. Prolonged hospitalisation also increases the risk of a hospital-acquired infection.

An alternative formulation of the drug is liposomal amphotericin B, branded as AmBisome by Gilead Sciences. Although its original patent expired in 2016, generic competition and access to the liposomal formulation remains limited in low- and middle-income countries.

The liposomal formulation of a drug is more easily absorbed by the body and better distributed in the target organ. Previous studies have shown that AmBisome is less toxic and can be given in large doses that remain in the brain for some time. A recent Phase II study with 80 HIV-positive patients found that given as a single high-dose, AmBisome is effective at clearing Cryptococcus from the brain. Therefore, this approach needed to be tested in a larger number of individuals.

Last week, in a late-breaker session of the 11th International AIDS Society Conference on HIV Science (IAS 2021), Dr David Lawrence from the London School of Hygiene and Tropical Medicine and the Botswana Harvard AIDS Institute Partnership presented the first data from the AMBITION-cm Phase III trial.

The trial’s objective was to evaluate whether only one 10mg/kg dose of AmBisome, paired with flucytosine plus fluconazole during 14 days, was non-inferior to the standard WHO-recommended first-line regimen in terms of reducing the number of deaths. For this, the two approaches were compared. The primary outcome was all-cause mortality at 10 weeks, with a non-inferiority margin at 10%. Major secondary outcomes included all-cause mortality at 2, 4 and 16 weeks; early fungicidal activity; and treatment safety.

Adults with a first episode of HIV-associated cryptococcal meningitis were enrolled from eight hospitals across Uganda, Malawi, Zimbabwe, Botswana and South Africa. After the induction phase, all participants received fluconazole 800mg/day for 8 weeks, then 200mg/day thereafter, as per the standard of care. Antiretroviral therapy was started, re-started or switched (around two-thirds of participants were using or had previously used antiretroviral therapy) four to six weeks after starting the treatments on study.

In total, 814 patients were enrolled and randomised to one of the two arms. The two study arms were well matched at baseline and represented the general population of patients with cryptococcal meningitis accurately: 60% were male, the median age was 37. The two arms were similar in terms of their level of consciousness as well as cerebrospinal fluid fungal count (measured from a lumbar puncture sample).

For all-cause mortality, results at 10 weeks are:

  • 101 deaths in the AmBisome arm, giving a mortality risk of 24.8% (95% CI; 20.7 to 29.3).
  • 117 deaths in the control arm, giving a mortality risk of 28.7% (95% CI; 24.4 to 33.4).
  • The risk difference between the former and the latter arms was -3.93% (90% CI; -9.0 to 1.2), which was well below the pre-specified 10% non-inferiority margin.

These results are from the intention-to-treat data analysis, i.e. when all trial participants were included in the final analysis, whether or not they fully completed their treatment course. Results from the per protocol analysis (removing data from patients not completing the treatment course) were similar. Comparable results were also found after adjusting for several factors that could affect the results including cerebrospinal fluid fungal count, age, sex and use of antiretrovirals, with a risk difference of -5.71% (-10.5 to -1.0).

Moving to the secondary outcomes, the mortality data at 2, 4 and 16 weeks showed liposomal amphotericin B to be non-inferior in all types of analysis.

Regarding the early fungicidal activity (clearance of Cryptococcus from the cerebrospinal fluid), the single high-dose of the liposomal formulation proved to be as efficient as the standard of care.

AmBisome had clear benefits in terms of safety. Within the first 21 days after randomisation, grade 3 or 4 adverse events occurred less frequently in the AmBisome arm (382 vs 579 in the control arm). Grade 3 or 4 anaemia was significantly less frequent in the AmBisome arm, occurring in 13% of participants, compared to 41% in the control arm. The mean haemoglobin drop over the first week was 0.3g/dL in the AmBisome arm and 1.9g/dL in the control arm, and as a result, more participants in the control arm needed blood transfusions. Meanwhile, the average creatinine increase from baseline to day 7 was 20.2% in the AmBisome arm and 49.7% in the control arm. Thrombophlebitis requiring antibiotic therapy occurred in 2% of AmBisome participants and in 7% of their control arm counterparts. Finally, a low frequency of grade 4 neutropoenia, thrombocytopaenia and elevated ALT was reported from the two arms.



Inflammation of the outer lining of the brain. Potential causes include bacterial or viral infections.


control group

A group of participants in a trial who receive standard treatment, or no treatment at all, rather than the experimental treatment which is being tested. Also known as a control arm.


A type of fungal infection usually affecting the membrane around the brain, causing meningitis. It can also affect the lungs and chest.

middle income countries

The World Bank classifies countries according to their income: low, lower-middle, upper-middle and high. There are around 50 lower-middle income countries (mostly in Africa and Asia) and around 60 upper-middle income countries (in Africa, Eastern Europe, Asia, Latin America and the Caribbean).


Refers to the mouth, for example a medicine taken by mouth.

Lawrence highlighted that AMBITION-cm is the largest cryptococcal meningitis study conducted to date, demonstrating that single, high-dose AmBisome given with flucytosine and fluconazole is as good as the current WHO-recommended standard of care in avoiding mortality at 10 weeks. Moreover, the alternative regimen effectively cleared Cryptococcus from around the brain and was associated with a significant reduction in adverse events, including significantly lower rates of anaemia, a reduced need for blood transfusions and a significantly smaller increase in creatinine levels.

Lawrence said that liposomal amphotericin B offers a practical, easier-to-administer and better tolerated treatment for HIV-associated cryptococcal meningitis in Africa that has the potential to reduce the length of hospital admissions.

“The real challenge begins now in ensuring that this regimen can be made available to the individuals that need it most,” he told the conference. The single high-dose requires around ten vials of AmBisome, which each cost between $80 and $200 USD in southern Africa. In 2018, Gilead promised to reduce the price per vial to $16 in 116 low- and middle-income countries. However, the lower price has so far been applied in just 48 of the 116 countries, of which only 22 have the product registered and available.

Production of the drug is limited, as is competition from generic manufacturers, in part due to Gilead considering their liposomal technology to be a trade secret. Last week Médecins Sans Frontières (MSF) called on the company to immediately ensure an adequate global supply of liposomal amphotericin B by expanding its ‘access’ price to all low-and middle-income countries.

For many people living with HIV in Africa, these results constitute a major breakthrough. Delegates called on WHO to update its guidelines on the treatment of cryptococcal meningitis without delay.


Lawrence DS et al. Single high-dose liposomal amphotericin based regimen for treatment of HIV-associated Cryptococcal Meningitis: results of the phase-3 Ambition-cm Randomised Trial. 11th IAS Conference on HIV Science, abstract OALB01LB03, 2021.

Update: Following the conference presentation, this study was published in a peer-reviewed journal:

Jarvis J et al. Single-Dose Liposomal Amphotericin B Treatment for Cryptococcal Meningitis. The New England Journal of Medicine, 386:1109-1120, March 2022.

DOI: 10.1056/NEJMoa2111904