In the time before HIV treatment, a crypto-coccal meningitis diagnosis was a death sentence for AIDS patients. Today, with access to anti-retrovirals and anti-fungal medicine, more patients are surviving – but not enough. It is still the second leading killer of people with HIV. A local screening programme is trying to plug the gap and catch the infection before it becomes deadly. But the gold-standard treatment, with the potential to save many more lives, is not registered in South Africa – or any African country.
Crypto-coccal meningitis is an infection of the lining of the brain and spinal cord caused by the crypto-coccus fungus. Despite the fact that crypto-coccal meningitis is the second leading killer of people living with HIV, it mostly flies under the radar. According to the World Health Organisation (WHO), 181,100 people with HIV die annually from this form of meningitis, making it responsible for more than 15% of all AIDS-related deaths. The vast majority of these deaths – 135,900 – occur in sub-Saharan Africa.
Cryptococcus is a fungus found in the environment, for example in soil with pigeon droppings and sometimes trees. In people with weak immune systems, particularly those who have problems with certain immune cells called lymphocytes, the fungus reactivates or “wakes up”. From the lungs, the fungus spreads to the bloodstream and, if not caught, eventually reaches the brain where “it causes a devastating meningitis”.
Access to anti-retroviral treatment (ART), which strengthens immune systems and increases people’s CD4 cell counts, makes HIV-infected patients much less vulnerable to developing crypto-coccal meningitis. Many South Africans are still only starting ART when they are already very ill, while an “increasing number” stop taking their ART, causing their CD4 cell counts to drop again. A study conducted in Uganda before access to ART or anti-fungal treatments found that, for HIV patients, crypto-coccal meningitis “was universally fatal with a 100% mortality rate”.
In routine care, in most sub-Saharan African settings, including in South African hospitals, in about three months over 60% of patients with crypto-coccal meningitis are dead,” said Govender. This is even with access to ART and appropriate anti-fungal treatment in the form of amphotericin B and fluconazole. A more effective drug called flucytosine has the potential to bring the death rate down dramatically, but few people have access to it because it is not registered in South Africa, or any other African country – the region where it is most needed.
In a clinical trial setting where patients with cryptococcal meningitis were treated with combination flucytosine plus amphotericin B, the mortality was as low as 25% at 10 weeks. The fact that flucytosine is not registered is a major problem. WHO guidelines in 2018 recommend flucytosine as the first-option treatment for crypto-coccal meningitis.
In 2018, the Southern African HIV Clinicians Society called on the Department of Health to update its outdated guidelines in line with the WHO’s recommendations. They also called on the drug manufacturer to urgently register the drug in sub-Saharan African countries. The South African Health Products Regulatory Authority (SAHPRA) confirmed that flucytosine is not registered in South Africa – and no application for its registration has been submitted. It seems a flucytosine product was previously registered with the old Medicines Control Council, but its registration was not maintained.
Currently, the only way for patients in South Africa to get the drug is if their clinician files a section 21 application to SAHPRA, a special mechanism that allows the importation of medicines that are not registered in South Africa, providing certain conditions are met. Infectious diseases specialist at Helen Joseph Hospital in Johannesburg, Dr Jeremy Nel, described this process as “cumbersome and slow”. It often takes weeks, making it “pointless” in many cases because patients may have already died by the time the drug arrives at the hospital. If there is such a pressing need for this drug, which has been shown to reduce mortality rates by a substantial 40%, why is it not registered?
According to Dr Laura Trivino-Duran, medical co-ordinator for Doctors Without Borders (MSF) in South Africa, it boils down to a “market failure” and procedural barriers on the part of the only WHO pre-qualified manufacturer of flucytosine, the pharmaceutical company Mylan. There are other suppliers of flucytosine, but they do not yet have WHO pre-qualification. Pre-qualification indicates that a medicine meets certain standards regarding quality, safety and efficacy – but WHO pre-qualification does not replace the need for registration in South Africa and is not sufficient to allow a drug to be sold or distributed here.
The situation is compounded by a lack of awareness (including among HIV clinicians), resulting in a failure to create demand, as well as inaction on the part of the government. Mylan has indicated that it “lacked certain data”, particularly related to programmatic implementation, as its main reason for failing to register the drug locally, according to Trivino-Duran. The pharmaceutical giant has reportedly also raised concerns that it has just one manufacturing site for flucytosine, in Poland, which is unable to produce the volumes required in the region. Mylan spokesperson Ritika Verma, said in August 2016, Mylan added flucytosine, Ancotil, to its portfolio through the Meda acquisition. It is manufactured by a third party and is available for lawful sale under the waiver in South Africa and is WHO pre-qualified. To meet the increasing demand and ensure access to this important product, Mylan has prioritised the development of its own generic flucytosine to be manufactured by Mylan, and anticipates being able to supply countries like South Africa – under waiver – by the end of this year. “We plan to file with the WHO Pre-qualification Programme within the same time-frame.
Mylan is planning to file Mylan’s generic flucytosine with the South African Health Products Regulatory Authority (SAHPRA) early next year. Flucytosine is available for a hefty price the US and the UK and costs around R3,200 for a two-week course when imported under the section 21 system. Trivino-Duran said there is no clear reason why generic players have not entered the market in Africa considering the drug is relatively old and off-patent. In an effort to create demand and programmatic data, MSF, with funding from the Clinton Health Access Initiative, launched a programme in November 2018 to deliver the drug to 15 hospitals, mostly in South Africa. This was facilitated through a bulk section 21 application.
Since the programme began, 372 courses have reached patients – with dramatic results. Nel said that since the MSF programme, death rates for crypto-coccal meningitis at his hospital have fallen from more than 20% to under 10%. But the programme is ending in a matter of weeks and clinicians will be forced to revert to the situation where only one in 20 patients who need flucytosine will actually receive it. “And then avoidable mortality will presumably rise again. As a treating clinician, this is very disheartening,” said Nel.
Further bulk section 21 applications are not considered to be a sustainable solution, with Trivino-Duran arguing that it is the government’s responsibility to ensure access to the drug. A life-saving screening programme In the interim, a screening programme launched by the NICD in 2016 is trying to catch cases of crypto-coccal disease in patients before it becomes meningitis, which significantly betters the chances of survival for patients. If the fungus is identified in the blood, before it reaches the brain and treated, the death rate plummets from 60% to around 30%. I It takes an average of three weeks for the fungus to spread to the brain causing meningitis, irreparable damage or death. The NICD, in collaboration with the Department of Health and the National Health Laboratory Service (NHLS), implemented a routine screening programme in October 2016. According to national guidelines, patients who test positive for HIV in local facilities are asked to give blood so that their CD4 cell count can be determined. Under the programme, CD4 counts that fall below 100 are flagged. The NHLS then uses a simple dipstick test on the same blood sample to detect the fungus. The results of this crypto-coccal antigen test are then sent back to the facility and when the patient next visits the clinic they should be immediately started on antifungal treatment if they tested positive.
Since the programme started, more than 600,000 patients blood samples have been screened, with 34,500 testing positive for the antigen. Many of these patients had not yet developed symptoms of meningitis and were started on treatment earlier than would have ordinarily been possible. The true impact on lives saved is unknown and this data will only be available in 2021. While the programme is undoubtedly a good move, many more lives could be saved if it was expanded to include people with CD4 counts below 200, as recommended by the WHO (as opposed to the current cut-off of 100). However, the laboratory system may not be able cope with doubling the volumes to 1.4 million. A costing analysis should be done with the Department of Health.
Making a lot more noise could maybe wake up the Department of Health, but will there be enough patients left alive to do so? – DM
Edited for style and length.