News feature: Malaria treatment and HIV: co-artemether is needed, but supply and cost remain as barriers

A six-dose course of the combination of two drugs - artemether and lumefantrine (coartum or Riamet)- is a highly effective treatment for malaria in the areas of Africa where resistance to frequently used malaria drugs is common, according to two randomised controlled studies published in the April 22nd issue of The Lancet.

A Tanzanian study reported that the combination was superior to several other commonly used treatments. The other study, conducted in Uganda, found that adherence to the combination’s complex dosing regimen, although perhaps not perfect, was good enough in a 'real world African setting' to remain highly effective. However, cost and supply issues continue to keep this combination beyond the reach of most African public health systems.

Malaria increasingly resistant to inexpensive remedies

Malaria kills around two million children each year, and 90% of the burden of disease is in Africa — including many regions also afflicted by HIV/AIDS.

Malaria control is important for the health of people with HIV because acute malaria infection increases HIV viral load, and so may increase the risk of HIV transmission and disease progression.

Pregnant women with HIV are at increased risk of developing malaria even if they have previous immunity to the infection. It also increases the risk of mother to child HIV transmission and malaria transmission if placental malaria develops.

Despite ambitious efforts to “roll back malaria” the disease continues unabated, partly because in many areas the parasite that causes the disease has become increasingly resistant to inexpensive treatments, such as chloroquine and sulfadoxine-pyrimethamine (SP).

Borrowing a page from tuberculosis (TB) and HIV treatment in 2001, disease control experts issued recommendations to use combination therapy for malaria.

Artemisinin combination therapy

The most promising combinations are anchored by the ancient Chinese herbal Artemisia annua. Several ‘artemisinins’ derived from the herb have potent activity against the malarial parasite in the blood. In fact, they are the most rapidly effective antimalarial drugs known.

However, they have a very short half-life in the blood (between 30 minutes and two to three hours). Although the activity against the parasite lasts longer than that, the disease often mounts a comeback. So researchers combined the artemisinins with longer acting anti-malarials.

There are several other drugs that could be combined with the ACTs, but each has its faults. A number had already been used in some settings (with increasing resistance), or were somewhat cross-resistant with those drugs. Others are too expensive or have dangerous side effects.

The tale of co-Artemether

A drug called lumefantrine also comes from the Chinese formulary, having been synthesized at the Academy of Military Medical Sciences in Beijing. It has a similar structure and activity to mefloquine, an antimalarial widely used in Southeast Asia, but it must be administered with food that is high in fat content in order to be absorbed. This drawback is also shared by one of the artemisinins called artemether. As a result, the Chinese researchers combined the two as ‘co-artemether’, which was registered for the oral treatment of malaria in China in 1987.

Clinical trials using a four-dose course of co-artemether, which was effective in much of Asia, were disappointing in Africa. Studies in Thailand found that for the treatment to be effective in more difficult settings, a six-dose regimen was necessary and the combination needed to be dosed in a very particular way.

Firstly, the co-artemether must be taken with fatty food (at least 23 grams of fat per dose), which could be difficult for anyone (usually a child) suffering from fever and nausea. In addition, the dosing interval could be confusing.

The second dose must be taken precisely eight hours after the first dose to maximise the blood levels of lumefantrine (which have been shown to be vital for a sustained response). The third dose must be taken 24 hours after the second dose and the three subsequent doses must be taken every twelve hours.

These dosing requirements provoked questions as to the practicality of co-artemether without clinical supervision in many African settings. Therefore, to make things simpler, the World Health Organisation developed a blister pack for the combination, with dose adjusted by weight. The blister packs have images on them in order to illustrate to the parents how and when the tablets should be given.

Even though co-artemether was still relatively untested in 'real world' African settings, the general consensus is that using it (or other ACTs) would be preferable to using chloroquine or SP where they were failing. There were calls from many quarters, including WHO, for countries to switch to ACTs.

Still, few African countries could afford the new drugs (at present, co-artemether is one of the most expensive treatment options available costing 10 to 20 times more than the old drugs — see discussion below).

If countries could not afford ACTs, WHO recommended using a combination of amodiaquine plus SP as a much cheaper, though perhaps a short-term option. Amodiaquine is a potent malarial still used as a monotherapy in some settings. Still, it is quite similar to chloroquine, and although it is active against some chloroquine-resistant strains, there is some cross-resistance.

The Tanzanian study

To determine which of these options worked best in a ‘real world’ African setting, researchers conducted a randomised controlled study in children with malaria from the Muheza district, in northeastern Tanzania. Muheza has the highest recorded levels of resistance to sulfadoxine-pyrimethamine and chloroquine.

A total of 1811 children with symptoms and serological evidence of non-severe malaria were randomised either to receive amodiaquine monotherapy, amodiaquine plus sulfadoxine-pyrimethamine, or to one of two ACT combinations: co-artemether (six dose regimen) or amodiaquine/artesunate (4 mg/kg artesunate given for three days).

After the start of treatment, parents or guardians were instructed to bring participating children back to clinic if the child’s health worsened, and on days 14 and 28 for routine follow-up.

Monotherapy with amodiaquine was a dramatic failure (with continued lab evidence of the parasite in 103 of 248 (42%) by day 14. Enrollment to the arm was stopped early by the data and safety monitoring board.

Co-artemether was significantly more effective than the other arms. By day 28, there were signs of parasites in the blood for 103 of 485 (21%) of those on co-artemether, compared to 182 of 239 (76%) for amodiaquine, 282 of 476 (61%) for amodiaquine plus SP, and 193 of 472 (40%) for amodiaquine/artesunate. However, in most cases, these represented re-infections from new exposures. Using a PCR based technique to differentiate between a rebound of the old treated infection versus a new infection, co-artemether was found to have cleared the original infection in all but 2.8% of patients, compared to 48·4%, 34·5%, and 11·2% on the other arms, respectively.

The Ugandan study

The second study published in the same issue of The Lancet reported almost identical activity for co-artemether in over 900 patients, whether the treatment was administered under supervision or given to families to self-administer at home.

There were indications, however, that adherence at home was not perfect Lumefantrine blood concentrations were about a third lower in patients receiving unsupervised treatment. Nevertheless, parasite clearance rates were 97.7% (296 of 303) and 98.0% (603 of 615) at day 28 in the supervised and unsupervised groups, respectively.

Data analysis issues

There could be a few reasons why co-artemether was superior to artesunate/amodiaquine in the Tanzanian study. One could be the WHO packaging, which is likely to increase adherence. Another reason could be lower susceptibility to amodiaquine because of high levels of chloroquine-resistance, and the fact that the amodiaquine has been used (sub-optimally) in Tanzania for several years now.

However, Dr. Patrice Piola, lead author of Ugandan study notes that other studies have shown the 28 days not being the best time to measure whether co-artemether has cleared the original infection. Lumefantrine has a much longer half-life in the blood than amodiaquine, and experts have reported that clearance of the malaria parasite cannot really be assessed for co-artemether until six weeks after the treatment starts (see here).

Nonetheless, lumefantrine’s long half-life also seemed to prevent a substantial number of new infections — which in itself is a significant finding. However, it could also increase the risk of resistance to treatment developing more rapidly (see below).

Cost and supply issues

Despite WHO’s recommendations and the recent findings, cost and supply issues continue to keep co-artemether beyond the reach of many in Africa.

Even though discovered and improved in a developing nation, China, the licence to market co-artemether to the rest of the world is held by the Swiss pharmaceutical giant Novartis. And although WHO performs audits to verify that the pricing is fair, it is questionable whether leaving the production of this life-saving drug in the hands of just one company is the best approach to making it more widely available. Some advocacy groups have called for international donors to subsidise the cost of the drug.

At present, Artemisia is only commercially grown in Vietnam and China and the demand for the herbal ingredient is high. As a result, Novartis has been unable to meet demand for the drug, leading to rationing.

Resistance?

The high cost and rarity of co-artemether does not merely limit access to treatment now, but they could limit also its usefulness in the future.

The supply problems could lead to misuse in African settings. Once knowledge spreads about the treatment’s value and effectiveness, families may hold some of the treatment in reserve in case other family members get malaria. Failure to complete the prescribed course could lead to the development of resistance and the loss of this treatment option — particularly in the light of lumefantrine’s long half-life.

After the initial dosing in combination with artemether, new infections will essentially occur in the presence of ever decreasing blood concentrations of lumefantrine — the very recipe for the development of resistance. Indeed, a study in Zanzibar (Sisowath), which recently began using co-artemether, has already shown that the malarial parasite is mutating in response to the treatment.

How best to use co-artemether?

Until issues regarding supply and cost are resolved, a rush to switch to co-artemether or other ACTs as frontline therapy may be wrong-headed. It may be best to hold co-artemether and the other ACTs in reserve for severe life-threatening cases of malaria. Although combination treatment with regimens such as amodiaquine/SP may not be as effective in totally clearing the parasite, they may induce a clinical response and should be used as a temporary frontline defence.

However, the Lancet studies should increase pressure on multilateral organizations to start treating malaria as a crisis and marshal their resources to make sure that ACTs become much more widely available. For each year they wait, another 2 million lives are at stake.