INTRODUCTION
Antimalarial drugs are used for the treatment and prevention of malaria infection. Most antimalarial drugs target the erythrocytic stage of malaria infection, which is the phase of infection that causes symptomatic illness. The extent of preerythrocytic (hepatic stage) activity for most antimalarial drugs is not well characterized.
Treatment of the acute blood stage infection is necessary for malaria caused by all malaria species. In addition, for infection due to Plasmodium ovale or Plasmodium vivax, terminal prophylaxis is required with a drug active against hypnozoites (which can remain dormant in the liver for months and, occasionally, years after the initial infection).
The mechanisms of action, resistance, and toxicities of antimalarial drugs will be reviewed here. Use of these agents for prevention and treatment of malaria is discussed in detail separately.
QUINOLINE DERIVATIVES
Quinoline derivatives include chloroquine, amodiaquine, quinine, quinidine, mefloquine, primaquine, lumefantrine, and halofantrine. These drugs have activity against the erythrocytic stage of infection; primaquine also kills intrahepatic forms and gametocytes. The drugs act by accumulating in the parasite food vacuole and forming a complex with heme that prevents crystallization in the Plasmodium food vacuole. Heme polymerase activity is inhibited, resulting in accumulation of cytotoxic-free heme.
4-aminoquinolines
Chloroquine Chloroquine was the first drug produced on a large scale for treatment and prevention of malaria infection. Chloroquine has activity against the blood stages of P. ovale, P. malariae, and susceptible strains of P. vivax and P. falciparum. Widespread resistance in most malaria-endemic countries has led to decline in its use for the treatment of P. falciparum, although it remains effective for treatment of P. ovale, P. malariae, and, in most regions, P. vivax.
Medical Definition of Antimalarial
Antimalarial: A drug directed against malaria. The original antimalarial agent was quinine which took its name from the Peruvian Indian word “kina” meaning “bark of the tree.” A large and complex molecule, quinine is the most important alkaloid found in cinchona bark. Until World War I, it was the only effective treatment for malaria. In fact, quinine was the first chemical compound to be successfully used to treat an infectious disease.
Quinine was isolated in crystalline form in 1820 by J.B. Caventou and P.J. Pelletier. In one of the classical achievements of synthetic organic chemistry, R.B. Woodward and W. Doering first made synthetic quinine in 1944.
Quinine acts by interfering with the growth and reproduction of the Plasmodium, the malarial parasite that lives within the victim’s red blood cells. Quinine causes the parasites to disappear from the blood and the symptoms of the disease are thereby alleviated. However, when quinine treatment ends, many patients relapse. They suffer another attack of malaria due to the failure of quinine to kill the malarial parasites in cells of the body other than the red blood cells. These parasites persist and, after a time, they reinvade the red blood cells and precipitate the relapse.
Since quinine does not permanently cure malaria, better drugs were sought. A number were discovered that replaced quinine during and after World War II. Some of these drugs (such as chloroquine and chloroguanide) are more effective than quinine in suppressing the growth of the blood forms of the malarial parasite. Others (such as primaquine and pyrimethamine) act upon both the blood and tissue phases of the parasite, producing a complete cure and preventing a relapse. Quinine was once used for the treatment of leg cramps, but it is no longer FDA approved for this use, due to serious side effects, including low platelet counts and death.
Clinical Pharmacodynamics
Patients with acute malaria may have up to 1012 parasites circulating in their blood. Even with killing rates of 99.99 percent per cycle, complete eradication of the parasite load requires at least three life cycles (6 days); therefore, therapeutic drug concentrations should be present for 4 cycles to effect a cure (White, 1997, 1998). Simply put, patients taking rapidly eliminated drugs must continue treatment for a full week. Treatment responses are always better in patients with some immunity (York and Macfie, 1924) because the immune response kills parasites in much the same way that a drug does. In endemic areas, the worst treatment results are seen in young children who have little immunity. In contrast, although their degree of immune protection cannot be quantitated or assumed, older children or adults in high-transmission areas may do surprisingly well with failing drugs because much of their therapeutic response stems from immunity rather than antimalarial drug action.
In severe falciparum malaria, the stage at which an antimalarial drug acts is especially important since the ultimate goal of treatment is to halt parasite maturation to late-stage, cytoadherent parasites (i.e., mature schizonts that attach to endothelial cells lining small blood vessels), which are primarily responsible for life-threatening complications. The artemisinin derivatives are advantageous because they prevent parasites from maturing to these more pathological stages, whereas quinine and quinidine do not affect parasites until they have already cytoadhered. The antifols act even later in the cycle, and are not recommended for severe malaria (Yayon et al., 1983; ter Kuile et al., 1993). None of the drugs will prevent rupture of infected erythrocytes and reinvasion once a schizont has formed. Young ring forms (i.e., early asexual parasites) also are relatively drug resistant, especially to quinine and pyrimethamine.
Artemesinin derivatives offer the broadest antimalarial action against the range of developmental stages, and the most rapid in vivo activity (ter Kuile et al., 1993; White, 1997). These compounds (and, to a lesser extent, chloroquine) prevent ring stages from maturing, hastening their clearance, and preventing end-organ pathology that would otherwise occur if cytoadherence progressed unchecked (Chotivanich et al., 2000).
MECHANISMS OF ACTION AND DRUG RESISTANCE
Antifolate Drugs
Pyrimethamine, and biguanides such as cycloguanil interfere with folic acid synthesis, inhibiting the parasite enzyme known as dihydrofolate reductase-thymidilate synthase (DHFR). Sulfonamides act at the previous step in the folic acid pathway, inhibiting the parasite enzyme dihydropteroate synthase (DHPS). There is marked synergy between these two classes of drugs when they are taken together. However, resistance to pyrimethamine in P. falciparum developed within a few years of its introduction (Peters, 1987) due to point mutations in the DHFR gene, which cause 100- to 1,000-fold reduced affinity of the enzyme complex to the drug. Progressive mutations in the DHFR gene of P. falciparum further decreased efficacy. Triple mutant infections are relatively resistant to antifolate treatment; with a fourth mutation within the malaria parasite, antifolate drugs become completely ineffective.
Quadruple mutant P. falciparum strains are now prevalent in parts of Southeast Asia, and South America (Imwong et al., 2001). Resistance to partner antifols sulfonamide and sulfone results from progressive acquisition of mutations in the P. falciparum gene encoding the target enzyme DHPS.
Chloroquine
One of chloroquine’s most dramatic characteristics is its ability to concentrate itself from nanomolar (10-9) levels outside the parasite to levels one million times higher (millimolar levels, 10-3) in the acid food vacuole of the parasite inside a red blood cell (Krogstad and Schlesinger, 1987). This action in itself does not explain chloroquine’s antimalarial activity, however. Chloroquine works by interfering with heme dimerization, the detoxifying biochemical process within the malaria parasite that normally yields malaria pigment (hemozoin).
Reduced intracellular drug concentrations accompany chloroquine resistance because resistant parasites expel chloroquine from their acid food vacuoles 40-50 times faster than do drug-sensitive parasites (Bray et al., 1998). Such accumulation deficits were once attributed to changes in pH gradient, or to altered membrane permeability, or both (Le Bras and Durand, 2003). However, chloroquine resistance was then found to be reversible by verapamil, a drug which also modulates resistance in multidrug resistant (MDR) mammalian cancer cells. This discovery led to the identification of the protein Pgh1 (an analog to overexpressed glycoproteins that expel cytotoxic drugs in cancer cells) in the digestive vacuole membrane of P. falciparum. Genes encoding MDR proteins have been identified in P. falciparum (pfmdr1); amplification of these “wild type” MDR genes has recently been shown to cause mefloquine resistance (Price et al., 1999b).
Point mutations in the gene encoding a food vacuole transporter protein (pfcrt) have been linked to chloroquine resistance (Durand et al., 2001; Warhurst, 2001) and correlate with reduced in vivo chloroquine efficacy (Djimde et al., 2001). In the presence of pfcrt mutations, mutations in the second transport genes (Pfmdr1) further modulate resistance in vitro although the role of Pfmdr1 mutations in determining in vivo responses to chloroquine treatment is still unclear. Additional unlinked mutations are probably involved in the development of chloroquine resistance, some of which have not yet been discovered.
In the laboratory, the efflux mechanism seen in chloroquine-resistant P. falciparum parasites can be inhibited by several unrelated drugs (calcium channel blockers such as verapamil as well as tricyclic antidepressants, phenothiazines, and antihistamines), whereas mefloquine resistance can be reversed by penfluridol (Martin et al., 1987; Oduola et al., 1993). Clinical applications of these findings are few to date, although chloroquine plus high doses of chlorpheniramine (an antihistamine) did show improved efficacy against chloroquine-resistant P. falciparum in Nigerian children (Sowunmi et al., 1997). Whether general use of resistance reversers will be safe and feasible in the future remains an open question (Personal communication, N. White, Mahidol University, March 2004).
Other Antimalarial Drugs
In general, antimalarial drug resistance to mefloquine, quinine, lumefantrine, and halofantrine is linked, whereas chloroquine, and mefloquine resistance are not. Cross-resistance between antimalarials is related to common aspects of their modes of action as well as their resistance mechanisms. Parasites with high-level chloroquine resistance (present in Southeast Asia), are generally resistant to amodiaquine as well; in residents of Southeast Asia, amodiaquine may thus fail as a back-up treatment (Le Bras and Durand, 2003). The same relationship holds true for halofantrine, and mefloquine. On the other hand, there may be an inverse correlation between chloroquine and mefloquine sensitivity: in Africa, for example, chloroquine-sensitive strains are substantially less sensitive to mefloquine or halofantrine, and vice versa (Oduola et al., 1987; Simon et al., 1988).
Atovaquone is a component of Malarone®, a new combination drug (consisting of atovaquone and proguanil) used for treatment and prevention of chloroquine-resistant P. falciparum. Atovaquone interferes with mitochondrial electron transport, and also blocks cellular respiration (Srivastava et al., 1997). High levels of atovaquone resistance result from single-point mutations in a gene encoding cytochrome b found on a small, extrachromosomal DNA-containing element in the parasite (Korsinczky et al., 2000).
ARTEMISININS
Of the available antimalarials, the artemisinins are effective at killing the broadest range of asexual stages of the parasite, ranging from medium-sized rings to early schizonts; they also produce the most rapid therapeutic responses by accelerating clearance of circulating ring-stage parasites (ter Kuile et al., 1993).
Qinghaosu, or artemisinin, is a sesquiterpene lactone peroxide extracted from the leaves of the shrub Artemisia annua (qinghao). Three derivatives are widely used: the oil-soluble methyl ether, artemether (artemotil [arteether] is a closely related compound); the water soluble hemi-succinate derivative, artesunate; and dihydroartemisinin (DHA). Artesunate, artemether, and arteether are all synthesized from DHA, and they are converted back to it within the body. Artemisinin itself is available in a few countries in Asia. It is 5-10 times less active than the derivatives, and it is not metabolized to DHA.
Artemisinin is available as capsules of powder, or as suppositories. Artemether is formulated in peanut oil, and arteether in sesame seed oil, for intramuscular injection, and in capsules or tablets for oral use. Artesunate is formulated either as tablets, in a gel enclosed in gelatin for rectal administration (called a rectocap™), or as dry powder of artesunic acid for injection, supplied with an ampoule of 5 percent sodium bicarbonate. The powder is dissolved in the sodium bicarbonate to form sodium artesunate, and then diluted in 5 percent dextrose or normal saline for intravenous or intramuscular injection. Artelinic acid is a water-soluble second-generation compound under long-term development. It has not yet been used in treatment. The majority of clinical data pertain to the most widely used derivative, artesunate.
Botanical Properties
Artemisinin was first isolated from the stems, leaves, and flowers of Artemisia annua by Chinese scientists (Anonymous, 1982; Klayman et al., 1984), but details of the process were not released. Researchers at the Walter Reed Army Institute of Medical Research (WRAIR) successfully isolated artemisinin derivatives from air-dried parts of plants growing in the wild near Washington, D.C., using petroleum ether extraction (Klayman, 1985). The plant grows easily in temperate areas, and has become naturalized in many countries. It can attain a height of two meters or more, appearing as an erect specimen with a woody stem. Artemisinin accumulates in all parts of A. annua except for the roots (Abdin et al., 2003). Artemisinin content in flowers is 4-5 times higher than in leaves. Plant age correlates with artemisinin yield, presumably due to a progressive increase in leaf yield and artemisinin content with plant growth. In agricultural settings in Asia, artesunate production has varied from 5 kg/hectare to 50 kg/hectare (Personal communication, J-M. Kindermans, Médecins Sans Frontières, February 2004).
MECHANISM OF ACTION
Artemisinin’s chemical structure is unlike any other known antimalarial. It includes an endoperoxide bridge necessary for its antimalarial action (Brossi et al., 1988). Artemisinin treatment of membranes, especially in the presence of heme, causes lipid peroxidation (Scott et al., 1989; Wei and Sadrzadeh, 1994; Berman and Adams, 1997); this event may occur as a result of the drug’s interaction with intracellular heme or iron (Meshnick et al., 1991). With respect to artemisinin’s direct effect on the malaria parasite, recent work suggests that artemisinin specifically inhibits PfATP6, the SERCA orthologue of Plasmodium falciparum, a calcium ATPase (Eckstein-Ludwig et al., 2003).
In vivo, artemisinins kill malaria parasites within host erythrocytes, after which dead parasites are culled by the spleen, leaving formerly infected red blood cells intact and circulating (Chotivanich et al., 2000). It is not yet clear which asexual parasite life-cycle stages are most sensitive to artemisinin derivatives: late rings and early trophozoites (ter Kuile et al., 1993) versus trophozoites (Geary et al., 1989). There is clear consensus, however, that artemisinin derivatives kill early-stage gametocytes and are more active over a broader range of the parasite life cycle than any other antimalarial drug currently in use.
General Clinical Experience with Artemisinins and ACTs
More randomized clinical trials have been published regarding the effects of artemisinins than any other individual or class of antimalarial drug (Myint et al., 2004). The artemisinins’ pharmacodynamic effects are due to their rapid absorption and activity against many stages of the malaria life cycle, from young asexual forms (rings) to early sexual forms (gametocytes) (Kumar and Zheng, 1990). Their half-lives are short (<1 hour for artesunate) which protects them from resistance. They reduce gametocyte carriage, thus decreasing infectiousness following treatment as witnessed by the dramatic fall in malaria transmission on the Thai-Burma border (Price et al., 1996; Nosten et al., 2000). Tolerability of the drugs is excellent (White and Olliaro, 1998).
Administered alone, artemisinin derivatives require at least 7 days of treatment. When combined with other drugs, however, artemisinin combination therapies (ACTs) given in 3-day regimens can eradicate parasites quickly and protect against the development of resistance to both drugs. In addition, artemisinin derivatives are the only first-line malaria treatments to act on gametocytes (early-stage).
On February 20, 2002, the World Health Organization (WHO) released a statement recommending that “Governments … rapidly adopt more effective treatments [for malaria]. The aim is to provide effective treatment against malaria and to slow the spread of drug resistance. … In particular, WHO recommends the use of artemisinin-based combination therapy” (WHO, 2001).
Current Evidence Base for ACTs
Through the largest randomized controlled trials ever conducted on antimalarials in Africa, a considerable body of evidence has now been collected showing that artemisinin-based combinations improve cure rates, decrease gametocyte carriage, and are well tolerated with few serious adverse effects. In the first of these multicenter trials to be published, 941 children who had uncomplicated P. falciparum malaria were randomly assigned 3 days treatment with amodiaquine plus artesunate, or amodiaquine plus placebo. Both regimens were well tolerated. The combination of artesunate and amodiaquine significantly improved treatment efficacy in Gabon (85 versus 71 percent, p=0.02) and in Kenya (68 vs. 41 percent, p<0.0001). In Senegal, however, the two regimens were equivalent: day-28 cure rates for amodiaquine-artesunate versus amodiaquine were 82 versus 79 percent (p=0.5) (Adjuik et al., 2002).
The efficacy and safety of artesunate in combination with sulfadoxine-pyrimethamine (SP) has been evaluated in randomized controlled trials involving 2,865 patients in sub-Saharan Africa. Results from the first study published from The Gambia (von Seidlein et al., 2000) (where the cure rate with SP monotherapy was then 93 percent), showed that both cure rate and parasite clearance were significantly higher in patients who received 3 days of artesunate plus a single dose of SP compared with those who received SP alone. Gametocyte carriage was 68 percent following solo SP treatment in comparison with 21 percent following the artesunate-SP combination (p=0.001).
In contrast, underlying SP resistance in Uganda led to unacceptable rates of late recrudescence when artesunate-SP was used there (Dorsey et al., 2002). This finding underscores that combining an artemisinin with a longer-acting drug without any underlying resistance may prove the optimal regimen in many areas. On the other hand, in Thailand, where drug resistance is particularly severe, artesunate plus mefloquine was highly effective, even in areas where mefloquine resistance was previously quite common (Price et al., 1997). In the largest-ever series of therapeutic efficacy studies with ACTs, artesunate plus mefloquine produced a sustained, increased cure rate (almost 100 percent from 1998 onward) despite established resistance to high-dose mefloquine alone seen between 1990 and 1994 (Nosten et al., 2000). Cure rates with other ACTs (atovaquone-proguanil-artesunate, artemether-lumefantrine) in Asia have also been consistently above 90 percent (van Vugt et al., 1998, 1999).
Coartem (artemether-lumefantrine) is the first fixed-dose ACT whose two antimalarial components were not widely used prior to marketing. When given in a six-dose regimen over 3 days, Coartem is effective against all P. falciparum and extremely well tolerated. Coartem’s principal drawback is its twice daily dosing and lumefantrine’s variable absorption linked to dietary fat intake (if patients are unable to eat, or consume a very low fat diet while taking the drug, low cure rates can result).
In randomized trials comparing a six-dose regimen of orally administered artemether-lumefantrine with mefloquine-artesunate for multidrug-resistant P. falciparum malaria, artemether-lumefantrine was as effective (94-100 percent) and overall better tolerated than mefloquine-artesunate (van Vugt et al., 1999, 2000; Lefevre et al., 2001).
Artesunate
Artesunate can be administered as a tablet, an injection, or as a rectal suppository. A comparison of intravenous artesunate and quinine in 113 adults with severe malaria reported mortality of 12 percent with artesunate, and 22 percent with quinine (p=0.22) (Newton et al., 2003). In patients with hyperparasitemia who had no other features of severe malaria but were at an increased risk of developing severe malaria, oral artesunate was found to be superior to intravenous quinine in reducing both the clinical symptoms and parasites (Luxemburger et al., 1995).
Rectally administered artesunate has been shown to be safe and highly effective in children and adults with uncomplicated or moderately severe falciparum malaria (Sabchareon et al., 1998; Karunajeewa et al., 2003; Barnes et al., 2004). A number of recent randomized controlled and open-label studies of rectal artesunate in Africa and Asia have demonstrated the rapid antimalarial efficacy of a single dose of rectal artesunate (10 mg/kg) in moderately severe falciparum malaria in both children and adults prior to referral for definitive treatment. All patients had evidence of adequate absorption of the drug. Clearance of malaria parasites from the peripheral blood was consistently more rapid with rectal artesunate than with quinine injection. There also are a number of small open label studies, some of which were randomized, demonstrating the clinical and parasitological efficacy of rectal artesunate in adults with severe P. falciparum infections (Awad et al., 2003) where rectal artesunate was administered repeatedly, and combined with a second oral antimalarial to prevent recrudescence.
These clinical and parasitological responses suggest that rectal artesunate could prove highly beneficial in the initial management of acute malaria in patients who cannot take medication by mouth (and for whom parenteral medication is not immediately available due to limited local resources). In Ghana, artesunate suppositories have already been given by trained village volunteers; in the future, traditional healers also could be trained to use them as an emergency treatment for febrile seizures in malaria-enemic areas (Personal communication, B. Greenwood, London School of Hygiene and Tropical Medicine, April 2004). If used widely enough, rectal artesunate could have a major effect on malaria deaths in Africa.
TOXICITY OF ARTEMISININS
In the early 1970s, Chinese scientists characterized the antimalarial properties of artemisinins and their excellent tolerability and safety (Anonymous, 1982; Luo and Shen, 1987). A subsequent review of published and unpublished studies of the arteminisinin derivatives confirmed the earlier Chinese findings (Pukrittayakamee et al., 2000). hose adverse events that were reported were not specifically associated with a single form of artemisinin or route of administration. Neutropenia (but not agranulocytosis) and asymptomatic ECG abnormalities occurred in 1.3 percent of patients; reduced reticulocyte count, anemia, eosinophilia, and elevated aspartate aminotransferase (a liver enzyme) occurred in ≤ 1.0 percent (Taylor and White, 2004).
In western Thailand, Price and colleagues (1999a) conducted a detailed study of oral artesunate or artemether used as monotherapy or in combination with mefloquine. The artemisinin derivatives were associated with substantially fewer adverse effects than the mefloquine-containing regimens: acute nausea (16 vs. 31 percent), vomiting (11 vs. 24 percent), anorexia (34 vs. 52 percent), and dizziness (15 vs. 47 percent). Oral artesunate or artemether alone were well tolerated. Since these data were published two cases of acute hives and anaphylaxis developed in the same population following artesunate monotherapy, raising the total number of allergic reactions to 6 patients out of roughly 17,000, or 1 in 2,833 (Leonardi et al., 2001).
Artemether-Lumefantrine (Coartem)
Experience with Coartem is increasing rapidly. In large-scale clinical trials the combination was very well tolerated, whether taken as a four- or six-dose regimen. Coartem has been better tolerated than either mefloquine alone or artesunate plus mefloquine (van Vugt et al., 1998; Looareesuwan et al., 1999). Reported adverse effects have generally been mild, including gastrointestinal upset, headache, dizziness, fatigue, sleep disturbance, palpitations, maylgia, arthralgia, and rash (Bakshi et al., 2000). Detailed prospective studies have not demonstrated any cardiotoxicity (van Vugt et al., 1999; Bindschedler et al., 2000). Twenty severe adverse effects in 1,869 patients include 19 that could be explained on the basis of the malarial episode or a concurrent illness; artemether-lumefantrine may have contributed to the development of hemolytic anemia in one 3-year-old patient 13 days following discontinuation of the drug (Bakshi et al., 2000).
