With more than 300 million cases every year, malaria remains one of the major public health problems in many developing countries. Of the four species infecting humans, Plasmodium falciparum is responsible for potentially fatal infections. Until a vaccine is found, malaria control will have to deal with drug-resistant parasites and insecticide-resistant vectors. Children suffer most from malaria. One to two million deaths are estimated every year, especially in Africa where 90% of cases are reported. This means a death due to malaria occurs every 30 seconds. In absolute terms, malaria kills 3,000 children under five years old every day. P. vivax is re-emerging in several countries, especially in Europe, with incipient evidence of reduced chloroquine sensitivity in South East Asia and Western Pacific regions.

Malaria resistance is defined as the "ability of a parasite strain to survive and/or to multiply despite the administration and absorption of a drug given in doses equal or higher than those usually recommended but within the limits of tolerance of the subject." This definition has been modified to specify that the drug "must gain access to the parasite or the infected red blood cell for the duration of time necessary for its normal action"(1).

The demonstration of resistance has been based on the results of in vivo tests or the description of prophylactic failure in travellers. In 1996, WHO replaced the in vivo test with the therapeutic efficacy test which, unlike the 1973 in vivo test, is based on clinical and parasitological criteria(2). The old classifica-tion scheme of S, RI, RII and RIII has been replaced with the 3-group classification: adequate clinical response, early and late therapeutic failure. In vitro testing can still be conducted in parallel with the therapeutic efficacy test, but only the latter is used in advocating any change to national drug policy. The measurement of blood levels of anti-malarials is a very useful way of confirming effective concentration in the blood and ruling out therapeutic failure due to poor drug absorption. Molecular biology can be used to distinguish true resistance from reinfection. It can also be used to find the mutations responsible for resistance to antimalarials. Measurement of blood levels of antimalarials and molecular biology are mainly research tools, and they cannot be used systematically in national malaria control program.

Resistance of P. falciparum to chloroquine appeared almost simultaneously in Colombia in 1960 and on the frontier between Thailand and Cambodia. In Asia, chloroquine-resistance was confined to Indochina until the 1970s, when it extended to the west and towards the neighboring islands to the south and east. The first case of chloroquine resistance in India was described in 1973(3). Today, only few countries in Central America north of the Panama Canal, including Haiti and Dominican Republic, do not report chloroquine-resistant falciparum malaria. Amodiaquine remains useful in areas where there is a moderate resistance to chloroquine, inspite of the results of some studies that suggest its low efficacy, perhaps because insufficient dosages were involved. Studies in Central Africa have shown that amodiaquine was most effective when doses were increased from 25 mg per kg to 35 mg per kg(4).

The sulfadoxine-pyrimethamine combi-nation was used to replace chloroquine. At the beginning of the 1980s, that combination became almost totally ineffective in Thailand and neighboring countries. In India, the first cases of resistance to pyrimethamine in combination with sulfalene were reported in New Delhi. Other foci were also described in Bombay, in the north-east and in the south of the country(5,6). Resistance to the drug combination spread rapidly in Central America. In East Africa, faced with the spreading resistance to chloroquine, Malawi was the first country to change its policy by recommending the sulfadoxine-pyrimethamine combination to be used as the first line drug. Other African countries followed that example, but because of massive utilization, resistance seems to be spreading in East Africa(7).

Resistance to quinine and mefloquine is found mostly in Thailand and Cambodia. Sporadic cases of prophylactic failure of mefloquine in travellers and therapeutic failure with amino-alcohols have been reported in Africa, South America, and in other Asian countries(8). Several studies have noted a diminution in in vitro sensitivity to quinine throughout the world, and in West Africa, in vitro studies have shown strains presenting decreased sensitivity to mefloquine even before its therapeutic use. Resistance to quinine is often overestimated since the dosage of 24 mg base per kg for seven days is rarely respected, and the threshold for in vitro resistance has not been clearly defined.

It is well known that primaquine efficacy against the hepatic forms of P. vivax varies with the origin of the strain, dosage and duration of treatment(89). The description of chloroquine-resistant P. vivax is more recent. In 1989, the first cases appeared in Papua New Guinea. Other cases of resistance or decreased sensitivity were reported from Irian Jaya and other Indonesian islands, Myanmar, the Solomon Islands, India and, more recently, Brazil and Guyana(10). The main problem in the evaluation of the sensitivity of P. vivax is the distinction between reappearance and relapse caused by the hypnozoites. As with P. falciparum, the measurement of the blood chloroquine level can give an individual confirmation that an effective concentration of the drug has been achieved. Chloroquine-resistant P. vivax infection could become a serious therapeutic problem since the sulfadoxine-pyrimethamine combination is not fully effective against this species.

The appearance of resistance to anti-malarials has increased the global cost of the disease. Therapeutic failure means consul-ting a health facility for further diagnosis and treatment resulting in a loss of working days for adults and absence from school for children. Studies in East Africa suggest that ineffective treatment causes anemia, which renders children’s health more fragile(11). In Central Africa, the appearance of chloroquine resistance led to an increase in hospital admissions because of the severe attacks of malaria(12). Similar results in increasing mortality trends were found at the community level in Senegal(13). The impact of drug resistance can also be illustrated by the modification of species composition. In India. P. falciparum accounts for about 40% after the advent of drug resistance instead of 15% usually reported(6).

The genetic mechanisms of drug resistance in P. falciparum are not yet fully understood. Several genes that seems to play a role in regulating resistance have been identified for the three main classes of antimalarial in common use: 4-aminoquinolines and amino-alcohols which act on digestion of hemo-globin, and antifolates which inhibit the enzymes dihydrofolate reductase (DHFR) and dihydrop-teroate synthase (DHPS).

The crucial importance of the mutation of codon 108 (Ser®Asn) in the dhfr gene in the pyrimethamine-resistant P. falciparum has been clearly demonstrated. Further mutations at codons 51, 59 and/or 164 increase the level of resistance to DHFR inhibitors. Not only is there an almost perfect correlation between the presence of the mutant codon in 108 and in vitro resistance to pyrimethamine, but the level of in vitro resistance increases with the number of mutations. Resistance to cycloguanil has been linked to Ser®Thr mutation in codon 108 associated with Ala®Val mutation in codon 16. Ser®Asn mutation alone at position 108 is not a good marker for cycloguanil resistance. One or two additional mutations at positions 51 and 59 seem to be associated with resistance to cycloguanil(14).

Sulfadoxine resistance is also related to mutations at specific codons. Five mutations in codons 436, 437, 540, 581 and 613 of the dhps gene have been reported. Asn®Gly mutations in codon 437 is the key mutation associated with sulfadoxine-resistance. For treatment, sulfadoxine is always prescribed with pyrimethamine. The synergistic anti-malarial action of this drug combination is based on the specific inhibition of two successive enzymes in the folate metabolic pathway. Several mutations are, therefore, associated with therapeutic failure of the sulfadoxine-pyrimethamine combination. Paradoxically, the presence of three mutations on the dhfr gene alone seems to be sufficient to lead to resistance to the combination(15).

Three genes have been identified for chloroquine resistance. The pfmdr1 gene has been proposed as a candidate gene of chloroquine resistance, with a mutation at codon 86 and/or due to gene amplification. Studies involving isolates from patients seem to show that there is no relationship between pfmdr1 mutations and in vitro resistance to chloroquine and the clinical response of patients treated with chloroquine. Genetic analysis of cg2 gene from reference clones has shown a genotype related to chloroquine resistance in vitro, characterized by mutations on 12 codons, the number of repetitive units in three repetitive sequence, k; g, and w, and changes in the length of a central poly-Asn tract(16). Recently, a novel gene, pftcr has been identified as being potentially involved in chloroquine resistance (cited in 17). Predictive value of genetic markers for chloroquine resistance needs more studies to evaluate their in vivo prognostic value.

The problem of resistance is exacerbated by the shortage of new antimalarials under development. Many pharmaceutical companies companies have ceased research and development of drugs used in tropical medicine, in particular for malaria treatment.

There are reasons to recommend the use of chlorproguanildapsone combination instead of sulfadoxine-pyrimethamine, because the for-mer has greater activity and shorter half-life(18). Nevertheless, the mode of action of chlorproguanil and dapsone is similar to that of pyrimethamine and sulfadoxine, respect-ively. The existence of a quadruple mutation on the dhfr gene has already rendered that combination ineffective in South-East Asia. Two 8-aminoquinolines are scheduled to be marketed soon: tafenoquine, developed by the Walter Reed Army Institute of Research in the United States, and 80/53, produced by the Central Drug Research Institute of Lucknow, India. The drugs are indicated for either prevention of P. falciparum infection or radical cure of P.vivax(19). Glaxo Wellcome laboratory has initiated a donation program of a combination of atovaquone and progunail. These drugs act in synergy in vitro, and the combination is effective for both treatment and prevention of P. falciparum(20).

By analogy with the treatment of tuberculosis and human immunodeficiency virus infection, the novel antimalarial treat-ment strategy is the use of drug combinations. Combinations are usually more effective than monotherapy, and they also help to avoid or at least delay the emergence of resistance. Artemisinin derivatives seem to be particularly useful in these combinations(21). In Thailand, the efficacy of mefloquine has decreased considerably after several years of intensive use, but the combination of mefloquine and artesunate is more than 95% effective in multi-resistant areas. Studies evaluating the efficacy and tolerance of artesunate in combination with chloroquine, amodiaquine and sulfadoxine-pyrimethamine are under way in Africa and South America. Only one fixed combination that includes an artemisinin derivative for oral use is available: artemether-lumefantrine (an amino-alcohol related to halofantrine). For the moment, these combinations with artemisinin derivatives are costly.

Artemisinin derivatives should not be used in monotherapy to treat uncomplicated malaria infections, or ever they are prescribed, they should be administred for at least seven days. Artemether and arteether are available in injectable form for the treatment of severe malaria.

Given the spread of drug resistance, the solution is not a systematic use of the latest, costly antimalarials. Each national malaria control program should make a precise therapeutic efficacy test evaluation of the drug resistance situation with standardized protocols by using sentinel sites selected on the basis of different transmission areas and including borders with neighboring countries. Training and quality control of microscopic examina-tions of smears and of the drugs used will have to be provided to ensure good quality data. Decisions to change policy must be based on the results obtained in the light of the antimalarials available and their cost. Drugs for the replacement of chloroquine and pyrimethamine-sulfadoxine will be more expensive in the immediate future. The establishment of a new policy must entail a more rational use of drugs (for example, limited courses of treatment for confirmed malaria infections) and improved compli-ance(22). Nevertheless, policies will always be confronted with the dilemma of improving access to treatment for patients but in absence of correct diagnosis this would lead to an inappropriate use of the drugs which is one of the main causes of resistance.

I thank Rosamund Williams Leonardo K. Basco for critical comments on the text.

Pascal Ringwald,
World Health Organization,
Anti-infective Drug Resistance Surveillance and Containment (DRS),
1211 Geneva 27, Swtizerland.
E-mail: [email protected]