There are currently no bedside tests for determining the susceptibility of the malaria parasite to antimalarials. Monitoring is therefore needed to determine geographical trends in susceptibility and the emergence and spread of drug resistance. The information obtained will help guide treatment choices and predictions about future resistance patterns.
The greatest problem with drug resistance occur with P. falciparum. Resistance of P. falciparum is of particular concern because of the enormous burden of disease caused by this species, its lethal potential, the propensity for epidemics, and the cost of candidate replacement drugs for areas with established drug resistance. Chloroquine resistance does occur in P. vivax, especially in western Oceania, but there are very few reports of resistance in P. malariae or P. ovale (although there have also been very few studies).
This annex defines resistance, examines how it arises and spreads, summarizes its current global distribution and describes ways in which it can be monitored.
Antimalarial drug resistance is defined as the ability of a parasite strain to survive and/or multiply despite the proper administration and absorption of an antimalarial drug in the dose normally recommended. Drug resistance to an antimalarial compound results in a right shift in the concentration-effect (dose-response) relationship (Figure A6.1). As the pharmacokinetic properties of antimalarials vary widely in different individuals, the definition of resistance should probably also include a "normal" plasma concentration profile for the active drug concerned or, in the case of a prodrug (a drug that is not active in the ingested form and requires chemical conversion through metabolic processes to become pharmacologically active), a "normal" profile of the biologically active metabolite. Antimalarial drug resistance is not necessarily the same as malaria "treatment failure", which is a failure to clear malarial parasitaemia and/or resolve clinical symptoms despite the administration of an antimalarial. So while drug resistance may lead to treatment failure, not all treatment failures are caused by drug resistance. Treatment failure can also be the result of incorrect dosing, problems of treatment adherence (compliance), poor drug quality, interactions with other drugs, compromised drug absorption, or misdiagnosis of the patient. Apart from leading to inappropriate case management, all these factors may also accelerate the spread of true drug resistance by exposure of the parasites to inadequate drug levels.
Figure A6.1 Resistance is a rightward shift in the concentration-effect relationship for a particular parasite population. This may be a parallel shift (B) from the "normal" profile (A) or, in some circumstances, the slope changes, and/or the maximum achievable effect is reduced (C). The effect is parasite killing
A6.3 The emergence and spread of antimalarial resistance
The development of resistance can be considered in two parts: the initial genetic event, which produces the resistant mutant; and the subsequent selection process in which the survival advantage in the presence of the drug leads to preferential transmission of resistant mutants and thus the spread of resistance. In the absence of the antimalarial, resistant mutants may have a survival disadvantage. This "fitness cost" of the resistance mechanism may result in a decline in the prevalence of resistance once drug pressure is removed.
Resistance to one drug may select for resistance to another where the mechanisms of resistance are similar (cross-resistance). There are many parallels with antibiotic resistance, in particular resistance to antituberculosis drugs where, as for malaria, transferable resistance genes are not involved in the emergence of resistance (1-3). In experimental models, drug-resistant mutations can be selected without mosquito passage (i.e. without meiotic recombination) by exposure of large numbers of malaria parasites (either in vitro, in animals, or as was done in the past, in volunteers) to subtherapeutic drug concentrations (4).
Various factors determine the propensity for antimalarial drug resistance to develop (5):
• the intrinsic frequency with which the genetic changes occur,
• the degree of resistance (the shift in the concentration-effect relationship, (Figure A6.1) conferred by the genetic change,
• the fitness cost of the resistance mechanism,
• the proportion of all transmissible infections that are exposed to the drug (the selection pressure),
• the number of parasites exposed to the drug,
• the concentrations of drug to which these parasites are exposed ,
• the pharmacokinetic and pharmacodynamic properties of the antimalarial,
• individual (dosing, duration, adherence) and community (quality, availability, distribution) patterns of drug use,
• the immunity profile of the community and the individual,
• the simultaneous presence of other antimalarials or substances in the blood to which the parasite is not resistant.
The emergence of resistance can be thought of in terms of the product of the probabilities of de novo emergence (a rare event) and subsequent spread. Resistant parasites, if present, will be selected when parasites are exposed to "selective" (subtherapeutic) drug concentrations. "Selective" in this context means a concentration of drug that will eradicate the sensitive parasites but still allow growth of the resistant parasite population such that it eventually transmits to another person. Because de novo resistance arises randomly among malaria parasites, non-immune patients infected with large numbers of parasites who receive inadequate treatment (either because of poor drug quality, poor adherence, vomiting of an oral treatment, etc.) are a potent source of de novo resistance. This emphasizes the importance of correct prescribing, and good adherence to prescribed drug regimens, and also provision of treatment regimens that are still highly effective in hyperparasitaemic patients. The principle specific immune response that controls the primary symptomatic infection in falciparum malaria is directed by the variant surface antigen (PfEMP1). The parasite population evades this immune response by switching its surface antigen in a specific sequence of changes. The probability of selecting a resistant parasite from the primary infection is the product of the switch rate and the rate of formation of viable resistant parasites.
The subsequent spread of resistant mutant malaria parasites is facilitated by the widespread use of drugs with long elimination phases. These provide a "selective filter", allowing infection by the resistant parasites while the residual antimalarial activity prevents infection by sensitive parasites. Slowly eliminated drugs such as mefloquine (terminal elimination half-life (T1/2β 2-3 weeks) or chloroquine (T1/2β 1-2 months) persist in the blood and provide a selective filter for months after drug administration has ceased.
A6.3.1 Transmission intensity and the selection and spread of resistance
The recrudescence and subsequent transmission of an infection that has generated a de novo resistant malaria parasite is essential for resistance to be propagated (5). Gametocytes carrying the resistance genes will not reach transmissible densities until the resistant biomass has expanded to numbers close to those producing illness (>107 parasites) (6). Thus to prevent resistance spreading from an infection that has generated de novo resistance, gametocyte production from the recrudescent resistant infection must be prevented. There has been debate as to whether resistance arises more rapidly in low-or high-transmission settings (7, 8), but aside from theoretical calculations, epidemiological studies clearly implicate low-transmission settings as the source of drug resistance. Chloroquine resistance and high-level sulfadoxine-pyrimethamine resistance in P. falciparum both originated in South-East Asia and subsequently spread to Africa (9).
In low-transmission areas, the majority of malaria infections are symptomatic and selection therefore takes place in the context of treatment. Relatively large numbers of parasites in an individual usually encounter antimalarials at concentrations that are maximally effective. But in a variable proportion of patients, for the reasons mentioned earlier, blood concentrations are low and may select for resistance.
In high-transmission areas, the majority of infections are asymptomatic and infections are acquired repeatedly throughout life. Symptomatic and sometimes fatal malaria occurs in the first years of life, but thereafter it is increasingly likely to be asymptomatic. This reflects a state of imperfect immunity (premunition), where the infection is controlled, usually at levels below those causing symptoms. The rate at which premunition is acquired depends on the intensity of transmission. In the context of intense malaria transmission, people still receive antimalarial treatments throughout their lives (often inappropriately for other febrile infections), but these "treatments" are largely unrelated to the peaks of parasitaemia, thereby reducing the probability of selection for resistance.
Immunity considerably reduces the emergence of resistance (9). Host defence contributes a major antiparasitic effect, and any spontaneously generated drug-resistant mutant malaria parasite must contend not only with the concentrations of antimalarial present, but also with host immunity. This kills parasites regardless of their antimalarial resistance, and reduces the probability of parasite survival (independently of drugs) at all stages of the transmission cycle. For the blood stage infection, immunity acts in a similar way to anti-malarials both to eliminate the rare de novo resistant mutants and stop them being transmitted (i.e. like a combination therapy), and also to improve cure rates with failing drugs (i.e. drugs falling to resistance) thereby reducing the relative transmission advantage of resistant parasites. Even if a resistant mutant does survive the initial drug treatment and multiplies, the chance that this will result in sufficient gametocytes for transmission is reduced as a result of asexual stage immunity (which reduces the multiplication rate and lowers the density at which the infection is controlled) and transmission-blocking immunity. Furthermore, other parasite genotypes are likely to be present, competing with the resistant parasites for red cells, and increasing the possibility of outbreeding of multigenic resistance mechanisms or competition in the feeding anopheline mosquito (10).
A6.3.2 Antimalarial pharmacodynamics and the selection of resistance
The genetic events that confer antimalarial drug resistance (while retaining parasite viability) are spontaneous and rare. They are thought to be independent of the drug. The resistance mechanisms that have been described are mutations in genes or changes in the copy number of genes relating to the drug's target or pumps that affect intraparasitic concentrations of the drug. A single genetic event may be all that is required, or multiple unlinked events may be necessary (epistasis). P. falciparum parasites from South-East Asia seem constitutionally to have an increased propensity to develop drug resistance.
Chloroquine resistance in P. falciparum may be multigenic and is initially conferred by mutations in a gene that encodes a transporter (PfCRT). PfCRT may be an anion channel pumping chloroquine out from the food vacuole. The initial mutation, which confers a moderate level of chloroquine resistance, is replacement of a lysine with threonine at codon 76. Positions 72 to 76 are critical for the binding of desethylamodiaquine (the biologically active metabolite of amodiaquine) and also verapamil (which may reverse chloroquine resistance in vitro). Eleven other PfCRT mutations have been described to date. These additional mutations may contribute to aminoquinoline resistance, although the precise mechanisms have not yet been determined. Amodiaquine resistance is linked to chloroquine resistance, but is not well characterized. In the presence of PfCRT mutations, point mutations in a second transporter (PfMDR1) modulate the level of P. falciparum resistance in vitro. Parasites that are highly resistant to chloroquine often have Lys76Thr and Ala220Ser in PfCRT, and Asn86Tyr in PfMDR. The role of PfMDR1 mutations in determining the therapeutic response following chloroquine treatment is still unclear. The cause of chloroquine resistance in P. vivax has not been found yet.
Resistance to mefloquine and other structurally related aryl-amino-alcohols in P. falciparum results from amplifications (i.e. duplications not mutations) in Pfmdr, which encodes an energy-demanding p-glycoprotein pump. This explains approximately two-thirds of the variance in susceptibility. Interestingly, it appears that generally only the "wild type" (PfMDR Asn86) amplifies, so that in the transition from chloroquine resistance, back mutation from mutant to wild type precedes amplification. Gene duplication is particularly frequent in the P. falciparum genome. It is a much more common genetic event than mutation. The low background frequency of gene amplification suggests that it may well confer a fitness disadvantage in the absence of selective pressure.
The products of these various genetic events result in reduced intracellular concentrations of the antimalarial quinolines in the parasite (the relative importance of reduced uptake and increased efflux remains unresolved).
For the antifolate antimalarials (pyrimethamine, and the biguanides cycloguanil and chlorcycloguanil - the active metabolites of proguanil and chlorproguanil, respectively) resistance in P. falciparum and P. vivax results from the sequential acquisition of mutations in the gene (dhfr) that encodes dihydrofolate reductase (DHFR). Each mutation confers a stepwise reduction in susceptibility. In P. falciparum, the initial mutation is almost invariably at position 108 (usually serine to asparagine), which confers only a ten-fold reduction in drug susceptibility, and does not affect therapeutic responses to sulfadoxine-pyrimethamine. This has little clinical relevance initially, but then mutations arise at positions 51 and 59, conferring increasing resistance to pyrimethamine containing medicines. Infections with triple mutants are relatively resistant but some therapeutic response is usually seen. The acquisition of a fourth and devastating mutation at position 164 (isoleucine to leucine) renders the available antifolates completely ineffective (11). Interestingly, mutations conferring moderate pyrimethamine resistance do not necessarily confer cycloguanil resistance, and vice versa. For example, mutations at positions 16 (alanine to valine) plus 108 (serine to threonine) confer high-level resistance to cycloguanil but not to pyrimethamine.In general, the biguanides are more active than pyrimethamine against the resistant mutants (and they are more effective clinically too), but they are ineffective against parasites with the DHFR mutation at position 164. P. vivax shares similar antifolate resistance mechanisms through serial acquisition of mutations in PvDHFR. The sequence of acquisition associated with increasing resistance is usually mutation at position 117 or 58, followed by mutation at positions 57, 61 and then 13.
Sulfonamide and sulfone
The marked synergy with sulfonamides and sulfones is very important for the antimalarial activity of sulfa-pyrimethamine or sulfone-biguanide combinations. In P. falciparum, sulfonamide and sulfone resistance also develops by progressive acquisition of mutations in the gene encoding the target enzyme PfDHPS (which is a bifunctional protein with the enzyme PPPK). Specifically altered amino acid residues have been found at positions 436, 437, 540, 581 and 613 in the PfDHPS domain. The mutations at positions 581 and 631 do not occur in isolation, but always following an initial mutation (usually at position 437, alanine to glycine). Mutations in P. vivax DHPS (at positions 383 and 553) also appear to contribute to resistance.
Resistance to atovaquone results from point mutations in the gene (cytB), which encodes cytochrome b. In the atovaquone-proguanil combination, it is proguanil itself probably acting on the mitochondrial membrane rather than the dhfr-inhibiting proguanil metabolite cycloguanil that appears to be important in this combination. Whether and how resistance develops to the mitochondrial action of proguanil is not known.
Although a target for the artemisinins has recently been identified (PfATPase6), preliminary studies have not so far associated polymorphisms in the gene encoding this enzyme with reduced susceptibility of malaria parasites. Amplification in PfMDR does reduce artemisinin susceptibility in vitro, but not to a degree that causes in vivo resistance. This has lead to erroneous claims that artemisinin resistance was being selected by widespread use of artemisinin derivatives, whereas in fact the selection pressure came from mefloquine use.
The mutation frequencies derived from in vitro studies are often much higher than those derived from observations in vivo (12). The absence of host defences and differences in antimalarial concentration profiles contribute to this discrepancy. The highest rates of emergence of resistance in vivo are for pyrimethamine and atovaquone. In the case of atovaquone, it has been estimated that one in three patients with symptomatic falciparum "contained", at presentation, a spontaneously arising atovaquone-resistant mutant parasite (5). For drugs such as chloroquine or artemisinin, the genetic events conferring resistance are much rarer. These genetic events may result in moderate changes in drug susceptibility, such that the drug still remains effective (e.g. as in the 108AsnDHFR mutation for pyrimethamine resistance) or, less commonly, very large reductions in susceptibility such that achievable concentrations of the drug are completely ineffective (e.g. as the cytochrome B mutations giving rise to atovaquone resistance) (13-16).
A6.3.3 Antimalarial pharmacokinetics and the selection of resistance
Absorption and disposition
The probability of selecting a de novo mutation that is resistant to antimalarials during the initial phase of treatment depends on the per-parasite frequency of the genetic event, the number of parasites present, immunity in the infected individual, and the relationship between the drug levels achieved and the degree of resistance conferred by the mutant parasite. Obviously, if the range of blood concentrations achieved in the patient considerably exceeds the concentrations giving 90% inhibition of multiplication (IC90 values) for the most resistant mutant (IC90R), then resistance cannot be selected in the acute phase of treatment as even the resistant mutants are prevented from multiplying. Conversely, if the degree of resistance provided by the genetic event is very small, the window of opportunity for selection may be negligible. Provided that there is such a window of selection then the broader the range of peak antimalarial concentrations and the closer the median value approaches IC90R , the greater the probability of selecting a resistant mutant in a patient. Peak drug concentrations are determined by absorption, distribution volume and dose. Several antimalarials (notably lumefantrine, halofantrine, atovaquone and, to a lesser extent, mefloquine) are lipophilic, hydrophobic and very variably absorbed (interindividual variation in bioavailability up to 20-fold) (17, 18). Interindividual variation in distribution volumes tends to be lower (usually less than five-fold) but, taken together with variable absorption, the outcome is considerable interindividual variation in peak antimalarial blood concentrations. The main sources of underdosing globally are incorrect self-medication because of poor adherence to the correctly prescribed drug regimen, poor quality drugs, uncontrolled drug availability and purchase of incorrect dose regimens, use of substandard drugs purchased in shops or markets, and incorrect administration in the home. The acute infection is the principal source of de novo resistance selection. Quality assured drugs, education, correct prescribing, good adherence, and optimized packaging and formulations therefore play pivotal roles in preventing the emergence of antimalarial drug resistance.
Drug elimination rates
In some areas of the world, transmission intensities may be as high as three infectious bites per person per day. In this context, a person who takes antimalarial treatment for symptomatic malaria exposes not only the parasites causing that infection to the drug, but also any newly acquired infections that emerge from the liver during the drug's elimination phase; the longer the terminal elimination half-life, the greater the exposure. The length of the terminal elimination half-life is an important determinant of the propensity for an antimalarial to select for resistance (19-21). Some rapidly eliminated antimalarials (e.g. the artemisinin derivatives) never present an intermediate drug concentration to infecting malaria parasites because they are eliminated completely within the two-day life-cycle of the asexual parasite. Others (e.g. mefloquine, chloroquine) have elimination half-lives of weeks or months and present a lengthy selection opportunity.
With the exception of the artemisinin derivatives, maximum antimalarial parasite reduction ratios (kill rates) do not exceed 1000-fold per cycle (22). Following hepatic schizogony, exposure of at least two asexual cycles (4 days) to therapeutic drug concentrations is therefore required to eradicate the blood stage parasites emerging from the liver. Even with maximum kill rates in the sensitive parasites and maximum growth rates in the resistant parasites, the resistant parasites only "overtake" the sensitive parasites in the third asexual cycle. Thus rapidly eliminated drugs (such as the artemisinin derivatives or quinine) cannot select during the elimination phase. Obviously, the greater the degree of resistance conferred by the resistance mutation - i.e. the higher the IC90R relative to the IC90 for susceptible parasites (IC90S) - the wider is the window of selection opportunity.
Patent gametocytaemia is more likely in recrudescent than in primary infections. Therefore, if de novo resistance arose in an acute symptomatic treated infection, the transmission probability from the subsequent recrudescent infection (bearing the new resistance genes) would be higher than from an infection newly acquired during the elimination phase of the antimalarial given for a previous infection, even if it attained the same parasite densities (23).
A6.3.4 Spread of resistance
Several mathematical models have been devised to examine the spread of antimalarial drug resistance (10, 21, 24, 25). Spread of resistance is determined by the reproductive advantage conferred by the resistance mechanism. This derives from the increased gametocyte carriage associated with treatment failure (both from the primary infection and the subsequent recrudescences) - the "donors", and then the selective pressure from residual concentrations of slowly eliminated antimalarial in potential recipients. A long elimination half-life results in long periods of post-treatment chemoprophylaxis.
Resistance encoded by multiple mutations at a single locus may be considered in two overlapping phases. The first phase, in which the drug is better tolerated by the parasites but therapeutic doses still usually clear the infection, and the second phase, when clinical failures start to occur. This second phase is very rapid and it is essential that surveillance programmes are in place and capable of monitoring the change from the first to the second phase. In areas of high transmission, the first phase may occur faster, but the subsequent phase slower. Combination therapy significantly slows the rate of evolution of resistance, but it should be instigated before significant resistance to either component is present.
A6.3.5 Prevention of resistance by use of combination therapy
The theory underlying combination treatment of tuberculosis, leprosy and HIV infection is well known, and has recently been applied to malaria (4, 5, 24, 26-29). If two drugs with different modes of action, and therefore different resistance mechanisms, are used in combination, then the per-parasite probability of developing resistance to both drugs is the product of their individual per-parasite probabilities. For example, if the per-parasite probabilities of developing resistance to drug A and drug B are both 1 in 1012, then a simultaneously resistant mutant will arise spontaneously in 1 in 1024 parasites. As it is postulated that there are approximately 1017 parasites in the entire world, and a cumulative total of less than 1020 in one year, such a simultaneously resistant parasite would arise spontaneously roughly once every 10 000 years - provided the drugs always confronted the parasites in combination. Thus the lower the de novo per-parasite probability of developing resistance, the greater the delay in the emergence of resistance.
Stable resistance to the artemisinin derivatives has not yet been identified, and cannot yet be induced in the laboratory, which suggests that it may be very rare indeed. De novo resistance to chloroquine is also very rare, and appears to have arisen and spread only twice in the world during the first decade of intensive use in the 1950 s (30). On the other hand, resistance to antifolate and atovaquone arises relatively frequently (e.g. antifolate resistance rose to high levels within two years of the initial deployment of proguanil in peninsular Malaya in 1947) and can be induced readily in experimental models (14, 27). Against a background of chloroquine resistance, mefloquine resistance arose over a six-year period on the north-west border of Thailand (31).
The ideal pharmacokinetic properties for an antimalarial have been much debated. Rapid elimination ensures that the residual concentrations do not provide a selective filter for resistant parasites, but drugs with this property (if used alone) must be given for at least 7 days, and adherence to 7-day regimens is poor. In order to be effective in a 3-day regimen, elimination half-lives usually need to exceed 24 h. Artemisinin derivatives are particularly effective in combinations with other antimalarials because of their very high killing rates (parasite reduction rate around 10 000-fold per cycle), lack of adverse effects and absence of significant resistance (5). Combinations of artemisinin derivatives (which are eliminated very rapidly) given for 3 days, with a slowly eliminated drug such as mefloquine, provide complete protection against the emergence of resistance to the artemisinin derivatives if adherence is good, but they do leave the slowly eliminated "tail" of mefloquine unprotected. Perhaps resistance could arise within the residual parasites that have not yet been killed by the artemisinin derivative. However, the number of parasites exposed to mefloquine alone is a tiny fraction (less than 0.00001%) of those present in the acute symptomatic infection. Furthermore, these residual parasites "see" relatively high levels of mefloquine and, even if susceptibility was reduced, these levels may be sufficient to eradicate the infection (Figure A6.2). The long mefloquine tail does, however, provide a selective filter for resistant parasites acquired from elsewhere, and therefore contributes to the spread of resistance once it has developed. Yet on the north-west border of Thailand, an area of low transmission where mefloquine resistance had already developed, systematic deployment of the artesunate-mefloquine combination was dramatically effective in stopping resistance and also in reducing the incidence of malaria (31, 32). This strategy is thought to be effective at preventing the emergence of resistance at higher levels of transmission, where high-biomass infections still constitutes the major source of de novo resistance.
Figure A6.2 The artesunate + mefloquine combination. If no artesunate is given, then the number of parasites exposed to mefloquine alone is given by the area of A; with the combination administered for 3 days, the number of parasites exposed to mefloquine alone is given by the area of B (100 million times fewer). Furthermore, mefloquine levels are higher (m to n) when confronting B than when confronting the same number of parasites (B1) if no artesunate is given (x to y). If a parasite containing a de novo mefloquine-resistant mutation were to occur, then such a parasite should still be susceptible to artesunate. Thus the probability of selecting a resistant mutant is reduced by 100 million times, as only a maximum of 100 000 parasites are exposed to mefloquine alone after the fourth day (i.e. in the third cycle), and any artesunate-resistant parasite selected by artesunate initially would always be killed by the accompanying mefloquine. As a result, the combination is more effective, reduces transmission and prevents the emergence of resistance to both drugs.
A6.4 A summary of the global distribution of antimalarial drug resistance
Resistance to antimalarials has been a particular problem with P. falciparum, in which widespread resistance to chloroquine, sulfadoxine-pyrimethamine and mefloquine has been observed (Figure A6.3). Antifolate and chloroquine resistance has developed in P. vivax in several areas, and chloroquine resistance in P. malariae has also recently been reported. No significant resistance has yet been observed to artemisinin and its derivatives despite their extensive deployment in several parts of Asia.
Figure A6.3 Malaria transmission areas and the distribution of reported resistance or treatment failures with selected antimalarial drugs, September 2004 (mefloquine resistance in Africa is currently being further reviewed)
A6.4.1 Plasmodium falciparum resistance
The first reports of chloroquine resistance occurred in Thailand and Colombia in the late 1950s, around 12 years after the drug's introduction. By 1980, all endemic areas in South America were affected, and by 1989, most of Asia and Oceania. In Africa, chloroquine resistance emerged in 1978 in the east, and gradually spread westwards through the 1980s. Resistance has now been documented in all falciparum-endemic areas except Central America and the Caribbean (33). Recent molecular studies favour importation of chloroquine resistance to Africa from East Asia (34, 35). Chloroquine resistance has emerged independently less than ten times in the past 50 years (Figure A6.4).
Figure A6.4 Distribution of chloroquine resistance in Plasmodium falciparum
Resistance to pyrimethamine emerged rapidly after its deployment for treatment, prophylaxis and, in some areas, mass treatment in the 1950s. Resistance to both components of sulfadoxine-pyrimethamine was noted shortly after this drug was introduced over a decade later. In South-East Asia this occurred on the Thai-Cambodian border in the mid-1960s. Resistance became an operational problem in the same area within the few years of the introduction of sulfadoxine-pyrimethamine to the malaria control programme in 1975 (36). High-level resistance is found in many parts of South-East Asia, southern China and the Amazon basin, and lower levels of resistance are seen on the coast of South America and in southern Asia and Oceania. In eastern Africa, sulfadoxine-pyrimethamine sensitivity was observed to be declining in the 1980s and resistance has progressed westwards across Africa relentlessly over the last decade. Clinical failure rates of more than 25% have already been reported in Liberia (37), Guinea Bissau (38) and Malawi (39).
Many areas now have high-level resistance with high-treatment failure rates in children. Recent molecular evidence suggests a common South-East Asian origin of the resistant P. falciparum parasites now prevalent in much of southern and Central Africa (triple dhfr mutant) (9, 40-42) (Figure A6.5).
Figure A6.5 Distribution of sulfadoxine-pyrimethamine resistance in Plasmodium falciparum
Mefloquine resistance was first observed on the Thai-Cambodian and Thai-Burmese borders in the late 1980s (43, 44) and the monotherapy is no longer effective there. Migrant gem miners returning from Cambodia may have been the means of spread of mefloquine resistance to India and Bangladesh (45). Isolated cases of mefloquine resistance have also been reported from the Amazon basin, and in vitro studies in Africa have identified some P. falciparum strains with low mefloquine sensitivity. Overall, clinical mefloquine resistance outside South-East Asia is rare (Figure A6.6).
The main determinant of mefloquine resistance is amplification of the gene (Pfmdr) that encodes the multidrug transporter (46). Amplification occurs only for the "wild type" allele explaining the inverse relationship between sensitivity to chloroquine (the Pfmdr Tyr86 mutation is associated with reduced sensitivity) and to mefloquine (and to the structurally related drugs, quinine and halofantrine) (15).
Figure A6.6 Distribution of mefloquine resistance in Plasmodium falciparum
The first reports of possible quinine resistance occurred in Brazil almost 100 years ago. Even today, however, clinical resistance to quinine monotherapy is reported only sporadically in South-East Asia and western Oceania, and resistance in Africa and South America is much less frequent. Widespread use of quinine in Thailand in the 1980s led to significant reduction in its sensitivity (45). Quinine is therefore now used in combination with an antibiotic, usually tetracycline, doxycycline or clindamycin, and is reserved for cases of severe malaria. Pfmdr1 mutations associated with chloroquine resistance have been believed to be associated with reduced susceptibility to quinine (47, 48).
Except in an animal model, there have been no confirmed reports of artemisinin resistance in malaria parasites that infect humans. The pharmacological characteristics of the drug, namely short elimination half-life, rapidity of action and ability to reduce gametocyte carriage, should delay the onset of significant resistance. Artemisinin derivatives are associated with high recrudescence rates (~10%) after monotherapy, so are usually combined with longer-acting antimalarials for clinical treatment. These recrudescences, however, are not a result of resistance.
Multidrug resistance is generally defined as resistance to three or more antimalarial compounds from different chemical classes. Generally, the first two classes are 4-aminoquinolines (e.g. chloroquine) and antifolates (e.g. sulfadoxine-pyrimethamine). The precise amount of resistance needed in order for a drug to be considered as failing is not universally agreed. Some consider clinical cure rates of less than 75% to be the minimum required for classification as failure, while the current recommendations aim for cure rates over 90%.
Established multidrug resistance occurs in South-East Asia (particularly along the borders of Thailand with Burma and Cambodia) and in the Amazon basin. In Thailand, mefloquine monotherapy was replaced with the combination of high-dose mefloquine and artesunate given for 3 days. Mefloquine resistance has been reduced by the use of this combination, as cure rates of more than 95% have been sustained for over 10 years, and susceptibility to mefloquine has actually improved despite extensive deployment of the combination.
Several areas are at risk of multidrug resistance, as resistance to chloroquine and sulfadoxine-pyrimethamine is already widespread. Progressive loss of sulfadoxine-pyrimethamine efficacy should be taken as a warning sign.
A6.4.2 Plasmodium vivax resistance
Resistance of P. vivax is rare and generally limited to chloroquine resistance, which was first reported in the late 1980sin Papua New Guinea and Indonesia. Focal true chloroquine resistance (with whole blood chloroquine + desethylchloroquine concentrations of >100 ng/ml on the day of failure) or prophylactic and/or treatment failure not necessarily related to true resistance, have since also been observed in Brazil, Colombia, Ethiopia, Guatemala, Guyana, India, Republic of Korea, Myanmar, Solomon Islands, Thailand and Turkey.
A6.4.3 Plasmodium malariae resistance
Resistance of P. malariae to chloroquine was observed recently in Indonesia.
A6.5 Monitoring of antimalarial drug resistance
A6.5.1 Monitoring methods
The rapid spread of antimalarial drug resistance over the last few decades has increased the need for monitoring, in order to ensure proper management of clinical cases, allow for early detection of changing patterns of resistance, and suggest where national malaria treatment policies should be revised. The monitoring procedures available include therapeutic efficacy testing (also known as in vivo testing). This involves the repeated assessment of clinical and parasitological outcomes of treatment - during a fixed period of follow-up - in order to detect any reappearance of symptoms and signs of clinical malaria and/or parasites in the blood, which would indicate reduced parasite sensitivity with the particular drug. Other methods include in vitro studies of parasite susceptibility to drugs in culture, and studies of point mutations or duplications in parasite resistance genes with molecular methods (polymerase chain reaction, PCR). Animal models are also used, although not routinely.
In vivo tests
(a) Therapeutic efficacy testing and the WHO standard protocol for P. falciparum
From a programmatic point of view, data on therapeutic efficacy are most useful in deciding whether or not a drug is still appropriate as first-line treatment. Therapeutic efficacy studies are relatively simple to conduct, and the requirements in terms of training of staff and technical facilities are therefore limited. However, the results can be affected by misdiagnosis and incorrect drug administration. In order to interpret and allow for comparison of the results within and between regions, and to follow trends over time, studies need to be conducted according to similar procedures and standards. WHO therefore recommends the use of the WHO standard protocol, which provides guidance on how best to obtain the minimum necessary information about the therapeutic responses to an antimalarial so as to allow informed decision-making on its future use (49).
The protocol is designed for use in the assessment of antimalarial drugs or drug combinations used routinely for treatment of uncomplicated P. falciparum malaria (chloroquine, sulfadoxine-pyrimethamine, amodiaquine, artemisinin-based combination therapies and others). It comprises a simple, one-arm prospective evaluation of clinical and parasitological treatment responses in children aged 6-59 months, in whom the level of acquired immunity is relatively low and therefore has only a minor influence on the outcome of the test. To ensure a reasonable specificity of the malaria diagnosis in areas of intense transmission, only individuals with a parasite density ≥ 2000 asexual parasites/µl of blood should be included in studies. In areas of low to moderate transmission, individuals with ≥ 1000 asexual parasites/µl can be included. Further methodological and operational considerations in relation to case definition, sample size calculations, ethical concerns and the criteria for inclusion and exclusion, some of which some relate only to specific drugs, are explained in detail in the protocol.
The recommended duration of follow-up is ≥ 28 days in areas of intense as well as low to moderate transmission. As a significant proportion of treatment failures do not appear until after day 14, shorter observation periods lead to a considerable overestimation of the efficacy of the tested drug. This is a particular problem at low levels of resistance and with low failure rates (50). As the objective of treatment is cure of the infection, and cure rates of more than 90% are required, the cure rate must be adequately characterized. For relatively effective, slowly eliminated antimalarials, half the recrudescences may occur after 28 days. For treatment with drugs such as amodiaquine, chloroquine and sulfadoxine-pyrimethamine, a 28-day follow-up is considered appropriate; follow-up periods of 42 days and 63 days are recommended for artemether-lumefantrine and mefloquine, respectively (51). These follow-up periods will capture most but often not all recrudescent infections - particularly at low levels of resistance. Studies even of > 28 days of duration risk loss to follow-up and should be accompanied by molecular assessments (PCR genotyping) so as to distinguish recrudescence from reinfection. If surveillance programmes do not have access to molecular techniques, studies of 14 days of duration without PCR adjustments can still provide useful information on failing drugs (i.e. to justify their replacement) - but they cannot be used to justify inclusion or continued recommendation. In areas of low to moderate transmission, the use of molecular methods is recommended, but is not strictly essential if the likelihood of reinfection is relatively small. PCR genotyping involves comparison of polymorphic parasite genes, usually those encoding variable blocks within PfMSP2, and also sometimes PfMSP1 and PfGLURP, in whole blood samples taken during the acute and recurrent infections.
The WHO standard protocol classifies outcomes of efficacy studies into the following four categories: early treatment failure, late clinical failure, late parasitological failure, and adequate clinical and parasitological response. These classifications rely on the presence or absence of fever or other signs of clinical malaria and/or presence of parasitaemia during the course of follow-up (Table A6.1). The therapeutic response is classified as early treatment failure if the patient develops clinical or parasitological symptoms during the first 3 days of follow-up. The response is classified as late clinical failure if symptoms develop during the follow-up period (from day 4 to day 28), without previously meeting the criteria for early treatment failure. It is a late parasitological failure if only parasitaemia reappear without any symptom, in the period from day 7 to day 28. Adequate clinical and parasitological response is defined as the absence of symptoms and of parasitaemia on day 28, without any of the criteria for the other three categories having been met previously.
Table A6.1 Classification of treatment outcomes in studies of antimalarial drug efficacy in areas of low, moderate and intense transmission (49)
Symptoms and signs
Early treatment failure
• Development of danger signs or severe malaria on days 1-3 in the presence of parasitaemia
• Parasitaemia on day 2 higher than the day 0 count irrespective of axillary temperature
• Parasitaemia on day 3 with axillary temperature ≥ 37.5 °C
• Parasitaemia on day 3 that is ≥ 25% of count on day 0.
Late treatment failure
• Late clinical failure
• Development of danger signs or severe malaria after day 3 in the presence of parasitaemia, without previously meeting any of the criteria of early treatment failure
• Presence of parasitaemia and axillary temperature ≥37.5 °C (or history of fever) on any day from day 4 to day 28, without previously meeting any of the criteria of early treatment failure.
• Late parasitological failure
• Presence of parasitaemia on any day from day 7 to day 28 and axillary temperature < 37.5 °C, without previously meeting any of the criteria of early treatment failure or late clinical failure.
Adequate clinical and parasitological response
• Absence of parasitaemia on day 28 irrespective of axillary temperature without previously meeting any of the criteria of early treatment failure, late clinical failure or late parasitological failure.
For simplicity, the outcome of efficacy studies can be summarized as "clinical failure", which is equal to the sum of early treatment failure and late clinical failure, and as "total failure", which is equal to the sum of early treatment failure, late clinical failure and late parasitological failure. The rates of clinical failure and total failure are used to define cut-off points for drug policy change, using the standard WHO protocol. It should be noted that the most recent classification of therapeutic responses described above differs from that used previously; late parasitological treatment responses are now also considered as an indicator of drug efficacy, as persistent parasitaemia is associated with increased risk of clinical malaria, anaemia and increased gametocyte carriage (52). The protocol provides guidance on how to calculate and present efficacy test results.
If feasible, any judgement of the therapeutic efficacy of a drug should be accompanied by measurements of blood drug concentrations, to ensure that therapeutic drug levels were reached; subtherapeutic levels confound the efficacy result. With modern techniques, antimalarial drug concentrations can often be analysed in small samples of blood dried on filter paper; samples can be sent to a central pharmacological laboratory for analysis.
(b) In vivo assessment of resistance in P. malariae
Protocols similar to those used for P. falciparum can be used.
(c) In vivo assessment of resistance in P. vivax and P. ovale infections
Relapse and recrudescence cannot be distinguished reliably in these infections, as they will usually have the same genotype. Nevertheless the in vivo assessment of chloroquine susceptibility can be performed using the same format as for P. falciparum, with a follow-up period of 28 or preferably 35 days, and preferably accompanied by measurement of whole blood chloroquine and desethychloroquine levels. Recurrent infections within this period presenting with whole blood chloroquine + desethychloroquine concentrations exceeding 100 ng/ml can be considered as resistant whether they are a relapse, a recrudescence, or even a new infection, as this concentration should be suppressive (50, 53, 54).
In vitro resistance tests
To support evidence of a failing antimalarial, in vitro tests can be used to provide a more accurate measure of drug sensitivity under controlled experimental conditions. Parasites obtained from finger-prick blood are placed in microtitre wells, exposed to precisely known concentrations of a particular drug and examined for the inhibition of maturation into schizont parasite stages (55). This test overcomes some of the many confounding factors influencing the results of in vivo tests, such as subtherapeutic drug concentrations and the influence of host factors on parasite growth (e.g. factors related to acquired immunity), and therefore provide a more accurate picture of the "true" level of resistance to the drug. Multiple tests can be performed on parasite isolates, using several drugs and drug combinations simultaneously. New experimental drugs can also be tested in this way. However, partly because in vitro tests do not include host factors, the correlation between results of in vitro and in vivo tests is not consistent and is not well understood. Furthermore, different parasite isolates may adapt differently to culture, which may affect the test result. For example, if a resistant strain adapts less well to culture and therefore dies off earlier, the outcome is an overestimation of its susceptibility. Prodrugs such as proguanil, which require conversion into active metabolites in the human host, cannot be tested, and P. vivax, P. ovale and P. malariae cannot be evaluated in vitro owing to constraints in culturing these species (although this has now largely been overcome for P. vivax). In vitro testing is more demanding in terms of technology and resources, and is not ideal for routine drug efficacy evaluation under field conditions. It should therefore primarily be used to provide additional information to support clinical efficacy data at selected resistance-monitoring sites.
In recent years, molecular tests have been developed for the detection of parasite gene mutations or amplifications associated with resistance to antimalarials as an additional means of assessing levels of drug resistance. These methods are based on PCR, using small amounts of parasite DNA material in finger-prick blood dried on filter paper.
Information on the prevalence of gene mutations may give an indication of the level of drug resistance in an area, and relatively well-defined molecular markers of resistance have been established for pyrimethamine (Pfdhfr and Pvdhfr), sulfadoxine (Pfdhps) and chloroquine (Pfcrt1) (56, 57). Amplification of Pfmdr (for mefloquine resistance) is considerably more technically demanding, requiring validated real-time PCR. No markers are yet available for other antimalarials.
These methods have their disadvantages. The results are seldom available rapidly, and mutations and the measured therapeutic efficacy do not always correlate well, as many factors determine the therapeutic response in addition to parasite sensitivity to the antimalarial drug treatment. However, serial assessment of molecular markers can be a useful guide to the emergence of resistance, especially if used consistently over time in comparable study populations to detect trends. The methods may also provide useful guidance on choices of treatment during acute malaria epidemics, where time will not allow for clinical efficacy tests, but where it may be critical to avoid the use of a particular drug (58). The requirements for technical skills and laboratory facilities prevent the routine use of these methods in most drug efficacy testing sites, although they are becoming increasingly used in laboratories in endemic areas - particularly those supporting clinical trials. Moreover, the results are subject to error and must be considered carefully. For example, a patient may have an infection with two different genotypes on day 0 but, only one genotype is detected with PCR. If one genotype is sensitive and the other is resistant, the resistant genotype may persist until day 14 despite antimalarial treatment, while the sensitive genotype will be cleared; if detected, the presence of the resistant genotype on day 14 may then incorrectly be interpreted as reinfection and not as recrudescence (59).
While monitoring of molecular markers may only be possible at central laboratories, it can support monitoring programmes that rely on in vivo testing, and also play an important part in early-warning systems to guide treatment policies coordinated at national and regional levels. With newly developed high-throughput methods (60, 61), more comprehensive population-based analyses will also be possible, which may allow for a better understanding and prediction of the future spread of resistance (40, 42).
Efficacy testing in animal models
In addition to in vivo efficacy studies involving human participants, drug sensitivity can be tested in animal models. Although such models do not play an important role in routine efficacy-monitoring programmes, they may be useful in the testing of newly developed drugs, not yet approved for use in humans, or to minimize the influence of host immunity on drug efficacy, while retaining the influence of some of the extrinsic factors observed in in vivo studies.
A6.5.2 Reporting of treatment failures
Reports of cases of treatment failure and decreased drug sensitivity have often provided important first evidence for more widespread resistance in an area. Although such evidence is subject to bias, it can be collected without much effort at peripheral health centres. If standardized and registered, such reports can make a valuable contribution to national early-warning systems, facilitating cost-effective monitoring by national programmes.
A6.5.3 Criteria for antimalarial drug policy change
The WHO malaria treatment guidelines recommend that antimalarial treatment policy should be changed at treatment failure rates considerably lower than those recommended previously. This major change reflects the availability of highly effective drugs, and the recognition both of the consequences of drug resistance, in terms of morbidity and mortality, and of the importance of high cure rates in malaria control.
It is now recommended that a change of first-line treatment should be initiated if the total failure proportion exceeds 10%. However, it is acknowledged that a decision to change may be influenced by a number of other factors, including the prevalence and geographical distribution of reported treatment failures, health service provider and/or patient dissatisfaction with the treatment, political and economical contexts, and the availability of affordable alternatives to the commonly used treatment.
1. Bonhoeffer S, Lipsitch M, Levin BR. Evaluating treatment protocols to prevent antibiotic resistance. Proceedings of the National Academy of Sciences of the USA, 1997, 94:12106-12111.
2. Lipsitch M, Levin BR. The population dynamics of antimicrobial chemotherapy. Antimicrobial Agents and Chemotherapy, 1997, 41:363-373.
3. Austin DJ, Anderson RM. Studies of antibiotic resistance within the patient, hospitals and the community using simple mathematical models. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences, 1999, 354:721-738.
4. Peters W. Chemotherapy and drug resistance in malaria, 2nd ed. London, Academic Press, 1987.
5. White NJ. Antimalarial drug resistance and combination chemotherapy. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences, 1999, 354:739-749.
6. Jeffery GM, Eyles DE. Infectivity to mosquitoes of Plasmodium falciparum as related to gametocyte density and duration of infection. American Journal of Tropical Medicine and Hygiene, 1955, 4:781-789.
7. Mackinnon MJ, Hastings IM. The evolution of multiple drug resistance in malaria parasites. Transactions of the Royal Society of Tropical Medicine and Hygiene, 1998, 92:188-195.
8. Hastings IM, Mackinnon MJ. The emergence of drug-resistant malaria. Parasitology, 1998, 117:411-417.
9. Roper C et al. Intercontinental spread of pyrimethamine-resistant malaria. Science, 2004, 305:1124.
10. Dye C, Williams BG. Multigenic drug resistance among inbred malaria parasites. Proceedings of the Royal Society of London, Series B, Biological Sciences, 1997, 264: 61-67.
11. Gatton ML, Martin LB, Cheng Q. Evolution of resistance to sulfadoxine-pyrimethamine in Plasmodium falciparum. Antimicrobial Agents and Chemotherapy, 2004, 48:2116-2123.
12. Rathod PK, McErlean T, Lee PC. Variations in frequencies of drug resistance in Plasmodium falciparum. Proceedings of the National Academy of Sciences of the USA, 1997, 94:9389-9393.
13. Cowman AF, Galatis D, Thompson JK. Selection for mefloquine resistance in Plasmodium falciparum is linked to amplification of the pfmdr1 gene and cross-resistance to halofantrine and quinine. Proceedings of the National Academy of Sciences of the USA, 1994, 91:1143-1147.
14. Looareesuwan S et al. Clinical studies of atovaquone, alone or in combination with other antimalarial drugs, for treatment of acute uncomplicated malaria in Thailand. American Journal of Tropical Medicine and Hygiene, 1996, 54:62-66.
15. Reed MB et al. Pgh1 modulates sensitivity and resistance to multiple antimalarials in Plasmodium falciparum. Nature, 2000, 403: 906-909.
16. Korsinczky M. et al. Mutations in Plasmodium falciparum cytochrome b that are associated with atovaquone resistance are located at a putative drug-binding site. Antimicrobial Agents and Chemotherapy, 2000, 44: 2100-2108.
17. White NJ. Antimalarial pharmacokinetics and treatment regimens. British Journal of Clinical Pharmacology, 1992, 34:1-10.
18. White NJ, van Vugt M, Ezzet F. Clinical pharmacokinetics and pharmacodynamics of artemether-lumefantrine. Clinical Pharmacokinetics, 1999, 37:105-125.
19. Watkins WM, Mosobo M. Treatment of Plasmodium falciparum malaria with pyrimethamine-sulphadoxine: selective pressure for resistance is a function of long elimination half-life. Transactions of the Royal Society of Tropical Medicine and Hygiene, 1973, 87:75-78.
20. Nzila AM et al. Molecular evidence of greater selective pressure for drug resistance exerted by the long-acting antifolate Pyrimethamine/Sulfadoxine compared with the shorter-acting chlorproguanil/dapsone on Kenyan Plasmodium falciparum. Journal of Infectious Diseases, 2000, 181:2023-2028.
21. Hastings I, Watkins WM, White NJ. The evolution of drug-resistance in malaria: the role of the terminal elimination half-life. Philosophical Transactions of the Royal Society London B Biological Sciences, 2002, 357:505-519.
22. White NJ. Assessment of the pharmacodynamic properties of antimalarial drugs in vivo. Antimicrobial Agents and Chemotherapy, 1997, 41: 1413-1422.
23. Price RN et al. Risk factors for gametocyte carriage in uncomplicated falciparum malaria. American Journal of Tropical Medicine and Hygiene, 1999, 60: 1019-1023.
24. Curtis CF, Otoo LN. A simple model of the build-up of resistance to mixtures of antimalarial drugs. Transactions of the Royal Society of Tropical Medicine and Hygiene, 1986, 80: 889-892.
25. Hastings IM. The origins of antimalarial drug resistance. Trends in Parasitology, 2004, 20:512-518.
26. Peters W. Drug resistance - a perspective. Transactions of the Royal Society of Tropical Medicine and Hygiene, 1969, 63:25-45.
27. Peters W. The prevention of antimalarial drug resistance. Pharmacology and Therapeutics, 1990, 47:499-508.
28. Chawira AN et al. The effect of combinations of qinghaosu (artemisinin) with standard antimalarial drugs in the suppressive treatment of malaria in mice. Transactions of the Royal Society of Tropical Medicine and Hygiene, 1987, 81:554-558.
29. White NJ et al. Averting a malaria disaster. Lancet, 1999, 353:1965-1967.
30. Su, X et al. Complex polymorphisms in an approximately 330 kDa protein are linked to chloroquine-resistant P. falciparum in Southeast Asia and Africa. Cell, 1997, 91: 593-603.
31. Nosten F et al. Effects of artesunate-mefloquine combination on incidence of Plasmodium falciparum malaria and mefloquine resistance in western Thailand; a prospective study. Lancet, 2000, 356: 297-302.
32. Brockman A et al. Plasmodium falciparum antimalarial drug susceptibility on the northwestern border of Thailand during five years of extensive artesunate-mefloquine use. Transactions of the Royal Society of Tropical Medicine and Hygiene, 2000, 94, 537-544.
33. Wongsrichanalai C et al. Epidemiology of drug-resistant malaria. Lancet Infectious Diseases, 2002, 2:209-218.
34. Wellems TE, Plowe CV. Chloroquine-resistant malaria. Journal of Infectious Diseases, 2001, 184:770-776.
35. Vieira PP et al. pfcrt Polymorphism and the spread of chloroquine resistance in Plasmodium falciparum populations across the Amazon Basin. Journal of Infectious Diseases, 2004, 190:417-424.
36. Hurwitz ES, Johnson D, Campbell CC. Resistance of Plasmodium falciparum malaria to sulfadoxine-pyrimethamine ("Fansidar") in a refugee camp in Thailand. Lancet, 1981, 1:1068-1070.
37. Checchi F et al. High Plasmodium falciparum resistance to chloroquine and sulfadoxine-pyrimethamine in Harper, Liberia: results in vivo and analysis of point mutations. Transactions of the Royal Society of Tropical Medicine and Hygiene, 2002, 96:664-669.
38. Kofoed PE et al. Treatment of uncomplicated malaria in children in Guinea-Bissau with chloroquine, quinine, and sulfadoxine-pyrimethamine. Transactions of the Royal Society of Tropical Medicine and Hygiene, 2002, 96:304-309.
39. Plowe CV et al. Sustained clinical efficacy of sulfadoxine-pyrimethamine for uncomplicated falciparum malaria in Malawi after 10 years as first line treatment: five year prospective study. British Medical Journal, 2004, 328:545.
40. Anderson TJ. Mapping drug resistance genes in Plasmodium falciparum by genome-wide association. Current Drug Targets. Infectious Disorders, 2004, 4:65-78.
41. Nair S et al. A selective sweep driven by pyrimethamine treatment in southeast Asian malaria parasites. Molecular Biology and Evolution, 2003, 20:1526-1536.
42. Roper C et al. Antifolate antimalarial resistance in southeast Africa: a population-based analysis. Lancet, 2003, 361:1174-1181.
43. Nosten F et al. Mefloquine-resistant falciparum malaria on the Thai-Burmese border. Lancet, 1991, 337:1140-1143.
44. Fontanet AL et al. High prevalence of mefloquine-resistant falciparum malaria in eastern Thailand. Bulletin of the World Health Organization, 1993, 71:377-383.
45. Wernsdorfer WH. Epidemiology of drug resistance in malaria. Acta Tropica, 1994, 56:143-156.
46. Price RN et al. Mefloquine resistance in Plasmodium falciparum and increased pfmdr1 gene copy number. Lancet, 2004, 364:438-447.
47. Zalis MG et al. Characterization of Plasmodium falciparum isolated from the Amazon region of Brazil: evidence for quinine resistance. American Journal of Tropical Medicine and Hygiene, 1998, 58:630-637.
48. Duraisingh MT et al. Linkage disequilibrium between two chromosomally distinct loci associated with increased resistance to chloroquine in Plasmodium falciparum. Parasitology, 2000, 121:1-7.
49. Assessment and monitoring of antimalarial drug efficacy for the treatment of uncomplicated falciparum malaria. Geneva, World Health Organization, 2003 (document WHO/HTM/RBM/2003.50).
50. White NJ. The assessment of antimalarial drug efficacy. Trends in Parasitology, 2002, 18: 458-464.
51. Ringwald P. Monitoring antimalarial drug efficacy. Clinical Infectious Diseases, 2004, 38:1192-1193.
52. Assessment of therapeutic efficacy of antimalarial drugs for uncomplicated falciparum malaria in areas with intense transmission. Geneva, World Health Organization, 1996 (document WHO/MAL/96.1077).
53. Baird JK et al. Diagnosis of resistance to chloroquine by Plasmodium vivax: timing of recurrence and whole blood chloroquine levels. American Journal of Tropical Medicine and Hygiene, 1997, 56:621-626.
54. Baird JK. Chloroquine resistance in Plasmodium vivax. Antimicrobial Agents and Chemotherapy, 2004, 48:4075-4083.
55. Rieckmann K.H. et al. Drug sensitivity of Plasmodium falciparum. An in-vitro microtechnique. Lancet, 1978, 1:22-23.
56. Djimde A et al. A molecular marker for chloroquine-resistant falciparum malaria. New England Journal of Medicine, 2001, 344:257-263.
57. Kublin J.G. et al. Molecular markers for failure of sulfadoxine-pyrimethamine and chlorproguanil-dapsone treatment of Plasmodium falciparum malaria. Journal of Infectious Diseases, 2002, 185:380-388.
58. Djimde A et al. Molecular diagnosis of resistance to antimalarial drugs during epidemics and in war zones. Journal of Infectious Diseases, 2004, 190:853-855.
59. Snounou G, Beck HP. The use of PCR genotyping in the assessment of recrudescence or reinfection after antimalarial drug treatment. Parasitology Today, 1998, 14:462-467.
60. Pearce RJ et al. Molecular determination of point mutation haplotypes in the dihydrofolate reductase and dihydropteroate synthase of Plasmodium falciparum in three districts of northern Tanzania. Antimicrobial Agents and Chemotherapy, 2003, 47:1347-1354.
61. Alifrangis A et al. A simple high-throughput method to detect P. falciparum single nucleotide polymorphisms in the dihydrofolate reductase, dihydropterate synthase, and P. falciparum chloroquine resistance transporter genes using polymerase chain reaction- and enzyme-linked immunosorbent assay-based technology. American Journal of Tropical Medicine and Hygiene, 2005, 72:155-162.