160.fulljurnal

7/30/2019 160.fulljurnal

1/10

Monitoring and Deterring Drug-Resistant Malaria in the Era of Combination Therapy

Miriam K. Laufer, Abdoulaye A. Djimd, and Christopher V. Plowe*

Malaria Section, Center for Vaccine Development, University of Maryland School of Medicine, Baltimore, Maryland; MalariaResearch and Training Center, Department of Epidemiology of Parasitic Diseases, Faculty of Medicine, Pharmacy, and Dentistry,

University of Bamako, Bamako, Mali

Abstract. As chloroquine and sulfadoxine-pyrimethamine (SP) are replaced by more effective artemisinin-basedcombination therapies (ACTs), strategies for monitoring (and, if possible, deterring) drug-resistant malaria must beupdated and optimized. In vitro methods for measuring resistance will be critical for confirming and characterizingresistance to ACTs. Molecular markers are useful for tracking the emergence and dissemination of resistance andguiding treatment policy where resistance is low or moderate. Genomic approaches may help identify molecular markersfor resistance to artemisinins and their partner drugs. Studies of reported ACT treatment failure should include assessingfactors other than resistance that affect efficacy, including pharmacokinetics. Longitudinal clinical trials are particularlyuseful for comparing the benefits and risks of repeated treatment in high transmission settings. The malaria research andcontrol community should not fail to exploit this opportunity to apply the lessons of the last 50 years to extend the usefultherapeutic lives of ACTs.

INTRODUCTION

The landscape of antimalarial therapy is changing. With

new multilateral support for artemisinin-based combinationtherapies (ACTs), highly efficacious alternatives are becom-ing available to replace less effective drugs (chloroquine andsulfadoxine-pyrimethamine [SP]) that are still used widely de-spite their impaired efficacy. New drugs and treatment strat-egies will save many lives every day. Combination therapypresents new challenges for monitoring resistance and effi-cacy, as well as new prospects for deterring drug resistance.This review will explore the emerging and complex issuessurrounding monitoring of antimalarial drug resistance in theera of combination treatment ofPlasmodium falciparum ma-laria. We suggest that monitoring should go beyond passivedocumentation of the success or failure of antimalarial treat-ment. The influx of drugs, infrastructure, support, and enthu-siasm that is accompanying the transition to combinationtherapies presents opportunities to gain new understanding ofhow resistance emerges and spreads. This knowledge will inturn inform strategies for deterring, delaying, and containingresistance.

IN VITRO SUSCEPTIBILITY ASSESSMENT

Methods for measuring parasite growth in vitro in the pres-ence of increasing drug concentrations were developed forculture-adapted malaria parasites in controlled laboratorysettings.1In vitro tests determine the intrinsic susceptibility ofthe parasite to drugs without the confounding effects of host

factors such as immunity or pharmacokinetics, and thus, pro-vide a direct, quantitative assessment of drug resistance.Limitations of in vitro tests. Most discussions about strate-

gies for monitoring antimalarial drug efficacy start with thelimitations ofin vitro methods. Adapting these methods forfield testing of clinical isolates2 has been challenging for rea-sons reviewed elsewhere.3 In several cases, clinical trialsevaluating drugs to which good in vitro susceptibility had

been documented found unexpectedly high rates of treatment

failure.4,5 Factors other than parasite resistance can contrib-

ute to the failure of clinical isolates to grow in vitro, leading tofalse alarms for emerging resistance. The results of rapid invitro tests done in the field must be viewed with caution until

they are confirmed by repeated testing under carefully con-

trolled conditions in the laboratory. This requires preserva-tion and culture adaptation of field isolates.

At present, a clear definition of in vitro resistance to the

artemisinin derivatives is not available. Although decreased

in vitro susceptibility to the artemisinins has been reported in

small numbers of field isolates,6 bona fide, clinically signifi-

cant resistance to artemisinin drugs has not been convincingly

documented, and the significance of the variations that havebeen described in susceptibility to artemisinins in vitro insome isolates is unknown. Understanding the clinical rel-

evance of in vitro artemisinin susceptibility is further compli-cated by the fact that these drugs are almost always used incombination with other drugs.

Critical role for in vitro tests in the ACT era. Despite theselimitations, in vitro assays are increasingly important in theera of ACTs because of the inability to rely on molecularmethods for monitoring resistance and the absence to date ofclinically significant resistance to the artemisinins. The earlystages of parasite resistance to individual drugs used in com-bination therapy regimens may not be clinically apparent be-cause of the action of the partner drug(s). Clinical studies tomonitor efficacy may thus be relatively insensitive for herald-ing the impending failure of drug combinations. While can-didate molecular markers for resistance to artemisinins are

being studied6 and molecular correlates of responses to sev-eral ACT partner drugs have been described,7,8 validated mo-lecular markers for clinical ACT treatment failure are notavailable and will not become available unless true clinicalresistance to the artemisinins appears, for the simple reasonthat it is not possible to validate a marker for a phenomenonthat does not yet exist. For these reasons, in vitro tests are thefront line of resistance monitoring for artemisinins and ACTs.

As an early warning system, routine ongoing in vitro moni-toring of resistance to artemisinins and ACT partner drugsshould be performed, particularly in parts of the world whereACTs have been used the most. When reports of clinicaltreatment failures of ACTs arise, or where clinical trials show

* Address correspondence to Christopher V. Plowe, Center for Vac-cine Development, University of Maryland School of Medicine, 685W. Baltimore Street, HSF-1 Room 480, Baltimore, MD 21201.E-mail: [email protected]

Am. J. Trop. Med. Hyg., 77(Suppl 6), 2007, pp. 160169Copyright 2007 by The American Society of Tropical Medicine and Hygiene

160


2/10

higher than expected rates of treatment failure, rigorous invitro testing with cryopreservation of parasite isolates shouldbe done in a targeted fashion, modeled after outbreak studiesperformed by epidemic intelligence services (Figure 1). In astepwise fashion, factors including resistance to individualpartner drugs, pharmacological parameters, and host immu-nity can be assessed for their potential contributions to treat-

ment failure. If true resistance to artemisinins or partnerdrugs is documented, genomic analyses can be done to iden-tify molecular markers for resistance, as described below.

Clinical treatment failure can occur for reasons other thanintrinsic parasite resistance. If clinical failures of ACTs aredocumented, careful in vitro testing can help to determine ifthe failures were caused by resistance or to other causes, suchas poor drug quality, pharmacokinetic variability, or drug in-teractions. Aggressive study of possible resistance to artemisi-nins and their partner drugs should be a high priority as thesedrugs are rolled out. Capacity in malaria endemic areas mustbe expanded to enable reliable, high-quality in vitro suscep-tibility testing to occur at or near sentinel sites. If definitive invitro testing is not done expeditiously, repeated unsubstanti-

ated reports of clinical resistance may result in a

boy whocried wolf situation, in which the eventual emergence of trueresistance goes unrecognized until the opportunity to containit has passed.

MOLECULAR MARKERS FOR DRUG RESISTANCE

Molecular markers for resistance to monotherapies. In thelate 20th century, candidate molecular markers for antima-larial drug resistance were identified by cloning and sequenc-ing parasite genes homologous to those known to mediateresistance in other organisms911 and/or by using reverse ge-netics approaches to analyze the progeny of genetic crosses

between sensitive and resistant parasites.1214

Based on dif-ferences in DNA sequence between sensitive and resistantparasite clones, point mutations,1315 variable length repeat

sequences,16 and/or differences in copy number8,11,17 wereevaluated for their association with in vitro resistance pheno-types. In some cases, the link between molecular markers andin vitro resistance was proved (or disproved) in genetic trans-formation studies in which sensitive parasites were renderedresistant or vice-versa by precise substitutions of genetic se-quences in parasite clones maintained in laboratory culture18

21

or in model systems such as yeast.

22

Finally, molecular markers for resistance were assessed inecological studies23,24 and clinical trials2527 to establish theirassociation with, and ability to predict, treatment failure invivo. This process of identifying and evaluating molecularmarkers of resistance has been painstakingly slow, and rea-sonably reliable methods for using molecular markers tomonitor chloroquine and SP resistance were established onlyafter resistance to these drugs was so widespread that theinformation provided by molecular surveys was, with someexceptions described below, mainly of academic interest andof little immediate relevance to antimalarial treatment poli-cies.

Validation and application of molecular markers for chlo-

roquine and SP resistance. Because factors other than intrin-sic parasite resistance contribute to clinical and parasitologicoutcomes of antimalarial drug treatment, establishing the pre-dictive value of molecular markers for treatment outcomeshas been challenging. Even for markers with virtually abso-lute correlation between genotype and in vitro phenotype,other factors including acquired immunity25,28 and pharma-cokinetic parameters8,29 affect the clearance of drug-resistantparasites. In an attempt to use resistance markers to predicttreatment outcomes, immunity, the most important contribu-tor to clearance of resistant parasites in areas of high malariaendemicity,25,28,30 was accounted for by controlling for age ina simple model using the prevalence of molecular markers topredict treatment failure rates.27 In this model, the genotype-

failure index (GFI) is the ratio of the prevalence of resistantgenotypes to the rate of treatment failure in a population.Once established and adjusted for age, it was suggested that

FIGURE 1. Algorithm for studying reports of resistance to ACTs. PK, pharmacokinetics.

MONITORING AND DETERRING RESISTANCE 161


3/10

GFIs could be used to predict the treatment failure rate in a

population based on simple measurements of the prevalence

of a resistance marker in that population. This approach is

best suited to populations with stable malaria endemicity so

that a GFI calculated for a specific population at one point in

time remains valid for predicting failure rates in that popula-

tion as the prevalence of resistance markers changes over

time.The GFI model was developed and assessed as a tool for

predicting chloroquine treatment failure rates at four sites in

Mali,27 and has since been applied in several countries in

Africa.26,3137 Table 1 summarizes results of published papers

that both cite the paper proposing the GFI model27 and in-

clude data from which GFIs can be calculated directly. As

expected, at sites where the prevalence of the chloroquine

resistance marker approached 100%, GFIs varied widely.

This is because when the resistant genotype approaches fixa-

tion in a population, factors other than resistance must deter-

mine treatment outcomes.38 Thus, the GFI model is both

most useful and most relevant where resistance is still low to

moderate,30 but even so, a very high prevalence of molecular

markers for resistance is a reliable if imprecise indication thattreatment failure rates are unacceptably high. In settings with

treatment failure rates of 50% or higher, tools to predict pre-

cisely how far efficacy has fallen below acceptable thresholds

are not needed. Indeed, continuing large efficacy trials of

drugs where they are known to be poorly efficacious is ethi-

cally questionable.39 Especially now that more effective

ACTs are becoming widely available, such trials should not

be repeated.

Where chloroquine treatment failure rates were < 50%,

which is to say at sites where a molecular tool for predicting

clinical outcomes may provide information useful for guiding

treatment policy, GFIs were close to 2 at most sites. This

allows straightforward interpretation of molecular surveys for

the chloroquine resistance marker: the predicted rate of treat-ment failure is equal to about one half of the prevalence of

the marker.

Several studies have applied the GFI model to predict SP

failure rates.26,35,36,40 Mugittu and others36 in Tanzania sys-

tematically calculated GFIs at five sites using the DHFR

triple mutant as the genotype to predict SP treatment failure

at 14 days. At four sites with SP treatment failure rates rang-

ing from 7% to 14%, age-adjusted GFIs were between 2.0 and

2.1. At one site with an 80% prevalence of the resistance

marker, the GFI was 3.4, consistent with the idea that GFIs

are unreliable when the prevalence of the marker approaches

fixation in the population. Studies done in Malawi, Uganda,

and Nigeria also had GFIs for SP that were very close to 2,

despite using different sets of markers for SP resistance and

different definitions of treatment failure, suggesting that GFIs

are a robust tool for predicting SP treatment failure rates insettings of low to moderate resistance.

The use of molecular markers for predicting resistance

could be further improved with more sophisticated multivari-

ate models that take into account pharmacokinetic, host ge-

netics, and other factors that contribute to treatment out-

come, although such comprehensive models would be imprac-

tical for routine surveillance of resistance. Additional

refinements that might improve the GFI model include stud-

ies confirming that different sampling strategies such as cross-

sectional population surveys and collection of samples at sick

clinic visits yield comparable estimates of genotype preva-

lence and testing the suggestion to use calculated allele fre-

quencies rather than simple genotype prevalence.41

This GFI model has been used to guide malaria treatmentin an outbreak of malaria in Mali, where civil unrest pre-

cluded other means of assessing drug efficacy42 and to guide

national treatment policy in Tanzania.36 In addition to being

used to guide treatment policies, resistance markers have also

predicted the return of susceptible parasites after the removal

of drug pressure. In Malawi, a decrease in the prevalence of

the molecular marker for chloroquine resistance was detected

soon after the country switched from chloroquine to SP for

the first line therapy of malaria because of very high rates of

clinical failure. Within 8 years, parasites with the resistance

marker were undetectable.43 A clinical trial confirmed that

chloroquine again had excellent clinical efficacy, 12 years af-

ter it was removed from use.44 As ACTs replace chloroquine

and SP, monitoring the prevalence of molecular markers maysignal a time to consider reintroducing these drugs as com-

ponents of combination therapy or in targeted interventions

such as intermittent preventive treatment.

Molecular markers for resistance to ACTs in the genomic

era. In the search for molecular markers of drug-resistant ma-

laria, the low-hanging fruit has been picked. In the cases of

chloroquine and the antifolates, investigators were fortunate

that their heroic efforts using reverse genetics and candidate

TABLE 1

Published drug efficacy studies directly or indirectly assessing the GFI model for using molecular markers for drug-resistant malaria to predicttreatment outcomes

CountryNo.sites Drug Resistant genotype

Prevalenceof genotype

Treatmentfailure

rate GFIAge

adjusted Reference

Mali 4 Chloroquine PfCRT T76 1328% 4366% 1.62.7 Yes Djimd and others 200125

Ghana 1 Chloroquine PfCRT T76 81% 58% 1.4 NA Mockenhaup and others 200578

Congo 1 Chloroquine PfCRT T76 100% 96% 1.04 NA Mayengue and others 200579

Tanzania 1 Chloroquine PfCRT T76 100% 51% 2.0 No Khalil and others 200580

Sudan 1 Chloroquine PfCRT T76 100% 61% 1.6 No Khalil and others 200580

Burkina Faso 4 Chloroquine PfCRT T76 2265% 1537% 1.13.0 No Tinto and others 200581

Malawi 1 SP DHFR 59 + DHPS 540 65% 30% 2.2 NA Kublin and others 200226

Tanzania 5 SP DHFR triple mutant 1980% 724% 2.03.4% Yes Mugittu and others 200482

Uganda 1 SP DHFR 59 + DHPS 540 61% 33% 1.9 NA Kyabayinze and others 200335

Nigeria 1 SP DHFR triple mutant 46% 25% 1.8 NA Happi and others 200583

PfCRT, Plasmodium falciparum chloroquine resistance transporter; DHFR, dihydrofolate reductase; DHPS, dihydropteroate synthase; ETF, early treatment failure; LCF, late clinical failure;LPF, late parasitologic failure; NA, not applicable; SP, sulfadoxine pyrimethamine.

LAUFER AND OTHERS162


4/10

gene approaches eventually paid off with the identification ofkey resistance genes, which, even if they are not the solegenetic contributors to resistance, are clearly its primary de-terminants. In the ACT era, we cannot count on being sofortunate again. Candidate gene approaches based on non-malaria homologs or on suspected mechanisms of drug actionhave been used to study genetic determinants of resistance to

drugs included in ACTs. In an example of the homolog can-didate gene approach, in vitro and clinical evidence suggeststhat increased pfmdr1 copy number is associated withresistance to mefloquine, lumefantrine, and artemisinin, aswell as other antimalarial drugs.8,45 Based on studies showingthat artemisinins inhibit Ca2+ ATPase, polymorphisms inPfATPAse6 have been measured in isolates tested for in vitrosusceptibility to artemisinin derivatives in rapid in vitro fieldtests. PfATPAse6 mutations were reported to be associatedwith artesunate resistance in a small number of isolates fromFrench Guiana but not in samples from Senegal or Cambo-dia.6 Although such studies are useful for hypothesis genera-tion, carefully designed prospective studies with adequate sta-tistical power and population genetics analyses that account

for regional allelic diversity are needed for valid assessmentsof putative molecular markers for resistance to the artemisi-nins.46 More importantly, the candidate gene approach relieson what may be overly optimistic hope that a single gene canbe identified that encodes resistance to artemisinins. If, asseems likely, resistance to artemisinins and their partnerdrugs are mediated by multiple genes, a coordinated globalconsortium using genomic tools to identify molecular markersof ACT resistance is more likely to be profitable than aniterative gene-by-gene candidate gene approach with rivallaboratories each championing their favorite candidate.

The ability to study the molecular basis of antimalarial drugresistance has been revolutionized by the sequencing of the P.

falciparum genome. With the genome fully sequenced and

being interrogated with a variety of approaches,47,48

the op-portunity exists to accelerate the process of identifying mo-lecular markers for resistance to the drugs used in ACTs. Thechallenge for genomic scientists, molecular parasitologists,and clinical translational researchers is to link cutting edgegenomic technology with clinical research and public healthsurveillance in a worldwide effort to expeditiously identifymolecular markers of resistance to the new ACTs. We needto be poised to aggressively study reports of resistance, con-firm it with in vitro assays, and bring genomic tools to bear toelucidate the mechanisms and identify candidate molecularmarkers of artemisinin resistance (Figure 1).

Using microsatellites49 and single nucleotide polymor-phisms (SNPs),48 genome-wide maps of quantitative trait loci

associated with resistance to ACTs can be created and used todetect regions of the P. falciparum genome that are underselection by ACT treatment. This approach may provideclues to the molecular bases of resistance to ACTs even be-fore frank clinical resistance arises, but its real promise lies infocused application to parasite isolates corresponding to casesof clinical treatment failure confirmed to represent resistanceby careful, reproducible in vitro testing. Molecular evolution-ary approaches developed to measure selection in other ge-nomes, such as measures of extended haplotype homozygos-ity,50 may also detect areas of the P. falciparum genome thatare under recent positive selection by ACTs. Ultimately, itwill be necessary to resequence large numbers of samples

from initial infections that cleared or failed to clear with ACTtreatment and from post-treatment recrudescent infections,to identify specific polymorphisms associated with resistanceto and selected by the ACTs. Because gene expression mayalso contribute to resistance, DNA microarray analyses ofgene expression will also be needed.51 Such direct applicationof advanced genomic technologies to analyze clinical samples

from field trials in developing countries to develop new toolsfor public health surveillance will be unprecedented. Manytechnical and analytical problems will have to be solved forthis approach to work, requiring close collaboration betweengenomic scientists and researchers in the field, but the ratio-nale for solving these problems could hardly be stronger. Re-sistance to ACTs must be detected and confirmed as soon asit emerges, so that tools to monitor it can be developed, vali-dated, and applied in time to make rational policies and de-sign and deploy drug combinations to deter and contain thisresistance.

PHARMACOKINETICS

In clinical assessments of drug efficacy, apparent drug fail-ures interpreted as clinical resistance may reflect the bioavail-ability and metabolism of the drug rather than parasite resis-tance. For example, the bioavailability of lumefantrine ishighly variable and is influenced by a number of factors, in-cluding the amount of fat consumed at the time of adminis-tration.52 Lumefantrine drug level is the strongest predictorof the outcome of treatment with one of the leading ACTs,lumefantrine-artemether.8 Alternatively, higher drug levelsmay be associated with increased rates of clearance of drug-resistant parasites.29 Although it may be neither practical nornecessary to include measures of drug levels in all drug effi-cacy trials, pharmacokinetic analyses are a critical element instudying and confirming reports of resistance to ACTs (Fig-

ure 1). Differences in pharmacokinetic parameters may ex-plain a lack of correlation between in vitro, molecular, andclinical measures of resistance. In particular, if therapeuticdrug levels are not achieved or maintained for an adequatetime, treatment failures may turn out to have nothing to dowith resistance.

The possibility of pharmacokinetic interactions betweenantimalarial and other drugs must also be considered as ACTsare rolled out, although fortunately, clinically significant in-teractions have not yet been reported for common antima-larial drugs.53 In malaria-endemic regions, concomitant treat-ment is common for multiple illnesses including acute bacte-rial infections, HIV, and tuberculosis. The role of pharmacokinetic interactions in responses to treatment withantimalarial drugs is only beginning to be explored.5456 Ifimportant drugdrug interactions are identified, it will at aminimum be necessary to record drug use in clinical assess-ments of in vivo resistance.

CLINICAL ASSESSMENT OF DRUG EFFICACY

Different methods are used to measure antimalarial drugefficacy for different purposes. For ongoing monitoring of thetherapeutic efficacy of standard first- and second-line drugs,most malaria control programs and malaria researchers haverelied on protocols published by the World Health Organiza-



5/10

tion. More rigorous controlled clinical trials are used to assessthe safety and efficacy of investigational new drugs and drugcombinations. Longitudinal clinical trials represent a rela-tively new tool that is particularly well suited for comparingtreatment regimens with similar high efficacy as measured instandard trials to select those with the greatest health benefitin particular settings.

In vivoprotocols for monitoring therapeutic efficacy. Thestandard in vivo protocols used to monitor antimalarial drug

resistance have been updated repeatedly by expert panelsconvened by the World Health Organization. A 1973 protocolrelied strictly on the presence or absence of parasites at in-tervals after treatment to classify infections as sensitive or asresistant at the RI, RII, or RIII levels.57 A protocol publishedin 1996 did not consider recurrent parasitemia after post-treatment Day 3 to constitute treatment failure unless it wasaccompanied by a documented fever, resulting in many casesof RI and RII resistance being classified as adequate clinicalresponses and not treatment failures.58 A cynic might won-der whether this change reflected recognition that resistanceto chloroquine and SP had resulted in unacceptably low effi-

cacy and that, lacking more efficacious alternatives, the onlyway to improve the situation was to redefine treatment failureto raise point estimates of efficacy. The next iteration, pub-lished in 2003, in turn reflected new optimism based on theintroduction of ACTs in southeast Asia, as well as increasedappreciation of the potential deleterious effects of chronicmalaria infection, and returned to a definition of efficacy,adequate clinical and parasitological response, that re-quired the resolution of parasitemia.59 Present discussions fo-cus on the role of genotyping to distinguish reinfection fromrecrudescence and on whether the duration of the post-treatment parasite-free period should be extended to allowmore sensitive detection of resistance manifested by in-creased rates of late treatment failure.60 The evolution of the

tool used to monitor in vivo efficacy of antimalarial drugs thuscontinues to reflect the state of affairs of malaria treatmentas the first line increasingly shifts to highly efficacious ACTsfor which significant resistance has not yet arisen, the defini-tion of efficacy is changing to permit detection of the mostsubtle, early signs of resistance.

Controlled clinical trials. Controlled clinical trials are usedto assess the safety and efficacy of drugs and drug combina-tions being considered for regulatory registration and to com-pare treatment regimens in preparation for changes in policyrecommendations. These trials are distinguished from stan-dard in vivo therapeutic efficacy protocols by the inclusion ofdifferent study arms comparing treatment regimens, by thecollection of additional safety data, and in recent years,

by required adherence to rigorous Good Clinical Practices(GCP).

The requirement that regulatory clinical trials of malariadrugs be conducted in compliance with GCP has been de-scribed as so burdensome as to impede important clinicalresearch in poor countries.61 Phase I, II, and III clinical trialsof investigational new drugs and vaccines on the path to regu-latory registration can and should be conducted in compliancewith GCP, and this is now regularly done in the poorest Af-rican countries. Several malaria vaccine trials done underGCP in countries like Mozambique, Mali, and Kenya havebeen published,6264 and dozens of GCP-compliant malariadrug trials in developing countries are listed in clinical trial

registries on the internet. Following the strict standards ofGCP and reporting all adverse events in a trial assures thesponsors and the public of the validity and completeness ofthe data, but the level of regulatory oversight and monitoring(which are not the same thing as GCP) should be appropriatefor the purpose of the study: a study of a new drug requiresmore detailed safety review than one of a drug whose safety

profile is already well established. Indiscriminate applicationof the same maximum degree of regulatory oversight andsafety monitoring to all drug studies will curtail importantresearch needed to monitor the implementation of ACTs.

Some studies of drug combinations have been conductedusing a factorial design, permitting assessment of the specificcontribution of individual partner drugs to the outcome oftreatment with combinations of drugs. For example, a study inUganda compared SP alone with SP combined with eitherartesunate or amodiaquine, allowing straightforward determi-nation of the benefit of each of the partner drugs.60 Since theadoption of ACTs in Southeast Asia, such studies havelargely been replaced by comparisons of combinations with-out contemporaneous assessment of individual partner drugs.

Examples include comparisons of mefloquine-artesunate toartemether-lumefantrine in Thailand65 and to dihydroarte-misinin-piperaquine in Vietnam.66 In these trials, it difficult ifnot impossible to determine whether differences in efficacyare caused by differences in the performance of the short-acting artemisinins, the long-acting partner drugs, or both.This limitation of clinical trials of ACTs again highlights thecritical role for in vitro assays in studying possible resistanceand the need to develop reliable molecular markers for resis-tance to ACTs.

Longitudinal trials. In longitudinal trials, participants areenrolled at the time of presentation with acute illness causedby uncomplicated malaria and followed prospectively for aspecified time, usually 1 year. Participants are randomized to

receive a specified treatment regimen at the time of enroll-ment and get the same treatment every time they developclinical malaria during the study period. The most compellingargument for longitudinal trials is that they provide a realisticestimation of the true public health benefit of the study treat-ments, something that studies measuring the efficacy of singletreatment episodes cannot do. People in many malaria-endemic areas develop clinical malaria up to several timeseach year, and treatment policy in these areas should ideallybe based on sustained efficacy over time. The toxicity of drugsmay also accumulate with repeated administration, so assess-ment of safety with dosing intervals that reflect the actual useof the drug are more informative than single-episode studies.Longitudinal trials may also reveal seasonal differences in

how well drugs perform as transmission levels wax and wane.Where transmission is very low and few people experiencemore than one malaria episode per year, the usefulness oflongitudinal trials is limited to their ability to detect very laterecrudescences.

The primary outcome measure of longitudinal trials is thecumulative incidence of clinical episodes of malaria needingtreatment. Secondary outcome measures include adequateclinical and parasitological response at 28 days or at any de-sired interval, for the initial episode as well as for subsequentepisodes, to measure waning efficacy over time; incidence ofnew infections and recrudescent infections; incidence of ane-mia; incidence of severe malaria; and repeated measures of



6/10

safety, which can include laboratory measures of renal andhepatic function, neurologic abnormalities, or other adverseevents. More global measures of health in the different armscan be recorded such as hemoglobin level at the end of thestudy and pediatric growth and development. These measuresfurther reflect the impact of the treatment regimens on over-all health in malaria-endemic regions.

Associations of molecular markers with treatment out-comes and selection for resistance genotypes can be measuredin longitudinal trials, and this study design is particularly wellsuited for pharmacokinetic studies because of the ability tomeasure drug levels at any time before or after treatment.Longitudinal trials also allow for direct comparison of thebenefits of drugs with differing pharmacokinetic characteris-tics. In areas of high transmission, the prophylactic benefitprovided by a long-acting drug must be weighed against theincreased selection of drug-resistant parasites as new infec-tions are acquired during the extended period when sub-therapeutic drug levels are present in the blood.

The first longitudinal trial to study the effects of drugs withvery different pharmacokinetic profiles was a double-blind,

placebo-controlled trial of SP and chlorproguanil-dapsone inKenya and Malawi. These similar antifolate combinations aredistinguished by the much shorter half-life of chlorproguanil-dapsone (CD) and its higher efficacy in Africa, where it hasbeen shown to be highly efficacious even against infectionsresistant to SP.67 The study randomized children to receiveeither CD or SP for all uncomplicated falciparum malariaepisodes over the course of a year and compared the cumu-lative number of treatment episodes and rates of treatmentfailure and anemia.68 In Malawi, CD had an efficacy of 95%compared with only 80% for SP. However, there were nodifferences between the treatment groups in the incidence ofuncomplicated malaria episodes, anemia, or severe malaria,suggesting that the disadvantage of lacking a long prophylac-

tic effect caused by CDs short half-life was offset by its better

efficacy and possibly by less selection for antifolate-resistantparasites. Researchers in Uganda have also used this methodto show the superiority of SP plus amodiaquine over SP plusartesunate by comparing incidence density of malaria treat-ment episodes.60 The best drug combinations to adopt astreatment policy should be chosen based on their efficacy andsafety over time and their ability to decrease the overall bur-den of clinical malaria in the population.

Genotyping to distinguish recrudescence from new infec-

tions. In all types of clinical studies of antimalarial drug effi-cacy, there has been a recent emphasis on distinguishing be-tween new and recrudescent infection after treatment. Thestrategy for differentiating these outcomes has been to geno-

type highly polymorphic regions of the parasite genome todetermine if the parasite strains causing the post-treatmentinfection are the same as (indicating a recrudescent infection)or different from (indicating a new infection) those present atthe time of treatment. The most common method uses poly-merase chain reaction (PCR) to amplify specific regions ofpolymorphic genes encoding the antigens merozoite surfaceprotein-1 (MSP-1), MSP-2, and glutamine-rich protein(GLURP).6971

These methods for strain-specific genotyping have beencriticized. The selected targets are not ideal because they areunder immune selection pressure, which might bias the dis-tribution of polymorphic forms of these antigens in different

populations and age groups.72 Any genotyping strategy has alimit of detection and may fail to detect minority populationspresent initially at levels below this threshold. When these(possibly drug-resistant) minority parasites recrudesce later,they will be incorrectly labeled as a new infection and notincluded in PCR-corrected efficacy estimates. On the otherhand, limited discriminatory power, particularly in less geneti-

cally diverse parasite populations, will lead to the mistakenlabeling of new infections as recrudescent.The use of microsatellites instead of genes encoding poly-

morphic antigens to distinguish recrudescence from reinfec-tion may resolve some of these shortcomings. Microsatellitesare short, variable-length, non-coding sequences of DNA thatare presumably not under natural immune selection.73 Geno-typing using microsatellites can detect higher genetic diversitythan methods using MSP-1, MSP-2, and GLURP.74 Microsat-ellite methods have increased sensitivity for detecting newinfections75,76 and may be less likely to misclassify new infec-tions as recrudescences in polyclonal infections.76

Genotyping to distinguish between recrudescence and newinfections may not be necessary in all studies of antimalarial

efficacy, depending on the purpose of the study. BeforeACTs, all post-treatment recurrent infections were consid-ered to represent resistance. In most cases this was the correctinterpretation, because common drugs including chloroquine,SP, and mefloquine have such long durations of action thatany parasites detected in the 24 weeks after treatment aresurviving exposure to drug concentrations that should inhibitthe growth of sensitive parasites. Thus, for studies intended todetect emerging resistance after treatment with long-actingdrugs, it may be irrelevant whether recurrent infections arecaused by recrudescence or new infection. For studies de-signed to inform treatment policy decisions, the incidence ofmalaria treatment episodes and the risk of complications suchas anemia and severe malaria are the outcomes of interest,

and again differentiating recrudescence from reinfection maynot be relevantdrugs that reduce post-treatment infectionsand life-threatening complications are better than drugs thatdo not, whether these infections are caused by new or oldparasites.

As ACTs began to be compared with each other and tomonotherapies in high-transmission settings, it was felt to beimportant to distinguish between failure to clear primary in-fections and failure to prevent new infections in the post-treatment period. This distinction is important for uniformassessment of efficacy of new drugs across regions with vary-ing intensity of malaria transmission and may be required forregulatory approval. Essentially, the idea is that drugs withshorter durations of action should not be blamed for high

rates of late treatment failure caused by new infections, eventhough from the point of view of the patient experiencing therecurrent infection it may matter little whether the parasitesare old or new. In fact, little is known about the relative risksand benefits of recrudescence versus reinfection after antima-larial treatment, and it is conceivable that new infections carrya higher risk of causing illness than do recrudescences of para-sites to which an immune response has already been devel-oped. If this is the case, in high-transmission areas, a longer-acting drug with leaky initial efficacy might be preferable toa highly efficacious ACT with a short duration of action, eventhough the radical cure provided by the latter is clearly morebeneficial in low-transmission settings where the risk of new



7/10

infections is low. To our knowledge, no published studies

address the question of whether new and recrudescent infec-

tions carry different risks.

Genotyping recurrent infections in longitudinal trials

should make it possible to compare the time to next clinical

episode after post-treatment recrudescences and reinfections.

It would be valuable to do these analyses in trials of various

drugs in populations with different levels of acquired immu-nity. If clinically meaningful differences are shown for post-

treatment recrudescence and reinfection, this would provide a

stronger rationale for including these genotyping analyses in

drug efficacy trials. Until more is known about the relative

risks of different types of post-treatment infections, from a

public health perspective, all post-treatment infections should

be considered treatment failure in recognition that malaria

parasitemia is an undesirable outcome, regardless of the ori-

gin of the infection. Although PCR-corrected efficacy may

allow for easier comparisons between Africa, South America,

and Asia, in our view, it is not be the best way to measure

efficacy in high-transmission settings.

Distinguishing recrudescence from new infection in con-

junction with measurement of molecular markers for resis-tance may help to understand the propagation and spread of

resistance. Treatment failure with recrudescent parasites al-

lows for the persistent transmission of resistant parasites.

New infections may occur either in the face of inhibitory drug

levels, representing resistance, or they may occur when drug

levels are low or absent in the blood, representing parasites

that are fully susceptible to the drugs being used. Describing

the dynamics of selection of resistance and defining the se-

lection windows77 after treatment with different drug com-

binations will aid in the rational design of new combinations

of drugs based on consideration of their pharmacokinetic and

pharmacodynamic properties.

SUMMARY

Fatalism and fear prevented many countries from letting go

of chloroquine, despite its obvious failure. Now, after years of

continued use of failing drugs in many countries, a consensus

has formed to replace them with effective combination thera-

pies. ACTs, with their rapid action and excellent efficacy, are

being embraced by policy makers throughout Africa. With

multilateral support for procurement and infrastructure, the

plan is to distribute ACTs widely. However, the current gen-

eration of ACTs will not maintain their efficacy indefinitely.

Researchers and malaria control workers of all stripes must

join efforts to apply in vitro, molecular, genomic, pharmaco-

kinetic, and clinical methods in coordinated proactive moni-

toring programs to detect the emergence and deter the spread

of resistance to ACTs.

Received January 17, 2007. Accepted for publication April 2, 2007.

Acknowledgments: The authors thank Malcolm Molyneux for help-ful discussions and suggestions to improve the manuscript. CVP issupported by a Distinguished Clinical Scientist Award from the DorisDuke Charitable Foundation.

Authors addresses: Miriam K. Laufer and Christopher V. Plowe,Center for Vaccine Development, University of Maryland School ofMedicine, 685W. Baltimore Street, HSF-1 Room 480, Baltimore, MD21201, Telephone: 410-706-5328, Fax: 410-706-1204, E-mails:[email protected] and [email protected]

.edu. Abdoulaye A. Djimde, Malaria Research and Training Center,Department of Epidemiology of Parasitic Diseases, Faculty of Medi-cine, Pharmacy and Dentistry, University of Bamako, Point G, BP1805, Bamako, Mali, Telephone/Fax: 223-222-1809, E-mail:[email protected].

REFERENCES

1. Rieckmann KH, McNamara JV, Frischer H, Stockert TA, CarsonPE, Powell RD, 1968. Effects of chloroquine, quinine, andcycloguanil upon the maturation of asexual erythrocytic formsof two strains of Plasmodium falciparum in vitro. Am J TropMed Hyg 17: 661671.

2. Nguyen-Dinh P, Payne D, 1980. Pyrimethamine sensitivity inPlasmodium falciparum: determination in vitro by a modified48-hour test. Bull World Health Organ 58: 909912.

3. Plowe CV, 2003. Monitoring antimalarial drug resistance: makingthe most of the tools at hand. J Exp Biol 206: 37453752.

4. Aubouy A, Mayombo J, Keundjian A, Bakary M, Le BJ, DeloronP, 2004. Short report: lack of prediction of amodiaquine effi-cacy in treating Plasmodium falciparum malaria by in vitrotests. Am J Trop Med Hyg 71: 294296.

5. Borrmann S, Binder RK, Adegnika AA, Missinou MA, Issifou S,Ramharter M, Wernsdorfer WH, Kremsner PG, 2002. Reas-sessment of the resistance of Plasmodium falciparum to chlo-

roquine in Gabon: implications for the validity of tests in vitrovs. in vivo. Trans R Soc Trop Med Hyg 96: 660663.

6. Jambou R, Legrand E, Niang M, Khim N, Lim P, Volney B,Ekala MT, Bouchier C, Esterre P, Fandeur T, Mercereau-Puijalon O, 2005. Resistance of Plasmodium falciparum fieldisolates to in-vitro artemether and point mutations of theSERCA-type PfATPase6. Lancet 366: 19601963.

7. Price RN, Cassar C, Brockman A, Duraisingh M, van Vugt M,White NJ, Nosten F, Krishna S, 1999. The pfmdr1 gene isassociated with a multidrug-resistant phenotype in Plasmo-dium falciparum from the western border of Thailand. Anti-microb Agents Chemother 43: 29432949.

8. Price RN, Uhlemann AC, Van VM, Brockman A, Hutagalung R,Nair S, Nash D, Singhasivanon P, Anderson TJ, Krishna S,White NJ, Nosten F, 2006. Molecular and pharmacological de-terminants of the therapeutic response to artemether-lumefantrine in multidrug-resistant Plasmodium falciparum

malaria. Clin Infect Dis 42: 15701577.9. Bzik DJ, Li WB, Horii T, Inselburg J, 1987. Molecular cloning

and sequence analysis of the Plasmodium falciparum dihydro-folate reductase-thymidylate synthase gene. Proc Natl AcadSci USA 84: 83608364.

10. Brooks DR, Wang P, Read M, Watkins WM, Sims PF, Hyde JE,1994. Sequence variation of the hydroxymethyldihydropterinpyrophosphokinase: dihydropteroate synthase gene in lines ofthe human malaria parasite, Plasmodium falciparum, with dif-fering resistance to sulfadoxine. Eur J Biochem 224: 397405.

11. Foote SJ, Thompson JK, Cowman AF, Kemp DJ, 1989. Ampli-fication of the multidrug resistance gene in some chloroquine-resistant isolates of P. falciparum. Cell 57: 921930.

12. Wellems TE, Walker-Jonah A, Panton LJ, 1991. Genetic map-ping of the chloroquine-resistance locus on Plasmodium falci-parum chromosome 7. Proc Natl Acad Sci USA 88: 33823386.

13. Peterson DS, Walliker D, Wellems TE, 1988. Evidence that apoint mutation in dihydrofolate reductase- thymidylate syn-thase confers resistance to pyrimethamine in falciparum ma-laria. Proc Natl Acad Sci USA 85: 91149118.

14. Fidock DA, Nomura T, Talley AK, Cooper RA, Dzekunov SM,Ferdig MT, Ursos LM, Bir Singh SA, Naude B, Deitsch KW,Su X, Wootton JC, Roepe PD, Wellems TE, 2000. Mutationsin the P. falciparum digestive vacuole transmembrane proteinPfCRT and evidence for their role in chloroquine resistance.Mol Cell 6: 861871.

15. Foote SJ, Kyle DE, Martin RK, Oduola AM, Forsyth K, KempDJ, Cowman AF, 1990. Several alleles of the multidrug-resistance gene are closely linked to chloroquine resistance inPlasmodium falciparum. Nature 345: 255258.

16. Su X, Kirkman LA, Fujioka H, Wellems TE, 1997. Complexpolymorphisms in an 330 kDa protein are linked to chloro-



8/10

quine-resistant P. falciparum in Southeast Asia and Africa.Cell 91: 593603.

17. Cowman AF, Galatis D, Thompson JK, 1994. Selection for me-floquine resistance in Plasmodium falciparum is linked to am-plification of the pfmdr1 gene and cross-resistance to halofan-trine and quinine. Proc Natl Acad Sci USA 91: 11431147.

18. Wu Y, Kirkman LA, Wellems TE, 1996. Transformation ofPlas-modium falciparum malaria parasites by homologous integra-tion of plasmids that confer resistance to pyrimethamine. Proc

Natl Acad Sci USA 93: 11301134.19. Sidhu AB, Verdier-Pinard D, Fidock DA, 2002. Chloroquine re-

sistance in Plasmodium falciparum malaria parasites conferredby pfcrt mutations. Science 298: 210213.

20. Triglia T, Wang P, Sims PG, Hyde JE, Cowman AF, 1998. Allelicexchange at the endogenous genomic locus in Plasmodiumfalciparum proves the role of dihyropteroate synthase in sul-fadoxine-resistant malaria. EMBO J 17: 38073815.

21. Reed MB, Saliba KJ, Caruana SR, Kirk K, Cowman AF, 2000.Pgh1 modulates sensitivity and resistance to multiple antima-larials in Plasmodium falciparum. Nature 403: 906909.

22. CorteseJF, Plowe CV, 1998. Antifolate resistance due to newandknown Plasmodium falciparum dihydrofolate reductase mu-tants expressed in yeast. Mol Biochem Parasitol 94: 205214.

23. Plowe CV, Cortese JF, Djimd A, Nwanyanwu OC, WatkinsWM, Winstanley PA, Estrada-Franco JG, Mollinedo RE,Avila JC, Cespedes JL, Carter D, Doumbo OK, 1997. Muta-

tions in Plasmodium falciparum dihydrofolate reductase anddihydropteroate synthase and epidemiologic patterns of py-rimethamine-sulfadoxine use and resistance. J Infect Dis 176:15901596.

24. Wang P, Lee C-S, Bayoumi R, Djimd A, Doumbo O, SwedbergG, Dao LD, Mshinda H, Tanner M, Watkins WM, Sims PFG,Hyde JE, 1997. Resistance to antifolates in Plasmodium falci-parum monitored by sequence analysis of dihydropteroate syn-thase and dihydrofolate reductase alleles in a large number offield samples of diverse origins. Mol Biochem Parasitol 89:161177.

25. Djimd A, Doumbo OK, Cortese JF, Kayentao K, Doumbo S,Diourte Y, Dicko A, Su XZ, Nomura T, Fidock DA, WellemsTE, Plowe CV, Coulibaly D, 2001. A molecular marker forchloroquine-resistant falciparum malaria. N Engl J Med 344:257263.

26. Kublin JG, Dzinjalamala FK, Kamwendo DD, Malkin EM, Cor-

tese JF, Martino LM, Mukadam RA, Rogerson SJ, LescanoAG, Molyneux ME, Winstanley PA, Chimpeni P, Taylor TE,Plowe CV, 2002. Molecular markers for failure of sulfadoxine-pyrimethamine and chlorproguanil-dapsone treatment of Plas-modium falciparum malaria. J Infect Dis 185: 380388.

27. Djimd A, Doumbo OK, Steketee RW, Plowe CV, 2001. Appli-cation of a molecular marker for surveillance of chloroquine-resistant falciparum malaria. Lancet 358: 890891.

28. Djimd AA, Doumbo OK, Traore O, Guindo AB, Kayentao K,Diourte Y, Niare-Doumbo S, Coulibaly D, Kone AK, CissokoY, Tekete M, Fofana B, Dicko A, Diallo DA, Wellems TE,Kwiatkowski D, Plowe CV, 2003. Clearance of drug-resistantparasites as a model for protective immunity in Plasmodiumfalciparum malaria. Am J Trop Med Hyg 69: 558563.

29. Dzinjalamala FK, Macheso A, Kublin JG, Taylor TE, Barnes KI,Molyneux ME, Plowe CV, Smith PJ, 2005. Blood folate con-centrations and in vivo sulfadoxine-pyrimethamine failure inMalawian children with uncomplicated Plasmodium falci-parum malaria. Am J Trop Med Hyg 72: 267272.

30. Wellems TE, Plowe CV, 2001. Chloroquine-resistant malaria. JInfect Dis 184: 770776.

31. Mockenhaupt FP, Ehrhardt S, Eggelte TA, Gana-Nsiire P, Stoll-berg K, Mathieu A, Markert M, Otchwemah RN, Bienzle U,2005. Chloroquine-treatment failure in northern Ghana: rolesof pfcrt T76 and pfmdr1 Y86. Ann Trop Med Parasitol 99:723732.

32. Mayengue PI, Ndounga M, Davy MM, Tandou N, Ntoumi F,2005. In vivo chloroquine resistance and prevalence of thePfcrt codon 76 mutation in Plasmodium falciparum isolatesfrom the Republic of Congo. Acta Trop 95: 219225.

33. Khalil IF, Alifrangis M, Tarimo DS, Staalso T, Satti GMH, The-ander TG, Ronn AM, Bygbjerg IC, 2005. The roles of the pfcrt

76T and pfmdr1 86Y mutations, immunity and the initial levelof parasitaemia, in predicting the outcome of chloroquinetreatment in two areas with different transmission intensities.Ann Trop Med Parasitol 99: 441448.

34. Tinto H, Sanou B, Dujardin JC, Ouedragogo JB, Van OvermeirC, Erhart A, Van Marck E, Guiguemde TR, DAlessandro U,2005. Short report: usefulness of the Plasmodium falciparumchloroquine resistance transporter T76 genotype failure indexfor the estimation of in vivo chloroquine resistance in Burkina

Faso. Am J Trop Med Hyg 73: 171173.35. Kyabayinze D, Cattamanchi A, Kamya MR, Rosenthal PJ,

Dorsey G, 2003. Validation of a simplified method for usingmolecular markers to predict sulfadoxine-pyrimethaminetreatment failure in African children with falciparum malaria.Am J Trop Med Hyg 69: 247252.

36. Mugittu K, Ndejembi M, Malisa A, Lemnge M, Premji Z, MwitaA, Nkya W, Kataraihya J, Abdulla S, Beck HP, Mshinda H,2004. Therapeutic efficacy of sulfadoxine-pyrimethamine andprevalence of resistance markers in tanzania prior to revisionof malaria treatment policy: Plasmodium falciparum dihydro-folate reductase and dihydropteroate synthase mutations inmonitoring in vivo resistance. Am J Trop Med Hyg 71: 696702.

37. Happi CT, Gbotosho GO, Folarin OA, Akinboye DO, YusufBO, Ebong OO, Sowunmi A, Kyle DE, Milhous W, Wirth DF,Oduola AM, 2005. Polymorphisms in Plasmodium falciparum

dhfr and dhps genes and age related in vivo sulfadoxine-pyrimethamine resistance in malaria-infected patients from Ni-geria. Acta Trop 95: 183193.

38. Francis D, Nsobya SL, Talisuna A, Yeka A, Kamya MR, Mache-kano R, Dokomajilar C, Rosenthal PJ, Dorsey G, 2006. Geo-graphic differences in antimalarial drug efficacy in Uganda areexplained by differences in endemicity and not by known mo-lecular markers of drug resistance. J Infect Dis 193: 978986.

39. White N, 2004. Sulfadoxine-pyrimethamine for uncomplicatedfalciparum malaria: sulfadoxine-pyrimethamine is not workingin Malawi. BMJ 328: 1259.

40. Happi CT, Gbotosho GO, Folarin OA, Akinboye DO, YusufBO, Ebong OO, Sowunmi A, Kyle DE, Milhous W, Wirth DF,Oduola AM, 2005. Polymorphisms in Plasmodium falciparumdhfr and dhps genes and age related in vivo sulfadoxine-pyrimethamine resistance in malaria-infected patients from Ni-geria. Acta Trop 95: 183193.

41. Schneider AG, Premji Z, Felger I, Smith T, Abdulla S, Beck HP,Mshinda H, 2002. A point mutation in codon 76 of pfcrt of P.falciparum is positively selected for by Chloroquine treatmentin Tanzania. Infect Genet Evol 1: 183189.

42. Djimd AA, Dolo A, Ouattara A, Diakite S, Plowe CV, DoumboOK, 2004. Molecular diagnosis of resistance to antimalarialdrugs during epidemics and in war zones. J Infect Dis 190:853855.

43. Kublin JG, Cortese JF, Njunju EM, Ra GM, Wirima JJ, KazembePN, Djimde AA, Kouriba B, Taylor TE, Plowe CV, 2003. Re-emergence of chloroquine-sensitive Plasmodium falciparummalaria after cessation of chloroquine use in Malawi. J InfectDis 187: 18701875.

44. Laufer MK, Thesing PC, Eddington ND, Masonga R, Dzinjala-mala FK, Takala SL, Taylor TE, Plowe CV, 2006. Return ofchloroquine antimalarial efficacy in Malawi. N Engl J Med 355:19591966.

45. Price RN, Uhlemann AC, Brockman A, McGready R, Ashley E,Phaipun L, Patel R, Laing K, Looareesuwan S, White NJ,Nosten F, Krishna S, 2004. Mefloquine resistance in Plasmo-dium falciparum and increased pfmdr1 gene copy number.Lancet 364: 438447.

46. Anderson TJ, Nair S, Qin H, Singlam S, Brockman A, Paiphun L,Nosten F, 2005. Are transporter genes other than the chloro-quine resistance locus (pfcrt) and multidrug resistance gene(pfmdr) associated with antimalarial drug resistance? Antimi-crob Agents Chemother 49: 21802188.

47. Ferdig MT, Cooper RA, Mu J, Deng B, Joy DA, Su XZ, WellemsTE, 2004. Dissecting the loci of low-level quinine resistance inmalaria parasites. Mol Microbiol 52: 985997.

48. Volkman SK, Sabeti PC, Decaprio D, Neafsey DE, Schaffner SF,Milner DA Jr, Daily JP, Sarr O, Ndiaye D, Ndir O, Mboup S,



9/10

Duraisingh MT, Lukens A, Derr A, Stange-Thomann N, Wag-goner S, Onofrio R, Ziaugra L, Mauceli E, Gnerre S, Jaffe DB,Zainoun J, Wiegand RC, Birren BW, Hartl DL, Galagan JE,Lander ES, Wirth DF, 2007. A genome-wide map of diversityin Plasmodium falciparum. Nat Genet 39: 113119.

49. Mu J, Ferdig MT, Feng X, Joy DA, Duan J, Furuya T, Subrama-nian G, Aravind L, Cooper RA, Wootton JC, Xiong M, Su XZ,2003. Multiple transporters associated with malaria parasiteresponses to chloroquine and quinine. Mol Microbiol 49: 977

989.50. Sabeti PC, Schaffner SF, Fry B, Lohmueller J, Varilly P, Sham-

ovsky O, Palma A, Mikkelsen TS, Altshuler D, Lander ES,2006. Positive natural selection in the human lineage. Science312: 16141620.

51. Rathod PK, Ganesan K, Hayward RE, Bozdech Z, DeRisi JL,2002. DNA microarrays for malaria. Trends Parasitol 18: 3945.

52. White NJ, Van VM, Ezzet F, 1999. Clinical pharmacokinetics andpharmacodynamics and pharmacodynamics of artemether-lumefantrine. Clin Pharmacokinet 37: 105125.

53. Giao PT, de Vries PJ, 2001. Pharmacokinetic interactions of an-timalarial agents. Clin Pharmacokinet 40: 343373.

54. Khaliq Y, Gallicano K, Tisdale C, Carignan G, Cooper C, Mc-Carthy A, 2001. Pharmacokinetic interaction between meflo-quine and ritonavir in healthy volunteers. Br J Clin Pharmacol51: 591600.

55. Ridtitid W, Wongnawa M, Mahatthanatrakul W, Chaipol P, Sun-bhanich M, 2000. Effect of rifampin on plasma concentrationsof mefloquine in healthy volunteers. J Pharm Pharmacol 52:12651269.

56. Schippers EF, Hugen PW, Den HJ, Burger DM, HoetelmansRM, Visser LG, Kroon FP, 2000. No drug-drug interactionbetween nelfinavir or indinavir and mefloquine in HIV-1-infected patients. AIDS 14: 27942795.

57. WHO, 1973. Chemotherapy of Malaria and Resistance to Antima-larials. Geneva: World Health Organization.

58. WHO, 1996. Assessment of Therapeutic Efficacy of AntimalarialDrugs for Uncomplicated Falciparum Malaria in Areas withIntense Transmission. Geneva: World Health Organization,Division of Control of Tropical Diseases.

59. WHO, 2003. Assessment and Monitoring of Antimalarial Efficacyfor the Treatment of Uuncomplicate d Falciparum Malaria.Geneva: World Health Organization.

60. Dorsey G, Njama D, Kamya MR, Cattamanchi A, Kyabayinze D,Staedke SG, Gasasira A, Rosenthal PJ, 2002. Sulfadoxine/pyrimethamine alone or with amodiaquine or artesunate fortreatment of uncomplicated malaria: a longitudinal ran-domised trial. Lancet 360: 20312038.

61. White NJ, 2006. Editorial: clinical trials in tropical diseases: apolitically incorrect view. Trop Med Int Health 11: 14831484.

62. Alonso PL, Sacarlal J, Aponte JJ, Leach A, Macete E, Milman J,Mandomando I, Spiessens B, Guinovart C, Espasa M, BassatQ, Aide P, Ofori-Anyinam O, Navia MM, Corachan S, Ceup-pens M, Dubois MC, Demoitie MA, Dubovsky F, MenendezC, Tornieporth N, Ballou WR, Thompson R, Cohen J, 2004.Efficacy of the RTS,S/AS02A vaccine against Plasmodium fal-ciparum infection and disease in young African children: ran-domised controlled trial. Lancet 364: 14111420.

63. Thera MA, Doumbo OK, Coulibaly D, Diallo DA, Sagara I,Dicko A, Diemert DJ, Heppner DG, Stewart VA, Angov E,Soisson L, Leach A, Tucker K, Lyke KE, Plowe CV, 2006.Safety and allele-specific immunogenicity of a malaria vaccinein Malian adults: results of a phase I randomized trial. PLoSClin Trials 1: e34.

64. Withers MR, McKinney D, Ogutu BR, Waitumbi JN, Milman JB,Apollo OJ, Allen OG, Tucker K, Soisson LA, Diggs C, LeachA, Wittes J, Dubovsky F, Stewart VA, Remich SA, Cohen J,Ballou WR, Holland CA, Lyon JA, Angov E, Stoute JA,Martin SK, Heppner DG, 2006. Safety and reactogenicityof an MSP-1 malaria vaccine candidate: a randomized phase Ibdose-escalation trial in Kenyan children. PLoS Clin Trials 1:e32.

65. Hutagalung R, Paiphun L, Ashley EA, McGready R, BrockmanA, Thwai KL, Singhasivanon P, Jelinek T, White NJ, NostenFH, 2005. A randomized trial of artemether-lumefantrine ver-

sus mefloquine-artesunate for the treatment of uncomplicatedmulti-drug resistant Plasmodium falciparum on the westernborder of Thailand. Malar J 4: 46.

66. Tran TH, Dolecek C, Pham PM, Nguyen TD, Nguyen TT, Le HT,Dong TH, Tran TT, Stepniewska K, White NJ, Farrar J, 2004.Dihydroartemisinin-piperaquine against multidrug-resistantPlasmodium falciparum malaria in Vietnam: randomised clini-cal trial. Lancet 363: 1822.

67. Mutabingwa T, Nzila A, Mberu E, Nduati E, Winstanley P, Hills

E, Watkins W, 2001. Chlorproguanil-dapsone for treatment ofdrug-resistant falciparum malaria in Tanzania. Lancet 358:12181223.

68. Sulo J, Chimpeni P, Hatcher J, Kublin JG, Plowe CV, MolyneuxME, Marsh K, Taylor TE, Watkins WM, Winstanley PA, 2002.Chlorproguanil-dapsone versus sulfadoxine-pyrimethaminefor sequential episodes of uncomplicated falciparum malaria inKenya and Malawi: a randomised clinical trial. Lancet 360:11361143.

69. Farnert A, Arez AP, Babiker HA, Beck HP, Benito A, BjorkmanA, Bruce MC, Conway DJ, Day KP, Henning L, Mercereau-Puijalon O, Ranford-Cartwright LC, Rubio JM, Snounou G,Walliker D, Zwetyenga J, do Rosario VE, 2001. Genotypingof Plasmodium falciparum infections by PCR: a comparativemulticentre study. Trans R Soc Trop Med Hyg 95: 225232.

70. Viriyakosol S, Siripoon N, Petcharapirat C, Petcharapirat P, JarraW, Brown KN, Snounou G, 1995. Genotyping of Plasmodium

falciparum isolates by the polymerase chain reaction and po-tential uses in epidemiological studies. Bull World Health Or-gan 73: 8595.

71. Cattamanchi A, Kyabayinze D, Hubbard A, Rosenthal PJ,Dorsey G, 2003. Distinguishing recrudescence from reinfectionin a longitudinal antimalarial drug efficacy study: comparisonof results based on genotyping of msp-1, msp-2, and glurp. AmJ Trop Med Hyg 68: 133139.

72. Escalante AA, Lal AA, Ayala FJ, 1998. Genetic polymorphismand natural selection in the malaria parasite Plasmodium fal-ciparum. Genetics 149: 189202.

73. Su XZ, Wellems TE, 1996. Toward a high-resolution Plasmodiumfalciparum linkage map: polymorphic markers from hundredsof simple sequence repeats. Genomics 33: 430444.

74. Mwangi JM, Omar SA, Ranford-Cartwright LC, 2006. Compari-son of microsatellite and antigen-coding loci for differentiatingrecrudescing Plasmodium falciparum infections from reinfec-

tions in Kenya. Int J Parasitol 36: 329336.75. Nyachieo A, Van OC, Laurent T, Dujardin JC, DAlessandro U,

2005. Plasmodium falciparum genotyping by microsatellites asa method to distinguish between recrudescent and new infec-tions. Am J Trop Med Hyg 73: 210213.

76. Greenhouse B, Myrick A, Dokomajilar C, Woo JM, Carlson EJ,Rosenthal PJ, Dorsey G, 2006. Validation of microsatellitemarkers for use in genotyping polyclonal Plasmodium falci-parum infections. Am J Trop Med Hyg 75: 836842.

77. Watkins WM, Mosobo M, 1993. Treatment ofPlasmodium falci-parum malaria with pyrimethamine- sulfadoxine: selectivepressure for resistance is a function of long elimination half-life. Trans R Soc Trop Med Hyg 87: 7578.

78. Mockenhaupt FP, Ehrhardt S, Eggelte TA, Gana-Nsiire P, Stoll-berg K, Mathieu A, Markert M, Otchwemah RN, Bienzle U,2005. Chloroquine-treatment failure in northern Ghana: rolesof pfcrt T76 and pfmdr1 Y86. Ann Trop Med Parasitol 99:723732.

79. Mayengue PI, Ndounga M, Davy MM, Tandou N, Ntoumi F,2005. In vivo chloroquine resistance and prevalence of thePfcrt codon 76 mutation in Plasmodium falciparum isolatesfrom the Republic of Congo. Acta Trop 95: 219225.

80. Khalil IF, Alifrangis M, Tarimo DS, Staalso T, Satti GMH, The-ander TG, Ronn AM, Bygbjerg IC, 2005. The roles of the pfcrt76T and pfmdr1 86Y mutations, immunity and the initial levelof parasitaemia, in predicting the outcome of chloroquinetreatment in two areas with different transmission intensities.Ann Trop Med Parasitol 99: 441448.

81. Tinto H, Sanou B, Dujardin JC, Ouedragogo JB, Van OvermeirC, Erhart A, Van Marck E, Guiguemd TR, DAlessandro U,2005. Short report: Usefulness of the Plasmodium falciparumchloroquine resistance transporter T76 genotype failure index



10/10

for the estimation of in vivo chloroquine resistance in BurkinaFaso. Am J Trop Med Hyg 73: 171173.

82. Mugittu K, Ndejembi M, Malisa A, Lemnge M, Premji Z, MwitaA, Nkya W, Kataraihya J, Abdulla S, Beck HP, Mshinda H,2004. Therapeutic efficacy of sulfadoxine-pyrimethamine andprevalence of resistance markers in tanzania prior to revisionof malaria treatment policy: Plasmodium falciparum dihydro-folate reductase and dihydropteroate synthase mutations in

monitoring in vivo resistance. Am J Trop Med Hyg 71: 696702.

83. Happi CT, Gbotosho GO, Folarin OA, Akinboye DO, YusufBO, Ebong OO, Sowunmi A, Kyle DE, Milhous W, Wirth DF,Oduola AM, 2005. Polymorphisms in Plasmodium falciparumdhfr and dhps genes and age related in vivo sulfadoxine-pyrimethamine resistance in malaria-infected patients from Ni-geria. Acta Trop 95: 183193.


160.fulljurnal

Documents