• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of aacPermissionsJournals.ASM.orgJournalAAC ArticleJournal InfoAuthorsReviewers
Antimicrob Agents Chemother. Jan 2008; 52(1): 288–298.
Published online Oct 22, 2007. doi:  10.1128/AAC.00263-07
PMCID: PMC2223885

In Vitro Monitoring of Plasmodium falciparum Drug Resistance in French Guiana: a Synopsis of Continuous Assessment from 1994 to 2005[down-pointing small open triangle]


Implemented as one arm of the malaria control program in French Guiana in the early 1990s, our laboratory has since established in vitro profiles for parasite drug susceptibility to a panel of eight antimalarials for more than 1,000 Plasmodium falciparum isolates from infected patients. The quinine-doxycycline combination was introduced in 1995 as the first-line drug treatment against uncomplicated P. falciparum malaria, replacing chloroquine, and the first-line drug combination was changed to the artemether-lumefantrine combination in 2002. Resistance to chloroquine declined 5 years after it was dropped in 1995 as the first-line drug, but unlike similar situations in Africa, there was a rapid halt to this decline. Doxycycline susceptibility substantially decreased from 2002 to 2005, suggesting parasite selection under quinine-doxycycline drug pressure. Susceptibility to mefloquine decreased from 1997 onward. Throughout the period from 1994 to 2005, most isolates were sensitive in vitro to quinine, amodiaquine, and atovaquone. Susceptibility to amodiaquine was strongly correlated with that to chloroquine and to a lesser extent with that to mefloquine and halofantrine. Susceptibilities to mefloquine and to halofantrine were also strongly correlated. There were two alerts issued for in vitro artemether resistance in the period from 2002 to 2003 and again in 2005, both of which could be associated with the presence of an S769N polymorphism in the sarcoplasmic/endoplasmic reticulum calcium ATPase (SERCA)-type P. falciparum ATPase6 (PfATPase6) gene. Analysis of susceptibility to lumefantrine, conducted for the first time in 2005, indicates an alarming rate of elevated 50% inhibitory concentrations. In vitro monitoring of parasite drug susceptibility should be pursued to further document the consequences of specific drug policies on the local parasite population and, in particular, to establish profiles of susceptibility to individual components of drug combinations to provide early warning signs of emerging parasite resistance.

The Guyana Shield currently has the highest malaria rates in South America and is the only geographical area in the Americas where Plasmodium falciparum infection is diagnosed more frequently than P. vivax infection. In French Guiana, P. falciparum accounts for 60 to 70% of the 3,000 to 5,000 malaria cases reported every year (3). Malaria occurs in isolated geographical foci situated along the rivers in the Amazonian forest, and the population has little specific immunity. Malarial control involves spraying insecticides over the coastal region, where approximately 80% of the population resides, and using bed nets, an evidence-based drug policy, and deploying health services in remote endemic areas to provide prompt treatment by public primary health centers and/or private practitioners.

As in the neighboring countries of the Amazon basin (36), P. falciparum is multidrug resistant, with failures documented for chloroquine (18), the amodiaquine/sulfadoxine-pyrimethamine combination (22), halofantrine (9, 19), chloroquine-proguanil (19), and even quinine (6). In the last decade, the drug policy in French Guiana has been based on both an in vivo assessment of therapeutic efficacy and longitudinal monitoring of in vitro drug susceptibility. In 1995, the quinine-doxycycline combination was recommended as the first-line treatment. It was replaced by the artemether-lumefantrine combination in 2002.

Monitoring drug resistance is particularly important in this region, where illegal gold-mining activities in the malaria endemicity areas are associated with erratic consumption of antimalarials and frequently with self-medication, unclear compliance to recommended regimens, and the formulation and use of illegally imported molecules of unknown quality, which contribute to the selection of drug-resistant parasites in the region (11, 29). However, surveillance of the therapeutic efficacy of the recommended regimens is complicated by the difficulty of monitoring treatment efficacy in the scattered endemicity foci. In vitro susceptibility testing by a dedicated reference laboratory was established in the early 1990s to carry out systematic in vitro drug susceptibility testing in French Guiana. In vitro tests assess parasite susceptibility to the individual components of drug combinations and, as such, provide early evidence of emerging resistance before it becomes clinically apparent. A recent illustration was the first description by our group of parasites from French Guiana with a markedly reduced in vitro susceptibility to artemether (11, 17).

Our reference laboratory has monitored, on a longitudinal basis, the in vitro drug susceptibility of isolates collected across the endemicity areas over the last 11 years. We have assessed parasite susceptibility to a panel of eight antimalarials, including the recommended drugs and molecules that are not yet in use in the area or that have been withdrawn. We report here a synopsis of the temporal variations in in vitro susceptibility profiles over that time period. We looked for temporal trends for each individual molecule, for cross-resistance between drugs from the same chemical class or between drugs with similar modes of action, and for correlations between susceptibility to different drug classes. We discuss these data in light of accumulated drug pressure and changes in drug policies in French Guiana and in the Guyana Shield region.


Background information on the study area, clinical failure rates, and drug policy.

French Guiana (administratively a French Overseas Department) has nowadays an overall population of approximately 150,000 inhabitants. The main malaria endemicity areas are shown in Fig. Fig.1.1. Malaria occurs in isolated, remote geographical foci located mainly along the Maroni River to the west and along the Oyapock River to the east, which serve as natural frontiers with Suriname and Brazil, respectively. Annual parasitic indices increase from the delta region (72 and 64‰ for the lower Maroni and Oyapock rivers, respectively) to the upper-river regions (371 and 223‰ for the upper Maroni and Oyapock rivers, respectively), where illegal mining activities occur (8).

FIG. 1.
Map of French Guiana. The main endemicity areas are highlighted in red. Solid zones indicate areas of high-incidence transmission, and hatched zones are low-incidence transmission areas.

There was a 22% rate of level RIII chloroquine resistance in 1989 (9, 18). The first sulfadoxine-pyrimethamine-resistant parasites were identified in 1988 (22). The combination was rapidly discontinued thereafter. There was an 11% clinical failure rate for the amodiaquine/sulfadoxine-pyrimethamine combination in 1986 (22). Halofantrine treatment failures, reported in the Maroni River area since 1988 (19; C. Venturin, I. Jeanne, J. B. Duchemin, G. Desmarchelier, V. Pavec, S. Laventure, E. Nuiaouet, J. Sankale-Suzanon, and J. L. Sarthou, presentation at the 3rd Infectious Diseases Medical Conference in French Guiana, Cayenne, 1988) reached 14% in 2000 (9). Two cases of level RIII in vivo resistance to quinine have been identified (6). From the 1980s onward, quinine was always used in association with tetracycline (similar to protocols used in Brazil and Suriname) and not with sulfadoxine-pyrimethamine, as used in Guyana. Prophylaxis with chloroquine-proguanil remained clinically effective in French Guiana until 1998, when seven falciparum malaria attacks, including two severe forms, occurred on a military base where chloroquine-proguanil prophylaxis was given (19).

A “consensus conference” was organized at 5- to 7-year intervals in order to review in vivo and in vitro drug susceptibility in the area and to adjust the treatment policy accordingly. Table Table11 summarizes the treatments recommended for mild malaria, severe malaria, and prophylaxis from 1990 to 2006. From 1990 to 2005, the first-line and second-line treatment drugs were chloroquine and halofantrine or mefloquine, respectively. The 1995 consensus conference introduced the quinine-doxycycline association as the first-line treatment. The update at the end of 2002 replaced this combination with atovaquone-proguanil (mainly for prophylaxis) and artemether-lumefantrine (for therapy) (Table (Table1).1). The artemether-lumefantrine combination (Riamet; Novartis) is currently the recommended first-line drug for the treatment of uncomplicated falciparum malaria in adults in French Guiana, but it has not yet been approved for use in children.

Antimalarial regimens recommended for French Guiana by the 1990, 1995, and 2002 consensus meetings

The antimalarials used are prescribed by doctors when patients seek treatment (from private doctors, public health centers, or hospitals). The medical staff is urged to follow the regional guidelines of the Ministry of Health. However, an unknown proportion of malaria cases occur in the illegal immigrant population in the area, mainly gold miners operating in the jungle. These “illegal” cases are treated mostly by self-medication and are prevented by self-prescribed prophylaxis with cheap drugs from neighboring Brazil and Suriname.

Isolates: geographic origin and patients.

All isolates studied were collected from malaria patients referred to our reference laboratory by private practitioners, health centers, and/or hospitals, as recommended by the health authorities. A 2- to 5-ml blood sample was collected from each patient by venipuncture in EDTA tubes, before they received treatment administration/prescription. The tube was maintained at 4°C during shipment to the regional reference center in Cayenne (CNRCP), where it was immediately processed by using in vitro susceptibility assays.

Out of the 2,855 isolates originating from the peripheral health centers located along the Maroni (60% of the annual sampling) and the Oyapock rivers and from the main hospitals on the coast, mainly in Cayenne (30%) and Saint-Laurent (9%), 2,535 isolates were identified as a monospecific P. falciparum infection.

In vitro susceptibility assays.

The inclusion criteria for in vitro drug assays were (i) a microscopically diagnosed (thin smear) P. falciparum isolate-monospecific infection (i.e., the absence of concomitant P. vivax and P. malariae isolates), confirmed in some cases by an immunochromatographic dipstick test or PCR, and (ii) a specimen collected less than 3 days before laboratory processing and before treatment administration. The samples were processed on receipt at the CNRCP and tested for susceptibility to eight antimalarial drugs, using a modified [3H]hypoxanthine-uptake in vitro assay (14) and a computer-piloted (GX110 model Dell) multidetector scintillation counter (Microbeta Trilux; Wallac Oy, Turku, Finland).

Tests were carried out over a 48-h culture period, using the [3H]hypoxanthine incorporation microtest. Each isolate was tested once in triplicate against serial twofold drug dilutions over the concentration ranges of 3,750 to 7 nM chloroquine (diphosphate; Sigma catalog no. C6628), 3,680 to 7 nM quinine (sulfate; Sigma catalog no. Q1878), 793 to 1.54 nM mefloquine (Hoffman-La Roche Inc), 63.1 to 0.123 nM halofantrine (GlaxoSmithKline Inc.), 840 to 1.64 μM amodiaquine (dihydrochloride dihydrate; Sigma catalog no. A2799), 50 to 0.097 μM cycloguanil (Astra Zeneca), 1039 to 2.03 μM doxycycline (hydrochloride; Sigma catalog no. D9891), 98.5 to 0.19 nM artemether (Novartis Pharma Inc), and 32 to 0.00625 nM atovaquone (GlaxoSmithKline Inc). The various antimalarial molecules were tested from the beginning, except for amodiaquine (for which testing began in 1995), artemether (in 1997), atovaquone (in 2001), and lumefantrine (in 2005).

Control susceptible lines were included for quality control. The 3D7 clone was used from 1999 onward. In 2002, we also included HB3 and W2, and 7G8 was added to the panel of control lines in 2003. The four control lines were kindly provided by J. Le Bras (Centre National de Référence pour la chimiosensibilité du paludisme, Paris, France). P. falciparum reference lines were run in parallel assays for each batch of antimalarial used. Since 1999, 3D7 has been tested in 13 assays; and HB3, W2, and 7G8 were tested in 12, 7, and 5 independent assays, respectively. The 50% inhibitory concentration (IC50) was calculated using a probit/logit regression of the percentage of growth inhibition for each drug, and the geometric mean IC50s were compared by Student's t test. To comply with previous reports from the area and with the commonly used interpretation of susceptibility tests (5, 14, 15, 20, 26, 28, 29), we have used a cutoff value to classify isolates as susceptible, intermediate, or resistant (Table (Table2).2). The in vitro levels of chloroquine resistance versus sensitivity were determined at the laboratory level by analyzing isolates from clinical cases with documented therapeutic efficacy from the area. For the antimalarials other than chloroquine, there is no threshold established for these grounds. It is not possible to assign such a value for drugs not yet in use in the area (or no longer in use). The temporal variation of in vitro-resistant prevalence was followed in the various areas. Generally speaking, the Maroni River samples (from two sentinel sites, Maripasoula and Saint Laurent du Maroni), representing isolates from a region with a high prevalence of multidrug resistance, were compared with those from the rest of the department.

Cutoff used by the reference laboratory for interpretation of in vitro susceptibility assays

DNA extraction and microsatellite genotyping.

The artemether-resistant sample isolated in 2005 was cultured for 21 days with the absence of artemether. DNA was prepared from the ex vivo sample and, after 21 days of culture, as described previously (25), was genotyped for four microsatellites, DHFR, PFE14F, C3M35, and C4M79, as described previously (16). The full-length P. falciparum ATPase6 (PfATPase6) gene was amplified for both samples and was sequenced as described previously (11).

Statistical analysis.

The global trends were calculated using a Kruskal-Wallis test for the last 6 years. In addition, the year-to-year evolution was analyzed using Student's t test with a level of significance at a P value of <0.05. In vitro correlations between the mean IC50s of various drugs were tentatively searched by means of standard correlation analysis (Pearson's test, P < 0.05) on a significant sample of isolates (n = 171 and n = 511, collected during the periods from 1996 to 1999 and 2000 to 2005, respectively). All analyses were carried out by using specialized software (SAS version 9.1; Sas Institute Inc., Cary, NC).



Chloroquine was the first-line drug for treating P. falciparum malaria until 1995, when its use was discontinued. From 1994 to 1999, close to 100% of the isolates were classified as in vitro chloroquine resistant (Fig. (Fig.2A).2A). The geometric mean IC50 was high (≥400 nM). From 2000 on, there were abrupt reductions in both the proportion of chloroquine-resistant isolates and the geometric mean IC50s, which dropped from 371 nM in 1999 to 96 nM in 2000 (t test, P < 0.0001). However, there was no further decline in subsequent years. Geometric mean IC50s increased moderately from 98.1 nM in 2001 to 107 nM in 2002, 120.2 nM in 2003, 124.5 nM in 2004, and 100.1 nM in 2005 (Kruskal-Wallis test, P = 0.01).

FIG. 2.
Temporal evolution of the in vitro susceptibility to chloroquine (A), quinine (B), and amodiaquine (C) in French Guiana from 1994 to 2005. (A) In vitro susceptibility to chloroquine was obtained for 16, 22, 48, 68, 46, 44, 68, 42, 144, 159, 70, and 62 ...


Amodiaquinized salt (0.43%) was distributed to the whole population from 1967 to 1978 (and in neighboring Suriname from 1965 to 1973) (27) and sporadically on an individual level at later dates (8). Amodiaquine susceptibility was tested routinely from 1996 onward. Approximately 94% of 482 isolates tested from the period from 1996 to 2005 were sensitive to amodiaquine (Fig. (Fig.2B).2B). There has been, however, an increase in mean IC50s in recent years, rising from 11.3 nM in 2000 to 28.2 nM in 2004 and to 22.3 nM in 2005 (Kruskal-Wallis test, P < 0.0001).


Quinine has been used for the treatment of severe or complicated cases in combination with tetracycline since the 1980s. The quinine-doxycycline association was introduced as the first-line drug for treating P. falciparum malaria in 1995. An elevated IC50, classified as quinine resistance, was reported for 17% of isolates from 1983 to 1987 (5). This classification accounted for 9 to 12% of the isolates analyzed from 1995 to 1997 and was no longer observed for 487 isolates tested from 2000 to 2005, with the exception of two geographically localized alerts issued over the last 3 years (Fig. (Fig.2C).2C). All during the period surveyed, the yearly mean IC50 remained in the sensitive zone (196.8 nM for 1998 to 2005), albeit with a significant increase from 1995 to 2001, followed by a significant decrease from 2001 to 2005 (Kruskal-Wallis test, P < 0.0001).


As shown in Table Table1,1, mefloquine was used for prophylaxis and as a second-line treatment drug throughout the years 1990 to 2002. In the 1980s and until 1995, all isolates tested were susceptible to mefloquine. In 1996, 2% of the isolates were classified as in vitro mefloquine resistant. The prevalence of mefloquine resistance subsequently fluctuated between 0 and 29% (Fig. (Fig.3A).3A). From 1997 onward, isolates with “intermediate” levels accounted for 11 to 23% of the isolates tested. The mean IC50 increased during this time period, from 9.4 nM in 1998 to 20.9 nM in 2002 and to 13.5 nM in 2005 (Kruskal-Wallis test, P = 0.039). It is interesting to note that the temporal variations were reflected at both the global and the local (Maroni River area) levels.

FIG. 3.
Temporal evolution of in vitro susceptibility of French Guiana isolates to mefloquine (A) and halofantrine (B) from 1994 to 2005. (A) In vitro susceptibility to mefloquine was obtained for 16, 22, 43, 66, 45, 51, 66, 36, 142, 158, 70 and 58 isolates collected ...


From 1994 to 2005, most isolates were classified as sensitive, and a minor fraction (6 to 9%) presented intermediate levels of resistance. This proportion increased to 21% in the next 2 years and fluctuated thereafter between 14 and 34% (Fig. (Fig.3B).3B). Isolates classified as in vitro halofantrine resistant were observed in 1996 and in subsequent years, but their prevalence fluctuated markedly. There was a peak of 35% prevalence of resistance for isolates from 1997 and a second peak of 66% prevalence of resistance in isolates from 2000. It decreased to less than 3% in 2004. The mean IC50 increased from 3 nM in 1997 to 14.1 nM in 2000 and decreased significantly thereafter (Kruskal-Wallis test, P < 0.0001) to 3.8 and 4.1 nM in 2004 and 2005, respectively.


Doxycycline has been investigated since mid-1996, using standard 48-h tests, despite the fact that it is a delayed parasite death inducer. With this limitation in mind, we nevertheless reviewed the data available across the decade. As there was no global consensus for the threshold of susceptibility, a tentative cutoff level was set at 9.6 μM by Sarthou and Reynes (29), corresponding to the 95th percentile of a representative local sample. The prevalence of isolates with an IC50 above that threshold rose from 15 to 25% in the period from 1996 to 2001, to 51% in 2002, to 61.5% in 2003, and to more than 67% in 2005 (Fig. (Fig.4).4). Mean IC50s increased from 9.6 μM in 1996 to 1999 to 7.4 μM in 2000, 12.2 μM in 2002 to 2003, and 13.1 μM in 2005 (Kruskal-Wallis test, P < 0.0001).

FIG. 4.
Temporal evolution of in vitro susceptibility and mean IC50 values for doxycycline of French Guiana isolates assessed from 1996 to 2005. In vitro susceptibility to doxycycline was obtained for 34, 32, 45, 50, 69, 39, 134, 147, 65, and 55 isolates collected ...


Resistance to proguanil, based on in vitro testing of its major metabolite cycloguanil, had a very high prevalence (98% in 1997 and 100% in all subsequent years) and was at an elevated level (mean IC50s, >4, 100 nM) (data not shown). In vitro testing was stopped in 2000.

Artemisinin derivatives.

Artemether has been incorporated in our panel of antimalarials for in vitro testing since 1997. The great majority of 634 isolates assayed (95.5% to 98%) had a low IC50 value. The mean IC50 varied between 2.2 and 3.9 nM, except in 2002 when it was 8.4 nM (Fig. (Fig.5A),5A), due to the presence of isolates showing markedly decreased in vitro sensitivity.

FIG. 5.
Temporal evolution of in vitro susceptibility of French Guiana isolates to artemether from 1997 to 2005. In vitro susceptibility to artemether was obtained for 31, 42, 44, 65, 39, 136, 153, 66, and 58 isolates collected in 1997, 1998, 1999, 2000, 2001, ...

Figure Figure5A5A shows the yearly distribution of isolates grouped into two classes using the 12 nM threshold, which represents the lower border of the highest 5th percentile of IC50 distribution among African isolates, as proposed by Pradines et al. (26). Figure Figure5B5B shows the distribution into two classes using the 30 nM threshold, which has been associated with the presence of a S769N polymorphism in the putative target enzyme sarcoplasmic/endoplasmic reticulum calcium ATPase (SERCA)-type PfATPase6 gene (11). Nine isolates with an IC50 for artemether of ≥30 nM were observed in 2002, and one isolate was observed again in 2005 (17). In all these cases, the observation of isolates with elevated IC50s for artemether is unlikely attributable to experimental culture artifacts, since unrelated patient isolates assayed on the same day exhibited low IC50 values for artemether. Molecular typing of the isolate collected in 2005, which presented an IC50 of 127 nM for artemether, using microsatellite markers indicated that the ex vivo isolate contained two clonal types (Table (Table3)3) and two PfATPase6 alleles, an S769N mutant allele and a wild-type allele. After 3 weeks of in vitro cultivation under standard culture conditions, the mutant allele was no longer detected, and the IC50 for artemether had dropped to 5 nM (16), suggesting a poor fitness of the mutant in vitro.

Susceptibility test and genotype of artemether resistance isolate at day 0 and day 21a

The distribution of IC50s for artemether over the period from 2002 to 2005 is shown in Fig. Fig.5C.5C. The geometric mean IC50 ± standard deviation during 2002 to 2005 was 2.03 ± 13 (n = 413 isolates). The interquartile range was 0.98 to 3.74. Elevated IC50s for artemether were observed for a very small fraction of all isolates.


The investigation of lumefantrine susceptibility in vitro was initiated in 2005. The distribution of IC50s for lumefantrine, shown in Fig. Fig.6,6, spanned a 3 log concentration range. This limited sample showed a geometric mean IC50 ± standard deviation of 156.8 ± 534.6 nM. Of 36 isolates tested, 14 (38.8%), 4 (11.1%), and 18 (50%) isolates presented IC50 values of <100, 100 to 150, and >150 nM, respectively, for this molecule.

FIG. 6.
Frequency distribution of IC50 values for lumefantrine in French Guiana for 41 isolates collected during 2005.


Atovaquone has been used in the area in association with proguanil (Malarone). The 368 in vitro tests performed before its implementation in the area (which started in 2002) showed an excellent level of sensitivity (with a mean IC50 for atovaquone of 0.55 nM) during the period from 1998 to 2004 (data not shown). The first case of in vitro resistance to atovaquone was clinically and parasitologically documented in 2005 with parasites collected from a patient for whom treatment with second-line atovaquone-proguanil failed (15).

In vitro cross-reactivity.

In vitro correlations between the mean IC50s for drugs with similar modes of action or for those of the same chemical classes were tentatively sought from two groups of samples collected from 1996 to 1999 and from 2000 to 2005. A lack of correlation between chloroquine and quinine, mefloquine, or doxycycline was noted. In contrast, positive correlations were observed between amodiaquine and chloroquine (r = 0.75; P = 0.02), amodiaquine and halofantrine (r = 0.51; P <0.001), amodiaquine and mefloquine (r = 0.48; P = 0.004), and to a lesser extent amodiaquine and quinine (r = 0.25; P = 0.03).

There were highly significant correlations between mefloquine and halofantrine in isolates from 1996 to 1999 (r = 0.72; P = 0.01) and those from 2000 to 2005 (r = 0.63; P < 0.0001). Interestingly, similar correlations had been noticed during local surveys organized in the mid-Maroni River area (Papaichton, r = 0.74, P < 0.01, on 27 in vitro tests analyzed in 2000 [9]) and in the high-Maroni River area (Maripasoula, r = 0.73, P < 0.0001, on 29 tests in 1995 [C. Venturin et al., 3rd Infectious Diseases Medical Conference, 1988]). For lumefantrine, a significant correlation was noticed with halofantrine (r = 0.78; P < 0.0001), but there was no association with mefloquine.


Our data highlight significant temporal changes of in vitro sensitivity to most molecules included in the established systematic screening scheme over the 12 years surveyed. Chloroquine and quinine susceptibility increased significantly during this period. Susceptibility to doxycycline, mefloquine, and amodiaquine decreased, and there were alerts issued for halofantrine and artemether resistance. These changes can be interpreted in light of the drug pressure exerted on the parasites in French Guiana and in the surrounding countries. Drug pressure in the area has been quite high from very early on. Amodiaquinized salt (0.43%) was distributed to the whole population from 1967 to 1978 (and in neighboring Suriname from 1965 to 1973) (30) and distributed sporadically on an individual basis after that (8). Chloroquine, quinine, tetracycline, doxycycline, mefloquine, and halofantrine have been used during the last decade, and more recently, atovaquone, proguanil, artemether, and lumefantrine have been introduced.

The use of chloroquine for treating P. falciparum malaria was abandoned in French Guiana in 1995 because of a documented loss of therapeutic efficacy at a local level and across the Amazonian region. During the following years, all French Guiana isolates tested were chloroquine resistant, but there was a marked increase in chloroquine sensitivity in 2000 and thereafter. The decrease in chloroquine resistance was altogether abrupt and partial. The abruptness is difficult to interpret and may reflect a geographical sample bias, as some endemic foci may have been overrepresented during certain years. As shown in Fig. Fig.1,1, the endemic foci in French Guiana have a patchy distribution, are isolated, and are separated by large inhabited primary forests and/or river banks. This situation may be responsible for microgeographic parasite heterogeneity, including drug resistance, as well as heterogeneity of transmission intensity and hence for the spread of resistance. Whereas this may create distribution distortions during some periods, this proviso no longer holds for longer term trends, such as the shift from 100% chloroquine resistance in 1995 to 1999 to 50 to 60% resistance in 2000 to 2005. Similarly partial decreases in chloroquine-resistant P. falciparum isolates after chloroquine cessation were observed in China (21, 34) and in Cambodia (7, 20). These decreases differ from the consequence of chloroquine cessation in Malawi that was followed by a decrease in the Pfcrt 76T marker from 85% in 1993 to 13% in 2000 and 0% in 2001 (13, 23). The maintenance of substantial levels of chloroquine resistance in non-African settings after the discontinuation of chloroquine treatment against P. falciparum malaria reflects the fact that unlike the situation in Africa, not all chloroquine pressure has been removed from the region. In particular, chloroquine has been used continuously and is still being used for the treatment of P. vivax malaria, which accounts for a substantial fraction of malaria cases and is most probably used also against some P. falciparum-P. vivax mixed isolate infections.

However, French Guiana treatment differs from that in Southeast Asian countries and China in the replacement of chloroquine. As in many countries in this region, in French Guiana chloroquine was replaced as the first-line treatment drug for P. falciparum malaria by the quinine-doxycycline combination. Whether this has contributed to preventing further decreases in chloroquine resistance is unclear. The decrease in chloroquine resistance is usually interpreted as due to its fitness cost in the absence of chloroquine pressure (33). Molecular analysis using transgenic parasites mutated at the Pfcrt or the Pfmdr1 locus showed an inverse relationship of mutations governing in vitro susceptibility to chloroquine on the one hand and to quinine, mefloquine, and halofantrine on the other hand in some genetic backgrounds (12, 13, 27, 31, 32). No such association was observed for the panel of isolates studied from French Guiana. In vitro susceptibility to chloroquine was associated with no other response than amodiaquine, consistent with the amodiaquine and chloroquine cross-resistance reported in many settings. Interestingly, amodiaquine susceptibility was associated with halofantrine, mefloquine, and to a lesser extent quinine susceptibility. Molecular studies of Pfcrt, Pfmdr1, and other putative chloroquine and quinine transporters are under way to understand the molecular basis of such associations and to document the consequences of changes in first-line treatment drugs for French Guiana parasites.

In French Guiana, as in Brazil and Suriname, quinine had been used for the treatment of severe or complicated cases of malaria in combination with tetracycline since the 1980s. The quinine-doxycycline combination was introduced for uncomplicated P. falciparum malaria cases in 1995. This combination has been used extensively in South America during the last 10 years and is still used because of irregular supplies of artemether-lumefantrine and nonprescription of artemether-lumefantrine in pediatric malaria cases. Our data indicate that the use of the quinine-doxycycline combination has selected parasites with reduced in vitro susceptibility to doxycycline. This may be underestimated by the use of 48-h in vitro assays since doxycycline is a slow-acting antimalarial. A more accurate assessment of doxycycline inhibition would have required cultivation for 72 to 96 h. Plans are under way to establish these assays in the laboratory. The reduction observed for in vitro susceptibility to doxycycline may have had limited clinical impact so far, in particular because of the continuing efficacy of quinine. However, further extensive use of doxycycline or of any combination drug containing doxycycline is not desirable, since it will further increase resistance to doxycycline and, in turn, to the combination in quinine monotherapy, thereby increasing the selection pressure for quinine resistance.

Artemisinin derivatives have been introduced in French Guiana in combination with lumefantrine (Riamet), but as discussed above, the treatment is not yet approved for treating pediatric malaria in French Guiana or for treating pregnant women and is not widely available. Furthermore, its elevated price is an incentive for users to cross the border and purchase artemether-lumefantrine (Coartem) and/or other artemisinin-based combination therapy drugs in neighboring countries. In the years 2002 and 2005, we observed isolates with markedly increased IC50s for artemether and a PfATPase6 S769N mutation. The mean IC50 for artemether (± 95% confidence interval [CI]) with the PfATPase6 S769N mutant isolates was 82.6 nM (± 95% CI, 23.3), i.e., 25-fold above the upper border of the 95% CI value for artemether for all parasites studied in French Guiana (geometric mean = 2.03 nM; 95% CI = 1.26). There are two points of concern here. First, these isolates originated from areas quite distant from each other across French Guiana, namely from the town of Cacao, and from settings along the Maroni River, which are separated by inhabited primary forest (Fig. (Fig.1).1). Epidemiological investigations of these patients revealed that drugs labeled as containing an artemisinin derivative had been illegally imported from Southeast Asia into Cacao. The exact drug composition and concentration could not be determined by the expert chemistry laboratory at the Institut de Veille Sanitaire. Whether this finding indicates counterfeit, underdosed drugs, which dramatically plague the market nowadays (24), is unclear. Investigations focus on areas along the Maroni River with illegal gold mining activities and the importation of antimalarial drugs from Suriname and Brazil, in particular dihydroartemisinin-piperaquine (Artekin [Tonghe Pharmaceuticals]) and piperaquine-dihydroartemisinin-trimethoprim (Artecom [Tonghe Pharmaceuticals]). The quality of these drugs has not been assessed, and the possibility of counterfeits remains open.

A second point of concern is the observation of an additional case of in vitro resistance to artemether from the Maroni River area, which was associated with exactly the same mutation in the target gene 3 years after the first cases. Unfortunately, this isolate could not be maintained in culture and could not be cloned out under artemether pressure in vitro after freeze-thawing, an experience also reported by another group (10). This suggests that prolonged cultivation can select for a drug-sensitive parasite, possibly because current culture conditions may be suboptimal for certain parasite strains. Since many field isolates are from mixed infections, this raises a warning that caution should be used when cultivating field isolates for extended periods of time. The data indicate that the emergence of resistance to artemether in the region is indeed a threat. Artemisinin derivatives are also active on gametocytes, thereby reducing the probability of resistance spreading. Nevertheless, continued and increased surveillance is needed to prevent the establishment of isolates with additional mutations. The S769N polymorphism in the PfATPase6 gene may represent a first step in a process of accumulating mutations that may eventually result in a more stable artemether-resistant haplotype that may subsequently spread, as reported for resistance to chloroquine (7, 35), pyrimethamine, or sulfadoxine (4). In vitro resistance is not synonymous with clinical failure, but it is on the critical path to eventual clinical failure.

Of further concern is the observation of an elevated IC50 for lumefantrine. Lumefantrine is an aryl alcohol, and there was evidence for cross-resistance of lumefantrine with halofantrine in French Guiana. We have included lumefantrine in the panel of antimalarials to be systematically assessed in current and future work. Again, the significance of this observation in terms of therapeutic efficacy is unclear, but it is a second warning for the long-term sustainability of the combination in this region. An assessment of the therapeutic efficacy of artemether-lumefantrine is needed in the various endemicity areas across French Guiana.

This synopsis identified temporal trends pointing to changes in parasite in vitro susceptibility following changes in drug policy. Longitudinal assessment by the same laboratory with the same technology over a 12-year period provides a basis for deriving reasonably reliable temporal trends. The temporal dynamics of the IC50s have been somewhat variable from year to year, possibly reflecting some occasional geographic sample biases, but consistent trends are evidenced when a longer time frame is considered. In vitro resistance to multiple antimalarials is not specific to French Guiana. Multiple resistance has been reported in the neighboring countries of Suriname, Brazil, and Guyana. Our results highlight the usefulness of longitudinal in vitro susceptibility monitoring to document the consequences of a drug policy on resistance selection. In vitro assays provide invaluable information on emerging resistance against individual components of drug combinations such as quinine-doxycycline or artemether-lumefantrine. Surveillance capacities need to be strengthened urgently in the region in order to anticipate in good time the emergence of resistance and to prevent the spread of resistant lines across the region.


We thank the colleagues who implemented and/or worked for the reference laboratory (T. Fandeur, J.B. Duchemin, J.M. Reynes, S. Laventure, and J.L. Sarthou). We are grateful to Cayenne's central hospital team, particularly B. Carme (Parasitology Unit) and F. Djossou (Health Center coordination), for organizing the collection of specimens.

This study was supported by the Prix Louis D. of the French Academy of Sciences and by the European Commission (contract QLK2-CT20021-1503-ResMalChip).


[down-pointing small open triangle]Published ahead of print on 22 October 2007.


1. Brasseur, P., P. Bitsindou, R. S. Moyou, T. A. Eggelte, G. Samba, L. Penchenier, and P. Druilhe. 1993. Fast emergence of Plasmodium falciparum resistance to halofantrine. Lancet 341:901-902. [PubMed]
2. Brasseur, P., J. Kouamouo, R. S. Moyou, and P. Druilhe. 1990. Emergence of mefloquine-resistant malaria in Africa without drug pressure. Lancet 336:59. [PubMed]
3. Carme, B. 2005. Substantial increase of malaria in inland areas of eastern French Guiana. Trop. Med. Int. Health 10:154-159. [PubMed]
4. Cortese, J. F., A. Caraballo, C. E. Contreras, and C. V. Plowe. 2002. Origin and dissemination of Plasmodium falciparum drug-resistance mutations in South America. J. Infect. Dis. 186:999-1006. [PubMed]
5. Dedet, J. P., P. Germanetto, G. Cordoliani, O. Bonnevie, and J. Le Bras. 1988. In vitro activity of various antimalarials (chloroquine, amodiaquine, quinine and mefloquine) against 32 isolates of Plasmodium falciparum in French Guiana. Bull. Soc. Pathol. Exot. Filiales 81:88-93. (In French.) [PubMed]
6. Demar, M., and B. Carme. 2004. Plasmodium falciparum in vivo resistance to quinine: description of two RIII responses in French Guiana. Am. J. Trop. Med. Hyg. 70:125-127. [PubMed]
7. Durrand, V., A. Berry, R. Sem, P. Glaziou, J. Beaudou, and T. Fandeur. 2004. Variations in the sequence and expression of the Plasmodium falciparum chloroquine resistance transporter (Pfcrt) and their relationship to chloroquine resistance in vitro. Mol. Biochem. Parasitol. 136:273-285. [PubMed]
8. Esterre, P., G. Cordoliani, P. Germanetto, and Y. Robin. 1990. Epidemiology of malaria in French Guiana. Bull. Soc. Pathol. Exot. 83:193-205. (In French.) [PubMed]
9. Gruenfeld, J. 2001. Plasmodium falciparum chemioresistance in a village from French Guiana: comparison of in vivo tests on halofantrine-treated patients. M.D. thesis. Paris 7, Paris. (In French.)
10. Hunt, P., A. Afonso, A. Creasey, R. Culleton, A. B. Sidhu, J. Logan, S. G. Valderramos, I. McNae, S. Cheesman, V. D. Rosario, R. Carter, D. A. Fidock, and P. Cravo. 2007. Gene encoding a deubiquitinating enzyme is mutated in artesunate- and chloroquine-resistant rodent malaria parasites. Mol. Microbiol. 65:27-40. [PMC free article] [PubMed]
11. Jambou, R., E. Legrand, M. Niang, N. Khim, P. Lim, B. Volney, M. T. Ekala, C. Bouchier, P. Esterre, T. Fandeur, and O. Mercereau-Puijalon. 2005. Resistance of Plasmodium falciparum field isolates to in-vitro artemether and point mutations of the SERCA-type PfATPase6. Lancet 366:1960-1963. [PubMed]
12. Kublin, J. G., J. F. Cortese, E. M. Njunju, R. A. Mukadam, J. J. Wirima, P. N. Kazembe, A. A. Djimde, B. Kouriba, T. E. Taylor, and C. V. Plowe. 2003. Reemergence of chloroquine-sensitive Plasmodium falciparum malaria after cessation of chloroquine use in Malawi. J. Infect. Dis. 187:1870-1875. [PubMed]
13. Lakshmanan, V., P. G. Bray, D. Verdier-Pinard, D. J. Johnson, P. Horrocks, R. A. Muhle, G. E. Alakpa, R. H. Hughes, S. A. Ward, D. J. Krogstad, A. B. Sidhu, and D. A. Fidock. 2005. A critical role for PfCRT K76T in Plasmodium falciparum verapamil-reversible chloroquine resistance. EMBO J. 24:2294-2305. [PMC free article] [PubMed]
14. Le Bras, J., and P. Deloron. 1983. In vitro study of drug sensitivity of Plasmodium falciparum: evaluation of a new semi-micro test. Am. J. Trop. Med. Hyg. 32:447-451. [PubMed]
15. Legrand, E., M. Demar, B. Volney, M. T. Ekala, M. Quinternet, C. Bouchier, T. Fandeur, C. Rogier, B. Carme, O. Mercereau-Puijalon, and P. Esterre. 2007. A first case of Plasmodium falciparum atovaquone-proguanil treatment failure in French Guiana. Antimicrob. Agents Chemother. 51:2280-2281. [PMC free article] [PubMed]
16. Legrand, E., B. Volney, A. Lavergne, C. Tournegros, L. Florent, D. Accrombessi, M. Guillotte, O. Mercereau-Puijalon, and P. Esterre. 2005. Molecular analysis of two local falciparum malaria outbreaks on the French Guiana coast confirms the msp1 B-K1/varD genotype association with severe malaria. Malar. J. 4:26. [PMC free article] [PubMed]
17. Legrand, E., B. Volney, J. B. Meynard, P. Esterre, and O. Mercereau-Puijalon. 2007. Resistance to dihydroartemisinin em. Infec. Dis. 13:808-809. [PMC free article] [PubMed]
18. Lepelletier, L., F. Gay, M. Nadire-Galliot, J. P. Poman, S. Bellony, J. Claustre, B. M. Traore, and J. Mouchet. 1989. Malaria in Guiana. I. General status of the endemic. Bull. Soc. Pathol. Exot. Filiales 82:385-392. (In French.) [PubMed]
19. Lial, J. P. 1999. Health assessments of a military company stationed on the Maroni River in French Guiana. Med. Trop. 59:95-98. (In French.) [PubMed]
20. Lim, P., S. Chy, F. Ariey, S. Incardona, P. Chim, R. Sem, M. B. Denis, S. Hewitt, S. Hoyer, D. Socheat, O. Merecreau-Puijalon, and T. Fandeur. 2003. pfcrt polymorphism and chloroquine resistance in Plasmodium falciparum strains isolated in Cambodia. Antimicrob. Agents Chemother. 47:87-94. [PMC free article] [PubMed]
21. Liu, D. Q., R. J. Liu, D. X. Ren, D. Q. Gao, C. Y. Zhang, C. P. Qui, X. Z. Cai, C. F. Ling, A. H. Song, and X. Tang. 1995. Changes in the resistance of Plasmodium falciparum to chloroquine in Hainan, China. Bull. W. H. O. 73:483-486. [PMC free article] [PubMed]
22. Maisonneuve, H., F. Joly, M. John, G. Carles, and J. F. Rossignol. 1988. Efficacy of halofantrine in Plasmodium falciparum or Plasmodium vivax malaria in a resistance area (French Guiana). Presse Med. 17:99-102. (In French.) [PubMed]
23. Mita, T., A. Kaneko, J. K. Lum, B. Bwijo, M. Takechi, I. L. Zungu, T. Tsukahara, K. Tanabe, T. Kobayakawa, and A. Bjorkman. 2003. Recovery of chloroquine sensitivity and low prevalence of the Plasmodium falciparum chloroquine resistance transporter gene mutation K76T following the discontinuance of chloroquine use in Malawi. Am. J. Trop. Med. Hyg. 68:413-415. [PubMed]
24. Newton, P. N., R. McGready, F. Fernandez, M. D. Green, M. Sunjio, C. Bruneton, S. Phanouvong, P. Millet, C. J. Whitty, A. O. Talisuna, S. Proux, E. M. Christophel, G. Malenga, P. Singhasivanon, K. Bojang, H. Kaur, K. Palmer, N. P. Day, B. M. Greenwood, F. Nosten, and N. J. White. 2006. Manslaughter by fake artesunate in Asia-will Africa be next? PLoS Med. 3:e197. [PMC free article] [PubMed]
25. Noranate, N., R. Durand, A. Tall, L. Marrama, A. Spiegel, C. Sokhna, B. Pradines, S. Cojean, M. Guillotte, E. Bischoff, M.-T. Ekala, C. Bouchier, T. Fandeur, F. Ariey, J. Patarapotikul, J. Le Bras, J. F. Trape, C. Rogier, and O. Mercereau-Puijalon. 2007. Rapid dissemination of Plasmodium falciparum drug resistance despite strictly controlled antimalarial use. PLoS ONE 1:e139. [PMC free article] [PubMed]
26. Pradines, B., C. Rogier, T. Fusai, A. Tall, J. F. Trape, and J. C. Doury. 1998. In vitro activity of artemether against African isolates (Senegal) of Plasmodium falciparum in comparison with standard antimalarial drugs. Am. J. Trop. Med. Hyg. 58:354-357. [PubMed]
27. Reed, M. B., K. J. Saliba, S. R. Caruana, K. Kirk, and A. F. Cowman. 2000. Pgh1 modulates sensitivity and resistance to multiple antimalarials in Plasmodium falciparum. Nature 403:906-909. [PubMed]
28. Reynes, J. M., J. Fargette, P. Gaborit, and S. Yarde. 1997. In vitro responses of Plasmodium falciparum isolates to five antimalaria drugs in French Guiana during 1994 and 1995. Mem. Inst. Oswaldo Cruz 92:251-252. [PubMed]
29. Sarthou, J. L., and J. M. Reynes. 1996. Malaria in French Guiana: chemoresistance of Plasmodium falciparum isolates. Rev. Epidemiol. Sante Publique 44:42. (In French.)
30. Schaapveld, K. 1984. Integration on antimalaria activities into basic health services in Suriname. ICG Printing, Dordrecht, The Netherlands.
31. Sidhu, A. B., S. G. Valderramos, and D. A. Fidock. 2005. pfmdr1 mutations contribute to quinine resistance and enhance mefloquine and artemisinin sensitivity in Plasmodium falciparum. Mol. Microbiol. 57:913-926. [PubMed]
32. Sidhu, A. B., D. Verdier-Pinard, and D. A. Fidock. 2002. Chloroquine resistance in Plasmodium falciparum malaria parasites conferred by pfcrt mutations. Science 298:210-213. [PMC free article] [PubMed]
33. Walliker, D., P. Hunt, and H. Babiker. 2005. Fitness of drug-resistant malaria parasites. Acta Trop. 94:251-259. [PubMed]
34. Wang, X., J. Mu, G. Li, P. Chen, X. Guo, L. Fu, L. Chen, X. Su, and T. E. Wellems. 2005. Decreased prevalence of the Plasmodium falciparum chloroquine resistance transporter 76T marker associated with cessation of chloroquine use against P. falciparum malaria in Hainan, People's Republic of China. Am. J. Trop. Med. Hyg. 72:410-414. [PubMed]
35. Wootton, J. C., X. Feng, M. T. Ferdig, R. A. Cooper, J. Mu, D. I. Baruch, A. J. Magill, and X. Z. Su. 2002. Genetic diversity and chloroquine selective sweeps in Plasmodium falciparum. Nature 418:320-323. [PubMed]
36. Zalis, M. G., L. Pang, M. S. Silveira, W. K. Milhous, and D. F. Wirth. 1998. Characterization of Plasmodium falciparum isolated from the Amazon region of Brazil: evidence for quinine resistance. Am. J. Trop. Med. Hyg. 58:630-637. [PubMed]

Articles from Antimicrobial Agents and Chemotherapy are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...