Scandinavian Journal of Infectious Diseases, 2010; 42: 22–32

REVIEW ARTICLE

History and current status of Plasmodium falciparum antimalarial drug resistance in

VALÉRIE ANDRIANTSOANIRINA1 , DIDIER MÉNARD1, LUCIANO TUSEO2 & RÉMY DURAND3

From the 1Malaria Research Unit, Institut Pasteur de Madagascar, , Madagascar, 2World Health Organization Offi ce of Madagascar and La Réunion, Antananarivo, Madagascar, and 3Laboratoire de Parasitologie-Mycologie, Hôpital Avicenne, AP-HP, Bobigny,

Abstract Malaria remains a major health problem in Madagascar. Over past decades, the burden of malarial disease has fl uctuated over time, partly in line with the successes and failures of antimalarial policy. In the 1950s and 1960s, a sharp decline in malaria transmission was observed in the central highlands due to indoor spraying with DDT and to the massive use of chloroquine by the population. Following this, the discontinuation of the ‘nivaquinization’ policy was followed by devastat- ing outbreaks in the central highlands in the 1980s. Currently, the rate of in vitro chloroquine-resistant Plasmodium falci- parum isolates does not exceed 5%. This fi gure appears disconnected from the high level of clinical treatment failure (near 40%). pfcrt mutant isolates are found in less than 1% of isolates on the Island. Conversely, pfmdr 1 mutant isolates are found in more than 60% of isolates and may be responsible for the bulk of resistance to chloroquine in Madagascar. Other anti- malarials remain generally effective in Madagascar. Recent clinical and in vitro data support the complete effi cacy of the combination artesunate–amodiaquine in Madagascar. As such, this artemisinin combination therapy should play a central role in the control and possible elimination of P. falciparum malaria in Madagascar. For personal use only.

Introduction chloroquine as the fi rst-line treatment, whereas the sulfadoxine–pyrimethamine combination (SP) is given Malaria remains a major health problem in Mada- intermittently as preventive treatment in pregnant gascar [ 1,2]. Together with indoor spraying of DDT women. Following the global trend, the Malagasy (and the use of long lasting impregnated bed nets), government claims as its objective the elimination of the use of antimalarial drugs is central to the control malaria as a public health problem [10]. The aim of of malaria [3]. Malaria morbidity and mortality have this review is to present the history and current fl uctuated over time, partly in line with the successes status of P. falciparum antimalarial drug resistance Scand J Infect Dis Downloaded from informahealthcare.com by Institut Pasteur on 09/11/12 and failures of antimalarial policy. In the 1950s and in Madagascar and to consider its impact on malaria 1960s, a sharp decline in malaria transmission was control programmes. observed, mostly in the central highlands, due to indoor spraying with dichlorodiphenyltrichloroethane (DDT) and to the massive use of chloroquine by the population Context and malaria situation in Madagascar [4,5]. These results were obtained because of the well-organized healthcare system, which in particular Madagascar appears as an isolated fragment of the allowed the implementation of parasitological diag- ancient continent ‘Gondwana’, which was separated nosis. Subsequently, the discontinuation of the ‘niva- from Africa 65–100 million y ago [ 11]. Human set- quinization’ policy was followed by devastating tlement in Madagascar occurred relatively recently, outbreaks in the central highland areas in the 1980s with the population originating from overseas. The [6–8]. Meanwhile, Plasmodium falciparum clinical fi rst arrivals are thought to have settled during the resistance emerged and spread [9]. Today, artemisinin- fi rst centuries of the Christian era, reaching the central based combination therapy (ACT) has replaced highlands only in the 5th –7th centuries. Parasites and

Correspondence: R. Durand, Laboratoire de Parasitologie-Mycologie, AP-HP, Hôpital Avicenne, 125 rue de Stalingrad, 93009 Bobigny Cedex, France. E-mail: [email protected]

(Received 25 June 2009; accepted 24 August 2009) ISSN 0036-5548 print/ISSN 1651-1980 online © 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS) DOI: 10.3109/00365540903289670 Antimalarial drug resistance in Madagascar 23

2500000 12

10 2000000

8

1500000

6

1000000

4

500000 2

0 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

number of reported malaria cases incidence (%)

Figure 1. Malaria incidence and number of cases in the y 1999–2008 in Madagascar.

For personal use only. vectors were probably introduced to the Malagasy reported 1.01 million cases in 2006 [15]. Reported coasts by the Indonesian, Bantou and Arab migrations cases since 1999 are presented in Figure 1. More than of the 1st to 14th centuries [12]. Hence, all human 90% of malaria cases are due to P. falciparum. The malaria species except Plasmodium knowlesi (P. fal- prevalence of P. vivax has been estimated to be 6.3% ciparum, Plasmodium vivax, Plasmodium ovale, and and of P. malariae to be 1.1% [16,17]. The estimated Plasmodium malariae) are found in Madagascar. population of the whole island soared from 2.5 million Currently, the main vectors (Anopheles gambiae, in 1900 to 5.5 million in 1961, 9 million in 1980 and Anopheles funestus and Anopheles arabiensis) are 18 million in 2008 [10]. Children aged very similar to those of the African continent. The 5 y make up 16.7% of the population and half of

Scand J Infect Dis Downloaded from informahealthcare.com by Institut Pasteur on 09/11/12 symptoms of ‘tazomoka’ (mosquito-borne fever), the the population is aged 20 y. The Republic of Mada- Malagasy word for malaria, were known by the local gascar is one of the poorest countries in the world. inhabitants who treated them with traditional herbal Malaria is the second most important cause of mor- remedies. Europeans brought with them their anti- bidity (after acute respiratory infections) and mortal- malarial medication in the form of Cinchona bark ity (after diarrhoea). Malaria is responsible for powder (from which quinine was isolated in 1820). more than 30% of outpatient attendance at health At the end of the 19th century, French colonizers, centres [10]. lead by Gallieni, brought large quantities of quinine The 587,000 km2 of Madagascar can be divided to Madagascar [10]. Chloroquine (CQ), the fi rst into 5 geographical regions: the east coast, the Tsar- synthetic antimalarial compound, was introduced in atanana Massif, the central highlands, the west coast, 1945. Resistance to CQ (CQR) in Madagascar was and the southwest. For epidemiological studies and suspected in 1975 [13] and confi rmed in 1981–1983 malaria control purposes, the country has been strat- [9,14]. However the level of CQR remained low for ifi ed into 4 distinct zones (as there is no transmission many years and even today CQ is still used, though in the Tsaratanana Massif): equatorial in the east, it should be replaced by other antimalarials. austral in the central highlands, tropical in the west, Half of the population of Madagascar is at high risk and semi-arid in the south (Figure 2) [18]. In the east of malaria. The National Malaria Control Programme and in the west, transmission is stable – perennial in 24 V. Andriantsoanirina et al. For personal use only.

Figure 2. Map of Madagascar based on malarious epidemiological strata and 1982–2002 collection sites of isolates used in the in vitro studies. EIR: entomological inoculation rate. Scand J Infect Dis Downloaded from informahealthcare.com by Institut Pasteur on 09/11/12 the east and more seasonal on the west coast, with a travel and migration may also contribute to the emer- decrease in transmission in July and August. In both gence of epidemics. Madagascar is located between regions, immunity amongst adults is reported to be Asia and Africa, both of which are known to have a high and most of the severe cases affect children aged high prevalence of antimalarial resistance. Interna- 10 y and pregnant women. In the central highlands, tional travel and trade between Madagascar and malaria is stable up to 1000 m in altitude, unstable these 2 continents has resulted in an increase in imported with seasonal transmission from November to May malaria cases. A high prevalence of antimalarial resis- between 1000 and 1500 m, and rare above 1500 m. In tance is also present in the Comoros Islands, which the semi-desert of the south, malaria is unstable and are not far from Madagascar [20]. hypoendemic, with a short seasonal transmission. In both the central highlands and in the south, immunity is limited and all age groups are vulnerable to periodic History of antimalarial use in Madagascar epidemics, which are often associated with high levels of morbidity and mortality. Human activity such as the CQ was the fi rst 4-aminoquinoline introduced to the cultivating of rice fi elds is of paramount importance Island. In 1949, the Ministry of Health launched an for the proliferation of the vectors [19]. Transportation, ambitious campaign for malaria prevention in children, Antimalarial drug resistance in Madagascar 25 on the basis of a weekly administration of CQ [4,5]. First detection of antimalarial resistance A sharp decline in malaria prevalence was observed In 1975, Goasguen et al. reported 2 cases of suspected at the end of the 1950s in the central highlands and CQR in Madagascar [13]. The fi rst case was a on its fringes. The weekly administration of CQ was prophylactic failure and a treatment failure in a maintained at all primary schools and at the Red Britannic traveller. The second case was a treatment Cross centres on the Island until 1970. In addition, failure in a Malagasy woman. Parasitological diagnosis CQ was readily available in many groceries. A short- was performed and both cases were cured with SP, age in CQ supply then occurred, together with the however isolates were not sent to a reference centre end of indoor residual spraying with DDT. This was to validate the CQR. Five y later, Aronsson et al. followed by a large-scale epidemic in the late 1980s, observed 2 further cases of prophylactic and treat- killing an estimated 30,000–100,000 people per y ment failure with CQ in 2 Swedish women who had over the course of 3 y [ 11,18,21,22]. The outbreak stayed several months in [14]. Correct was brought under control by resuming indoor plasma drug concentrations were found and in vitro residual spraying and wide-scale access to CQ treat- tests showed CQR isolates. These cases showed that ment. CQ is still used, though in December 2005 prophylaxis failure in travellers may lead to the earlier the Ministry of Health recommended its replace- detection of resistance and at weaker levels than ment (and abandonment) as fi rst-line treatment for treatment failures within the local population, because uncomplicated cases, with the combination artesu- the doses used are lower and because travellers are nate plus amodiaquine (ASAQ), which is an ACT. usually non-immune subjects. The policy for the management of fever in children, launched in 2003, led to a wide-scale distribution of free pre-packaged CQ treatments, and very large Study and follow-up of antimalarial resistance quantities of CQ were ordered in 2007 by the health districts [ 23]. Thus, CQ remains easily available in In vivo studies Madagascar and its importation is, as yet, not for- Methodologies used to monitor antimalarial drug bidden. resistance have varied over time [25]. Until the end Amodiaquine, a 4-aminoquinoline, has been reg- of the 1980s, most in vivo studies focused on the istered and commercially available in Madagascar since parasitological response to a given drug, and malaria the 1950s. It has been used without discontinuation episodes were classifi ed as sensitive (S) or resistant for more than 20 y. (R) at 1 of 3 levels, R1, R2 or R3. An R1 response For personal use only. Sulfadoxine–pyrimethamine (SP) is an antifolate corresponded to an initial clearance of parasitaemia combination recommended as second-line treatment for followed by recrudescence 8 or more days after treat- uncomplicated malaria cases since 1997. It was already ment; an R2 response was the clearance or substantial in use on the Island many years before this date. It has reduction of parasitaemia with recrudescence of also been used since 2005 for the intermittent presump- parasitaemia on day 7; and an R3 response referred tive/preventive treatment of pregnant women [10]. to a situation in which there was no initial reduction Quinine has been used for many years in Mada- of asexual parasitaemia or even increase in level of gascar. It is still very effective, but should be reserved parasitaemia. Treated patients were assessed weekly for the treatment of severe malaria and cases of failure until day 7, day 14 or day 28, depending on the dif-

Scand J Infect Dis Downloaded from informahealthcare.com by Institut Pasteur on 09/11/12 with other antimalarial treatments. ferent studies. World Health Organization (WHO) Artemisinin is the active ingredient of an ancient protocols have been modifi ed and simplifi ed to facil- Chinese medicinal beverage used to treat fevers. It itate their use in high-transmission areas in Africa, is obtained from the leaves of the plant Artemisia where populations may have asymptomatic parasi- annua and it was rediscovered in the 1970s by taemia. Therapeutic responses were then classifi ed as scientists who called this crystalline compound either adequate clinical response (ACR) or early or ‘qinghaosu’ or ‘artemisinin’. Artesunate and other late treatment failure (ETF and LTF). The therapeutic compounds derived from artemisinin, such as arte- response is classifi ed as ETF if the patient develops mether, appear to be very effective at killing P. fal- clinical or parasitological symptoms during the fi rst ciparum. Artemether was introduced to the Malagasy 3 days of follow-up, and as LTF if the patient develops markets for the fi rst time in early 1990 [ 24]. ACTs symptoms during the follow-up period from day 4 to are today the basis of the malaria control/elimination day 14, without previously meeting the criteria for ETF. programme, together with the use of insecticide- ACR is defi ned as either the absence of parasitaemia impregnated bed nets and indoor spraying with on day 14 (irrespective of axillary temperature), or the DDT. However, in 2008, few of the health districts absence of clinical symptoms on day 14 (irrespective of Madagascar had adequate supplies of artesunate– of parasitaemia), in patients who did not meet the amodiaquine (ASAQ). criteria of ETF or LTF before. In vitro protocols 26 V. Andriantsoanirina et al. have also varied over time and between the different small series of uncomplicated cases collected in 2003 laboratories involved in phenotype determination of [40,41]. In a multi-site study performed in 2006– antimalarial resistance. 2007, SP showed a mean of near 97% effi cacy, with Chloroquine. Various studies have been performed some variation depending on site (range from 0% to to follow CQ treatment effi cacy. Since 1982, stan- 12.5%) [37]. dard protocols have been followed. In vivo tests were Amodiaquine. In 1983, on the east coast of performed using 25 mg/kg of CQ (10 mg/kg on the Island, Deloron et al. reported 10.6% R1 or day 0 and day 1 and 5 mg/kg on day 2). Thin R2 among uncomplicated cases (n 47) treated with and thick smears were performed on day 0, day 7 10 mg/kg [42]. In 1985–1986, on the east coast, and day 14. Marchais et al. reported 24.1% of SR1 and 3.4% Deloron et al. reported 20% of late treatment of R1 in uncomplicated cases (n 58) treated with failure among standard treatments ( n 40) per- 19 mg/kg [28]. Using 25 mg/kg or more, other formed in 1983 [26]. Biaud et al. reported 4.3% of authors obtained better results, indicating that treat- R1 treatment failure among standard treatments ment failure was dependent on the dose prescribed. (n 70) in 1986 by day 28 of follow up [27]. Mar- In 1984, on the east coast and in the central highlands, chais et al. reported 12.3% of R2 and 45.6% of R1 Biaud et al. reported complete effi cacy on day 14 in treatment failure among standard treatments by uncomplicated cases (n 64) treated with 25 mg/kg day 28 of follow up (n 57) in 1987 and concluded [27]. Lepers et al. reported a complete effi cacy during that CQ was ‘perfectly active’ in the 2 studied sites the 14-day follow-up in their study of 25 mg/kg in of Madagascar, because CQ retained its clinical uncomplicated cases (n 35) collected in 1988 in the effi cacy [28]. Until 1987, the rates of CQR studied central highlands [31]. Randriamanantena et al. reported in various points of the Island were always below a complete effi cacy on day 14 in uncomplicated cases 10%, with some rare cases of R2 [29]. In 1989, (n 25) treated with 30 mg/kg collected in 2004 on Lepers et al. reported a ‘serious degradation’ of the the east coast and in the central highlands [42]. In a effi cacy of CQ, with 10% of R1, 19% of R2 and multi-site study performed in 2006–2007, Ménard 32% of SR1 (parasitaemia recurring between days et al. reported near 99% effi cacy on day 14 (n 372) 7 and 14, corresponding to late recrudescence or and near 98% on day 28 (n 356) in those treated reinfection) among standard treatments (n 41) with 30 mg/kg [37]. performed on the island of Sainte-Marie (northeast Other antimalarials. Very few in vivo studies coast of Madagascar) [30]. Comparable results have been performed with other antimalarials in For personal use only. were found in 1988 (n 91) and in 1989 (n 123) Madagascar [43]. Quinine remained highly active in in the central highlands by the same team [31,32]. 2 studies performed in 1988 and 1997 [31,38,44]. However Lepers et al. continued to recommend The combination artesunate–amodiaquine showed CQ as fi rst-line treatment of malaria in Madagascar complete effi cacy on day 14 and less than 2% of for “economical reasons and for its good tolerance overall treatment failure on day 28 in a multi-site and ease of use” and because of the lack of reported study performed in 2006–2007 [37]. R3 resistance [32–34]. These authors also sug- gested the close monitoring of CQ effi cacy to detect the emergence of R3 in the future. Ringwald et al. In vitro studies

Scand J Infect Dis Downloaded from informahealthcare.com by Institut Pasteur on 09/11/12 reported a stability of CQR in the central highlands Various in vitro studies have been performed to in the y 1990–1991 [ 35]. R3 resistance was fi rst monitor CQ susceptibility [9,24,27,28,31,33,36,43, evidenced by Randrianarivelojosia et al. in 1997 46–53,55,57–60]. The fi rst in vitro studies performed [36]. A study performed in 2001 in Ankazobe in the 1980s reported variable rates of CQR, depend- reported 23.5% of treatment failure (n 46), ing on the site, ranging from 1% to 36% of CQR including 2 R3 resistance cases [ 1]. In the most isolates (Table I). The main event in those years was recent studies, treatment failure of the standard the sudden increase in CQR reported in 1988 in 2 regimen (25 mg/kg over 3 days) evaluated by day distinct areas (one located in the central highlands 28 of follow-up exceeded 40% ( n 300), but early and the other on the east coast) [33]. The treatment failure remained below 5% [37]. only study that was performed in the 1990s showed Sulfadoxine–pyrimethamine. A fi rst survey of SP 8.3% of CQR isolates (n 216) [ 60]. Results from effi cacy showed parasitaemia for 2 patients out of 26 the malaria resistance surveillance network (RER) on day 7 [38]. Another study performed in 1997 showed founded in September 1999 by the Ministry of Health R1 or R2 resistance for 2 patients out of 46 [36]. A lack and the Institut Pasteur de Madagascar, showed of pyrimethamine resistance was reported in 2002 by 4.6% of CQR isolates ( n 216) in isolates collected Randrianarivelojosia et al. [39]. Randrianasolo et al. in 2001 from different sites on the Island [47]. Today, showed the complete effi cacy of SP in a relatively the rate of in vitro CQR isolates remains surprisingly Antimalarial drug resistance in Madagascar 27

Table 1: In vitro susceptibilities and prevalence of resistant P. falciparum isolates to anti-malarial drugs, Madagascar, 1982-2002.

No. of Resistant Geometric mean IIC50% Years Collection sites isolates isolates (%) of IC50% (nM) range (nM) Ref.

Chloroquine 1982 Antananarivo, , Toamasina 34 20.6 - - 59 1983 Ifanadiana, Andekaleka, Ankazobe 23 4.3 - - 57 Manakara, Foulpointe 24 8.3 - - 26 Morondava, Nosy-Be 19 26.3 - - 58 Morondava, Nosy-Be, Manakara, 78 35.9 - - 9 Foulpointe, Tamatave, Ankazobe 1984 Alatsinainy 91 16.0 - - 33 Antananarivo, Nosy-Be, Andekaleka 91 1.1 - - 27 Morondava, Nosy-Be, Manakara 91 16.5 - - 9 1985 Central Highlands 91 16.5 - - 33 1985–1986 Foulpointe 40 5.0 35 - 28 Toamasina 16 0 17 - Foulpointe, Ankazobe 30 6.7 - - 9 East 56 8.9 - - 33 1987 Manarintsoa 139 5.8 - - 55, 33 - 84 9.0 - - 43 1988 Central Highlands 104 27.9 - - 33 Manarintsoa 104 29.8 78 10–364 31 1987–1989 Manarintsoa 243 15.6 - - 60 1990–1991 Ankazobe 42 9.5 39 - 35 1990–1993 Ankazobe 216 8.3 - - 60 1997 Marovoay, Nosy-Be, Sainte-Marie, 90 4.4 161.7 - 36 Toamasina 43.7 - 1998–1999 Ankazobe, Saharevo 46 11.0 171 109–245 24 18 0.4–79 Ankazobe, Saharevo, Toamasina, Tolagnaro 205 5.8 9.4 - 50 1999–2000 Ankazobe, Moramanga, Tolagnaro 190 7.9 52.6 0.2–283.4 48 2000 Ankazobe, Saharevo, Mahakary, Sainte-Marie, 255 4.8 - - 53 Toamasina, Marovoay, Nosy-Be, Tolagnaro For personal use only. Sainte-Marie 27 14.8 34.9 - 51 2001 Toamasina 18 16.6 66.3 42.6–90 52 25 0 22.6 16.8–28.7 East 96 5.2 39.7 4.1–206.6 49 West 31 3.2 31.6 0.2–152.1 Centre 19 4.5 33.3 0.2–292.8 2000–2001 17 sites of the RER 432 5.7 - - 47 2001–2002 Tsiroanomandidy, Sainte-Marie, Saharevo 74 1.3 29.9 - 46

Amodiaquine 1983 Ifanadiana, Andekaleka, Ankazobe 23 0 - - 58 Scand J Infect Dis Downloaded from informahealthcare.com by Institut Pasteur on 09/11/12 Manakara, Foulpointe 27 0 - - 26 18 5.5 - - 58 1984 - 84 0 - Max range 76 57 1985–1986 Foulpointe 26 0 18 - 28 Toamasina 11 0 42 - Foulpointe 25 0 19 - Toamasina 5 0 59 - 1999–2000 Ankazobe, Moramanga, Tolagnaro 118 0 15.3 0.2–52.5 48 2000 Sainte-Marie 13 0 9.4 - 9 2001 Toamasina 17 0 17.8 10.4–25.3 52 Mahajanga 24 0 7.9 5.5–10.4 East 96 0 12.7 0.4–57.9 49 West 31 0 9.3 0.01–27.5 Centre 86 0 9.9 1.4–64.2 2001–2002 Tsiroanomandidy, Sainte-Marie, Saharevo 71 0 10.8 9.2–12.4 45

(Continued) 28 V. Andriantsoanirina et al.

Table I. (Continued)

No. of Resistant Geometric mean IC50% Years Collection sites isolates isolates (%) of IC50% (nM) range (nM Ref.

Quinine 1988 Manarintsoa 64 0 157 25–440 31 84 0 - Max range 560 43 1998–1999 Ankazobe, Saharevo, Toamasina, Tolagnaro 199 0 26.8 50 1999–2000 Ankazobe, Moramanga, Tolagnaro 212 0 107.7 11.1–630.7 48 2000 Sainte-Marie 26 0 93.1 48.1–138.1 51 2001 Toamasina 17 0 194.2 125.6–262.9 52 Mahajanga 25 0 88.8 69–108.6 Antsiranana, Mahajanga, Ranohira, 215 0 133.7 - 46 Morondava, Toliary, Antananarivo, Saharevo, Sambava, Sainte-Marie, Toamasina, Esana East 96 0 108.1 4.2–584.2 49 West 31 0 109.9 32.5–214.7 Centre 87 0 85.1 8.5–280.4 2001–2002 Tsiroanomandidy, Sainte-Marie, Saharevo 71 0 70.6 59.2–82 45 3 sites 153 0 - - 54 Halofantrine

1984 Antananarivo, Nosy-Be, Andekaleka 91 0 12 -27 84 0 - - 44 3 sites 153 0 - - 54 1998–1999 Ankazobe, Saharevo, Toamasina, Tolagnaro 56 0 0.3 - 50 Mefl oquine 1988 Manarintsoa 23 0 12.5 2.1–33 31 -840--43 1998–1999 Ankazobe, Saharevo 46 0 10.5 - 24 Ankazobe, Saharevo, Toamasina, Tolagnaro 199 2 3.8 - 50 1999–2000 Ankazobe, Moramanga, Tolagnaro 139 0.71 11.57 2.9–124.6 48 2000 Ankazobe, Saharevo, Mahakary, Sainte-Marie 254 8.6 - - 53 Toamasina, Marovoay, Nosy-Be, Tolagnaro, 25 4.0 12.3 3.1–21.5 51 For personal use only. Sainte-Marie 2001 Toamasina 12 0 12.9 8.1–17.8 53 Mahajanga 15 0 11.5 7.2–15.9 East 46 2.17 12.1 1.7–53.2 49 West 18 0 10.2 3.4–20.4 Centre 46 0 8.8 1.3–24.1 Antsiranana, Mahajanga, Ranohira, 110 0.9 18.3 - 46 Morondava, Toliary, Antananarivo, Saharevo, Sambava, Sainte-Marie, Toamasina, Esana 2001–2002 Tsiroanomandidy, Sainte-Marie, Saharevo 59 1.7 - - 45

Scand J Infect Dis Downloaded from informahealthcare.com by Institut Pasteur on 09/11/12 Cycloguanil 1999–2000 Ankazobe, Moramanga, Tolagnaro 79 0 12.4 0.1–275.9 48 2001 17 sites RER 56 0 - - 49 Toamasina 10 0 9.8 2.7–16.8 52 Mahajanga 7 0 6.2 0.4–12.2 Antsiranana, Mahajanga, Ranohira, 56 0 50 46 Morondava, Toliary, Antananarivo, Saharevo, Sambava, Sainte-Marie, Toamasina, Esana Pyrimethamine 1997 Marovoay, Nosy-Be, Sainte-Marie, Toamasina 46 2.17 - - 36 1999–2000 - 58 0 30.5 - 39 Ankazobe, Moramanga, Tolagnaro 86 0 46.7 0.1–679.6 48 2001 Toamasina 12 0 - - 52 Mahajanga 9 0 - - 17 RER sites 98 0 - - 49

RER, malaria resistance surveillance network (Réseau d’Etude de la Résistance). Antimalarial drug resistance in Madagascar 29 low – near 5%. In vitro resistance to other antima- thresholds for in vitro resistance interpretation accord- larials is anecdotal [9,24,27,28,31,36,39,43,45,46, ing to the study have been noted. For this reason, the 48–54,56–58]. Most recent data on the sentinel implementation of regional (RER) or global (Obser- sites from the RER did not show any in vitro vatoire de la Resistance aux Antipaludiques, ORA, resistance to pyrimethamine, cycloguanil, quinine or www.oranet.fr or World Antimalarial Resistance Net- dihydroartemisinin. Rates of in vitro resistance to work, WARN) networks of surveillance for antima- amodiaquine and mefl oquine were below 1%. larial resistance, using comparable methodologies, represents major progress in the control of malaria. Molecular markers Though CQR was reported early in Madagascar in comparison with other African countries, and It has previously been shown that a single pfcrt mutant despite an important and continuous drug pressure, allele (T 76) associated with CQR emerged in South- the level of CQR remained low for many years on east Asia and invaded the African continent in the the Island, and even today the level of CQR is lower 1970s [ 61]. Molecular studies did not report the than that encountered in African countries. This presence of this pfcrt mutant allele in Madagascar CQR most often leads to late treatment failure; early until 2001–2002, when it was found in 3.3% of treatment failure is rare. This is in sharp contrast to isolates (n 183) collected from 2 sites (one in the reports from mainland Africa and the Comoros central highlands and the other on the northeast Islands, where early CQ treatment failure occurs at coast) [62,63]. A recent study has shown that para- a high rate, particularly in children [ 62]. Currently, sites having the pfcrt mutant allele were introduced the rate of in vitro CQR isolates does not exceed 5%. into Madagascar by travellers coming from the This fi gure appears disconnected from the level of Comoros Islands [20]. However, the rate of isolates CQ clinical failure (near 40%). It appears that CQR harbouring this mutant allele remained low in in Madagascar differs from other areas of the world comparison to the rate of CQ clinical treatment in that pfcrt mutant isolates are very uncommon. failure. Conversely, the pfmdr 1 Y86 mutation, which Conversely, pfmdr1 mutant isolates are frequently has been associated elsewhere with in vitro CQR [64 ], found and may be responsible for the bulk of CQR was present in 67.5% of isolates (n 117) collected in Madagascar. This may also be in relation to the at 3 sites in Madagascar [45]. In addition, pfmdr 1 fact that most clinical CQ treatment failures in Mad- mutant genotypes, including Y86 , F184 and Y1246 agascar are late clinical failures. mutations, have frequently been found on the Island CQ treatment failure reached an intolerable rate

For personal use only. (Andriantsoanirina et al., submitted). Randrianariv- around 2004, which promoted the Ministry of Health elojosia et al. reported a lack of mutant dihydrofolate to recommend that the fi rst-line treatment for uncom- reductase (DHFR) N108 allele in isolates (n 58) col- plicated malaria be changed to a drug with better effi - lected in 1999–2000 at 1 site in the central highlands cacy. CQ is no longer recommended for the treatment and at another site on the east coast [39]. However, of P. falciparum, except at the community level, in the parasites carrying DHFR mutant alleles are imported home management of fever until ACTs are available. from the Comoros Islands. Recent data show that the The problem is, that though often ineffective, treat- triple DHFR I51 R59N108 mutant allele has spread ment with CQ may be perceived as successful by the widely in Madagascar (Andriantsoanirina et al., sub- population [66]. Indeed, most acute fevers are treated mitted). In addition, unusual DHFR haplotypes, Scand J Infect Dis Downloaded from informahealthcare.com by Institut Pasteur on 09/11/12 with antimalarials without laboratory diagnosis. Two- including the L164 mutation, have recently been thirds of these fevers are not malarial in origin and reported in Madagascar [65]. However, the quadru- generally resolve spontaneously in a few days. In addi- ple mutant DHFR I 51R59N108L164 associated with tion, a proportion of malaria attacks resolve spontane- high levels of resistance to antifolates and resulting in ously due to the immunity of the subject. Moreover, a complete ineffi cacy of SP, has not been reported in as treatment failure with CQ is most often late treat- Madagascar so far. A recent longitudinal and multi- ment failure in Madagascar, the recrudescence of fever site study has shown that the dihydropteroate syn- 3 weeks later or more may not be perceived as a failure thase (DHPS) G437 mutant allele is increasingly being of the treatment. This misperception by the consumer reported (Andriantsoanirina et al., submitted). could delay the replacement of CQ by ACTs. The clinical effi cacy of SP remains high in Mad- agascar, but with variations according to site, and in Comments and concluding remarks vitro data show reassuring results. However, the tri- The evolution of in vivo protocols over time and a ple mutant DHFR allele and DHPS G437 mutant relative lack of standardization between the different alleles are present in a high proportion. The triple in vitro studies have rendered the comparison of mutant DHFR allele is probably the same as that results somewhat unreliable. In particular, variable originating from Southeast Asia, which has invaded 30 V. Andriantsoanirina et al. the African continent [61], but this has to be assessed [5] Joncour G. The fi ght against malaria in Madagascar. Bull by microsatellite genotyping studies. The SP drug World Health Organ. 1956;15:711–23. pressure has increased since the adoption of inter- [6] Blanchy S, Rakotonjanabelo A, Ranaivoson G, Rajaonarivelo E. Epidémiologie du paludisme sur les hautes terres mal- mittent preventive treatment for pregnant women. A gaches depuis 1878. Santé. 1993;3:155–61. survey of its effi cacy is required in order to continue [7] Lepers JP, Deloron P, Fontenille D, Coulanges P. Reappear- recommending SP for this indication. ance of falciparum malaria in central highland plateaux of Recent clinical and in vitro data support the com- Madagascar. Lancet. 1988;1:586. plete effi cacy of the ASAQ combination in Madagascar. [8] Mouchet J, Laventure S, Blanchy S, Fioramonti R, Rakoton- janabelo A, Rabarison P, et al. The reconquest of the Mada- As such, this ACT should play a central role in the gascar highlands by malaria. Bull Soc Pathol Exot. control of P. falciparum malaria. In the current efforts 1997;90:162–8. towards the elimination of P. falciparum from the Island, [9] Coulanges P, Le Bras J, Deloron P, Ramanamirija JA, ASAQ could play a role as important as CQ did in the Biaud JM, Marchais H. In vivo and in vitro study of the 1950–1970s. It is important to note that the re-emer- chemosensitivity of Plasmodium falciparum in Madagascar, 1982–1986. Arch Inst Pasteur Madagascar. 1987;53:63–76. gence of malaria in the central highlands in the 1980s [10]Ministère de la Santé, du Planning Familial et de la Protec- and the major outbreaks that followed were not due to tion Sociale. Plan stratégique de lutte contre le paludisme, CQR, which was moderate at that time, but to a short- du contrôle vers l’élimination du paludisme à Madagascar. age of CQ treatment following the deterioration of the Madagascar: Ministry of Health; 2007. p. 54. public health system. These past events highlight the [11] Mouchet J, Carnevale P, Coosemans M, Julvez J, Manguin necessity to durably sustain the purchase of ACTs at S, Richard-Lenoble D, et al. Iles du sud-ouest de l’océan Indien. In: Libbey J, editor. Biodiversité du paludisme dans subsidized or preferential prices and to distribute them le monde. Paris: Eurotext; 2004. pp. 167–74. to populations for an indefi nite period [ 67]. One can [12] Hewitt R, Krause A, Goldman A, Campbell G, Jenkins only hope that the current global fi nancial crisis will not T. b -Globin haplotype analysis suggests that a major source result in drastic cuts in the funding dedicated to the fi ght of Malagasy ancestry is derived from Bantu-speaking against malaria in developing countries and that ACTs negroids. Am J Hum Genet. 1996;58:1303–8. [13] Goasguen J, Gentelet B, Moreau J, Fouquet R, Coulanges P. will preserve their good effi cacy in the future. Plasmodium falciparum chloroquino-résistant? A propos des deux premiers cas malgaches. Arch Inst Pasteur Madagascar 1975;44:143–5. Acknowledgements [14] Aronsson B, Bengtsson E, Bjorkman A, Pehrson PO, Rombo L, Wahlgren M. Chloroquine-resistant falciparum malaria in The authors thank the staff of the Ministry of Health Madagascar and Kenya. Ann Trop Med Parasitol. 1981;75: of Madagascar. This work was supported by grants 367–73. For personal use only. from Natixis/Impact Malaria through the ‘Observa- [15] World Health Organization. World malaria report. Geneva: WHO; 2008. Available at: http://malaria.who.int/wmr2008/ toire de la Résistance aux Antipaludiques’ project. MAL2008-CountryProfi les/MAL2008-Madagascar-EN.pdf Valérie Andriantsoanirina is a graduate PhD student (accessed September 2009). funded by the Institut Pasteur de Madagascar (Bourse [16] Barnadas C, Tichit M, Bouchier C, Ratsimbasoa A, Randria- ‘Girard’) and the Direction des Affaires Internation- nasolo L, Raherinjafy R, et al. Plasmodium vivax dhfr and dhps ales (Institut Pasteur). mutations in isolates from Madagascar and therapeutic response to sulfadoxine–pyrimethamine. Malar J. 2008;7:35. [17] Barnadas C, Ratsimbasoa A, Ranaivosoa H, Ralaizandry D, Raveloariseheno D, Rabekotonorina V, et al. Short report: Declaration of interest: None declared. prevalence and chloroquine sensitivity of Plasmodium malar-

Scand J Infect Dis Downloaded from informahealthcare.com by Institut Pasteur on 09/11/12 iae in Madagascar. Am J Trop Med Hyg. 2007;77:1039–42. [18] Mouchet J, Blanchy S, Rakotonjabelo A, Ranaivoson G, Rajaonarivelo E, Laventure S, et al. Epidemiological stratifi - References cation of malaria in Madagascar. Arch Inst Pasteur Mada- gascar. 1993;60:50–9. [1] Willcox ML, Rakotondrazafy E, Andriamanalimanana R, [19] Blanchy S, Rakotonjanabelo A, Ranaivoson G. Surveillance Andrianasolo D, Rasoanaivo P. Decreasing clinical effi cacy épidémiologique du paludisme instable. Santé. 1993;3:247–55. of chloroquine in Ankazobe, Central Highlands of Madagas- [20] Ménard D, Randrianarivo-Solofoniaina AE, Ahmed BS, car. Trans R Soc Trop Med Hyg 2004;98:311–4. Jahevitra M, Andriantsoanirina V, Rasolofomanana JR, et al. [2] Robert V, Le Goff G, Andrianaivolambo L, Randimby FM, Drug-resistant malaria parasites introduced into Madagas- Domarle O, Randrianarivelojosia M, et al. Moderate trans- car from Comoros Islands. Emerg Infect Dis. 2007;13: mission but high prevalence of malaria in Madagascar. Int J 1759–62. Parasitol 2006;36:1273–81. [21] Razanamparany MS, Randriamiharisoa FA, Razanamparany [3] Romi R, Razaiarimanga MC, Raharimanga R, Rakotond- NJ, Ramialimanana V. The malaria epidemic in Antananarivo raibe EM, Ranaivo LH, Pietra V, et al. Impact of the malaria from 1983 to 1994 as seen through the Pediatric Service A control campaign (1993–1998) in the highlands of Madagas- in the Befelatanana General Hospital. Santé. 1995;5:382–5. car: parasitological and entomological data. Am J Trop Med [22] Razanamparany M, Bosman A, Ravelomanana N, Randriami- Hyg. 2002;66:2–6. harisoa F, Razafi mantsoa L. The malaria epidemic in Antan- [4] Bernard PM. Trois ans de lutte antipaludique à Madagascar anarivo observed from Pediatric Service ‘A’ of the Befelatanana (1950–1951–1952). Bull Madagascar. 1954;96:415–20. General Hospital. Parassitologia. 1989;31:89–99. Antimalarial drug resistance in Madagascar 31

[23] Ratsimbasoa A, Randrianarivelojosia M, Millet P, Soarès JL, [38] Rakoto SA, Raharimalala L, Jambou R. Etude de la Rabarijaona L, Rakotoson B, et al. Use of pre-packaged chlo- chimiosensibilité de Plasmodium falciparum à Madagascar: roquine for the home management of presumed malaria in 1 – situation des médicaments de seconde ligne à Taolagnaro Malagasy children. Malar J. 2006;5:79. (quinine, sulfadoxine–pyrimeéthamine). Arch Inst Pasteur [24] Randrianarivelojosia M, Raharimalala LA, Randrianasolo L, Madagascar. 1998;64:51–5. Ratsimbasoa A, Rason MA, Ariey F, et al. Madagascan isolates [39] Randrianarivelojosia M, Ariey F, Raharimalala LA, Parzy D, of Plasmodium falciparum showing low sensitivity to arte- Rogier C, Jambou R. Current absence of pyrimethamine mether in vitro. Ann Trop Med Parasitol. 2001;95:237–43. resistance of Plasmodium falciparum in Madagascar. Trans [25] Talisuna AO, Bloland P, D’Alessandro U. History, dynamics, R Soc Trop Med Hyg. 2002;96:557–9. and public health importance of malaria parasite resistance. [40] Randrianasolo L, Randriamanantena A, Ratsimbasoa A, Clin Microbiol Rev. 2004;17:235–54. Rakotoson JD, Randriambelosoa J, Raveloson A, et al. Surveil- [26] Deloron P, Ramanamirija JA, Verdíer F, Coulanges P. lance de l’effi cacité de sulfadoxine–pyriméthamine (Fansidar®, 4-Aminoquinoline-sensitivity of Plasmodium falciparum in Paludar®) pour le traitement de l’accés palustre à Madagascar Madagascar. II. Study in two regions of the east coast. Arch : proposition d’un protocole simplifi é. Arch Inst Pasteur Inst Pasteur Madagascar 1984;51:43–55. Madagascar. 2003;69:52–6. [27] Biaud JM, Ramanamirija JA, Ramiliarisoa O, Verdier F, Cou- [41] Randrianasolo L, Randriamanantena A, Ranarivelo L, langes P, Le Bras J. Drug sensitivity of malaria in Madagascar. Ratsimbasoa A, Domarle O, Randrianarivelojosia M. Monitor- Comparison in vivo and in vitro of the effects of chloroquine ing susceptibility to sulfadoxine–pyrimethamine among cases and amodiaquine on Pfalciparum in 1984. Arch Inst Pasteur of uncomplicated Plasmodium falciparum malaria in Saharevo, Madagascar. 1985;52:73–84. Madagascar. Ann Trop Med Parasitol. 2004;98:551–4. [28] Marchais H, Ramanamirija JA, Le Bras J, Coulanges P. In [42] Randriamanantena A, Randrianasolo L, Vonimpaisomihanta vivo and in vitro sensitivity to 4-aminoquinolines of Plasmo- JA, Tafangy PB, Bayant Z, Randrianarivelojosia M. Thera- dium falciparum in Madagascar: results of a study conducted peutic effi cacy of amodiaquine against uncomplicated malaria on the east coast (July 1985–July 1986). Arch Inst Pasteur in Madagascar. Santé. 2007;17:75–8. Madagascar. 1987;53:43–62. [43] Deloron P, Le Bras J, Ramanamirija JA, Coulanges P. Plas- [29] Coulanges P, Le Bras J, Deloron P, Ramanamirija JA. Resist- modium falciparum in Madagascar: in vivo and in vitro sen- ance of Plasmodium falciparum to 4-aminoquinolines. A sitivity to seven drugs. Ann Trop Med Parasitol. 1985;79: review based on the Madagascar experience. Arch Inst Pas- 357–65. teur Madagascar. 1987;53:77–103. [44] Deloron P, Lepers JP, Verdier F, Chougnet C, Remanamirija [30] Lepers JP, Ramanamirija JA, Randriasamimanana JR, Ram- JA, Andriamangatiana-Rason MD, et al. Effi cacy of a 3-day anandraibe LM, Rahelimalala RS, Rene JP, et al. Malaria in oral regimen of a quinine–quinidine–cinchonine association Ile de Sainte Marie in 1988. Epidemiological approach Chlo- (Quinimax) for treatment of falciparum malaria in Madagas- roquine sensitivity of Plasmodium falciparum. Arch Inst Pas- car. Trans R Soc Trop Med Hyg. 1989;83:751–4. teur Madagascar. 1989;56:131–42. [45] Rason MA, Andrianantenaina HB, Ariey F, Raveloson A, [31] Lepers JP, Deloron P, Andriamangatiana-Rason MD, Ram- Domarle O, Randrianarivelojosia M. Prevalent pfmdr 1 N86Y anamirija JA, Coulanges P. Sensitivity of Plasmodium falci- mutant Plasmodium falciparum in Madagascar despite absence

For personal use only. parum to different antimalarials. A study in 1988 in a village of pfcrt mutant strains. Am J Trop Med Hyg. 2007;76:1079–83. in the Hauts-Plateaux of Madagascar. Bull Soc Pathol Exot. [46] Randrianarivelojosia M, Randrianasolo L, Randremanana 1989;82:650–7. RV, Randriamanantena A, Ratsimbasoa A, Rakotoson JD. [32] Andriamangatiana-Rason MD, Lepers JP, Raharimalala Susceptibility of Plasmodium falciparum to the drugs used Andrianavalona L, Verdier F, Coulanges P. Aspects of in vivo to treat severe malaria (quinine) and to prevent malaria chemosensitivity of Plasmodium falciparum to chloroquine (mefl oquine, cycloguanil) in Comoros Union and Madagas- in a village of the highland plateaux of Madagascar in 1989. car. S Afr Med J. 2004;94:47–51. Arch Inst Pasteur Madagascar. 1990;57:93–104. [47] Randrianarivelojosia M, Sahondra-Harisoa JL, Raharimalala [33] Lepers JP, Deloron P, Mouden JC, Le Bras J, Coulanges P. LA, Raveloson A, Mauclère P, Ariey F. A network RER Sudden increase in number of isolates of Plasmodium falci- rooted on in vitro readout assays of Plasmodium falciparum parum resistant to chloroquine in Madagascar. Trans R Soc sensitivity to chloroquine in the region. Santé.

Scand J Infect Dis Downloaded from informahealthcare.com by Institut Pasteur on 09/11/12 Trop Med Hyg. 1989;83:491–2. 2003;13:95–100. [34] Blanchy S, Rakotonjanabelo A, Ranaivoson G. Evolution of [48] Randrianarivelojosia M, Harisoa JL, Rabarijaona LP, Raha- Plasmodium falciparum drug sensitivity in Madagascar from rimalala LA, Ranaivo L, Pietra V, et al. In vitro sensitivity of 1982 to 1993 and operational consequences. Bull Soc Pathol Plasmodium falciparum to amodiaquine compared with Exot. 1993;86:254–9. other major antimalarials in Madagascar. Parassitologia. [35] Ringwald P, Rabarison P, Raharimalala L, Rason MD, Lep- 2002;44: 141–7. ers JP, Le Bras J, et al. Stability and chemosensitivity of Plas- [49] Randrianarivelojosia M, Rakotonjanabelo LA, Mauclère P, modium falciparum to chloroquine in 1990 and 1991 in Ratsimbasoa A, Raharimalala LA, Ariey F. National network Ankazobe, a village in high plateau Madagascar. Bull Soc study to perpetuate the surveillance of Plasmodium falci- Pathol Exot. 1992;85:219–22. parum sensitivity to antimalarials in Madagascar. Arch Inst [36] Randrianarivelojosia M, Raharimalala L, Randriamanan- Pasteur Madagascar. 2002;68:73–8. tena A, Jambou R. Drug resistance of Plasmodium falci- [50] Randrianarivelojosia M, Ratsimbasoa A, Randrianasolo L, parum in coastal regions of Madagascar. Med Trop. 2000; Randrianarijaona A, Jambou R. In-vitro sensitivity of Plasmo- 60:243–9. dium falciparum to chloroquine, halofantrine, mefl oquine and [37] Ménard D, Ratsimbasoa A, Randrianarivelojosia M, Rabari- quinine in Madagascar. East Afr Med J. 2002;79:237–41. jaona LP, Raharimalala L, Domarle O, et al. Assessment of [51] Raharimalala AL, Randrianarivelojosia M, Randriamanan- the effi cacy of antimalarial drugs recommended by the tena A, Ranarivelo LA, Jaureguiberry S, Rason MA, et al. National Malaria Control Programme in Madagascar: up- Chemosensitivity of Plasmodium falciparum in Sainte Marie dated baseline data from randomized and multi-site clinical island, east coast of Madagascar: in vivo and in vitro studies. trials. Malar J. 2008;7:55. Arch Inst Pasteur Madagascar. 2000;66:26–31. 32 V. Andriantsoanirina et al.

[52] Rason MA, Ariey F, Rafi dimanantsoa L, Andrianantenaina BH, [60] Raharimalala L, Lepers JP, Lepers-Rason MD, Rabarison Sahondra Harisoa JL, Randrianarivelojosia M. Monitoring the P, Ramambanirina L, Roux J. Aspects of Plasmodium falci- drug-sensitivity of Plasmodium falciparum in coastal towns in parum sensitivity to chloroquine in Madagascar from 1982 Madagascar by use of in vitro chemosensitivity and mutation to 1993. Arch Inst Pasteur Madagascar. 1993;60:60–4. detection tests. Parasite. 2002;9:247–53. [61] Ariey F, Fandeur T, Durand R, Randrianarivelojosia M, [53] Milijaona R, Jambou R, Raharimalala L, Ranaivo L, Rason Jambou R, Legrand E, et al. Invasion of Africa by a single MA, Roux J. Mefl oquine-resistant strains of Plasmodium pfcrt allele of South East Asian type. Malar J. 2006;5:34. falciparum in Madagascar: impact on tourists and public [62] Ariey F, Randrianarivelojosia M, Duchemin JB, Rakotondra- health. Ann Trop Med Parasitol. 2000;94:313–7. marina D, Ouledi A, Robert V, et al. Mapping of a Plasmo- [54] Milijaona R, Raharimalala L, Ramambanirina L, Ranaivo dium falciparum pfcrt K76T mutation: a useful strategy for LH, Jambou R. Drug resistance of Plasmodium falciparum controlling chloroquine resistance in Madagascar. J Infect along the borders of the highlands in Madagascar: outlook Dis. 2002;185:710–2. for a national control program. Med Trop. 1998;58:261–5. [63] Randrianarivelojosia M, Fidock DA, Belmonte O, Valderra- [55] Lepers JP, Ramanamirija JA, Rason MD, Coulanges P. In mos SG, Mercereau-Puijalon O, Ariey F. First evidence of vitro sensitivity of Plasmodium falciparum to chloroquine in pfcrt mutant Plasmodium falciparum in Madagascar. Trans the highland region of Madagascar in 1987. Bull Soc Pathol R Soc Trop Med Hyg. 2006;100:826–30. Exot. 1988;81:338–44. [64] Picot S, Olliaro P, de Monbrison F, Bienvenu AL, Price RN, [56] Ramanamirija JA, Deloron P, Biaud JM, Le Bras J, Cou- Ringwald P. A systematic review and meta-analysis of evi- langes P. In vivo and in vitro sensitivity to 4-aminoquinolines dence for correlation between molecular markers of parasite of Plasmodium falciparum in Madagascar: results of 2 years’ resistance and treatment outcome in falciparum malaria study. Bull Soc Pathol Exot. 1985;78:606–14. Malar J. 2009;8:89. [57] Deloron P, Ramanamirija JA, Le Bras J, Coulanges P. 4-Ami- [65] Ménard D, Andriantsoanirina V, Jahevitra M, Barnadas C, noquinoline-sensitivity of Plasmodium falciparum in Mada- Tichit M, Bouchier C, et al. Dihydrofolate reductase I164L gascar III Studies in three regions of the interior. Arch Inst mutation in Plasmodium falciparum, Madagascar. Emerg Pasteur Madagascar. 1984;51:57–68. Infect Dis. 2008;14:1166–7. [58] Deloron P, Ramanamirija JA, Le Bras J, Verdier P, Coulanges [66] Hastings IM, Korenromp EL, Bloland PB. The anatomy P. 4-Aminoquinoline-sensitivity of Plasmodium falciparum of a malaria disaster: drug policy choice and mortality in Madagascar. I. Study in two regions of the west coast. Arch in African children. Lancet Infect Dis. 2007;7:739–48. Inst Pasteur Madagascar. 1984;51:29–42. [67] Randrianarivelojosia M, Raveloson A, Randriamanantena A, [59] Le Bras J, Dulat C, Coulanges P, Ralamboson D, Rakotoniri- Juliano JJ, Andrianjafy T, Raharimalala LA, et al. Lessons na-Randriambeloma PJ, Deloron P. Preliminary in vitro learnt from the six decades of chloroquine use (1945–2005) study of drug sensitivity of Plasmodium falciparum in Mada- to control malaria in Madagascar. Trans R Soc Trop Med gascar. Arch Inst Pasteur Madagascar. 1982;50:15–22. Hyg. 2009;103:3–10. For personal use only. Scand J Infect Dis Downloaded from informahealthcare.com by Institut Pasteur on 09/11/12