In vitro susceptibility of Plasmodium falciparum to monodesethylamodiaquine, quinine, mefloquine and halofantrine in Abidjan (Côte d'Ivoire)

Abstract

Background

Malaria is the primary cause of hospitalization in Côte d'Ivoire. Early treatment is one of the strategies to control this illness. However, the spread of resistance of Plasmodium falciparum to antimalarial drugs can seriously compromise this strategy.

Objectives

The aim of this study was to assess the in vitro susceptibility of P. falciparum to monodesethylamodiaquine and aminoalcohols in Abidjan (Côte d'Ivoire).

Methods

We assessed the in vitro susceptibility of isolates collected from patients with uncomplicated malaria by using the WHO optical microtest technique.

Results

The proportions of resistance to monodesethylamodiaquine, méfloquine and halofantrine were 12.5%, 15.6% and 25.9%, respectively. For quinine, none of isolates showed evidence of in vitro resistance. However, two isolates (6.1%) had IC₅₀ values above 300 nM. The IC₅₀ of each drug was positively and significantly correlated to that of the other three drugs, and the correlation was higher between halofantrine and mefloquine.

Conclusions

Our results showed that the in vitro chloroquine resistance reported in previous studies has been extended to other antimalarial drugs investigated in this study except for quinine. Therefore, it is necessary to implement a long-term monitoring system of antimalarial drug resistance.

Introduction

Malaria still remains a major public health problem in sub-Saharan African countries¹. Several strategies, such as rapid diagnosis and appropriate treatment, have been recommended to control malaria. However, faced with the growing inefficiency of monotherapies, as in most countries, the Ministry of Public Health in Côte d'Ivoire has adopted a novel strategy based on the use of drug combinations including artemisinin derivatives (artemisinin-based combination therapy, i.e ACT). Quinine is reserved for curative treatment in case of treatment failure of ACTs or severe and complicated malaria. Sulfadoxine-pyrimethanine (SP) is used to prevent malaria in pregnant women ^{2, 3}. Unifortunately, these recommendations are not always followed by drug prescriptors ⁴, thereby increasing the probability of selection and spread of drug-resistant Plasmodium falciparum strains.

Because of the presence of mefloquine and amodiaquine in some ACTs, it is necessary to assess the susceptibility of P. falciparum to these antimalarial drugs by in vitro and/or in vivo tests.

Moreover, a decrease in the sensitivity of P. falciparum to quinine has been reported in Southeast Asia, East Africa and South America ^{5, 6, 7}. As quinine has been used for decades to treat severe and complicated malaria until now, resistance to quinine could lead to a public health disaster. It is therefore necessary to implement an improved program for monitoring drug-resistant malaria in order to plan and adopt appropriate strategies to control this disease. Several methods can be used to evaluate the susceptibility of P. falciparum to antimalarial drugs. Laboratory tools, such as in vitro drug sensitivity assays, can provide an early warning to orient therapeutic efficacy studies and antimalarial treatment policy ⁸. Furthermore, cultivation of clinical isolates and measurement of their susceptibility to antimalarial compounds in vitro remove host-related variables, such as patients' compliance, nutritional status, immune status, re-infection and pharmacokinetics, thereby providing a powerful technique for detecting the emergence of drug-resistant parasites ⁹. The aims of the present study were: a) to assess the in vitro susceptibility of clinical isolates of P. falciparum to monodesethylamodiaquine, quinine, mefloquine, and halofantrine, and b) analyse the potential for cross-resistance between these drugs.

Methods

Study area

The study was carried out between February 2006 and February 2007 in the district of Abobo, situated in the north of Abidjan (the economic capital city). In this area, malaria is hyperendemic with seasonal transmission. The most common vectors are Anopheles gambiae ss and A. funestus.

Isolates of P. falciparum

Patients aged between 2 to 45 years presenting signs and symptoms of uncomplicated malaria were recruited at El Rapha and Anokoua Kouté, two health centers of Abobo area. Informed consent was obtained from the patients or guardian accompanying the sick children. The study was approved by the Ethics Committee of the Ivorian National Institute of Public Health (NIPH).

Venous blood samples were collected in EDTA-coated Vacutainer® tubes (Terumo Europe N.V., Leuven, Belgium) before treatment. They were transported at 4°C to NIPH within 6 h, if the parasitemia was at least 4,000 asexual parasites/µl of blood.

Parasitized erythrocytes were washed three times in RPMI 1640 medium (Invitrogen, UK), and Giemsa-stained thin blood smears were examined under the microscope to determine the parasite density and confirm the Plasmodium species (P. falciparum monoinfections).

Samples with parasitemia ranging from 0.1% to 0.25% were used directly to test drug susceptibility. If parasitemia exceeded 0.25%, infected erythrocytes were diluted to this parasitemia range with uninfected erythrocytes. Patients were treated with amodiaquine-artesunate or artemether-lumefantrine according to the recommended national therapeutic regimens.

Drugs

The test compounds were obtained from the following sources: monodesethylamodiaquine (TDR/World Health Organization [WHO] Drug Discovery Research), quinine chlorhydrate (Sanofi-Aventis, Antony, France), mefloquine hydrochloride (Roche, Mannheim, Germany), and halofantrine (Glaxo Smith Kline, Evreux, France). Stock solutions of each drug were prepared in 70% methanol. Twofold serial dilutions were prepared in RPMI 1640 medium and distributed in triplicate into 96-well culture plates.

In vitro assay

The WHO microtest technique was used to measure the inhibition of schizont maturation by microscopy¹⁰. Washed infected erythrocytes were suspended in RPMI 1640 with 10% human serum, 25 mM HEPES, and 25 mM NaHCO₃ at a hematocrit of 1.5%. Fifty microliters of the blood-medium mixture were distributed into each well of the predosed 96-well tissue culture plates and incubated at 37°C in candle jars for 42 h according to standard methodology. Final concentrations were ranged from 3.125 to 400 nM for monodesethylamodiaquine and mefloquine, from 12.5 to 1600 nM for quinine and from 0.25 to 32 nM for halofantrine. After incubation period, parasites were harvested and Giemsa stained thick blood films were prepared. The number of schizonts, defined as schizonts with more than 3 nuclei, was counted per 200 asexual parasites. Isolates with less than 20% of schizonts in drug-free control well were excluded. The results were expressed as 50% inhibitory concentration values (IC₅₀).

The cut-off values for in vitro resistance to monodesethylamodiaquine, quinine, mefloquine, halofantrine were fixed at 60 nM ¹¹, 800 nM ¹², 30 nM ¹³, 6 nM ¹⁴ respectively.

Statistical analysis

The IC₅₀ values were determined by nonlinear regression analysis of the plot of logarithm of concentration against growth inhibition. Data were adapted to fit the logprobit model (Excel; Microsoft, Redmond, WA). The in vitro response was expressed as the geometric mean IC₅₀ values with 95% confidence intervals.

The degree of correlation between different antimalarial drugs was estimated by the Spearman correlation coefficient (rho) and the coefficient of determination (r²). The level of significance was set at 0.05.

Results

Forty three isolates of P. falciparum were collected. In this study, asexual parasite densities ranged from 0.1% to 13.5%. The following proportions of isolates were successfully cultured for each drug tested: 74.4% (32/43) for monodesethylamodiaquine, 76.7% (33/43) for quinine, 76.7% (33/43) for mefloquine and 62.8% (27/43) for halofantrine. The geometric mean IC₅₀s of four antimalarial drugs tested are summarized in Table 1.

Table 1.

The geometric mean IC₅₀s of four antimalarial drugs tested against wild isolates of P. falciparum in Abidjan

Drugs	No. of isolates	Geometric Mean	95% Confidence	Range (nM)
	tested	(nM) IC50	interval (nM)
				Minimum	Maximum
Monodesethylamodiaquine	32	13.3	12 – 14.6	2.04	269
Quinine	33	60.6	59.6 – 61.7	4.31	486
Mefloquine	33	7.45	6.41 – 8.49	1.11	64.2
Halofantrine	27	1.64	0.93 – 2.35	0.17	20.4

Four (12.5%) isolates were resistant to monodesethylamodiaquine, and five (15.6%) monodesethylamodiaquine-sensitive isolates showed IC₅₀ up to 25 nM. Five (15.2%) and seven (25.9%) isolates showed in vitro resistance to mefloquine and halofantrine, respectively. For each antimalarial drug, there were three sensitive isolates which showed borderline sensitivity (i.e IC₅₀ > 20 nM but < 30 nM for mefloquine and > 4 nM but < 6 nM for halofantrine). There was no resistance to quinine (Figure 1). Two isolates (6.1%) presented quinine IC₅₀ up to 300 nM.

Figure 1.

Rate of in vitro resistance of Plasmodium falciparum

Concerning cross-resistance, one isolate was resistant in vitro to monodesethylamodiaquine, mefloquine and halofantrine, three isolates were resistant to monodesethylamodiaquine and halofantrine, and additional three isolates were resistant in vitro to halofantrine and mefloquine.

The IC₅₀ of each drug was positively and significantly correlated to that of the other three drugs, and correlation was highest between halofantrine and mefloquine. Mefloquine and quinine IC₅₀s were weakly correlated (Table 2).

Table 2.

Correlation between the IC₅₀s values of the four antimalarial drugs tested

Drug pair	No. of	r	r2	p*
	isolates
	tested
Monodesethylamodiaquine	27	0.52	0.27	0.0003
- Halofantrine
Monodesethylamodiaquine	32	0.56	0.32	<0.0001
- Quinine
Monodesethylamodiaquine -	32	0.24	0.06	<0.0001
Mefloquine
Halofantrine - Quinine	27	0.32	0.10	< 0.0001
Halofantrine -Mefloquine	27	0.67	0.46	0.0039
Mefloquine - Quinine	33	0.16	0.03	< 0.0001

^*Student Fisher test

Discussion

In our study, the proportion of monodesethylamodiaquine-resistant isolates was higher than that described in previous studies in Africa, which was between 2 and 7% ^{15, 16, 17}. Our result can be explained by the high rate of chloroquine resistance in Abidjan area^{18, 19} and the similar chemical structure between amodiaquine and chloroquine. In Cameroon, IC₅₀ values ranging from 25.6 to 115 nM were reported for most of the isolates collected at the time of treatment failure with amodiaquine, indicating that the threshold for monodesethylamodiaquine resistance in vitro might be lower than the usual value of more or equal to 60 nM ²⁰. On this basis, we can say that the rate of decreased sensitivity to monodesethylamodiaquine (i.e IC₅₀s > 25 nM) was 28.1% in our study. If the increase in clinical resistance to amodiaquine is confirmed, this situation could compromise the current efficacy of ACT which contains amodiaquine. Indeed, while currently employed ACTs demonstrate excellent clinical efficacy, the history of antimalarial chemotherapy predicts that it is only a matter of time before parasite resistance emerges ²¹.

All isolates tested in our study were sensitive to quinine, as in the previous in vitro susceptibility tests carried out in Côte d'Ivoire ^{18, 22}. These data suggest that quinine still highly effective and confirm the choice to treat severe malaria or treatment failures with this drug. However, we must monitor quinine susceptibility of P. falciparum isolates because of the increasing use of quinine as presumptive treatment for uncomplicated malaria, often without respecting the recommended therapeutic protocol and dosage ²³.

This raises the question as to whether drug pressure due to quinine use in urban areas selects parasites with decreased sensitivity to quinine ²⁴. In Senegal, the prevalence of in vitro resistance to quinine was 5% ²⁵, while it was 3% in Comoros ¹⁷, 6% in Congo ¹⁵ and 8% in Rwanda ¹⁶ with intermediate susceptibility to quinine. In Guyana, a reduced P. falciparum sensitivity to quinine was observed in 6/14 isolates tested ²⁶. In Asia, where decreased in vitro susceptibility to quinine was first reported at the beginning of the 1980s in patients living near the Thai-Cambodia border ²⁷ treatment failures with this drug occurred subsequently ⁵. Thus, it is necessary to evaluate the therapeutic efficacy of quinine in patients.

Despite the uncommon use of mefloquine compared to other antimalarial drugs in Côte d'Ivoire, 15.2% of mefloquine-resistant isolates were observed in our study. The presence of isolates that are naturally less sensitive to mefloquine could partially explain this proportion of resistant isolates. In Senegal, where there was 13% of isolates with reduced susceptibility to mefloquine, prophylactic failures with this drug were previously described ^{25, 28}. The same situation could exist in Côte d'Ivoire. Indeed, in this country, mefloquine is one of the drugs recommended to prevent malaria in nonimmune populations such as tourists ³. Elsewhere in Africa, in particular in Madagascar and Central African Republic, there were only 2% of in vitro resistance to mefloquine ^{29, 30}.

The prevalence of in vitro halofantrine resistance was the highest in our study. In 2002–2003, we found 3 / 11 (27.3%) isolates tested resistant in vitro to halofantrine ³¹. The data reported in this current study indicate that P. falciparum susceptibility to halofantrine has been stable. From 1994 to 2005, there was an alert issued on halofantrine resistance in French Guiana with a peak of 66% of prevalence of resistance in isolates from 2000 ³². In Burkina Faso, where the rate of in vitro resistance to halofantrine was 11.2%, the authors attributed this rate to the presence of isolates naturally resistant, as with mefloquine ²³. Indeed, we observed a strongly positive correlation between halofantine and mefloquine, more than with the other drugs. This positive correlation between two aminoalcohols may be partly explained by their similar chemical structure ^{15, 33, 34}. The correlation between monodesethylamodiaquine and aminoalcohols has been previously described^{16, 34}.

A positive correlation between the IC₅₀s of two antimalarial drugs may suggest in vitro crossresistance ³⁵ although we did not observe cross-resistance with quinine in our study.

Conclusion

In conclusion, our results showed that the in vitro P. falciparum resistance already observed with chloroquine has extended to other antimalarial drugs investigated in this study except for quinine. For quinine, the presence of isolates with reduced susceptibility and correlation with other antimalarial drugs need further investigations.