In vitro Anti-Malarial Drug Susceptibility of Temperate Plasmodium vivax from Central China

Abstract

In the face of recent increase of Plasmodium vivax malaria in central China, we conducted a study to evaluate in vitro susceptibility of temperate-zone P. vivax parasites to antimalarial drugs. During 2005–2006, in vitro drug susceptibility was measured for 42 clinical P. vivax isolates by using a schizont maturation inhibition technique. Geometric means of 50% inhibitory concentrations (IC₅₀s) and 95% confidence intervals (CIs) were 10.87 (4.50–26.26) ng/mL for chloroquine, 4.21 (1.88–9.42–8) ng/mL for mefloquine, 11.82 (6.20–22.56) ng/mL for quinine, 0.13 (0.09–0.20) ng/mL for artesunate, 18.32 (8.08–41.50) ng/mL for pyrimethamine, and 17.73 (10.29–30.57) ng/mL for piperaquine. The IC₅₀ for chloroquine was lower than those obtained from isolates from Thailand and South Korea, suggesting that chloroquine remained effective against P. vivax malaria in central China. The results further indicated that temperate-zone P. vivax isolates from China were more susceptible to chloroquine, quinine, and mefloquine than isolates from Thailand.

Introduction

Of the four human Plasmodium species, Plasmodium vivax has the widest global distribution and accounts for most malaria infections outside Africa.¹ Currently, it is estimated that 130–145 million P. vivax malaria cases occur annually.² The recent reports of P. vivax infections that produce severe clinical manifestations in malaria-endemic areas of Asia and South America have dramatically changed the view of P. vivax malaria as benign tertian malaria.^3–7 Furthermore, the ability of the parasite to produce multiple relapses in patients after the initial episode causes increased morbidity.⁸

In spite of its significance in public health, first-line therapies for P. vivax malaria, chloroquine (CQ) and primaquine (PQ), have remained unchanged for the past 50 years. The emergence of resistance to CQ and the failing efficacy of PQ regimens in preventing relapses have raised great concerns in the control of P. vivax malaria.^9,10 Clinical CQ-resistant (CQR) P. vivax was first reported from Papua New Guinea in 1989,^11,12 followed by multiple reports from Indonesia,^13–15 Myanmar,^16,17 India,^18,19 Guyana,²⁰ Brazil,^21,22 Colombia,²³ and more recently in Ethiopia²⁴ and South Korea.²⁵ Continued surveillance of P. vivax drug resistance may detect CQR P. vivax parasites in most of its geographic range. In view of this critical scenario, it is well justified to consider P. vivax malaria as the most neglected, highly prevalent, and potentially dangerous disease.^26,27

Malaria in China is concentrated in two subtropical provinces Yunnan and Hainan, where P. vivax and P. falciparum coexist with P. vivax as the predominant species. The malaria eradication campaign in the 1950s had significantly reduced the range and prevalence of malaria in China. This reduction is especially evident in the central region, where P. falciparum has been largely eradicated. Although malaria is hypoendemic to central China, this region has experienced malaria resurgence since 2000; there have been frequent focal outbreaks in areas along the Huai River. Anhui Province was most seriously affected; malaria incidence increased in 2006 in this province to the highest level of any province in China.²⁸ Choroquine–primaquine is still the first-line treatment for P. vivax malaria in China. Although there was a report of chloroquine treatment failure in four cases of P. vivax malaria in Yunnan,²⁹ no assessment of the CQ-PQ efficacy against temperate-zone P. vivax strains has been carried out in central provinces. Therefore, there is an urgent need to monitor drug sensitivity of P. vivax for currently used antimalarial drugs in China during the malaria elimination phase.

Most reports of CQR P. vivax were based on observations of failures after the standard CQ treatment despite adequate blood levels of CQ and desethylchloroquine (> 100 ng/mL).⁹ This finding was largely caused by difficulties involved in culturing P. vivax parasites. In recent years, renewed interests in culturing P. vivax led to improvement of an ex vivo procedure and adaptation of this technique for in vitro drug assays.^30–32 These assays, based on schizont maturation, have been used to evaluate drug sensitivity of tropical P. vivax isolates from southeast Asia and temperate-zone isolates from South Korea.^33–35 We present the first in vitro evaluation of drug sensitivity of P. vivax clinical isolates from a temperate malarious region of central China.

Materials and Methods

Study site and sample collection criteria.

This study was carried out in Bengbu, Anhui Province, China, during the malaria transmission seasons (July–September) of 2005 and 2006. Ethical approval was obtained from the ethical review committees at the National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention and Walter Reed Army Institute of Research. A total of 75 patients ≥ 18 years of age attending county hospitals and diagnosed with P. vivax infections were recruited into this study. Informed consent was obtained from all participants before 2 mL of venous blood was obtained from each patient by venipuncture into a heparinized tube. Parasitemia was determined by microscopic examination of Giemsa-stained smears. Blood was stored at 37°C and transported to the laboratory for in vitro assays within 8 hours of collection. Only samples from patients with parasitemias ≥ 0.01% and who had not received antimalarial treatment in the preceding three weeks were chosen for in vitro drug assays.

In vitro drug assay.

Parasite ex vivo culture was performed essentially as described.³⁰ Briefly, 2 mL of P. vivax-infected blood was centrifuged at 2,000 rpm for 5 minutes, and the pellet was resuspended in two volumes of serum-free RPMI 1640 medium (ICM). Leukocytes were removed by passing the blood through a Plasmodipur^® filter (Clonagen, Paris, France). Erythrocytes were washed twice with serum-free RPMI 1640 medium ICM and resuspended in McCoy's 5A medium containing 25% AB⁺ serum to make a 2% cell suspension. A 100-μL erythrocyte suspension was dispensed into wells of a pre-dosed antimalarial microtiter plate.

The drugs assayed and their concentration ranges were chloroquine (0.1 ng/mL–100 μg/mL), mefloquine (0.1 ng/mL–100 μg/mL), pyrimethamine (0.1 ng/mL–100 μg/mL), quinine (0.1 ng/mL–100 μg/mL), artesunate (0.006–6 ng/mL), piperaquine (0.6 ng/ml–100 μg/mL), and primaquine (8–500 ng/mL). Each drug concentration was tested in triplicate. Control wells without drugs were included in all tests. The test plates were incubated at 37°C in a gas chamber containing 5% CO₂ for 12–40 hours depending on the stage of the parasites on admission. Parasite development in drug-free wells was monitored every 6 hours. When parasites in control wells developed into mature schizonts, all wells were harvested to prepare thick and thin blood films. Blood films were stained with Giemsa and examined under a microscope. Only schizonts with at least three well-defined chromatin dots were counted. Parasite preparations with > 80% of parasites matured to schizonts in control wells were included in the analysis.

Data analysis.

The percentage of parasites matured to schizonts were determined for each drug concentration and compared with those from controls. Data were compared and analyzed by using WinNonlin software version 4.1 (Pharsight Corporation, Mountain View, CA) to determine the 50% inhibitory concentration (IC₅₀) of each drug among different isolates. Geometric mean IC₅₀s and 95% confident intervals (CIs) of each drug were obtained by using SPSS 11.0. (SPSS, Inc., Chicago, IL). Spearman's nonparametric correlation was used to evaluate correlations between IC₅₀s of parasite isolates and different drugs.

Results

In the malaria transmission seasons of 2005–2006, we recruited 75 P. vivax-infected patients for in vivo drug sensitivity study. Parasitemia on admission varied between 0.01% and 1.8% and showed a geometric mean of 0.08% (95% CI = 0.06–0.10%). Among these clinical isolates tested, 42 successfully developed into schizonts in control wells. All cases had mixed stages of parasites before culture. Twenty-five isolates, with mostly rings, were cultured for 24–35 hours, and the other 17 isolates were cultured less than 24 hours. Seven commonly used antimalarial drugs listed above were assayed in parallel and the IC₅₀ for each drug was calculated.

Numbers of isolates that were assayed successfully were 29, 31, 31, 38, 40, and 35 for CQ, mefloquine, pyrimethamine, quinine, artesunate, and piperaquine, respectively. Numbers of isolates that developed under different culture times and drugs are shown in Table 1, and IC₅₀s and 95% CIs are shown in Table 2. Parasites were highly sensitive to mefloquine and artesunate and had IC₅₀s in the lower nanomolar range. For CQ, a recent study showed that P. vivax trophozoites were insensitive to CQ.^36,37 Consistent with this finding, we found that most parasite isolates at the trophozoite stage were significantly less sensitive to CQ; four isolates had IC₅₀s > 100 μg/mL. The mean in vitro IC₅₀s to pyrimethamine, quinine, and piperaquine were 18.32, 11.82, and 17.73 ng/mL, respectively.

Table 1.

Number of Plasmodium vivax isolates included in drug susceptibility assays for each drug, China

Treatment	No. isolates
	Culture time < 24 hours	Culture time ≥ 24 hours	Total
Control	17	25	42
Chloroquine	10	19	29
Mefloquine	13	18	31
Pyrimethamine	13	18	31
Quinine	16	22	38
Artesunate	17	23	40
Piperaquine	14	21	35
Primaquine	0	2	2

Table 2.

50% inhibitory concentrations and 95% confidence intervals of antimalarial drug susceptibility patterns of Plasmodium vivax, China^*

Drug	Culture time < 24 hours		Culture time ≥ 24 hours		All cases
	IC50, ng/mL	95% CI	IC50, ng/mL	95% CI	IC50, ng/mL	95% CI
Chloroquine	28.64	6.27–130.75	6.52	1.98–21.45	10.87	4.50–26.26
Mefloquine	4.15	1.05–16.38	4.25	1.31–13.79	4.21	1.88–9.42
Pyrimethamine	17.26	3.44–86.57	19.13	6.67–54.86	18.32	8.08–41.50
Quinine	20.63	7.53–56.58	7.89	3.14–19.79	11.82	6.20–22.56
Artesunate	0.20	0.12–0.33	0.10	0.05–0.18	0.13	0.09–0.20
Piperaquine	27.40	10.35–72.52	13.27	6.44–27.35	17.73	10.29–30.57
Primaquine	ND	ND	69.75	ND	69.75	ND

^*IC₅₀ = 50% inhibitory concentration; CI = confidence interval; ND = not done.

Primaquine was a poor blood stage schizonticide, and the IC₅₀ could be obtained only from two isolates (mean = 69.75 ng/mL). For the remaining isolates, PQ only partially inhibited schizogony at the highest concentration tested. Positive correlations of IC₅₀ values were detected between quinine and CQ (r = 0.568, P = 0.002), quinine and mefloquine(r = 0.438, P = 0.016), quinine and artesunate (r = 0.408, P = 0.013), piperaquine and mefloquine (r = 0.426, P = 0.030), artesunate and CQ (r = 0.424, P = 0.027) and artesunate and piperaquine (r = 0.416, P = 0.014).

Discussion

In recent years, the resurgence of P. vivax malaria in central China has raised doubts about efficacy of the first-line treatment CQ-PQ. However, no stringent clinical observations on the efficacy of this drug combination have been conducted to date. We tested the sensitivity of P. vivax to a number of commonly used antimalarial drugs by using an in vitro assay based on schizont maturation. The lower success rate (69%) of the assay in our study may have been caused by the extended transportation time (> 6 hours) for some samples from hospitals to the laboratory and suspected use of antibiotics by some patients before admission to the study. Nevertheless, our success using parasite isolates with parasitemias as low as 0.015% demonstrates applicability of this assay to a wide range of parasitemias.

Although assay conditions may have had some slight variations between different studies and may not be directly comparable with results from other studies,^31,32,34 the same technique has been used to evaluate in vitro drug susceptibility of P. vivax parasites in Thailand and South Korea.^33,35 These studies detected similar sensitivities to some drugs (artesunate and CQ) but dramatic differences to others (e.g., quinine, pyrimethamine, and mefloquine), which could be explained by difference in malaria epidemiology and antimalarial drug use history in these countries.

Temperate P. vivax strains are endemic to central China and South Korea. These strains have a long interval between primary infection and relapse in these two countries. However, tropical P. vivax strains have short relapse intervals in Thailand. The intrinsic differences in parasite strains may be partially responsible for the observed differences in drug sensitivities, but different drug histories in these countries may be the major contributor to the difference. In Thailand, P. vivax and P. falciparum co-exist, and cryptic infections with one parasite species often occur in clinical malaria.³⁸ Because drug treatment is primarily dependent on diagnosis by microscopy, which often misses cases of mixed-species infections,³⁹ drugs used to treat one parasite species will inevitably influence sensitivity of the other species.⁴⁰ As a result, P. vivax parasites in these countries have been subjected to different drug pressures, even though CQ-PQ is the drug combination used to treat P. vivax malaria in all countries. In Thailand, melfoquine has been used extensively for more than 20 years to treat P. falciparum malaria,⁴⁰ and P. falciparum has developed resistance to this drug.⁴¹ Plasmodium vivax has been selected for amplification of the P. vivax multidrug resistance 1 gene under pressure of the mefloquine treatment for P. falciparum malaria in the area that these two species were found.^42,43 Quinine had been used for approximately 5 years in the early 1980s, and is often used to treat severe P. falciparum malaria.⁴⁰ Although development of resistance in P. falciparum to quinine is slow, reduced susceptibility has already been detected in clinical studies.⁴⁴ In comparison, mefloquine has not been used in China and South Korea, and quinine has been used for treating severe P. falciparum malaria in southern China. Therefore, these different drug histories are consistent with the observed higher sensitivity of the temperate-zone P. vivax strains to these two drugs.

We detected reduced in vitro sensitivity of P. vivax to pyrimethamine than in isolates from Thailand,³³ but greater sensitivity to this drug than in isolates from South Korea.³⁵ It was suggested the in vitro assay performed in the presence of folic acid may not reflect the in vivo sensitivity of the parasites,³⁵ but values obtained in this study may represent the relative sensitivity of the parasites to this antifolate drug. Pyrimethamine, a component of the two combination regimens Maloprim (plus dapsone) and Fansidar (plus sulfadoxine), has been widely used for malaria prophylaxis in China. This use has led to reduced efficacy of pyrimethamine for prophylaxis and clinical resistance in approximately 35% of patients treated for P. vivax malaria in temperate-zone provinces in the late 1970s.^45,46 Extensive use of pyrimethamine may have posed substantial selection pressure on P. vivax and is probably associated with the reduced in vitro sensitivity to pyrimethamine. These in vitro studies indicate that parasites of different origins displayed great variations in sensitivity to pyrimethamine, which might be correlated with different prevalences of resistance-conferring dihydrofolate reductase alleles.^47,48 The in vitro assay technique we used may be a useful tool for studying relationships between antifolate resistance and mutations in target genes.

The in vitro IC₅₀s of the P. vivax isolates from China (10.87 ng/mL) for CQ were lower than those of isolates from South Korea (39 ng/mL) and Thailand (50 ng/mL). Because in vivo studies showed that the current treatment regimen of CQ-PQ was still effective in Thailand,^49,50 our results suggested that this first-line treatment for P. vivax malaria may also be effective in central China. However, although we did not monitor clinical efficacy of CQ in this study, we observed that one of the four patients enrolled in this study who had parasite isolates with higher in vitro IC₅₀s had recurrent P. vivax infections on day 16, suggesting clinical resistance to CQ in this area.

In addition, the observation of clinical failures after standard CQ treatment in Yunnan Province, where P. vivax and P. falciparum co-exist, raised further concerns.²⁹ Therefore, clinical efficacy of CQ in treating temperate-zone P. vivax malaria in central China needs to be evaluated. Moreover, studies in other P. vivax-endemic regions found great variations in the prevalence of CQR P. vivax parasites. Multiple cross-sectional and longitudinal surveys performed in different regions of Indonesia showed considerable geographic variations in the proportion of CQR P. vivax parasites.^51–57 In Vietnam, CQR P. vivax infection was not found 10 years ago,⁵⁸ but a more recent survey found 25% of patients in a study had recurrent P. vivax infections after standard treatment, suggesting of emerging CQ resistance.⁵⁹ These studies highlight the geographic heterogeneity and temporal dynamics of drug resistance in P. vivax, which further emphasizes the importance of continued surveillance efforts.

Currently, mechanism of CQ resistance in P. vivax is not well understood, but does not appear to be associated with the ortholog of the P. falciparum CQ-resistant transporter gene. Although certain polymorphisms in the P. vivax multidrug resistance 1 gene have been shown to be associated with reduced sensitivity to CQ in vitro³⁴ and in vivo,⁶⁰ this association needs to be further evaluated. Therefore, with the lack of convenient molecular surveillance tools, the ex vivo assay is useful for monitoring drug sensitivity of P. vivax. Standardization of the assay conditions among investigators is needed so that data from different geographic regions can be compared.

Disclaimer: The view of the authors does not purport to reflect the position of the U.S. Department of the Army or Department of Defense.