Clinical and molecular surveillance of drug resistant vivax malaria in Myanmar (2009–2016)

The study included the retrospective analysis of previously conducted therapeutic efficacy studies (TES) of chloroquine in uncomplicated vivax malaria in Myanmar and prospective multi-site, longitudinal study by clinical and molecular markers analysis. From 2009 to 2012, TES on vivax malaria was conducted in nine sentinel sites (Fig. 1) which covered most of the malaria-endemic areas in Myanmar. In 2012, Shwegyin, in the southern part of central Myanmar, was selected for TES of chloroquine as this study site has a high burden of both falciparum and vivax malaria among migrant goldmine workers. In 2015–2016, TES of chloroquine on vivax malaria was conducted in Buthidaung, in the western border area and Kawthaung, southern Myanmar.

All procedures on case selection, recruitment and follow-up were carried out according to the standardized protocol recommended by WHO [4]. Briefly, uncomplicated vivax mono-infected patients over 6 years old with fever or history of fever within previous 48 h, were recruited and treated with chloroquine standard dose calculated by body weight, followed by observation with follow-up schedule, i.e., day 3, 7, 21 and 28 after treatment. Clinical and parasitological assessment was done on each follow-up day. All anti-malarials used in this study were provided by the national malaria control programme.

Blood samples were taken on day 0 and all follow-up days for microscopic examination and molecular analysis. Blood film examination on peripheral blood smear was carried out after 3% Giemsa stain for 45 min. Blood film examination was done as described [4] and calculated as parasite count per µL of blood. For molecular analysis, finger-prick blood samples were collected on filter paper (Whatman©), dried and stored in plastic bags with desiccant until analysis. Molecular markers analysis was conducted only in prospective study sites such as Kawthaung, Shwegyin and Buthidaung.

All day-0 samples were subjected to analysis for distribution and spread of chloroquine and antifolate drug resistance markers. Filter papers were prepared and extracted for parasite genomic DNA using QIAamp DNA Blood Mini Kit (QIAGEN) according to manufacturer’s instruction. The target genes were amplified by using the specific pair of primers (Table 1). In this study, chloroquine resistance marker, ‘AAG’ insert in pvcrt-O; multi-drug resistance marker, mutations in pvmdr1; antifolate resistance makers, mutations in pvdhps and pvdhfr were amplified and analysed.

PCR reactions were performed in a reaction mixture that contained 0.25 mM of each dNTP, 10 mM Tris–HCl (pH 9.0), 30 mM KCl, 1.5 mM MgCl₂, 1.0 units of Taq polymerase (Accupower^© premix, Bioneer, Seoul, Korea), 0.02 µM primers, and 2 µL of genomic DNA.

For target gene amplifications, initial denaturation at 95 °C for 10 min was followed by 35 cycles of 95 °C for 30 s, 58 °C (pvdhps) or 60 °C (pvcrt-O) or 62 °C (pvdhps and pvdhfr) for 45 s, 72 °C for 1 min, and a final extension of 72 °C for 10 min. Amplified products were checked by 1% agarose gel electrophoresis stained with Red safe^© (iNtRON, Seongnam, Republic of Korea). The PCR clean-up was proceeded by MEGAquick-spin DNA fragment purification Kit (iNtRON, Republic of Korea) and sequencing. The nucleotide and amino acid sequences were aligned and analysed by Lasergene^® software (DNASTAR, Madison, WI, USA) using the reference strain of Sal-1 retrieved from Plasmodium data base [5]. The nucleotide sequences were submitted to GenBank under accession numbers KX000945–KX000959.

Participation in this study was entirely voluntary. Written consent was taken from all participants. This project obtained ethical clearance from the ethical committee of the Department of Medical Research, Republic of the Union of Myanmar (Approval No-52/Ethics, 2012 and 31/Ethics, 2015) and institutional ethical committee of the Kangwon National University, Republic of Korea (KWNUIRB-2016-04-005).

During 2009–2012, TES of chloroquine on 906 uncomplicated vivax malaria infected cases was conducted in nine sentinel sites in Myanmar (Fig. 1). LTF cases, according to the WHO definition [4], were reported in Kawthaung (southern Myanmar), Sagaing and Buthidaung (western Myanmar) sites only while there was no treatment failure cases until day 28 follow-up in remaining study sites. The highest failure rate was noted in Kawthaung in 2012 (18/60, 30%). Kawthaung and Buthidaung site showed failure cases in every study period. There was no early treatment failure case in all study sites.

Kawthaung (southern Myanmar), Shwegyin (central Myanmar) and Buthidaung (western Myanmar) were selected for prospective longitudinal study after treatment with chloroquine regimens. A total of 208 uncomplicated vivax cases were included in this study with mean age of 24.2 (±9.03) years and geometric mean of the parasite density of 4160 with 95% CI (3567–4851) (Table 2). There was no clinical failure case in Shwegyin, but one LTF case in Kawthaung at day 28 (1/60, 1.7%) and two in Buthidaung at days 21 and 28 (2/60, 3.3%).

All day-0 samples collected from three sentinel sites were analysed for molecular markers (Fig. 2; Table 3). More than half of the samples (133/204, 63.9%) showed K10 ‘AAG’ insertion in chloroquine resistance transporter gene, pvcrt-O. Among them, Shwegyin showed the highest mutant rate (64/88, 72.7%) followed by Kawthaung (40/60, 66.7%) and Buthidaung (29/60, 48.3%). Similarly, pvmdr1 (Y976F) was found in Kawthaung, Shwegyin and Buthidaung as 16/60, 26.7%; 15/88, 17.0% and 1/60, 1.7%, respectively. Interestingly, significant highest mutant rate of F1076L was observed in Buthidaung (38/60, 63.3%) (p = 0.067). To estimate the chloroquine resistance status, pvcrt-O (K10 insertion) and pvmdr1 mutations (Y976F and F1076L) were analysed together. Only K10 insertion was lowest in Buthidaung (29/60, 48.3%) with highest rate of F1076L alone or with K10 insertion together. However, the number of isolates showing K10 insert in pvcrt-O gene as well as both pvmdr1 mutations (Y976F and F1076L) was highest in Kawthaung (11/60, 16.7%) followed by Shwegyin (11/88, 12.5%) and no isolate in Buthidaung (Tables 3, 4). Similarly, all mutations of pvdhps (S382A, A383G, K512M, A553G) showed the highest in Kawthaung (17/60, 28.3%) followed by Shwegyin (6/88, 6.8%) and Buthidaung (2/60, 3.3%) (Table 3). Moreover, half of the samples showed the wild type pvdhps alleles while no wild type in Kawthaung and one only (1/88, 1.1%) in Shwegyin (Table 4).

Furthermore, mutations in pvdhfr gene were analysed and all mutations, F57L/I, S58R, T61M, and S117T/N, showed the highest mutant rate in Shwegyin (70/88, 79.5%) followed by Kawthaung (45/60, 75.0%). Wild type alleles were found only in Buthidaung (5/60, 8.3%). Triple or quadruple mutant alleles of pvdhps was observed and accounted for more than 50% of samples in all study sites (Table 4). A combined analysis of all mutations in 208 samples totaled 80 different genotypes (Additional file 1).