Emergence of chloroquine-sensitive Plasmodium falciparum and rising resistance to first-line artemisinin partner drugs in Malawi

ABSTRACT

The emergence and spread of Plasmodium falciparum resistance to malaria drugs pose a major threat to malaria control efforts. This study assessed the prevalence of molecular markers associated with resistance to key antimalarials in P. falciparum clinical isolates from Mzuzu and Lilongwe in Malawi. These two regions have high human mobility and are strategically located near the border with Zambia and Tanzania, respectively. A total of 1582 blood samples were collected from individuals who visited hospitals for diagnosis between December 2020 and June 2021. P. falciparum infections were confirmed using nested and quantitative PCR, and drug resistance marker genes (pfmdr1, pfcrt, pfk13, pfatp6, pfdhfr, and pfdhps) were sequenced by Sanger sequencing. No resistance-associated mutations were detected in pfk13 and pfcrt genes, supporting continued susceptibility to artemisinin derivatives and chloroquine (CQ). However, the pfmdr1-NFD haplotype, linked to reduced lumefantrine (LUM) susceptibility, was present in 159/371 (42.9%) isolates. Notably, the quadruple pfdhfr-pfdhps mutant haplotype (AIRNVI-SGEAA), associated with high-level sulfadoxine-pyrimethamine (SP) resistance, was found in 287/328 (87.5%) of isolates. These findings highlight the ongoing risk of declining efficacy of LUM partner drugs in artemisinin-based combination therapies (ACT) and reduced SP effectiveness for intermittent preventive treatment in pregnancy (IPTp). The absence of pfcrt mutations, together with the presence of wild-type pfmdr1 alleles lacking CQ-related mutation, suggests the re-emergence of CQ-sensitive parasites. Continuous molecular surveillance, alongside clinical efficacy studies, is essential to inform treatment policies and prevent the spread of drug-resistant malaria in Malawi.

Introduction

Despite the global decline in malaria incidence over the past decade, the disease remains one of the leading causes of mortality, particularly in sub-Saharan African countries. An estimated 263 million malaria cases and 597,000 deaths were documented in 2023, with WHO African countries accounting for more than 94% of both cases and fatalities [1]. Plasmodium falciparum is responsible for over 94% of global malaria cases and deaths caused by Plasmodium parasites. The development of resistance to antimalarial drugs is a significant obstacle to efforts aimed at controlling and ultimately eradicating malaria [2, 3].

In 1993, Malawi shifted its malaria treatment policy from using chloroquine (CQ) as a first-line drug to sulfadoxine-pyrimethamine (SP), and then in 2007 it adopted lumefantrine (LUM) in combination with artemether (ATM) [4]. Like in many countries in Sub-Saharan Africa, sulfadoxine-pyrimethamine is still in use for the intermittent preventive treatment in pregnancy (IPTp) [5, 6]. Although resistance of artemether-lumefantrine therapy (ALU) has not been reported in Malawi, monitoring of its efficacy is crucial as selection of resistance to the partner drugs in artemisinin combination therapy (ACT) (e.g. LUM) is anticipated [7]. This may relatively reduce ACT cure rates and, consequently, therapeutic outcome in malaria-endemic countries. Although artemisinin-resistant (ART-R) parasites and those with reduced susceptibility to partner drugs have been reported in neighbouring Tanzania [8–10] and Zambia [11, 12], surveillance in Malawi remains limited, highlighting the importance of assessing the prevalence of these resistance markers locally.

The burden of malaria in Malawi has been substantially reduced following the implementation of vector control measures and the use of ACT for the treatment of uncomplicated malaria [13]. This reduction in parasite burden has changed the malaria landscape to pockets of highest transmission and low transmission strata within the same geographical units [14]. Under conditions of intense drug pressure, such heterogeneity may favour the selection and spread of drug resistance parasites. The rapid emergence of resistance to SP prompted the shift in malaria treatment policy from SP to ALU in 2007. This points to the critical need to systematically monitor mutations associated with ALU resistance across the region [15–17].

Polymorphisms in the P. falciparum chloroquine resistance transporter (pfcrt) gene are key predictors of CQ resistance and treatment failure [18]. Mutations in pfcrt are also associated with resistance to other 4-aminoquinoline drugs, including amodiaquine (AQ) and piperaquine (PPQ) [19]. In addition, mutations in the P. falciparum multidrug resistance 1 (pfmdr1) and the kelch 13 (pfk13) gene have been identified as validated molecular markers for resistance to endoperoxides such as artemisinin (ART), artesunate (AS), artemether (ATM), and dihydroartemisinin (DHA) [20, 21]. While earlier studies suggested the role of P. falciparum sarco/endoplasmic reticulum Ca2+-ATPase ortholog (pfatp6) in artemisinin resistance, subsequent research has not confirmed this association, and it is no longer considered a reliable marker for artemisinin resistance [22]. Furthermore, polymorphism in pfmdr1 has been shown to confer resistance to aryl-amino alcohols, including lumefantrine (LUM), mefloquine (MFQ), and quinine (QN) [23]. Resistance to SP is associated to point mutations in the P. falciparum dihydropteroate synthase (pfdhps) and dihydrofolate reductase (pfdhfr) genes [24]. This study investigated the prevalence and profiles of polymorphisms in pfmdr1, pfcrt, pfk13, pfatp6, pfdhfr, and pfdhps in clinical isolates collected from Mzuzu and Lilongwe in Malawi. It aimed to assess antimalarial resistance using these molecular markers and provide baseline data to inform evidence-based treatment policies.

Materials and methods

Population and sampling

This study was conducted in two specific geographic locations in Malawi, with Mzuzu located in the northern region and Lilongwe in the central region. These sites were strategically selected due to their positions along major cross-border transport and trade corridors. Mzuzu lies near the border with Tanzania, while Lilongwe is connected to Zambia. Both regions experience significant formal and informal human movement across the border, which increases the risk of importation and may facilitate the spread of P. falciparum resistant variants across national borders. Monitoring these locations is therefore essential for early detection of antimalarial drug resistance.

This cross-sectional study was conducted from December 2020 to June 2021 at Mzuzu and Lilongwe referral hospitals in Malawi. Blood samples were collected via finger-prick from all patients aged six months or older who attended the health facilities and provided informed consent. Participants who declined consent or were in critical condition and unable to comply with study procedures were excluded. Approximately 50 µL of blood was collected from each participant and spotted onto Whatman 903 filter papers (Whatman, Maidstone, UK). After drying, the dried blood spots (DBS) were individually stored in desiccant plastic bags and transported to Kangwon National University (KNU) for laboratory analysis. Malaria infection was initially assessed at the hospitals using microscopy and rapid diagnostic tests (RDTs), and all samples were subsequently reconfirmed for malaria positivity at KNU by PCR-based detection. Drug resistance genotyping for P. falciparum was performed on all confirmed positive samples.

Ethical considerations

The protocols for this study were approved by the National Health Science Committee (NHSRC), a department of the Ministry of Health (MoH) in Malawi (IRB00003905) and the Ethical Review Boards at Kangwon National University (KWNUIRB-2023-05-008). All participants provided consent prior to their participation in the study. Blood specimens were collected only when participants (parents or guardians for children) consented and agreed to participate.

Preparation of genomic DNA from positive controls and patient samples

Genomic DNA from the positive control was obtained using P. falciparum 3D7 strain parasites cultured in vitro as described previously [25]. Briefly, P. falciparum 3D7 parasites were adapted and maintained in O+ human erythrocytes at 4% haematocrit in RPMI 1640 medium (Gibco: Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 25 mM HEPES, 0.2% sodium bicarbonate, 0.5% Albumax II, 50 µg/mL hypoxanthine, and 50 µg/mL gentamicin. Cultures were incubated at 37°C under a gas mixture of 5% CO₂, 5% O₂, and 90% N₂, with daily medium changes. Ring-stage parasites at approximately 5% parasitaemia (corresponding to 1.25×10⁵ parasites/µL) were harvested for genomic DNA extraction.

Genomic DNA from P. falciparum 3D7 culture and DBS from patient samples was extracted using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. DNA concentration and purity were measured using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific) at 260/280 nm. Samples with an A260/280 ratio between 1.8 and 2.0 were considered acceptable and used for downstream nested PCR (nPCR) and probe-based quantitative PCR (qPCR) analyses.

Diagnosis of Plasmodium falciparum infections by nested PCR

A set of specific primers rPLU_F: TCAAAGATTAAGCCATGCAAGTGA, rPLU_R: CCTGTTGTTGCCTAAAACTTC were used for the first PCR and rFAL_F: TTAAACTGGTT GGGAAAACCAAATATATT, rFAL_R: ACACAATGAACTCAATCATGACTACCCGTC for the nested diagnosis PCR [26]. PCR mixture was prepared in a total volume of 20 µL comprising AccuPower PCR Premix (Bioneer, Daejeon, Republic of Korea), 1 µL of template DNA extracted from collected samples, 17 µL of DDW, and 1 µL of each 10 µM primer. For both primary and nested PCR, conditions entailed an initial denaturation step at 95°C for 10 min, followed by 35 cycles of 95°C for 30 s, 56°C for 30 s, and 72°C for 1 min, with a final extension at 72°C for 10 min. The DNA amplicons were run on 2% agarose (Invitrogen, Carlsbad, CA, USA) gels in 1× TBE buffer (Invitrogen) and visualized using a UV trans-illuminator. P. falciparum 3D7 cultured parasite gDNA and sterile distilled water (DDW) were used as positive and negative controls for all runs, respectively.

Diagnosis of P. falciparum infections by qPCR

Primers and probe sequences specific to P. falciparum for the gene encoding the 18S rRNA gene were designed and qPCR was performed as described previously (Forward: ATTGCTTTTGAGAGGTTTTGTTACTTT; reverse: GCTGTAGTATTCAAACACAATGAACTCAA; probe: CATAACAGACGGGTAGTCAT, labelled with 5′-6-carboxyfluorescein [FAM]) [26]. Each qPCR was carried out in a total volume of 25 µL, consisting of 12.5 µL of 2X PrimeTime Gene Expression Master Mix with ROX reference dye (Integrated DNA Technologies, Coralville, IA, USA), 2 µL of genomic DNA, 10 µM species-specific primers, and 5 µM of a TaqMan probe. Reactions were performed on an AriaMx Real-Time PCR System (Agilent Technologies, Santa Clara, CA, USA) with cycling conditions of hot-start polymerase activation at 95°C for 10 min, followed by 40 cycles of denaturation at 95°C for 10 s and annealing-elongation at 60°C for 65 s. Amplification curves were evaluated using Agilent AriaMx software (Agilent Technologies) and visualized with GraphPad Prism 8.0.2. P. falciparum positivity was determined using a standard curve generated from 10-fold serial dilutions of cultured 3D7 genomic DNA, ranging from 125,000 to 0.125 parasites/µL, with all samples analysed in duplicate and DDW serving as the negative control [25].

Amplification and sequencing of drug resistance marker

Genotyping was performed on any PCR-positive P. falciparum samples. The nPCR was carried out to amplify the pfmdr1 (PF3D7_0523000), pfcrt (PF3D7_0709000), pfk13 (PF3D7_1343700), pfatp6 (PF3D7_0106300), pfdhps (PF3D7_0810800), and pfdhfr (PF3D7_0417200) genes, with sequences retrieved from PlasmoDB (https://plasmodb.org/). The pfmdr1 was divided into two fragments (pfmdr1_F1 and pfmdr1_F2), pfcrt into three fragments (pfcrt_F1, pfcrt_F2, and pfcrt_F3), and the pfatp6 into two fragments (pfatp6_F1 and pfatp6_F2) while other genes were amplified as single fragments (Supplementary Table 1). PCR amplification for each fragment was performed separately in 20 µL reactions using AccuPower PCR Premix (Bioneer), 1 µL of the template DNA, 1 µL of each 10 µM forward and reverse primers (Supplementary Table 1), and 17 µL of DDW. PCR conditions were similar for all fragments, except for fragment-specific annealing temperatures (Supplementary Table 1) and elongation times based on target size (1 min per 1 kbp). Briefly, PCR included an initial denaturation at 95°C for 10 min, followed by 35 cycles of denaturation at 95°C for 30 s, annealing at the optimized temperature for 30 s, elongation at 72°C based on fragment size, and a final extension at 72°C for 10 min. PCR products were visualized on 2% agarose gels in 1× TBE buffer and purified using the PureLink™ Quick Gel Extraction Kit (Invitrogen).

For each target gene, PCR was validated using the P. falciparum 3D7 reference strain as a positive control and DDW as a negative control. Sequencing primers were first validated on 3D7 to confirm amplification and sequencing conditions. Purified PCR amplicons were then subjected to Sanger sequencing using an ABI3730xl DNA analyzer (Nbit, Chuncheon, Republic of Korea). Sequencing data quality was carefully assessed by visual inspection of chromatograms to ensure accurate base calling and reliable estimation of multiplicity of infection (MOI).

Data analysis

Sanger sequencing data were analysed and visualized using SnapGene® 2.3.2 software. Sequence alignments were performed against the P. falciparum 3D7 reference genome obtained from the National Center for Biotechnology Information (NCBI) database. The prevalence of each point mutation was calculated as a percentage. Differences in haplotype and mutation frequencies between two study sites were assessed using the Fisher's exact test in GraphPad Prism version 8.0.2. The detailed gene sequences are available at GenBank (accession numbers: PX227571 – PX229438).

Results

Prevalence of Plasmodium falciparum

A total of 1582 blood samples were collected, comprising 1186 from Mzuzu and 396 from Lilongwe. According to the 18S rRNA gene nPCR and qPCR, 132 (11.1%) and 283 (23.9%) isolates from Mzuzu, and 162 (40.9%) and 214 (54.0%) from Lilongwe were positive for P. falciparum, respectively (Figure 1). Any isolate testing positive by either nPCR or qPCR was selected for sequencing of drug resistance markers in pfmdr1, pfcrt, pfk13, pfatp6, pfdhfr, and pfdhps. Sequencing analysis was performed only on amplicons that yielded high-quality sequences suitable for analysis, which varied among the different target genes. The overall process of sample screening, inclusion, and genotyping is summarized in the brief flow chart (Figure 1).

Figure 1.

Location of sampling sites, distribution of positive diagnosed samples, and resistance markers of Plasmodium falciparum in Malawi (n = 1582). Colours indicate district-level Plasmodium falciparum predicted risk (PfPR) in children aged 2–10 years in Malawi. Map was generated using QGIS 3.34.1.

Drug resistance marker gene polymorphisms

pfmdr1

We successfully genotyped pfmdr1 gene and sequenced 227 isolates from Mzuzu and 144 from Lilongwe (Figure 1). The pfmdr1-184F mutant was found in a total of 102 (44.9%) and 57 (39.6%) isolates from Mzuzu and Lilongwe, respectively (Table 1). Similarly, a synonymous mutation GGT to GGG coding for glycine at the 182 position was observed in 10 (6.9%) from Lilongwe and 2 (0.9%) from Mzuzu and all isolates with this mutation had the 184F mutant polymorphism. Although the 184F mutant appeared more frequently in Lilongwe than Mzuzu, the overall distribution of wild-type haplotypes (N86, Y184, D1246) and mutant haplotypes (N86, 184F, D1246) did not differ significantly between the two sites (Fisher's exact test, p = 0.334) (Table 1 and Figure 2(A)).

Table 1.

Frequency of pfmdr1, pfcrt, pfk13, pfatp6, pfdhfr, pfdhps, and pfdhfr-pfdhps haplotypes.

Drug resistance marker (n)	Location (n)		Number of mutations	Mutant haplotypes	Location, n (%)		Total, n (%)
	Mzuzu	Lilongwe			Mzuzu	Lilongwe
pfmdr1 (371)	227	144	None	Wild-type HNYSND	125 (55.1)	87 (60.4)	212 (57.1)
			Single	HNFSND	102 (44.9)	57 (39.6)	159 (42.9)
pfcrt (316)	187	129	None	Wild-type CKTHCFIFVMGI	187 (100.0)	129 (100.0)	316 (100.0)
pfk13 (250)	160	90	None	Wild-type MYNFRINAPCQV	160 (100.0)	90 (100.0)	250 (100.0)
pfatp6 (266)	170	96	None	Wild-type HSLAAS	134 (78.8)	70 (72.9)	204 (76.7)
			Single	YSLAAS	4 (2.4)	3 (3.1)	7 (2.6)
			Single	HNLAAS	0 (0.0)	2 (2.1)	2 (0.8)
			Single	HSLKAS	32 (18.8)	21 (21.9)	53 (19.9)
pfdhfr (339)	221	118	None	Wild-type ANCSVI	17 (7.7)	4 (3.4)	21 (6.2)
			Single	ANCNVI	1 (0.5)	0 (0.0)	1 (0.3)
			Double	AIRSVI	0 (0.0)	1 (0.8)	1 (0.3)
			Double	AICNVI	1 (0.5)	1 (0.8)	2 (0.6)
			Triple	AIRNVI	201 (91.0)	109 (92.4)	310 (91.4)
			Quadruple	AIRNVL	1 (0.5)	3 (2.5)	4 (1.2)
pfdhps (372)	225	147	None	Wild-type SGKAA	22 (9.8)	1 (0.7)	23 (6.2)
			Single	SGEAA	199 (88.4)	143 (97.3)	342 (91.9)
			Double	SGEGA	4 (1.8)	3 (2.0)	7 (1.9)
pfdhfr-pfdhps (328)	211	117	None	Wild-type ANCSVI-SGKAA	8 (3.8)	0 (0.0)	8 (2.4)
			Single	ANCSVI-SGEAA	7 (3.3)	4 (3.4)	11 (3.4)
			Triple	AIRNVI-SGKAA	10 (4.7)	0 (0.0)	10 (3.0)
			Triple	AIRSVI-SGEAA	0 (0.0)	1 (0.9)	1 (0.3)
			Triple	AICNVI-SGEAA	1 (0.5)	1 (0.9)	2 (0.6)
			Quadruple	AIRNVI-SGEAA	182 (86.3)	105 (89.7)	287 (87.5)
			Quintuple	AIRNVL-SGEAA	1 (0.5)	3 (2.6)	4 (1.2)
			Quintuple	AIRNVI-SGEGA	2 (0.9)	3 (2.6)	5 (1.5)

Figure 2.

Prevalence and distribution of polymorphism in targeted genes in Lilongwe and Mzuzu. (A) Prevalence of pfmdr1-184 mutant codon, which modulates susceptibility to multiple drugs including LUM (the partner drug in ALU, Malawi's first-line ACT). (B) Prevalence of pfcrt polymorphisms. Only wild-type codons were observed, indicating full CQ susceptibility. (C) pfk13 polymorphism showing the prevalence of only wild-type allele to indicate that the isolates were susceptible to ART drugs. (D) Prevalence of polymorphism in pfatp6 gene showing several mutant haplotypes. Mutant codons have been suggested to reduce parasite susceptibility to ART, although further in vitro and in vivo validation is required. (E) Polymorphism in pfdhfr gene responsible parasites resistance to pyrimethamine and piperaquine drugs. High prevalence of resistance-associated polymorphisms indicates reduced susceptibility to pyrimethamine and piperaquine. (F) Polymorphism in the pfdhps gene which indicate P. falciparum resistance to sulfadoxine drugs. Over 90% of isolates expressed resistant polymorphisms to the drug. (G) Prevalence of combined haplotypes of pfdhfr and pfdhpa in Lilongwe and Mzuzu associated with SP resistance. Haplotypes were categorized as *partial resistance, **full resistance, and ***super resistance.

pfcrt

DNA from 316 isolates was successfully amplified, 187 (59.2%) from Mzuzu and 129 (40.8%) from Lilongwe, for analysis of the pfcrt gene at amino acid positions 72, 74, 76, 93, 97, 101, 145, 146, 158, 159, 343, 353, and 356 (Figure 1). No polymorphism was detected in pfcrt gene in all isolates (Figure 2(B)).

pfk13

We sequenced 90 and 160 isolates from Lilongwe and Mzuzu, respectively, for the pfk13 gene SNPs polymorphisms (Figure 1). We analysed the pfk13 gene from the L444 to D680 codons and there was no mutation throughout (Figure 2(C)).

pfatp6

Ninety-six isolates from Lilongwe and 170 from Mzuzu were sequenced and analysed for polymorphism in the pfatp6 gene targeting mutation at H243Y, S250N, L263D, A569 K, A623E, and S769N codons. It was found that 3 (3.1%) of isolates from Lilongwe and 4 (2.4%) from Mzuzu had a mutation at pfatp6-H243Y codon. Of the isolates from Lilongwe, 2 (2.1%) had a mutation at pfatp6 S250N. A high mutation rate was found at the pfatp6-A569 K, where 21 (21.9%) and 32 (18.8%) were detected from Lilongwe and Mzuzu, respectively. The distribution of the pfatp6 haplotypes, including the wild-type haplotype HSLAAS and single mutants YSLAAS, HNLAAS, and HSLKAS, did not differ significantly between the two sites (Fisher's exact test, p = 0.236) (Table 1 and Figure 2(D)).

pfdhfr and pfdhps

For mutation at pfdhfr gene, 118 isolates from Lilongwe and 221 from Mzuzu were sequenced and analysed at the A16 V, N51I, C53R, S108N, V140L and I164L codons. The results showed that there was no mutation observed at the 16 and 140 codons of the pfdhfr gene. However, isolates from Lilongwe contained mutant allele at the pfdhfr 51, 59, 108 and 164 codons at 114 (96.6%); 113 (95.8%); 113 (95.8%) and 3 (2.5%) respectively. Likewise, isolates from Mzuzu had 203 (91.9%); 202 (91.4%); 204 (92.3%) and 1 (0.5%) mutant allele at respective codons. The pfdhfr triple mutant AIRNVI haplotype was significantly more frequent than the other haplotypes at both sites (Fisher's exact test, p = 0.002). However, when considering all haplotypes together, there was no significant difference in their distribution between the two sites (Fisher's exact test, p = 0.839) (Table 1 and Figure 2(F)).

A total of 225 samples from Mzuzu and 147 samples from Lilongwe were successfully genotyped for the pfdhps gene targeting S436A, G437A, K540E, A581G, and A613S mutation (Figure 1). Among all isolates sequenced for pfdhps, mutation was detected at 540E; 581G in 199 (88.4%); 4 (1.8%) from Mzuzu and 143 (97.3%); 3 (2.0%) from Lilongwe. In both sites, prevalence of a single mutant pfdhps SGEAA haplotype was significantly higher than the wild-type and double SGEGA haplotypes (Fisher's exact test, p = 0.002) (Table 1 and Figure 2(E)).

The quadruple pfdhfr-pfdhps mutant haplotype AIRNVI-SGEAA was the most prevalent in both Lilongwe (105/117, 89.7%) and Mzuzu (182/211, 86.3%) (Fisher's exact test, p < 0.01) (Figure 2(G)). However, when considering all haplotypes together, their overall distribution did not differ significantly between the two sites (Fisher's exact test, p = 0.390).

Prevalence and impact of resistance markers to antimalarials

To understand the impact of observed polymorphisms on different antimalarials, the polymorphisms were compared with previous reports on their effects on the susceptibility of P. falciparum parasites to the four antimalarial groups: endoperoxides, 4-aminoquinolines, antifolates, and aryl-amino alcohols (Supplementary Table S2). The absence of mutant markers on the pfcrt and pfk13 genes indicated that parasites were fully susceptible to endoperoxides, CQ and AQ. Polymorphism in the pfmdr1, particularly the N86 wild-type allele and 184F mutant codon, which are associated with reduced susceptibility to aryl-amino-alcohol (QN, LUM, and MFQ), was observed at varying frequencies across the isolates. The pfdhfr-pfdhps markers that associate with parasite resistance to antifolates (PYR, PG, and SDX) were highly prevalent in both sites (Figure 3).

Figure 3.

Prevalence and association of resistance markers to antimalarials. Association of genotyped markers with parasite susceptibility across four antimalarial drug classes. Antifolates were less susceptible followed by the aryl-amino alcohols. No resistance-associated markers were observed for endoperoxides and the 4-amino quinolines, except that PND susceptibility may be influenced by the pfmdr1-184F mutant codon.

Discussion

This study provides an updated molecular surveillance of P. falciparum antimalarial resistance markers in isolates from two key border regions in Malawi, Mzuzu and Lilongwe. The findings demonstrate notable differences in the prevalence of resistance-associated polymorphisms, with implications for the continued efficacy of current treatment strategies. Among the six genes analysed, significant variation was observed in pfmdr1, pfatp6, pfdhfr, and pfdhps, whereas no resistance-associated mutations were detected in pfcrt or pfk13.

The pfmdr1 gene, which modulates susceptibility to partner drugs in ACT [27, 28], exhibited considerable polymorphism. The 184F mutant was present at both study sites, with a higher frequency in Mzuzu than in Lilongwe. The NFD haplotype (N86/184F/D1246) was the most common and has previously been linked to reduced susceptibility to lumefantrine (LUM) [29–31], the partner drug in ALU, Malawi's first-line therapy. Although this study did not assess clinical or in vitro drug efficacy, the high prevalence of the NFD haplotype likely reflects ALU-driven selection. These observations are consistent with previous reports from Tanzania [9, 32], Zambia [31], and elsewhere [30], highlighting the need for longitudinal surveillance to detect emerging tolerance before clinical failures occur. By providing the first comprehensive molecular data on pfmdr1 haplotypes in Malawi, this study establishes a baseline for future monitoring and informs strategies to preserve ACT efficacy.

No mutations were detected in pfcrt, including the canonical K76 T position associated with CQ resistance [33, 34]. All isolates harboured wild-type pfcrt and pfmdr1-N86, consistent with full restoration of CQ sensitivity following drug withdrawal [35]. This genotype confers CQ sensitivity but has been associated with reduced LUM susceptibility, particularly in the context of the NFD haplotype, underscoring the need for ongoing monitoring of ALU efficacy [36]. While the genotype suggests restored CQ sensitivity, any consideration of CQ reintroduction must be approached with caution. Historical evidence from Zambia demonstrates that CQ reintroduction rapidly led to re-selected resistant strains due to residual alleles [35]. Therefore, continuous molecular surveillance and clinical efficacy monitoring are essential to guide antimalarial policy and prevent the resurgence of resistance.

The pfk13 gene, a validated marker of ART resistance [37], showed no resistance-associated mutations, consistent with sustained ART efficacy in Malawi [38]. Although pfatp6 is no longer considered a validated marker of ART resistance [22], it was analysed for exploratory purposes due to prior associations with ART resistance [39]. The A569 K mutation was observed at notable frequencies at both sites, warranting further investigation to clarify its potential impact on drug response. While direct evidence linking A569 K to ART tolerance is lacking, previous studies suggest that pfatp6 mutations may influence ART binding and functional activity, particularly when combined with other polymorphisms [21, 40, 41].

High frequencies of pfdhfr-pfdhps mutant haplotypes were observed, particularly the quadruple AIRNVI-SGEAA haplotype, which exceeded 86% in both sites. This haplotype includes the critical pfdhfr-S108N and pfdhps-K540E mutations and has been associated with reduced efficacy of SP [42]. The observed prevalence surpasses the WHO threshold beyond which SP is unlikely to remain effective for IPTp [7]. While SP continues to be recommended for IPTp even in high-resistance settings [42]. This policy is partly because it may confer some protection against adverse pregnancy outcomes, including low birth weight and maternal anaemia, potentially via non-parasitic mechanisms [43, 44]. However, the increasing prevalence of resistance alleles suggests diminishing protective benefits. Alternatives such as dihydroartemisinin-piperaquine (DHA-PQ), which has demonstrated superior efficacy in clinical trials [7], as well as other candidates including mefloquine or doxycycline, should be evaluated in local contexts. Such evaluations are particularly warranted given that the declining efficacy of IPTp may compromise maternal and neonatal health outcomes in endemic regions. However, the widespread adoption of alternatives such as DHA-PQ faces challenges related to cost, availability, adherence, and safety, which may constrain their immediate implementation. Ongoing multi-country trials and WHO policy reviews are currently assessing these regimens, but further evidence from Malawi and similar high-resistance settings will be critical to guide IPTp strategies.

While these findings provide important insights into the current landscape of antimalarial resistance in Malawi, their interpretation should be tempered by several limitations. Sampling was restricted to two urban hospitals with limited geographic coverage, which may not fully represent national resistance dynamics, and the cross-sectional design also precludes assessment of temporal trends. Moreover, reliance solely on molecular markers without in vivo or in vitro assays limits direct inference of clinical efficacy.

Finally, the border location of the study sites increases the likelihood of cross-border parasite movement, the importance of coordinated surveillance with neighbouring countries such as Tanzania, Mozambique, and Zambia. Integration of molecular surveillance with population genetics approaches, including microsatellite or whole-genome sequencing, could provide deeper insight into parasite population structure and gene flow.

Conclusion

This study confirms a sustained absence of pfcrt and pfk13 resistance markers in Malawi, supporting the ongoing effectiveness of CQ and ART derivatives. However, the elevated prevalence of the pfmdr1-NFD haplotype and widespread pfdhfr-pfdhps mutant combinations raise concerns about the long-term efficacy of LUM and SP, respectively. These findings underscore the importance of sustained molecular surveillance, complemented by clinical efficacy testing, to guide evidence-based antimalarial policy within the National Malaria Control Program (NMCP) and preserve treatment efficacy in Malawi.

Authors’ contributions

EM and HJ led all experiments, data cleaning and analysis, and wrote the first draft of the manuscript. J-HH contributed to data analysis and the manuscript drafting. W-JL, JML, JHS, FF, MAA, FM, LF, J-HH, T-HK, SJL, SN, WC, WSP, E-TH, and WK contributed to significant experiments and results interpretation. All authors contributed to the article and approved the submitted version.

Funding Statement

This study was supported by the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare [HI22C0820], and the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning and Ministry of Education [NRF-2021R1A4A1031574 and RS-2023-00240627] (J-H.H).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplemental Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/22221751.2025.2572679.