Rapid multiplexed nanopore amplicon sequencing to distinguish Plasmodium falciparum recrudescence from new infection in antimalarial drug trials

The emergence of artemisinin-resistant Plasmodium falciparum parasites and subsequent treatment failure of artemisinin-based combination therapies (ACTs) are a significant public health concern threatening global efforts to control and eliminate malaria. Delayed clearance following treatment with artemisinin derivatives, also referred to as artemisinin partial resistance (ART-R), was first described in 2008¹ and was initially confined to Southeast Asia. However, the recent independent emergence of ART-R in several countries from East and Horn of Africa^2,3,4,5 is challenging ACT and malaria control in sub-Saharan Africa. Widespread ART-R combined with resistance to partner drugs (i.e., ACT failure) in sub-Saharan Africa would have a catastrophic public health impact, as the region bears the heaviest malaria burden⁶. Regular reporting of drug efficacy across sites in Africa is essential, particularly in areas where P. falciparum kelch13 mutations have been reported, the key mediator of ART-R⁷.

Therapeutic efficacy studies (TES), conducted every 2–3 years in malaria-endemic regions, are considered the gold standard to assess antimalarial drug efficacy. In these trials, patients with microscopy-confirmed P. falciparum infection are treated with antimalarials and followed up at regular intervals to evaluate their clinical and parasitological response. During follow-up, some patients may experience recurrent parasitemia, which can arise from recrudescence – the original infection not being fully cleared (true treatment failure)—or from a new infection (new mosquito inoculation), or a mixture of both. To distinguish between these, molecular correction, also referred to as PCR-correction, is performed^8,9. This involves genotyping parasites from paired blood samples collected on day zero (before treatment) and at the day of treatment failure, to distinguish recrudescence from new infection. The therapeutic efficacy (genotype-corrected efficacy) estimate, the primary outcome of TES, reflects the ability of the treatment to fully clear initial infections and potentially provide prophylactic effects by preventing new infections⁸. Various genotyping methods and molecular markers are in use, along with differing algorithms for results interpretation^9,10, making it challenging to directly compare drug efficacy estimates across studies and regions. The World Health Organization (WHO) currently recommends genotyping three size polymorphic markers – P. falciparum merozoite surface protein 1 and 2 (msp1 and msp2) and microsatellites—using capillary electrophoresis⁸.

Recent advances in next-generation sequencing (NGS) techniques have facilitated the development of targeted amplicon sequencing (AmpSeq) protocols of highly polymorphic markers for genotyping malaria parasites in clinical trials of antimalarial drugs^11,12,13. By including short, highly diverse loci with abundant single nucleotide polymorphisms (SNPs), such high heterozygosity microhaplotypes can greatly improve discrimination between different parasite strains^{14,15,16,17,18}. Detection of minority clones is possible at frequencies as low as 0.1% in polyclonal infections^11,12,18,19, and data can be analyzed using inferential methods considering allele sharing probabilities based on population frequencies^20,21.

Nanopore sequencing is increasingly used to generate genetic data on malaria parasites, including in endemic regions, for applications such as detecting antimalarial drug resistance markers, identifying variations in vaccine targets, characterizing hrp2/3 gene deletions, and studying parasite population dynamics^{22,23,24,25,26,27,28}. Oxford Nanopore Technologies (ONT) offers a low-cost, portable, and scalable alternative to traditional sequencing methods, potentially addressing the need to expand and decentralize sequencing capacity in resource-limited settings^29,30. Workflows with ONT are straightforward, offer fast turnaround times, and are more easily deployed in endemic settings than other platforms^23,24,26,29. Thus, nanopore AmpSeq could provide rapid corrected estimates of drug efficacy, for example if a TES is needed in response to rapid emergence of potentially drug-resistant parasites. While nanopore sequencing used in previous malaria studies primarily focused on genomic surveillance^23,24,26, few studies have used nanopore AmpSeq to explore within-host diversity and detection of minority clones in polyclonal infections²⁶.

Here, we evaluated the feasibility of a nanopore AmpSeq assay for genotyping P. falciparum to distinguish recrudescence from new infection in clinical trials by assessing sensitivity, specificity, robustness and genetic marker diversity using the current latest ONT chemistry (kit 14/R10.4.1 flow cells) and comparing genotyping results to data generated with Illumina AmpSeq¹³. We used a multiplex AmpSeq panel targeting short, highly informative microhaplotype loci that are amenable to multiplexing, exhibit high amplification efficiency, and are compatible with streamlined, field-adaptable workflows.

Laboratory strain mixtures

Laboratory P. falciparum strain mixtures were prepared as previously described¹³. Briefly, parasitemia of synchronized ring-stage parasite strains 3D7, K1, HB3, and FCB1 was quantified by qPCR. The four strains were then combined in mixtures ratios ranging from 1:1:1:1 to 1:100:100:100 and diluted in human DNA from malaria-negative donors to mimic clinical samples. The concentration of the minority clone was always 10 parasites/µL. Twenty-four different ratios were used for this study, and each mixture was tested in triplicate in each of the two sequencing runs. Parasitemia and ratios of all strain mixtures is available in Supplementary Table 1.

Patient samples

Twenty well-characterized paired frozen whole blood samples (n = 40) from patients with recurrent infections were used from a previously conducted randomized, phase 2 clinical trial evaluating the efficacy and safety of cipargamin in adults with uncomplicated P. falciparum malaria in Mali, Gabon, Ghana, Uganda, and Rwanda. The protocol and all amendments were reviewed by the Independent Ethics Committee or Institutional Review Board for each center. All approvals were obtained in 2017 and 2018. A list of the institutions is provided in the Supplementary Table 2. The trial was conducted according to International Conference on Harmonization (ICH) E6 Guidelines for Good Clinical Practice. Informed consent was obtained from all subjects and/or their legal guardian(s). The trial is registered with ClinicalTrials.gov (NCT03334747). The trial was conducted between November 16, 2017, and November 23, 2019. Details of the study protocol and patient recruitment have been published elsewhere^31,32,33. Parasite counts (Giemsa-stained thick and thin films) were made per 200 white blood cells (or if the count was < 100 parasites, counting was continued for up to 500 white blood cells). Parasitemia of the twenty paired samples ranged from 31 to 33,930 parasites/µL (Supplementary Table 3). The samples had previously been analyzed as part of a comparative evaluation of genotyping methods, including AmpSeq using the Illumina platform¹³.

Multiplex PCR

Six polymorphic microhaplotype loci (ama1, celtos, cpmp, cpp, csp, and surfin1.1; Supplementary Table 4) were selected based on previously demonstrated high genetic diversity and discriminatory power^14,15,17,26. Additionally, the genetic diversity of the corresponding genes has been extensively characterized across multiple studies^{11,12,14,18,34,35,36}. We leveraged an existing 6-plex PCR panel targeting these loci²⁶ and optimized it to ensure uniform amplification across all targets. All six amplicons use previously published primer sequences^14,15,26. Details on primer pool concentrations and reaction conditions are provided in Supplementary Tables 5–7. An overview of A + T content and homopolymer regions (≥ 3 bases) within each amplicon is shown in Supplementary Fig. 1.

Library Preparation and sequencing

Nanopore sequencing was performed using the Native Barcoding Kit 96 V14 (SQK-NBD114.96) on the MinION Mk1C platform with MinKNOW software (v24.06.15) and R10.4.1 flow cells according to the manufacturer’s instructions with some modifications (Supplementary Text). Each run included three negative controls, consisting of nuclease-free water, which underwent the entire workflow, including PCR and library preparation. A positive control (strain FCB1) was included alongside the paired patient samples for quality control. For each sequencing run, the target sequencing depth was approximately 25,000 reads per marker per sample, or 150,000 reads total, to compensate for downstream filtering of low-quality reads. Sequencing was stopped once the desired number of reads was reached. An overview of the complete workflow is shown in Supplementary Fig. 2 and a laboratory protocol, including materials and primer sequences, is available online at protocols.io (https://www.protocols.io/) as “Rapid multiplexed nanopore amplicon sequencing of Plasmodium falciparum microhaplotype loci”³⁷.

Bioinformatics pipeline

Raw nanopore data was simplex basecalled and double-ended demultiplexed with dorado (v0.8.2) using the super-accurate (sup) model with the minimum q-score for passing reads set to 20 (accuracy of ≥ 99%) to minimize erroneous reads, as previously described (Supplementary Text)²⁶. Haplotypes were inferred using R packages HaplotypR (v0.5)¹¹ and DADA2 (v1.26.0)³⁸ as previously described with minor modifications (https://github.com/lerch-a/HaplotypR/releases/tag/v0.5; Supplementary Text)²⁶. To remove false-positive haplotypes and artefacts, a number of cutoffs were implemented: (1) each marker required a minimum of 1000 total reads per sample, (2) haplotype calls required a minimum coverage of ≥ 20 reads per haplotype, and (3) a within-sample allele frequency (WSAF) of ≥ 0.1%. Additionally, reads with ambiguous base calls (e.g., N) and incorrect sequence length (e.g., to remove amplification and/or sequencing errors caused by homopolymer-rich regions) were removed. An overview of the bioinformatics workflow is shown in Supplementary Fig. 3.

Data analysis

The limit of detection (LOD) of minority clones was defined as the lowest ratio of laboratory strains having at least two out of three replicates positive. To assess the intra-assay variability, we calculated the percentage of samples that returned the three technical replicates with the same result, and the inter-assay variability as the percentage of pairs of technical replicates giving the same LOD between two runs. Two algorithms were used to classify genotyping results for the six microhaplotype markers. The WHO algorithm defines sample pairs as recrudescent when all markers used show recrudescence⁸, while the two out of three algorithm classifies sample pairs as recrudescent when the majority of markers used (i.e., a minimum of 4/6) show recrudescence³⁹. Allelic frequency was assessed by calculating the number of times, each haplotype occurred in all pre-treatment samples divided by the number of times all haplotypes occurred in all pre-treatment samples. Heterozygosity (H_E) for each microhaplotype marker, the probability that two alleles will be different, in the twenty pre-treatment patient samples was determined using the equation (:{text{H}}_{text{E}}=:frac{n}{n-1}[1-:sum:{p}_{i}^{2}]), where n = the number of samples analyzed, p_i = the allele frequency of the i^th allele in the population. The combined probability of discerning two clones using any of the microhaplotypes was estimated using (:{P}_{diff}=1-:prod:_{i=1}^{n}(1-{H}_{E,:i})), where H_{E, i} = expected heterozygosity for microhaplotype i. Multiplicity of infection (MOI) was estimated in two ways: (i) naively, where MOI was determined by the locus with the greatest number of detected alleles, and (ii) using SNP data from the microhaplotypes to estimate MOI using THE REAL McCOIL⁴⁰, run in categorical method. Pairwise genetic relatedness based on identity-by-descent (IBD) in the paired patient samples was estimated using Dcifer (v1.2.1) with default settings, which also provides 95% confidence intervals (95%CI) via likelihood ratios²¹.

Sequencing quality and coverage

We sequenced triplicates of 24 control mixtures across two independent MinION runs (n = 72 in each run), and 20 paired patient samples (n = 40) in a third run. All samples were successfully amplified and sequenced. The q-scores of reads from both control mixtures and patient samples were comparable across all three sequencing runs (Fig. 1a-c, Supplementary Table 8). PCR optimization (Supplementary Fig. 4, Supplementary Table 9) improved coverage uniformity across all amplicons in both control mixtures (Fig. 1d-e) and patient samples (Fig. 1f). After filtering out reads with q-scores ≤ Q20, the median coverage across control mixtures was 85,789× per sample (interquartile range (IQR): 68,172×–107,275×) in run 1 and 58,591× (IQR: 45,819×–73,139×) in run 2. The median coverage for the 20 paired patient samples was 55,105× (IQR: 40,435×–72,904×). The median fold-difference in coverage among amplicons was 1.4 (IQR: 1.3–1.7) in run 1 and 1.5 (IQR: 1.3–1.7) in run 2 of control mixtures, and 1.7 (IQR: 1.5–2.2) in patient samples.

Quality metrics and read coverage profiles of three MinION sequencing runs. Distribution of q-scores for the control mixtures (a) run 1 (n = 72, median q-score 21.6)) and (b) run 2 (n = 72, median q-score 19.2), and (c) paired patient samples (n = 40, median q-score 20.8). The dotted grey line indicates the minimum q-score of Q20 for passing reads and the dotted colored lines indicate the median q-scores of each sequencing run. Coverage profiles of amplicon targets of run 1 (d) and run 2 (e) for the control mixtures and paired patient samples (f). Each dot represents a sample. The y-axis shows number of reads (log10) covering each amplicon target per sample for each of the three MinION runs. In all three figures, positive and negative controls were excluded. Median coverage for all amplicons is indicated on top of the graph. The box bounds the IQR divided by the median, and Tukey-style whiskers extend to a maximum of 1.5 × IQR beyond the box.

Full size image

Specificity, limit of detection, and reproducibility using control mixtures

Haplotype calling showed high specificity. Only one false-positive haplotype passed the cutoff criteria (≥ 0.1% WSAF and ≥ 20 reads) in a single replicate (sample S11) for marker surfin1.1 at 1.02% WSAF (Supplementary Fig. 5). In control mixtures containing only one of the four P. falciparum strains (S01–S04), trace contaminant haplotypes from other samples were observed in 19/92 instances in run 1 (median reads: 2, range: 2–5) and 15/88 in run 2 (median reads: 2, range: 2–4), all below the defined cutoff. Additional false-positive haplotypes, likely due to amplification or sequencing error, occurred in 0.28% of amplicon–sample instances in run 1 (4/1,415; median 9 reads, range 7–229) and 0.07% in run 2 (1/1,408; 12 reads). Overall, contaminant or false-positive haplotypes represented < 0.01% of total reads (302/6,133,887 in run 1; 50/4,160,477 in run 2), with all but one occurring below 0.1% WSAF. Based on these results, a threshold of ≥ 20 reads and WSAF ≥ 0.1% was selected to minimize false positives while maintaining sensitivity. No marker in any negative control (n = 9 across three runs) exceeded 12 reads, precluding haplotype calling.

The LOD for minority clones was consistent across markers (Fig. 2, Supplementary Table 10). For two microhaplotypes, the LOD could not be determined for all four control mixtures due to identical alleles in ama1 (FCB1 and K1) and csp (FCB1 and HB3). Detection of minority clones for these markers was assessed using only the strains with distinguishable alleles (3D7 and HB3 for ama1; 3D7 and K1 for csp), when one strain was in majority and the other in minority. At a WSAF of ≥ 0.1% in both runs, all six markers reliably detected minority clones even when majority clones were present at a 100-fold higher concentration (1:100:100:100), corresponding to a minority clone WSAF of 0.3% (Fig. 2, Supplementary Table 10). The assay also demonstrated robust performance and high sensitivity at low parasite density; in control mixture S05, containing all four strains at 10 parasites/µL each (40 parasites/µL total), all strains were consistently amplified and detected across all replicates in both sequencing runs. All markers exhibited high intra-assay and inter-assay reproducibility, ranging from 94.7% to 100% (Supplementary Fig. 6a). A strong correlation between expected and observed WSAF was observed in the control mixtures in both sequencing runs (R² = 0.97, Supplementary Fig. 6b).

Limit of detection (LOD) for minority clones in control mixtures. Haplotypes detected in the control mixtures of four P. falciparum strains across the six microhaplotypes in Run1 (left panel) and Run2 (right panel). Every control mixture was sequenced in triplicate in both runs. The ratios represent the ratio of the four P. falciparum laboratory strains assessed in this study: 3D7:K1:HB3:FCB1, with the concentration of the minority clone always being 10 parasites/µL. The ratios in parentheses are for markers ama1 and csp, as two strains had identical alleles (FCB1 and K1 for ama1; FCB1 and HB3 for csp) and LOD could only be determined for two strains instead of four. Y-axis: proportion of reads (log10). Dashed horizontal lines: WSAF cut-off for minority clones set at 0.1% of total reads. The black cross indicates a false-positive haplotype.

Full size image

Haplotype diversity and allelic frequency in paired patient samples

The genetic diversity of the 6 microhaplotype markers was assessed in the 20 pre-treatment samples. Haplotype diversity, defined as the number of distinct haplotypes, was high but varied across the six microhaplotypes (median: 20, range: 14–28; Fig. 3; Table 1). Marker cpmp exhibited the highest diversity, while surfin1.1 showed the lowest number of distinct haplotypes. Allelic frequencies also varied by marker (Fig. 3). For cpmp, no haplotype had an allelic frequency > 10%. Markers ama1, cpp and csp each had one haplotype with allelic frequencies > 10%, with no clearly dominant haplotype. In contrast, celtos had one haplotype at a frequency of > 15% and surfin1.1 had three dominant haplotypes, collectively accounting for over 50% of all reads (Fig. 3). The number of SNPs ranged from 13 for celtos and csp to 39 for cpmp (Table 1). SNP frequencies and positions are shown in Supplementary Fig. 7 and Supplementary Table 11. The discriminatory power (i.e., the probability that two alleles will be different; H_E) was > 0.9 for all markers (Table 1). The combined probability of discerning two clones using all six microhaplotypes was estimated to be > 99.99%. Estimates of MOI determined naively and using THE REAL McCOIL were comparable, with no significant differences observed (P = 0.77, Wilcoxon signed rank test; Supplementary Fig. 8).

Diversity of the six microhaplotypes in the 20 pre-treatment patient samples. Each pie chart shows the distribution of haplotypes for each marker across the pre-treatment patient sample pool (n = 20). The total number of haplotypes found is shown at the bottom right of each microhaplotype pie chart. Haplotype frequencies greater than 5% are indicated.

Full size image

Direct haplotype comparison for markers cpmp, cpp and csp between nanopore and Illumina sequencing was possible as the shorter nanopore amplicons were nested within the amplicons previously sequenced by Illumina in the same 20 paired patient samples (n = 40)¹³. Corresponding regions from Illumina data were extracted to determine the proportion of haplotypes identified by Illumina that were also detected using nanopore sequencing. Haplotype concordance between the two platforms was high: 100% (30/30) for cpmp, 100% (19/19) for cpp, and 95% (19/20) for csp (Supplementary Fig. 9a-c).

Distinguishing recrudescence from new infection in an antimalarial drug trial

The ability of nanopore AmpSeq to reliably distinguish recrudescence from new infection was assessed using 20 paired patient samples. Concordance across the six microhaplotype markers was high, with all markers providing the same outcome for 85% (17/20) of patients (Table 2). Fleiss’ Kappa indicated almost perfect agreement (κ = 0.825, z = 13.9, p < 0.001). Comparisons with AmpSeq data from Illumina sequencing¹³ yielded nearly identical results (Table 2). One patient sample pair (P09) failed to amplify marker cpp, likely due to mutations in the primer binding site, which was also observed previously using Illumina sequencing¹³. The final genotyping outcome was determined using two algorithms: the WHO algorithm (i.e., all markers show recrudescence) and the two out of three algorithm (i.e., ≥ 4/6 markers show recrudescence) (Table 3). Both algorithms yielded concordant results for all 20 patient samples, with 100% agreement with Illumina AmpSeq genotyping outcomes. To further confirm these results, we estimated pairwise genetic relatedness for each of the 20 paired patient samples (baseline and recurrent) using an IBD-based approach²¹. As expected, the four samples classified as new infections showed low to no genetic relatedness (arbitrary IBD threshold < 0.25), while the 16 samples classified as recrudescence showed clonal relatedness (IBD = 1; Table 3, Supplementary Table 12).

Molecular correction is critical for accurate estimation of antimalarial drug efficacy. Genotyping assays must be sensitive enough to detect minority clones present at low frequency in polyclonal infections, reproducible, and use markers with high diversity⁹. Previous studies have shown that AmpSeq provides superior results in distinguishing recrudescence from new infections compared to other genotyping methods, reinforcing its potential as the standard for molecular correction^12,13,20. However, many smaller laboratories in malaria-endemic countries still rely on shipping samples internationally for sequencing. Although capacity is gradually improving, efficient, cost-effective, reliable and accessible tools are needed to strengthen the genomic capacity of research and public health institutions in malaria-endemic countries before AmpSeq can be adopted for routine TES.

In this study, we optimized a multiplexed nanopore AmpSeq panel for P. falciparum microhaplotypes. Our panel accurately detects minority clones in polyclonal infections and is thus a promising tool for distinguishing recrudescence from new infection in clinical trials. The assay achieved high and uniform coverage across all targeted microhaplotypes – critical for accurate variant calling, particularly when identifying minority alleles. The method proved highly sensitive, specific and reproducible, detecting minor alleles (present at 10 parasites/µL) at WSAF as low as 0.3% (i.e., 1:100:100:100). This sensitivity is comparable to that of P. falciparum AmpSeq panels developed for Illumina platforms^14,15,16. Assay performance was robust, with high intra- and inter-assay reproducibility. To ensure high specificity while maintaining sensitivity, we excluded reads with < 99% accuracy (Q < 20) and applied a WSAF threshold of ≥ 0.1%, resulting in only one false-positive haplotype.

While our study and others^11,35 have shown that a WSAF threshold as low as ≥ 0.1% enables robust detection of minority clones, the optimal cutoff may vary depending on the study context. A stringent cutoff is required for excluding sequencing artefacts, whereas a less stringent cutoff increases sensitivity to detect minority clones. In regulatory clinical trials, where robustness and reproducibility are paramount, a conservative cutoff that errs on the side of overestimating treatment failure may be preferable. Importantly, cutoff selection should be guided by appropriate internal controls within each sequencing experiment. In our study, the chosen threshold was supported by results from control mixtures of P. falciparum strains, which defined the technical detection limit for minority clones. Several published P. falciparum AmpSeq studies have adopted WSAF cutoffs of ~ 1–2%, striking an effective balance between minimizing false positives and retaining sensitivity^{12,14,15,16,18,19,41}. The assay was effective on strain mixtures with parasitemias as low as 40 parasites/µL and maintained high coverage in paired patient samples, including nine with low parasitemia levels below 500 parasites/µL (31–462 parasites/µL). In previous work, we also demonstrated that the assay performs robustly on dried blood spots with parasitemias as low as 250 parasites/µL²⁶.

The entire workflow – from DNA extraction to results – can be completed in approximately three to four days. Estimated running costs are approximately US$25 per sample when multiplexing 96 samples (Supplementary Table 13). Further cost reductions may be possible with higher degrees of multiplexing as the number of barcodes increases and through flow cell reuse following washing. In addition to the 96 native barcodes currently available, custom barcodes could be added to the amplicons to generate dual-barcoded libraries, thereby enabling multiplexing of > 96 samples per flow cell.

To fully leverage the information content from diverse microhaplotype loci, particularly in polyclonal infections, it is critical to use haplotype inference tools capable of accurately identifying complex infections with unknown haplotype proportions. While such tools exist for short-read Illumina data, only our previously developed workflow effectively captures all information from polyclonal infections using nanopore data²⁶. We adapted existing short-read haplotype inference tools, HaplotypR¹¹ and DADA2 ³⁸, using only high-quality reads and appropriate cutoffs which are required by DADA2³⁸. Given the short amplicon length (~ 200 bp), sequencing errors are minimal, allowing for reliable haplotype inference and distinguishing true variation from errors. However, this approach may not be suitable for longer amplicons (> 1 kb), where error accumulation complicates the differentiation between true variants and sequencing errors. Hence, nanopore AmpSeq of short microhaplotypes is more robust against sequencing errors, compared to the previously utilized longer amplicons^23,24, while maintaining high discriminatory power.

Our panel demonstrated high genetic diversity (H_E > 0.9 for all markers) and low allelic frequencies in patient samples, with up to 28 distinct haplotypes for cpmp, providing high discriminatory power. For example, based on the H_E of all six microhaplotypes across all 20 pre-treatment samples, there would be a > 99.99% chance of determining that strains are different. MOI estimates in patient samples were highly similar whether based on the locus with the greatest number of alleles or using THE REAL McCOIL. Concordance between markers cpmp, cpp, and csp and Illumina results was very high. However, two of the six markers in our panel – ama1 (FCB1 and K1) and csp (FCB1 and HB3) – were unable to distinguish between two well-characterized laboratory strains. These findings highlight the need of validating new genotyping methods across geographically diverse strains and evaluating marker diversity and allele frequency in TES study site samples, as marker performance may vary substantially between regions.

Previous studies have shown that molecular correction outcomes can vary widely depending on the algorithm used^13,20. In this study, the six markers consistently yielded concordant results in 85% of the samples analyzed, and more importantly, the final corrected outcomes were 100% identical across all samples, regardless of the algorithm applied. The results were identical to those previously obtained with longer amplicons sequenced on the Illumina platform¹³. The high concordance between the two methods supports multiplexed nanopore AmpSeq as a promising tool for distinguishing different parasite clones. Additionally, using Dcifer, a relatedness-based approach that accounts for population allele frequencies and provides uncertainty measures²¹, yielded the same results as the WHO and two out of three algorithms. Such novel approaches could potentially provide even more robust estimates of interhost relatedness by using a larger set of microhaplotypes and accounting for missing alleles, thus providing a more accurate corrected estimate.

ONT devices provide a cost-effective alternative to traditional sequencing methods, offering rapid data generation and enabling in-country sequencing, thus minimizing result turnaround times. Despite its apparent benefits and continuous improvements in sequencing performance, nanopore sequencing of P. falciparum remains underutilized^22,23,24,26. Previous malaria studies have demonstrated the feasibility of nanopore AmpSeq for various applications^{22,23,24,26,27,28}, including sequencing in malaria-endemic countries^23,26; however, this is the first study to use nanopore sequencing to distinguish recrudescence from new infection in antimalarial drug efficacy studies. Although nanopore sequencing allows long-read AmpSeq, we opted for short amplicons to distinguish recrudescence from new infection for several reasons. While longer microhaplotype reads may offer greater discriminatory power in resolving haplotypes by capturing more SNPs and thus increasing genetic diversity, shorter amplicons generally exhibit higher amplification efficiency and can be designed to target the most informative regions, avoiding error-prone homopolymers or tandem repeats. Indeed, nanopore sequencing struggles to accurately resolve low-complexity regions^24,42, which are abundant in the P. falciparum genome⁴³. Previous studies using nanopore AmpSeq did not attempt to infer individual haplotypes or detect minority clones in polyclonal infections. Rather, they focused on the prevalence of drug-resistance mutations^22,23,24,28.

Our study has several limitations. First, for two markers, ama1 and csp, the LOD could only be determined for two out of four strains due to identical alleles. Second, we did not assess inter-assay reproducibility across different operators or laboratories. Third, the small number of patient samples, with a high proportion of recrudescence, limits the generalizability of the results. Fourth, while our assay was designed to be user-friendly with high and uniform coverage, it was not a direct adaptation of the previously described Illumina AmpSeq protocol¹³. Rather, we employed a multiplex PCR approach targeting short microhaplotypes with optimized amplification efficiency that does not require a prior pre-amplification step (i.e., selective whole genome amplification). As a result, direct comparisons with the Illumina protocol were limited; only three of our amplicons overlapped with regions sequenced by Illumina, allowing for restricted, locus-specific comparisons. Direct cross-platform comparisons will require the use of an identical AmpSeq panel across platforms in future work. Finally, this study did not evaluate the practical utility or in-country implementation of the assay for use in antimalarial drug trials. However, in a previous pilot study, we have successfully used the microhaplotype nanopore AmpSeq assay outside of an advanced laboratory in a malaria-endemic country²⁶.

In conclusion, our study demonstrates the feasibility of using rapid, multiplexed nanopore AmpSeq to genotype P. falciparum and distinguish recrudescence from new infection, expanding the current applications of nanopore sequencing in malaria research and molecular surveillance. We show that detecting minority clones in polyclonal infections and characterizing within-host diversity is feasible using nanopore AmpSeq targeting short, highly diverse microhaplotypes. Although further studies from clinical drug trials with larger sample sizes are needed to validate the assay’s ability to distinguish recrudescence from new infection, this work represents an important initial proof-of-principle for this approach. ONT sequencing platforms are affordable, portable, and scalable, making them suitable for resource-limited endemic settings. This technology offers a promising solution to allow for in-country sequencing and rapid corrected estimates of drug failure using AmpSeq.

Raw sequence data is available for download from NCBI’s Sequence Read Archive (https://www.ncbi.nlm.nih.gov/sra) under the accession PRJNA1188139. Haplotype data can be found at (https://doi.org/10.5281/zenodo.14176815) . All other data can be found in the supplementary files. A complete laboratory protocol, including materials and primer sequences, is available online at protocols.io (https://www.protocols.io) as ” [Rapid multiplexed nanopore amplicon sequencing of Plasmodium falciparum microhaplotype loci] (https://www.protocols.io/view/rapid-multiplexed-nanopore-amplicon-sequencing-of-81wgbrqnqlpk/v1)”³⁷.

1. Dondorp Arjen, M. et al. Artemisinin resistance in plasmodium falciparum malaria. N. Engl. J. Med. 361, 455–467 (2009).

   Article  CAS  PubMed  PubMed Central  Google Scholar 

2. Uwimana, A. et al. Association of plasmodium falciparum kelch13 R561H genotypes with delayed parasite clearance in rwanda: an open-label, single-arm, multicentre, therapeutic efficacy study. Lancet. Infect. Dis. 21, 1120–1128 (2021).

   Article  CAS  PubMed  PubMed Central  Google Scholar 

3. Balikagala et al. Evidence of Artemisinin-Resistant malaria in Africa. N. Engl. J. Med. 385, 1163–1171 (2021).

   Article  CAS  PubMed  Google Scholar 

4. Mihreteab, S. et al. Increasing prevalence of Artemisinin-Resistant HRP2-Negative malaria in Eritrea. N. Engl. J. Med. 389, 1191–1202 (2023).

   Article  CAS  PubMed  PubMed Central  Google Scholar 

5. Ishengoma, D. S. et al. Evidence of Artemisinin partial resistance in Northwestern tanzania: clinical and molecular markers of resistance. Lancet Infect. Dis. S1473-3099 (24), 00362–00361. https://doi.org/10.1016/S1473-3099(24)00362-1 (2024).

   Article  Google Scholar 

6. WHO. World malaria report 2023. Geneva: World Health Organization. (2023).

7. Ariey, F. et al. A molecular marker of artemisinin-resistant plasmodium falciparum malaria. Nature 505, 50–55 (2014).

   Article  ADS  PubMed  Google Scholar 

8. WHO. Methods and Techniques for Clinical Trials on Antimalarial Drug Efficacy: Genotyping To Identify Parasite Populations : Informal Consultation Organized by the Medicines for Malaria Venture and Cosponsored by the World Health Organization, 29–31 May 2007, Amsterdam, the Netherlands (World Health Organization, 2008).

9. Felger, I., Snounou, G., Hastings, I., Moehrle, J. J. & Beck H.-P. PCR correction strategies for malaria drug trials: updates and clarifications. Lancet. Infect. Dis. 20, e20–e25 (2020).

   Article  CAS  PubMed  Google Scholar 

10. Plucinski, M. M. et al. Variation in calculating and reporting antimalarial efficacy against plasmodium falciparum in Sub-Saharan africa: A systematic review of published reports. Am. J. Trop. Med. Hyg. 104, 1820 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

11. Lerch, A. et al. Development of amplicon deep sequencing markers and data analysis pipeline for genotyping multi-clonal malaria infections. BMC Genom. 18, 864 (2017).

    Article  Google Scholar 

12. Gruenberg, M., Lerch, A., Beck, H. P. & Felger, I. Amplicon deep sequencing improves plasmodium falciparum genotyping in clinical trials of antimalarial drugs. Sci. Rep. 9, 1–12 (2019).

    Article  CAS  Google Scholar 

13. Schnoz, A. et al. Genotyping methods to distinguish plasmodium falciparum recrudescence from new infection for the assessment of antimalarial drug efficacy: an observational, single-centre, comparison study. The Lancet Microbe 0, (2024).

14. Tessema, S. K. et al. Sensitive, highly multiplexed sequencing of microhaplotypes from the plasmodium falciparum heterozygome. J. Infect. Dis. https://doi.org/10.1093/infdis/jiaa527 (2020).

    Article  PubMed Central  Google Scholar 

15. Holzschuh, A. et al. Multiplexed ddPCR-amplicon sequencing reveals isolated plasmodium falciparum populations amenable to local elimination in Zanzibar, Tanzania. Nat. Commun. 14, 3699 (2023).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

16. LaVerriere, E. et al. Design and implementation of multiplexed amplicon sequencing panels to serve genomic epidemiology of infectious disease: A malaria case study. Mol. Ecol. Resour. 22, 2285–2303 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

17. Holzschuh, A. et al. Plasmodium falciparum transmission in the highlands of Ethiopia is driven by closely related and clonal parasites. Mol. Ecol. 33, e17292 (2024).

    Article  CAS  PubMed  Google Scholar 

18. Aranda-Díaz, A. et al. Sensitive and modular amplicon sequencing of plasmodium falciparum diversity and resistance for research and public health. Sci. Rep. 15, 10737 (2025).

    Article  ADS  PubMed  PubMed Central  Google Scholar 

19. Early, A. M. et al. Detection of low-density plasmodium falciparum infections using amplicon deep sequencing. Malar. J. 18, 219 (2019).

    Article  PubMed  PubMed Central  Google Scholar 

20. Jones, S. et al. Should Deep-Sequenced amplicons become the new gold standard for analyzing malaria drug clinical trials? Antimicrob. Agents Chemother. 65 https://doi.org/10.1128/aac.00437-21 (2021).

21. Gerlovina, I., Gerlovin, B., Rodríguez-Barraquer, I. & Greenhouse, B. Dcifer: an IBD-based method to calculate genetic distance between polyclonal infections. Genetics iyac126 https://doi.org/10.1093/genetics/iyac126 (2022).

22. Runtuwene, L. R. et al. Nanopore sequencing of drug-resistance-associated genes in malaria parasites, plasmodium falciparum. Sci. Rep. 8, 8286 (2018).

    Article  ADS  PubMed  PubMed Central  Google Scholar 

23. Girgis, S. T. et al. Drug resistance and vaccine target surveillance of plasmodium falciparum using nanopore sequencing in Ghana. Nat. Microbiol. 8, 2365–2377 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

24. de Cesare, M. et al. Flexible and cost-effective genomic surveillance of P. falciparum malaria with targeted nanopore sequencing. Nat. Commun. 15, 1413 (2024).

    Article  ADS  PubMed  PubMed Central  Google Scholar 

25. Razook, Z. et al. Real time, field-deployable whole genome sequencing of malaria parasites using nanopore technology. bioRxiv 2020.12.17.423341 (2020). https://doi.org/10.1101/2020.12.17.423341

26. Holzschuh, A. et al. Using a mobile nanopore sequencing lab for end-to-end genomic surveillance of plasmodium falciparum: A feasibility study. PLOS Global Public. Health. 4, e0002743 (2024).

    Article  PubMed  PubMed Central  Google Scholar 

27. Sabin, S. et al. Portable and cost-effective genetic detection and characterization of plasmodium falciparum hrp2 using the minion sequencer. Sci. Rep. 13, 2893 (2023).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

28. Hosch, S. et al. PHARE: a bioinformatics pipeline for compositional profiling of multiclonal plasmodium falciparum infections from long-read nanopore sequencing data. J. Antimicrob. Chemother. 79, 987–996 (2024).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

29. Hamilton, W. L., Ishengoma, D. S., Parr, J. B., Bridges, D. J. & Barry, A. E. Nanopore sequencing for malaria molecular surveillance: opportunities and challenges. Trends Parasitol. 39, 996–1000 (2023).

    Article  CAS  PubMed  Google Scholar 

30. Zerebinski, J. & Plaza, D. F. Progress in malaria genomic surveillance using long-read sequencing. Trends Parasitol. 40, 956–958 (2024).

    Article  CAS  PubMed  Google Scholar 

31. Schmitt, E. K. et al. Efficacy of Cipargamin (KAE609) in a Randomized, phase II Dose-Escalation study in adults in Sub-Saharan Africa with uncomplicated plasmodium falciparum malaria. Clin. Infect. Dis. 74, 1831–1839 (2022).

    Article  CAS  PubMed  Google Scholar 

32. Straimer, J., Gandhi, P., Renner, K. C. & Schmitt, E. K. High prevalence of plasmodium falciparum K13 mutations in Rwanda is associated with slow parasite clearance after treatment with Artemether-Lumefantrine. J. Infect. Dis. 225, 1411–1414 (2022).

    Article  CAS  PubMed  Google Scholar 

33. Ndayisaba, G. et al. Hepatic safety and tolerability of Cipargamin (KAE609), in adult patients with plasmodium falciparum malaria: a randomized, phase II, controlled, dose-escalation trial in sub-Saharan Africa. Malar. J. 20, 478 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

34. Miller, R. H. et al. A deep sequencing approach to estimate plasmodium falciparum complexity of infection (COI) and explore apical membrane antigen 1 diversity. Malar. J. 16, 490 (2017).

    Article  PubMed  PubMed Central  Google Scholar 

35. Lerch, A. et al. Longitudinal tracking and quantification of individual plasmodium falciparum clones in complex infections. Sci. Rep. 9, 1–8 (2019).

    Article  ADS  CAS  Google Scholar 

36. Neafsey, D. E. et al. Genetic diversity and protective efficacy of the RTS,S/AS01 malaria vaccine. N. Engl. J. Med. 373, 2025–2037 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

37. Holzschuh, A. Rapid multiplexed nanopore amplicon sequencing of Plasmodium falciparum microhaplotype loci protocol metrics. (2025).

38. Callahan, B. J. et al. DADA2: High-resolution sample inference from illumina amplicon data. Nat. Methods. 13, 581–583 (2016).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

39. Messerli, C., Hofmann, N. E., Beck, H. P. & Felger, I. Critical evaluation of molecular monitoring in malaria drug efficacy trials and pitfalls of Length-Polymorphic markers. Antimicrob. Agents Chemother 61, (2017).

40. Chang, H. H. et al. THE REAL mccoil: A method for the concurrent Estimation of the complexity of infection and SNP allele frequency for malaria parasites. PLoS Comput. Biol. 13, e1005348 (2017).

    Article  PubMed  PubMed Central  Google Scholar 

41. Sadler, J. M. et al. Application of a new highly multiplexed amplicon sequencing tool to evaluate plasmodium falciparum antimalarial resistance and relatedness in individual and pooled samples from Dschang, Cameroon. Front. Parasitol. 3, 1509261 (2025).

    Article  PubMed  PubMed Central  Google Scholar 

42. Delahaye, C. & Nicolas, J. Sequencing DNA with nanopores: troubles and biases. PLOS ONE. 16, e0257521 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

43. Gardner, M. J. et al. Genome sequence of the human malaria parasite plasmodium falciparum. Nature 419, 498–511 (2002).

    Article  ADS  CAS  PubMed  Google Scholar 

Download references

Authors and Affiliations

1. Swiss Tropical and Public Health Institute, Kreuzstrasse 2, 4123, Allschwil, Switzerland

   Aurel Holzschuh & Christian Nsanzabana

2. University of Basel, Basel, Switzerland

   Aurel Holzschuh & Christian Nsanzabana

3. Department of Biological Sciences, Eck Institute for Global Health, University of Notre Dame, Notre Dame, IN, USA

   Anita Lerch

Contributions

A.H.: Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Visualization, Writing – Original Draft. A.L.: Formal Analysis, Methodology, Software, Writing – Review & Editing. C.N.: Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing – Review & Editing. All authors reviewed and approved the final version of the manuscript.

Corresponding author

Correspondence to Aurel Holzschuh.

Competing interests

A.H. and C.N. have provided genotyping services to Novartis. A.L. declares no competing interests.

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions