Evolution of genetic markers for drug resistance after the introduction of dihydroartemisinin–piperaquine as first-line anti-malarial treatment for uncomplicated falciparum malaria in Indonesia

Background Dihydroartemisinin–piperaquine has been Indonesia’s first-line anti-malarial treatment since 2008. Annual therapeutic efficacy studies (TES) done in the last 12 years showed continued high treatment efficacy in uncomplicated Plasmodium falciparum malaria. Although these studies did not show evidence for artemisinin resistance, a slight increase in Late Treatment Failure was observed over time. It is highlight to explore the evolution of genetic markers for ACT partner drug resistance since adopting DHA–PPQ. Methods Dry blood spots were identified from a mass blood survey of uncomplicated falciparum malaria patients (N = 50) in Sumba from 2010 to 2018. Analysis of genotypic profile (N = 51) and a Therapeutic Efficacy Study (TES) from Papua (N = 142) from 2020 to 2021, 42-day follow-up. PCR correction using msp1, msp2, and glurp was used to distinguish recrudescence and reinfection. Parasite DNA from DBSs was used for genotyping molecular markers for antimalaria drug resistance, including in Pfk13, pfcrt, and pfmdr1, as well as gene copy number variation in pfpm2/3 and pfmdr1. Results The study revealed the absence of SNPs associated with ART resistance and several novel SNPs such as L396F, I526V, M579I and N537S (4.25%). In Sumba, the mutant haplotype SDD of pfmdr1 was found in one-third of the isolates, while only 8.9% in Papua. None of the pfcrt mutations linked to piperaquine resistance were observed, but 71% of isolates had pfcrt I356L. Amplification of the pfpm2/3 genes was in Sumba (17.02%) and Papua (13.7%), while pfmdr1 copy number prevalence was low (3.8%) in both areas. For the TES study, ten recurrences of infection were observed on days 28, 35, and 42. Late parasitological failure (LPF) was observed in 10/117 (8.5%) subjects by microscopy. PCR correction revealed that all nine cases were re-infections and one was confirmed as recrudescence. Conclusion This study revealed that DHA–PPQ is still highly effective against P. falciparum. The genetic architecture of the parasite P. falciparum isolates during 2010–2021 revealed single copy of Pfpm2 and pfmdr1 were highly prevalent. The slight increase in DHA–PPQ LTF alerts researchers to start testing other ACTs as alternatives to DHA–PPQ for baseline data in order to get a chance of achieving malaria elimination wants by 2030. Graphical Abstract Supplementary Information The online version contains supplementary material available at 10.1186/s12936-023-04658-4.


Background
In Indonesia, increasing treatment failure rates in the treatment of uncomplicated falciparum malaria with chloroquine (CQ) and sulfadoxine-pyrimethamine (SP) has prompted a change to artemisinin-based combination therapy (ACT) as first-line anti-malarial treatment since 2004 [1].ACT combines a potent but short-acting artemisinin (ART) component with a less potent but long-acting partner drug [2,3].Artesunate-amodiaquine (AS-AQ) was introduced first, but reports on poor tolerability and increasing treatment failure rates led to changing it to DHA-PPQ in 2008 [4][5][6][7].
Since 2010, this well-tolerated and effective regimen has also become the first-line treatment of other human malaria species, including Plasmodium vivax, due to increasing CQ resistance in this species [8].Studies conducted between 1995 and 2002 in North Sumatra, West Kalimantan, North Sulawesi, West Nusa Tenggara, East Nusa Tenggara, and Papua have consistently demonstrated treatment failure for CQ in P. vivax infections.Since then, CQ has not been an effective treatment for acute vivax malaria [9][10][11][12][13][14][15].Several studies have [4,9,16] documented excellent effectivity and tolerability for the DHA-PPQ treatment of uncomplicated malaria in Indonesia, including in Papua, Indonesia [17].Artemisinin resistance in Plasmodium falciparum has not been established in Indonesia until now (Fig. 1).
Therapeutic efficacy studies (TES) performed between 2011 and 2018 in South Sumatra, Central Kalimantan, West Kalimantan, North Sulawesi, Central Sulawesi, North Maluku, and East Nusa Tenggara have shown consistently high 42-day cure rates with DHA-PPQ for the treatment of uncomplicated falciparum malaria.No evidence was found for delayed parasite clearance, the hallmark of ART resistance [17].From 2017 to 2018, molecular surveillance of K13 in Papua New Guinea samples revealed that all P. falciparum isolates carried the wild-type allele of K13 [18,19].Some other mutations, such as G453W (20%), V454C (20%), E455K (20%), and T474A (2.6%), were also observed at low frequencies [20].An increased copy number of the pfpm2/3 gene has been detected in some Papua isolates surviving the DHA-PPQ [5].
Even though DHA-PPQ are still efficacious in treating malaria patients and there was no evidence of treatment failure, evaluation study for genotypic profile and additional measures, such as parasite clearance time (PCT) and parasite density should be regularly monitored.The interval between the patient's first dose to the time of the first negative blood slide was called the PCT.According to a study in Papua, from April 2017 to April 2018, recurrent P. falciparum parasites were detected in 7 out of 102 cases that completed the 42-day follow-up and were classified as LTF at days 21, 35, and 42.Of the 7 LTF cases, one was re-infected with P. vivax, 2 were confirmed as recrudescent infections, and the remaining 4 were re-infections.No delay in parasite clearance or severe adverse reaction was observed in any study participant [5].
The efficacy of ACT depends on the sensitivity of the parasites for both components of the combination.Artemisinin resistance can be monitored by assessing the parasite clearance rate and the presence of SNPs in the Pfk13 gene, a well-established marker for ART resistance.Resistance to some of the ACT partner drugs can be monitored through molecular surveillance.These include: for PPQ SNPs in the pfcrt gene (position 343, 350, 353) and copy number variations (CNVs) of the P. falciparum Plasmepsin2/3 gene (pfpm2/3) and pfmdr1 gene.In addition for ART SNPs in the pfcrt gene (356).
The present study aims to assess the evolution of genetic markers for anti-malarial drug resistance following the adoption of DHA-PPQ as the first-line antimalarial drug in 2010 for any uncomplicated malaria cases, the temporal dynamics trends of the evolution of pfk13, pfcrt, pfmdr1 genes also a copy number of pfpm2/3 and pfmdr1 from Sumba and Papua and observation of PCT including parasite density.This risk arises as more parasites may develop resistant to PPQ.

Study sites and sample collection
Plasmodium falciparum parasite DNA from filter paper blood spots was obtained from two different sample set studies as described in Fig. 2. The selected area for sample set studies was based on the fact that both sites had high annual parasite incidence (API).The total API in Jayapura and Keerom, Papua, from 2019 until 2021 was 95. 43 [21].

Sample set 1 malariometric surveys
Fifty blood blots on filter paper (3 MM; Whatman, Hillsboro, OR, USA) containing approximately 25-50 μL blood equivalent were collected from Sumba during 2010-2018 through active case detection.Sample selection was based on the availability of the existing archived sample.

Sample set 2 therapeutics efficacy study (TES)
A therapeutic efficacy study was conducted in Papua Province from 2020 to 2021 by the Eijkman Institute for Molecular Biology, National Research and Innovation Agency, Cibinong, Indonesia, serves as part of DBS's analysis for the current study.TES procedures followed World Health Organization (WHO) criteria.Briefly, smears and blots on filter paper (Whatman International Ltd., Maidstone, UK) were collected from finger pricks on days before enrolment, then on days 1, 2, 3, 7, 14, 21, 28, 35 and 42.The blood-spotted filter papers were allowed to dry completely, transferred into individual plastic bags, labelled, and stored at room temperature in a silica gel desiccator until further processing.2749 subjects were screened through passive and active case detection, and 42% (1156/2749) were positive for malaria.One hundred forty-two who met the inclusion criteria samples were obtained from 768, a pool of P. falciparum malaria-infected individuals.The respondents were aged between 1 and 65 years, weighed more than 5 kg, and had a fever or history of fever in the preceding 24 h, with slide-confirmed malaria with a parasitemia of ≥ 500/ µL asexual parasites for P. falciparum.Meanwhile, they were excluded with the following exclusion criteria: pregnant, had a history of allergy to the study drugs or studied drug's derivative, had previously completed treatment with an anti-malarial drug in the preceding 2 weeks, or had a medical history of untreated hypertension or chronic heart, kidney, or liver disease [5].All study participants were given a supervised treatment of DHA-PPQ from a primary health center containing 40 mg DHA and 320 mg PPQ per tablet and were administered once a day for three days, as a weight per dose regimen of 2.25 and 18 mg/kg of DHA-PPQ [26].Treatment outcomes and a new infection were classified according to the WHO criteria [16,17,22].
In this study, fifty-one archived samples, consisting of 41 successful treatments and 10 treatment failure, were selected from one hundred forty-two subjects to analyze the genotypic profile of molecular markers associated with DHA-PPQ resistance treatment.As part of the Papua TES study, 142 subjects were evaluated for clinical and parasitological efficacy of DHA-PPQ during days 0, 1, 2, 3, 7, 14.21, 28, 35, 42 [22].

Genomic DNA preparation
DNA from all samples (including sample set 2, on the day of enrolment and day of recurrence) were extracted from the DBS samples using a Chelex-100 ion exchanger (Biorad Laboratories, Hercules, CA, USA) [22].The genomic DNA obtained was purified following the Qiagen procedures.

Evaluation of mutations in Pfk13, Pfcrt and Pfmdr1
Polymorphisms in the Pfk13 gene were investigated using nested PCR amplification covering the gene's propeller region [23], followed by sequencing with an ABI sequencer (Macrogen Inc, South Korea).The sequencing results were then aligned against the reference strain 3D7's Pfk13 gene (PF13 0238) (NCBI reference sequence no.XM 001350122.1).The analysis was carried out using the BioEdit software (Abbott, CA, USA).
PCR: pfcrt was amplified from the DNA template to assess pfcrt mutations linked to PPQ resistance identified in a previous study [24,25].These were codons 343, 350, 353 and 356.On a 2% agarose gel, PCR product were visualized.
Pfmdr1 was amplified from the DNA template using nested PCR to assess pfmdr1 mutations, including the following SNPs: 1034, 1042 and 1246.The PCR amplicon was analysed on a 3% agarose gel under ultraviolet illumination.All PCR products were sent for DNA sequencing at 1 st Base Inc. in Singapore for quality control [26,27].Details of amplification primers sequences and PCR product results are available in Additional file 1: Table S1, Fig. S1.

Assessment of PfPlasmepsin2/3 and pfmdr1 gene amplification
Relative quantitative real-time PCR (TaqMan real-time PCR) on an Applied Biosystems 7500 quantified pfpm 2/3 and pfmdr1 copy numbers (Roche Molecular Systems, Inc., USA).Previously disclosed primers and probes [28,29], a BioRad CFX 96 thermocycler was used to amplify 20 μL in triplicate.Copy number estimates were 2−∆∆CT, where CT is the difference between the unknown sample's threshold cycle (CT) and the reference sample's CT.Runs are not interpretable of Ct values > 33 for pfpm2/3 or pftub or sample with a copy number estimate of < 0.5; reactions were repeated.As in previous studies, the main analysis used a cut-off copy number estimate of 1.5 to distinguish single-copy from multiplecopy pfpm2/3 and pfmdr1 gene carriage [28,29].

Method for distinguishing between recrudescence and re-infection
The Plasmodium speciation and genotype of P. falciparum were determined using PCR.Genotypic analyses of the parasites at day 0 and the day of recurrence were conducted using the three markers recommended by the WHO: merozoite surface protein 1 (MSP1), MSP2, and glutamate-rich protein (GLURP) genes [26,27].Cases were categorized as re-infections as the genotypes of the parasites found on the day of recurrence differed from those found on day 0 (pre-treatment).The identical genotypes for the three markers could be either recrudescent [25,28].

Analysis
Analysis was performed using Microsoft Office Excel basic functions and open-source software, RStudio version 2022.07.2+576, based on R version 4.2.2 [30,31].Significant differences in SNPs prevalence proportions each year during the study period were analyzed using the Fischer exact test for categorical variables or the Mann-Whitney U test for nonparametric comparisons.This study used the Excel Kaplan-Meier analysis template provided by the WHO.The results are expressed as success and failure cumulative incidence, with 95% CI.

Results
Demographic characteristics for TES samples as shown in Table 1.Dry blood spots from a mass blood survey of uncomplicated falciparum malaria patients (N = 50) in Sumba from 2010 to 2018 and fifty-one from TES from 2020 to 2021 in Papua to analyze the genotypic profile of molecular markers associated with DHA-PPQ resistance treatment.As part of the Papua TES study, one hundred forty-two subjects were evaluated for clinical and parasitological efficacy of DHA-PPQ during a 42-day follow-up (Table 2).No early treatment failure (ETF) was observed in Papua.However, ten patients out of 117 (8.5%) had a recurrent infection on days 28, 35 and 42 as late treatment failure (LTF) (Table 3).

Single nucleotide polymorphisms in the Pfk13 gene
Of the 101 DNA specimens analysed, 94 gave full amplicons.None of the 20 SNPs previously reported to be associated with ART resistance were found.As shown in Fig. 3, the overall prevalence of the non-synonymous new mutant allele in BTB/POZ and the propeller domain was found in Sumba with a percentage of 4/94 (4.25%, 95% CI 0.94-1; see Additional file 1: Fig. S2, Table S2; Table 4) at positions L396F, I526V, N537S and M579I.
Eighty samples were successfully measured for pfmdr1 copy numbers, of which 3/80 (3.8%) were multiple copies.In Sumba and Papua, parasites with pfmdr1 amplification tend to disappeared after AS-AQ was replaced with DHA-PPQ in 2010 (Table 4).This study did not observe the concomitant amplification of both pfpm2/3 and pfmdr1 (Table 5).

Single nucleotide polymorphisms in the pfcrt gene
Only 31 of the 101 specimens were successfully PCR amplified for the pfcrt gene due to a lack of parasite DNA.None of the pfcrt mutations linked to piperaquine resistance were observed.Although the number of samples was small, approximately 71% (22/31) of all isolates from both study areas had pfcrt I356L, as shown in Table 4 and Additional file 1: Table S2.S2 show that approximately 79.6% (74/93) of isolates had the wild-type SND haplotype.In Sumba, the mutant haplotype SDD was found in onethird of the isolates (15/48; 31.3%); in Papua, the haplotype was found in 8.9% (4/45).In Papua, during 2 years of observation, prevalence decreased from 10.3% (3 of 29) to 6.3% (1 of 16) (Table 4).As both alleles (1034 and 1246) were observed at multiple locations in Indonesia across the region [32,33], similar observations in pfmdr1 (1034C and 1246Y) were not found in any of the isolates examined.

Clinical and parasitological characteristics of Papuan TES
Analysis of one hundred forty-two P. falciparum-infected subjects (Table 1) 117 cases completed the 42 day-followup, and 25 cases were either lost to follow-up (LFU) or withdrawn (WTH).The classification of the treatment outcomes by PCR correction is presented in Table 3.At day 42, ACPR was noted in 91.5% (95% CI 84.8-95.8).Of the 10 LTF cases, nine were re-infections and one was confirmed as recrudescent (Table 3).Therefore, the PCRcorrected DHA-PPQ efficacy for falciparum was 99.1% (95% CI 94.9-100.0).No delay in parasite clearance at day 3 was observed in any isolates.

Discussion
In Indonesia, AS-AQ was introduced in 2004 and poor tolerability was reported.Increasing treatment failure rates of AS-AQ led to drug policy change to DHA-PPQ in 2008.DHA-PPQ resistance has emerged in South-East Asia, posing a significant threat to malaria control and elimination efforts [34][35][36].DHA-PPQ is well tolerated, with a faster ART derivative clearing parasites and the active compound PPQ removing the remaining parasites more slowly [37].TES conducted in several parts of Indonesia [17,19] revealed that DHA-PPQ is still highly effective, evidenced by the absence of delayed parasite clearance and low cases of parasite recurrence after 42-day observations.Analysis of the pfk13 gene of P. falciparum isolates collected from 2010 to 2021 revealed the absence of mutations associated with ART resistance.However, several polymorphisms, such as L396F, I526V, N537S, and M579I, were found.Although not associated with ART resistance, these SNPs are newly found and have never been described in any endemic area [16, 18-20, 23, 24, 34-50].It is still important to describe further the roles of the four SNPs.The average PCT result of 1.5 days and no delay in parasite clearance at day 3 was observed in any isolates.The TES results from Papua do not meet the WHO criteria for suspected ART resistance [51].
Resistance to PPQ has been associated with the increasing copy number of the pfpm2 gene [28,52,53] and more recently with mutant alleles of pfcrt [54][55][56][57][58].This research result on 10 recurrent infection cases revealed no association of PPQ resistance with pfpm2/3 copy number (Table 5).Prior TES analysis in the southern region of Papua also did not identify any P. falciparum isolates carrying multiple copies of the pfpm2/3 gene [18].Other studies proposed pfcrt mutations associated with PPQ resistance, namely 343, 350, or 353 [24,25].Those mutations were not observed in this study.After the deployment of DHA-PPQ, the prevalence of parasites with multiple copies of pfpm2/3 slowly increased from 2016 to 2018 and seemed to be relatively prevalent in no recurrent isolates.All this happened probably due to the selective exposure of DHA-PPQ over more than 10 years of adoption in Indonesia.The limited number of samples used in this study could also be a determining factor, so it will need further investigation to confirm.Although there was no evidence of an increased copy number of pfpm2/3 in recrudescence cases, the PPQ treatment still failed to eradicate the parasite from the blood and prevent reinfection during the follow-up period.Surveillance in areas relying on DHA-PPQ as firstline anti-malarial treatment indicated that pfpm2/3 amplification was not the sole factor rendering PPQ resistance.The genetic background of circulating field isolates appeared to play a role in drug susceptibility [36,37].It is also supported by Fidock et al. [42] and Iwanaga et al. [59], specifically revealing that a drugresistant strain's successful production is directly generated in a drug-sensitive strain via in vitro study or genome-wide functional screening of drug resistance.These transformations were influenced by geographical origin (South East Asia, Africa and South America) and genetic background (haplotype allele or genotype of all other related genes), supporting the population survey results that the mutated pfcrt was possibly sufficient to confer resistance [60].It was also suggested that the initial selection of pfpm2/3 and pfmdr1 CNVs, although a PPQ-resistant phenotype does not emerge, it developed a genetic background for novel pfcrt mutations [61].
Pfcrt is a 13-exon gene with several point mutations 74,75,76,220,271,326,356, 371 that connect with CQR [42] and 93, 97, 145, 218, 343, 350, 353 exclusively associated with PPQ resistance [36,60,62] located on chromosome 7 spanning from 36 kb segment.The overlapped region was correlated to the pfcrt gene and its regulatory elements such as the promoter and 3′ untranslated region (3′ UTR) responsible for regulating and activating the coding region [59].The mutations in pfcrt might interfere with transporting the natural substrates out of the digestive vacuole, resulting in increased osmotic pressure.This phenotype was also observed in Dd2 parasites expressing the pfcrt mutations F145I, M343L, and G353V [60].Not all novel pfcrt mutations exhibit a swollen DV phenotype, depending on the location of the mutated amino acids [63].
Intriguingly, pfcrt I356T/L mutation also increased ART IC50 values and resistance [38,39], emphasizing the recent correlation of the I356T mutation in Southeast Asia with ART-resistant parasite populations in the pfk13 mutation.[40,41].Several pfcrt haplotype lines from many geographic regions serve as genetic backgrounds, with 356 alleles as one of them was associated with the development of ART resistance in P. falciparum parasites [38][39][40]42].The presence of moderate frequencies of pfcrt I356L 22/31 (71%) in the study site was possibly associated with long-time drug-selected pressure from DHA-PPQ treatment used and the availability of access to CQ in the private health sectors (non-malaria purposes) that could facilitate the evolution of ART resistance pfcrt alleles [16,39].
Pfmdr1 was also proposed as a modulator for PPQ resistance.[28,34] Resistance to PPQ was also associated with the amplification of pfmdr1.Another in-vitro study reported a correlation between a single copy number of pfmdr1 and P. falciparum isolates resistance to PPQ [35,36,43].Plasmodium falciparum parasites might suffer a fitness disadvantage or reduced transmissibility if pfmdr1 gene is amplified more frequently [47].By contrast, the multiple copy number of pfmdr1 was associated with MQ resistance [29,37,44,64].This study revealed that P. falciparum isolates in Indonesia predominantly posed single copy of pfmdr1 95%, suggesting reduced PPQ effectiveness.Previous studies observed a reduced prevalence of multicopy pfmdr1 since adopting DHA-PPQ [28,37,45].However, the role of pfmdr1 remained controversial [43,46].
Besides decreasing CNVs of pfmdr1, polymorphism in the pfmdr1 gene, such as N1042D increased by 76.9% before the introduction of PPQ treatment.The result in another eastern part of Indonesia SNP mutation S1034C was observed to reach 100% in 2010 [49].After PPQ was introduced, the pfmdr1 S1034C, N1042D and D1246Y decreased significantly by 20.4%.There was a notable

Fig. 1
Fig. 1 Flowchart for samples set study

Fig. 3 4
Fig. 3 Map of Indonesia indicating molecular marker of resistance and sampling locations of the P. falciparum field isolates during observation study.Map source from Natural Earth (https:// www.natur alear thdata.com) and modified according to data from the references

Table 1
Demographic characteristics from Papua TES samples

Table 2
Treatment outcome from Papua TES during the 42 day of follow-up ETF early treatment failure, LCF late clinical failure, LPF late parasitological failure, LFU lost to follow up, WTH withdrawn a Kaplan-Meier analysis