A likelihood approach to estimate the number of co-infections

Review of MrsFreqPhase methods: methods designed to estimate statistically malaria parasite multiplicity of infection, relatedness, frequency and phase

Malaria parasites are haploid within humans, but infections often contain genetically distinct groups of clonal parasites. When the per-infection number of genetically distinct clones (i.e., the multiplicity of infection, MOI) exceeds one, and per-infection genetic data are generated in bulk, important information are obfuscated. For example, the MOI, the phases of the haploid genotypes of genetically distinct clones (i.e., how the alleles concatenate into sequences), and their frequencies. This complicates many downstream analyses, including relatedness estimation. MOIs, parasite sequences, their frequencies, and degrees of relatedness are used ubiquitously in malaria studies: for example, to monitor anti-malarial drug resistance and to track changes in transmission. In this article, MrsFreqPhase methods designed to estimate statistically malaria parasite MOI, relatedness, frequency and phase are reviewed. An overview, a historical account of the literature, and a statistical description of contemporary software is provided for each method class. The article ends with a look towards future method development, needed to make best use of new data types generated by cutting-edge malaria studies reliant on MrsFreqPhase methods.

Estimating multiplicity of infection, allele frequencies, and prevalences accounting for incomplete data

BACKGROUND

Molecular surveillance of infectious diseases allows the monitoring of pathogens beyond the granularity of traditional epidemiological approaches and is well-established for some of the most relevant infectious diseases such as malaria. The presence of genetically distinct pathogenic variants within an infection, referred to as multiplicity of infection (MOI) or complexity of infection (COI) is common in malaria and similar infectious diseases. It is an important metric that scales with transmission intensities, potentially affects the clinical pathogenesis, and a confounding factor when monitoring the frequency and prevalence of pathogenic variants. Several statistical methods exist to estimate MOI and the frequency distribution of pathogen variants. However, a common problem is the quality of the underlying molecular data. If molecular assays fail not randomly, it is likely to underestimate MOI and the prevalence of pathogen variants.

METHODS AND FINDINGS

A statistical model is introduced, which explicitly addresses data quality, by assuming a probability by which a pathogen variant remains undetected in a molecular assay. This is different from the assumption of missing at random, for which a molecular assay either performs perfectly or fails completely. The method is applicable to a single molecular marker and allows to estimate allele-frequency spectra, the distribution of MOI, and the probability of variants to remain undetected (incomplete information). Based on the statistical model, expressions for the prevalence of pathogen variants are derived and differences between frequency and prevalence are discussed. The usual desirable asymptotic properties of the maximum-likelihood estimator (MLE) are established by rewriting the model into an exponential family. The MLE has promising finite sample properties in terms of bias and variance. The covariance matrix of the estimator is close to the Cramér-Rao lower bound (inverse Fisher information). Importantly, the estimator's variance is larger than that of a similar method which disregards incomplete information, but its bias is smaller.

CONCLUSIONS

Although the model introduced here has convenient properties, in terms of the mean squared error it does not outperform a simple standard method that neglects missing information. Thus, the new method is recommendable only for data sets in which the molecular assays produced poor-quality results. This will be particularly true if the model is extended to accommodate information from multiple molecular markers at the same time, and incomplete information at one or more markers leads to a strong depletion of sample size.

The diversity of Plasmodium falciparum isolates from asymptomatic and symptomatic school-age children in Kinshasa Province, Democratic Republic of Congo

BACKGROUND

Understanding Plasmodium falciparum population diversity and transmission dynamics provides information on the intensity of malaria transmission, which is needed for assessing malaria control interventions. This study aimed to determine P. falciparum allelic diversity and multiplicity of infection (MOI) among asymptomatic and symptomatic school-age children in Kinshasa Province, Democratic Republic of Congo (DRC).

METHODS

A total of 438 DNA samples (248 asymptomatic and 190 symptomatic) were characterized by nested PCR and genotyping the polymorphic regions of pfmsp1 block 2 and pfmsp2 block 3.

RESULTS

Nine allele types were observed in pfmsp1 block2. The K1-type allele was predominant with 78% (229/293) prevalence, followed by the MAD20-type allele (52%, 152/293) and RO33-type allele (44%, 129/293). Twelve alleles were detected in pfmsp2, and the 3D7-type allele was the most frequent with 84% (256/304) prevalence, followed by the FC27-type allele (66%, 201/304). Polyclonal infections were detected in 63% (95% CI 56, 69) of the samples, and the MOI (SD) was 1.99 (0.97) in P. falciparum single-species infections. MOIs significantly increased in P. falciparum isolates from symptomatic parasite carriers compared with asymptomatic carriers (2.24 versus 1.69, adjusted b: 0.36, (95% CI 0.01, 0.72), p = 0.046) and parasitaemia > 10,000 parasites/µL compared to parasitaemia < 5000 parasites/µL (2.68 versus 1.63, adjusted b: 0.89, (95% CI 0.46, 1.25), p < 0.001).

CONCLUSION

This survey showed low allelic diversity and MOI of P. falciparum, which reflects a moderate intensity of malaria transmission in the study areas. MOIs were more likely to be common in symptomatic infections and increased with the parasitaemia level. Further studies in different transmission zones are needed to understand the epidemiology and parasite complexity in the DRC.

Drought correlates with reduced infection complexity and possibly prevalence in a decades-long study of the lizard malaria parasite Plasmodium mexicanum

Microparasites often exist as a collection of genetic 'clones' within a single host (termed multi-clonal, or complex, infections). Malaria parasites are no exception, with complex infections playing key roles in parasite ecology. Even so, we know little about what factors govern the distribution and abundance of complex infections in natural settings. Utilizing a natural dataset that spans more than 20 years, we examined the effects of drought conditions on infection complexity and prevalence in the lizard malaria parasite Plasmodium mexicanum and its vertebrate host, the western fence lizard, Sceloporus occidentalis. We analyzed data for 14,011 lizards sampled from ten sites over 34 years with an average infection rate of 16.2%. Infection complexity was assessed for 546 infected lizards sampled during the most recent 20 years. Our data illustrate significant, negative effects of drought-like conditions on infection complexity, with infection complexity expected to increase by a factor of 2.27 from the lowest to highest rainfall years. The relationship between rainfall and parasite prevalence is somewhat more ambiguous; when prevalence is modeled over the full range in years, a 50% increase in prevalence is predicted between the lowest and highest rainfall years, but this trend is not apparent or is reversed when data are analyzed over a shorter timeframe. To our knowledge, this is the first reported evidence for drought affecting the abundance of multi-clonal infections in malaria parasites. It is not yet clear what mechanism might connect drought with infection complexity, but the correlation we observed suggests that additional research on how drought influences parasite features like infection complexity, transmission rates and within-host competition may be worthwhile.

The many definitions of multiplicity of infection

The presence of multiple genetically different pathogenic variants within the same individual host is common in infectious diseases. Although this is neglected in some diseases, it is well recognized in others like malaria, where it is typically referred to as multiplicity of infection (MOI) or complexity of infection (COI). In malaria, with the advent of molecular surveillance, data is increasingly being available with enough resolution to capture MOI and integrate it into molecular surveillance strategies. The distribution of MOI on the population level scales with transmission intensities, while MOI on the individual level is a confounding factor when monitoring haplotypes of particular interests, e.g., those associated with drug-resistance. Particularly, in high-transmission areas, MOI leads to a discrepancy between the likelihood of a haplotype being observed in an infection (prevalence) and its abundance in the pathogen population (frequency). Despite its importance, MOI is not universally defined. Competing definitions vary from verbal ones to those based on concise statistical frameworks. Heuristic approaches to MOI are popular, although they do not mine the full potential of available data and are typically biased, potentially leading to misinferences. We introduce a formal statistical framework and suggest a concise definition of MOI and its distribution on the host-population level. We show how it relates to alternative definitions such as the number of distinct haplotypes within an infection or the maximum number of alleles detectable across a set of genetic markers. It is shown how alternatives can be derived from the general framework. Different statistical methods to estimate the distribution of MOI and pathogenic variants at the population level are discussed. The estimates can be used as plug-ins to reconstruct the most probable MOI of an infection and set of infecting haplotypes in individual infections. Furthermore, the relation between prevalence of pathogenic variants and their frequency (relative abundance) in the pathogen population in the context of MOI is clarified, with particular regard to seasonality in transmission intensities. The framework introduced here helps to guide the correct interpretation of results emerging from different definitions of MOI. Especially, it excels comparisons between studies based on different analytical methods.

A maximum-likelihood method to estimate haplotype frequencies and prevalence alongside multiplicity of infection from SNP data

The introduction of genomic methods facilitated standardized molecular disease surveillance. For instance, SNP barcodes in Plasmodium vivax and Plasmodium falciparum malaria allows the characterization of haplotypes, their frequencies and prevalence to reveal temporal and spatial transmission patterns. A confounding factor is the presence of multiple genetically distinct pathogen variants within the same infection, known as multiplicity of infection (MOI). Disregarding ambiguous information, as usually done in ad-hoc approaches, leads to less confident and biased estimates. We introduce a statistical framework to obtain maximum-likelihood estimates (MLE) of haplotype frequencies and prevalence alongside MOI from malaria SNP data, i.e., multiple biallelic marker loci. The number of model parameters increases geometrically with the number of genetic markers considered and no closed-form solution exists for the MLE. Therefore, the MLE needs to be derived numerically. We use the Expectation-Maximization (EM) algorithm to derive the maximum-likelihood estimates, an efficient and easy-to-implement algorithm that yields a numerically stable solution. We also derive expressions for haplotype prevalence based on either all or just the unambiguous genetic information and compare both approaches. The latter corresponds to a biased ad-hoc estimate of prevalence. We assess the performance of our estimator by systematic numerical simulations assuming realistic sample sizes and various scenarios of transmission intensity. For reasonable sample sizes, and number of loci, the method has little bias. As an example, we apply the method to a dataset from Cameroon on sulfadoxine-pyrimethamine resistance in P. falciparum malaria. The method is not confined to malaria and can be applied to any infectious disease with similar transmission behavior. An easy-to-use implementation of the method as an R-script is provided.

Bias-corrected maximum-likelihood estimation of multiplicity of infection and lineage frequencies

Background

The UN’s Sustainable Development Goals are devoted to eradicate a range of infectious diseases to achieve global well-being. These efforts require monitoring disease transmission at a level that differentiates between pathogen variants at the genetic/molecular level. In fact, the advantages of genetic (molecular) measures like multiplicity of infection (MOI) over traditional metrics, e.g., R₀, are being increasingly recognized. MOI refers to the presence of multiple pathogen variants within an infection due to multiple infective contacts. Maximum-likelihood (ML) methods have been proposed to derive MOI and pathogen-lineage frequencies from molecular data. However, these methods are biased.

Methods and findings

Based on a single molecular marker, we derive a bias-corrected ML estimator for MOI and pathogen-lineage frequencies. We further improve these estimators by heuristical adjustments that compensate shortcomings in the derivation of the bias correction, which implicitly assumes that data lies in the interior of the observational space. The finite sample properties of the different variants of the bias-corrected estimators are investigated by a systematic simulation study. In particular, we investigate the performance of the estimator in terms of bias, variance, and robustness against model violations. The corrections successfully remove bias except for extreme parameters that likely yield uninformative data, which cannot sustain accurate parameter estimation. Heuristic adjustments further improve the bias correction, particularly for small sample sizes. The bias corrections also reduce the estimators’ variances, which coincide with the Cramér-Rao lower bound. The estimators are reasonably robust against model violations.

Conclusions

Applying bias corrections can substantially improve the quality of MOI estimates, particularly in areas of low as well as areas of high transmission—in both cases estimates tend to be biased. The bias-corrected estimators are (almost) unbiased and their variance coincides with the Cramér-Rao lower bound, suggesting that no further improvements are possible unless additional information is provided. Additional information can be obtained by combining data from several molecular markers, or by including information that allows stratifying the data into heterogeneous groups.

Distribution of clones among hosts for the lizard malaria parasite Plasmodium mexicanum

Background

Malaria parasites reproduce asexually, leading to the production of large numbers of genetically identical parasites, here termed a clonal line or clone. Infected hosts may harbor one or more clones, and the number of clones in a host is termed multiplicity of infection (MOI). Understanding the distribution of parasite clones among hosts can shed light on the processes shaping this distribution and is important for modeling MOI. Here, I determine whether the distribution of clones of the lizard malaria parasite Plasmodium mexicanum differ significantly from statistical distributions commonly used to model MOI and logical extensions of these models.

Methods

The number of clones per infection was assessed using four microsatellite loci with the maximum number of alleles at any one locus used as a simple estimate of MOI for each infection. I fit statistical models (Poisson, negative binomial, zero-inflated models) to data from four individual sites to determine a best fit model. I also simulated the number of alleles per locus using an unbiased estimate of MOI to determine whether the simple (but potentially biased) method I used to estimate MOI influenced model fit.

Results

The distribution of clones among hosts at individual sites differed significantly from traditional Poisson and negative binomial distributions, but not from zero-inflated modifications of these distributions. A consistent excess of two-clone infections and shortage of one-clone infections relative to all fit distributions was also observed. Any bias introduced by the simple method for estimating of MOI did not appear to qualitatively alter the results.

Conclusions

The statistical distributions used to model MOI are typically zero-truncated; truncating the Poisson or zero-inflated Poisson yield the same distribution, so the reasonable fit of the zero-inflated Poisson to the data suggests that the use of the zero-truncated Poisson in modeling is adequate. The improved fit of zero-inflated distributions relative to standard distributions may suggest that only a portion of the host population is located in areas suitable for transmission even at small sites (<1 ha). Collective transmission of clones and premunition may also contribute to deviations from standard distributions.

Malaria in Venezuela: changes in the complexity of infection reflects the increment in transmission intensity

Background

Malaria incidence has reached staggering numbers in Venezuela. Commonly, Bolívar State accounted for approximately 70% of the country cases every year. Most cases cluster in the Sifontes municipality, a region characterized by an extractive economy, including gold mining. An increase in migration to Sifontes, driven by gold mining, fueled a malaria spillover to the rest of the country and the region. Here samples collected in 2018 were compared with a previous study of 2003/2004 to describe changes in the parasites population structures and the frequency of point mutations linked to anti-malarial drugs.

Methods

A total of 88 Plasmodium falciparum and 94 Plasmodium vivax isolates were collected in 2018 and compared with samples from 2003/2004 (106 P. falciparum and 104 P. vivax). For P. falciparum, mutations linked to drug resistance (Pfdhfr, Pfdhps, and Pfcrt) and the Pfk13 gene associated with artemisinin delayed parasite clearance, were analysed. To estimate the multiplicity of infection (MOI), and perform P. falciparum and P. vivax population genetic analyses, the parasites were genotyped by using eight standardized microsatellite loci.

Results

The P. falciparum parasites are still harbouring drug-resistant mutations in Pfdhfr, Pfdhps, and Pfcrt. However, there was a decrease in the frequency of highly resistant Pfdhps alleles. Mutations associated with artemisinin delayed parasite clearance in the Pfk13 gene were not found. Consistent with the increase in transmission, polyclonal infections raised from 1.9% in 2003/2004 to 39% in 2018 in P. falciparum and from 16.3 to 68% in P. vivax. There is also a decrease in linkage disequilibrium. Bayesian clustering yields two populations linked to the time of sampling, showing that the parasite populations temporarily changed. However, the samples from 2003/2004 and 2018 have several alleles per locus in common without sharing multi-locus genotypes.

Conclusions

The frequency of mutations linked with drug resistance in P. falciparum shows only changes in Pfdhps. Observations presented here are consistent with an increase in transmission from the previously circulating parasites. Following populations longitudinally, using molecular surveillance, provides valuable information in cases such as Venezuela with a fluid malaria situation that is affecting the regional goals toward elimination.

Limited differentiation among Plasmodium vivax populations from the northwest and to the south Pacific Coast of Colombia: A malaria corridor?

Background

Malaria remains endemic in several countries of South America with low to moderate transmission intensity. Regional human migration through underserved endemic areas may be responsible for significant parasite dispersion making the disease resilient to interventions. Thus, the genetic characterization of malarial parasites is an important tool to assess how endemic areas may connect via the movement of infected individuals. Here, four sites in geographically separated areas reporting 80% of the malaria morbidity in Colombia were studied. The sites are located on an imaginary transect line of 1,500 km from the northwest to the south Pacific Coast of Colombia with a minimal distance of 500 km between populations that display noticeable ethnic, economic, epidemiological, and ecological differences.

Methodology/Principal findings

A total of 624 Plasmodium vivax samples from the four populations were genotyped by using eight microsatellite loci. Although a strong geographic structure was expected between these populations, only moderate evidence of genetic differentiation was observed using a suite of population genetic analyses. High genetic diversity, shared alleles, and low linkage disequilibrium were also found in these P. vivax populations providing no evidence for a bottleneck or clonal expansions as expected from recent reductions in the transmission that could have been the result of scaling up interventions or environmental changes. These patterns are consistent with a disease that is not only endemic in each site but also imply that there is gene flow among these populations across 1,500 km.

Conclusion /Significance

The observed patterns in P. vivax are consistent with a “corridor” where connected endemic areas can sustain a high level of genetic diversity locally and can restore parasite-subdivided populations via migration of infected individuals even after local interventions achieved a substantial reduction of clinical cases. The consequences of these findings in terms of control and elimination are discussed.

The regional movements of infected individuals that connect suitable transmission areas make malaria resilient to control efforts. Those movements are expected to leave genetic signatures in the parasite populations that can be detected using analytical tools. In this study, the genetic makeups of Plasmodium vivax populations were characterized to assess whether the most endemic areas in Colombia were connected. Samples were collected from passive surveillance studies in four locations across an imaginary transect line of 1,500 km from the northwest to the south Pacific Coast of Colombia (South America). Considering the distance, and contrary to expectations, we found weak levels of genetic differentiation between these parasite populations with no evidence indicating that their genetic diversity has been eroded as expected whenever the prevalence of the disease is successfully reduced, e.g., through control programs or environmental changes. Although the sampling lacks the geographic and temporal detail to describe how the dispersion of parasite lineages occurred, the observed patterns are consistent with a series of infected populations that are connected in space by human movements allowing the parasite to diffuse across this 1,500 km transect. This malaria corridor needs to be characterized to achieve elimination.

Large and finite sample properties of a maximum-likelihood estimator for multiplicity of infection

Reliable measures of transmission intensities can be incorporated into metrics for monitoring disease-control interventions. Genetic (molecular) measures like multiplicity of infection (MOI) have several advantages compared with traditional measures, e.g., R0. Here, we investigate the properties of a maximum-likelihood approach to estimate MOI and pathogen-lineage frequencies. By verifying regulatory conditions, we prove asymptotical unbiasedness, consistency and efficiency of the estimator. Finite sample properties concerning bias and variance are evaluated over a comprehensive parameter range by a systematic simulation study. Moreover, the estimator's sensitivity to model violations is studied. The estimator performs well for realistic sample sizes and parameter ranges. In particular, the lineage-frequency estimates are almost unbiased independently of sample size. The MOI estimate's bias vanishes with increasing sample size, but might be substantial if sample size is too small. The estimator's variance matrix agrees well with the Cramér-Rao lower bound, even for small sample size. The numerical and analytical results of this study can be used for study design. This is exemplified by a malaria data set from Venezuela. It is shown how the results can be used to determine the necessary sample size to achieve certain performance goals. An implementation of the likelihood method and a simulation algorithm for study design, implemented as an R script, is available as S1 File alongside a documentation (S2 File) and example data (S3 File).

Extensive diversity in the allelic frequency of Plasmodium falciparum merozoite surface proteins and glutamate-rich protein in rural and urban settings of southwestern Nigeria

Background

Nigeria carries a high burden of malaria which makes continuous surveillance for current information on genetic diversity imperative. In this study, the merozoite surface proteins (msp-1, msp-2) and glutamate-rich protein (glurp) of Plasmodium falciparum collected from two communities representing rural and urban settings in Ibadan, southwestern Nigeria were analysed.

Methods

A total of 511 febrile children, aged 3–59 months, whose parents/guardians provided informed consent, were recruited into the study. Capillary blood was obtained for malaria rapid diagnostic test, thick blood smears for parasite count and blood spots on filter paper for molecular analysis.

Results

Three-hundred and nine samples were successfully genotyped for msp-1, msp-2 and glurp genes. The allelic distribution of the three genes was not significantly different in the rural and urban communities. R033 and 3D7 were the most prevalent alleles in both rural and urban communities for msp-1 and msp-2, respectively. Eleven of glurp RII region genotypes, coded I–XII, with sizes ranging from 500 to 1100 base pairs were detected in the rural setting. Genotype XI (1000–1050 bp) had the highest prevalence of 41.5 and 38.5% in rural and urban settings, respectively. Overall, 82.1 and 70.0% of samples had multiclonal infection with msp-1 gene resulting in a mean multiplicity of infection (MOI) of 2.8 and 2.6 for rural and urban samples, respectively. Msp-1 and msp-2 genes displayed higher levels of diversity and higher MOI rates than the glurp gene.

Conclusion

Significant genetic diversity was observed between rural and urban parasite populations in Ibadan, southwestern Nigeria. The results of this study show that malaria transmission intensity in these regions is still high. No significant difference was observed between rural and urban settings, except for a completely different msp-1 allele, compared to previous reports, thereby confirming the changing face of malaria transmission in these communities. This study provides important baseline information required for monitoring the impact of malaria elimination efforts in this region and data points useful in revising current protocols.

Multiplicity of Infection and Disease Severity in Plasmodium vivax

Background

Multiplicity of infection (MOI) refers to the average number of distinct parasite genotypes concurrently infecting a patient. Although several studies have reported on MOI and the frequency of multiclonal infections in Plasmodium falciparum, there is limited data on Plasmodium vivax. Here, MOI and the frequency of multiclonal infections were studied in areas from South America where P. vivax and P. falciparum can be compared.

Methodology/Principal Findings

As part of a passive surveillance study, 1,328 positive malaria patients were recruited between 2011 and 2013 in low transmission areas from Colombia. Of those, there were only 38 P. vivax and 24 P. falciparum clinically complicated cases scattered throughout the time of the study. Samples from uncomplicated cases were matched in time and location with the complicated cases in order to compare the circulating genotypes for these two categories. A total of 92 P. vivax and 57 P. falciparum uncomplicated cases were randomly subsampled. All samples were genotyped by using neutral microsatellites. Plasmodium vivax showed more multiclonal infections (47.7%) than P. falciparum (14.8%). Population genetics and haplotype network analyses did not detect differences in the circulating genotypes between complicated and uncomplicated cases in each parasite. However, a Fisher exact test yielded a significant association between having multiclonal P. vivax infections and complicated malaria. No association was found for P. falciparum infections.

Conclusion

The association between multiclonal infections and disease severity in P. vivax is consistent with previous observations made in rodent malaria. The contrasting pattern between P. vivax and P. falciparum could be explained, at least in part, by the fact that P. vivax infections have lineages that were more distantly related among them than in the case of the P. falciparum multiclonal infections. Future research should address the possible role that acquired immunity and exposure may have on multiclonal infections and their association with disease severity.

Previous studies on rodent malarias and mathematical models have postulated a link between multiclonal infections and disease severity. This association has been tested in Plasmodium falciparum mostly in Africa with limited information on P. vivax. Furthermore, there is a paucity of information from areas with low transmission. Here, we used samples available from a passive surveillance carried out in Colombia, South America. We found an association between multiclonal infections and disease severity in P. vivax but not in P. falciparum. Although the number of complicated malaria cases is low, the contrasting pattern between these two species emphasizes their epidemiological differences. We discuss how this pattern could be the result of a higher divergence among the P. vivax lineages co-infecting a patient. We hypothesize that low levels of acquired immunity may play a role in the association between multiclonal infections and disease severity.

Malaria Molecular Epidemiology: Lessons from the International Centers of Excellence for Malaria Research Network

Molecular epidemiology leverages genetic information to study the risk factors that affect the frequency and distribution of malaria cases. This article describes molecular epidemiologic investigations currently being carried out by the International Centers of Excellence for Malaria Research (ICEMR) network in a variety of malaria-endemic settings. First, we discuss various novel approaches to understand malaria incidence and gametocytemia, focusing on Plasmodium falciparum and Plasmodium vivax. Second, we describe and compare different parasite genotyping methods commonly used in malaria epidemiology and population genetics. Finally, we discuss potential applications of molecular epidemiological tools and methods toward malaria control and elimination efforts.

Persistence of Sulfadoxine-Pyrimethamine Resistance Despite Reduction of Drug Pressure in Malawi

BACKGROUND

In 2007, Malawi replaced sulfadoxine-pyrimethamine (SP) with an artemisinin-based combination therapy as the first-line treatment for uncomplicated Plasmodium falciparum malaria in response to failing SP efficacy. Here we estimate the effect of reduced SP pressure on the prevalence of SP-resistant parasites and the characteristics of the associated selective sweeps flanking the resistance loci.

METHODS

Samples obtained from individuals with clinical malaria during a period of high SP use (1999-2001), a transitional period (2007-2008), and a period of low SP use (2012) were genotyped for resistance markers at pfdhfr-ts codons 51, 59, and 108 and pfdhps codons 437, 540, and 581. Expected heterozygosity was estimated to evaluate the genetic diversity flanking pfdhfr-ts and pfdhps.

RESULTS

An increase in the prevalence of the resistance haplotypes DHFR 51I/59R/108N and DHPS 437G/540E occurred under sustained drug pressure, with no change in haplotype prevalence 5 years after reduction in SP pressure. The DHPS 437G/540E/581G haplotype was observed in 2007 and increased in prevalence during a period of reduced SP pressure. Changes to the sweep characteristics flanking pfdhfr-ts and pfdhps were minimal.

CONCLUSIONS

In contrast to the rapid and complete return of chloroquine-susceptible falciparum malaria after chloroquine was withdrawn from Malawi, a reemergence of SP efficacy is unlikely in the near future.