Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Infect Dis. Author manuscript; available in PMC 2010 August 15.
Published in final edited form as:
PMCID: PMC2761972

Misclassification of Drug Failures in Plasmodium falciparum Clinical Trials in Southeast Asia


Most trials of antimalarials occur in areas where reinfections are possible. For Plasmodium falciparum, reinfections are distinguished from recrudescences by PCR analysis of 3 polymorphic genes. However, the validity of this approach has never been rigorously tested. We tested for misclassification in 6 patients from clinical trials in Thailand and Cambodia who were classified as reinfected by the standard PCR protocol. Using heteroduplex tracking assays and direct DNA sequencing, we found that 5 of 6 (83%) patients were misclassified. Misclassification in this manner overestimates the efficacy of antimalarials and delays recognition of decreasing therapeutic efficacy, thus delaying potential policy changes.

Keywords: Plasmodium falciparum, Malaria, Clinical trial, Heteroduplex tracking assay

Drug-resistant Plasmodium falciparum malaria continues to be a growing health problem through most of the world, thus making the development of new antimalarial drugs of vital importance [1]. Antimalarial clinical trials are often carried out in areas where malaria is endemic and patients who are successfully treated can potentially be reinfected during follow-up. Thus, antimalarial efficacy can only be evaluated if reinfections can be distinguished from recrudescences, a process complicated by the fact that these patients are usually infected with a mixture of genetically distinct variants [14].

A recent expert panel proposed distinguishing recrudescence from reinfection using PCR amplification and agarose gel electrophoresis of 3 polymorphic genes: merozoite surface protein 1 (msp-1), merozoite surface protein 2 (msp-2) and glutamine-rich protein (glurp) [5]. The report suggested that “a new infection is defined as subsequently occurring parasitemia in which all alleles in the post treatment sample of a patient are completely different from those in the admission sample for one or more loci tested [5].” It also proposes that a recurrent parasitemia be classified as a recrudescence if at least one allele at each locus is common to paired pre and post-treatment samples. However, there are few well controlled evaluations of these definitions. We have previously shown that this assumption may be incorrect in high transmission areas where a small number of variants predominate [4].

The currently recommended strategy involves sequential analysis of the three genes from most polymorphic, msp-2 or glurp, to least polymorphic, msp-1 [5]. However, it is clear that the discriminatory ability of this procedure to identify unique variants is questionable [4, 6]. The use of three markers is meant to overcome these issues. However, it has been suggested that increasing the number of markers used may increase that chance that a given variant might not be amplified at one of the markers [7]. Thus, using fewer sites with methods that better describe the diversity at each site could be beneficial [5].

Within any patient, the complexity of infection (COI) can be high, with as many as 13 variants seen in a given sample [4]. Many variants may represent <20% of the total parasite population (minority variants). It has been shown that nested PCR and PCR analysis with sequencing are insensitive to these minority variants in malaria and other polyclonal infections, such as HIV [6, 810]. Therefore, shared msp-1, msp-2 or glurp variants could potentially be missed leading to misclassification of recrudescences as reinfections, thus over-estimating drug efficacy.

Heteroduplex tracking assays (HTAs), an alternative genotyping strategy, are able to detect minority variants and both size and sequence polymorphism [4, 6, 11]. We have previously used a HTA of msp-1 block 2 to describe COI in areas of low and high malaria transmission and describe a new msp-2 HTA assay here [4, 11].

In order to determine if standard genotyping methods are misclassifying recrudesences as reinfections, we evaluated paired enrollment and recurrent parasitemia samples from 12 patients in two studies: 1) a Phase II clinical trial of DB-289 at the Hospital for Tropical Diseases in Bangkok, Thailand and 2) an in vivo efficacy study of artesunate-mefloquine in Chumkiri, Cambodia. All of these patients had little to no chance of reinfection clinically. We compared two criteria for the classification of enrollment and recurrent parasitemias: 1) classification of the samples by msp-1, msp-2, and glurp PCR, as recommended by the MMV/WHO report [5] or 2) by HTA analysis of the samples at two sites, msp-1 and msp-2. Using both methods, recurrences were classified as reinfections if shared bands were absent from any gel comparing enrollment and recurrent infections as recommended by the WHO/MMV report. Pre- and post-treatment samples from 6 patients classified as reinfection by standard PCR, as well as a pre and post treatment samples from 6 patients classified as recrudesences, were studied.

Patients, materials, and methods

P. falciparum DNA was extracted from paired de-identified dried blood spots at enrollment and recurrence of parasitemia. Informed consent was received from all patients in both trials. Twenty eight patients in the Phase II clinical trial (Thailand) and 31 patients in the in vivo clinical efficacy study (Cambodia) experienced recurrent parasitemia. Details of the studies are included in the on-line supplemental material and will be published elsewhere. PCR analysis of msp-1, msp-2 and glurp was completed on all samples. The PCR methods have been described previously [12]. Gels were reviewed independently by two authors.

In total, six patients from these two trials were classified as reinfection by PCR (Table 1). In addition to the six sample pairs classified as reinfection, six sample pairs classified as recrudesences were also selected for analysis by HTA. There were no significant differences in the characteristics of the patients classified as reinfection or recrudescence, or between the patients from Thailand and Cambodia (see on-line supplemental material). Of note, all four patients from the Phase II trial in Bangkok never left the hospital before the recurrence of parasitemia. Bangkok is not a transmission zone for malaria and therefore these patients had no risk of reinfection. The patients from the in vivo clinical efficacy trial in Cambodia did not remain in the hospital during their follow-up and were therefore at risk for reinfection. However, the risk of local reacquisition was low.

Table 1
Genotyping Results by Nested PCR and Heteroduplex Tracking Assay

The msp-1 HTA was carried out as previously described in the literature [11]. For this study, a new msp-2 HTA was developed (assay development and validation are described in the Supplementary Material). Each sample was analyzed in duplicate to ensure the accuracy of the results and reliability of the assay. Bands were interpreted based on criteria established in the literature and considered significant only if they were present in both replicates, represented >3% of the total parasite population, and were visible to the eye [4, 13].


Standard PCR methods for the three markers classified six patients as reinfections and six patients as recrudescences based upon the MMV/WHO criteria. Of the six patients classified as reinfections, two lacked shared bands at msp-2, two lacked shared bands at glurp, and two lacked shared bands in both. However, HTAs detected shared bands in 5 of the 6 patients suggesting that only one patient was reinfected (Table 1). All patients classified as recrudescences by nested PCR were also classified as recrudescences by HTA.

This difference is due to the ability of HTAs to detect variants that are missed by standard PCR methods [4, 6, 11]. The msp-1 HTA detected an average of 3.1 bands, as compared to 1.8 bands by PCR (paired t-test, p=0.003). At msp-2, the HTA and PCR detected on average 4.5 bands compared to 1.8 bands, respectively (paired t-test, p=0.0001).

The shared variants which are missed by standard PCR are “minority variants” representing <20% of the parasite population. For example, in one patient sample pair (Figure 1, Lanes A-D), the shared variant represented only 14.5% of the initial parasite population and 10.6% of the recurrent parasitemia parasite population. This phenomenon was seen in all three sample pairs that were misclassified at msp-2 (Legend of Figure 1).

Figure 1
msp-2 HTA of Three Sample Pairs Misclassified by Standard PCR

Two of the patients classified as reinfections by PCR, Thailand patients 14 and 29, could not have been reinfected, since they never left the hospital in Bangkok. Sequence analysis of variants from Thailand patient 29 confirmed a shared variant (Genbank Accession Number: EU331092). The details of this analysis are provided in the Supplementary Material.


In this study, we show that standard PCR methods for genotyping recurrent parasitemias can misclassify recrudescent parasitemias as reinfections. We found that 5 of 6 (83%) of samples classified as reinfection by the WHO/MMV scheme in this study were more likely to be recrudesences. Even if the two patients who were classified as reinfection because of not sharing a band at glurp by standard PCR are removed, 3 of the other 4 (75%) samples would still be misclassified. This misclassification is an inherent problem of the standard PCR assays due to their inability to detect minority variants. This is important because misclassification in this manner overestimates drug efficacy. Thus, this type of misclassification might delay needed switches in first-line antimalarial policy, which the WHO recommends when 28 day efficacy falls below 90% [14].

This study involves a relatively small number of patients. However, the patients from the Phase II clinical trial are unusual in that all those presented here failed while remaining in the hospital in Bangkok without risk of reinfection. Thus, they provide a unique opportunity for determining misclassification of recrudescences as reinfections. The expense of keeping patients in hospital during clinical trials means this method cannot be employed by most trials to prevent reinfection and therefore the availability of this type of clinical sample is limited.

The use of multiple markers in a step wise approach may increase the probability of a recrudescence being misclassified as a reinfection due to the increased chance that at least one marker will fail to detect a shared minority variant [4, 15]. If assays that are sensitive to minority variants are used, such as HTAs, fewer loci may be needed to accurately classify samples. However, defining how many markers are appropriate would require further investigation.

The results of this study show that minority variants occur frequently, even in regions of relatively low transmission, and may cause a significant amount of misclassification of recurrent parasitemias with the WHO/MMV recommended methods. It suggests that using traditional PCR correction methods in regions of low transmission may be detrimental as clinical efficacy of a drug may be overestimated. Assays sensitive to minority variants, such as HTAs, capillary electrophoresis and other fragment analysis techniques, may therefore be better suited for correcting clinical trials in areas of low transmission. Currently, it is unclear if this same problem exists in regions with high transmission and further investigation of this is warranted.

In addition to the problem of misclassifying reinfections describe here, we also need to worry about misclassifying recrudescences [4]. In some areas, specific variants are present at a high prevalence. Because of this, patients who are reinfected by the same variant that was present before treatment could be misjudged as recrudescences. This can be adjusted for statistically if the prevalence of the various variants are measured [4].

The msp-2 HTA developed for this study provided an accurate and quantitative representation of the variants within the clinical sample. Nevertheless, it clear that this method still has limitations. Within some recurrent parasitemias, major variants can exist which were not detected in the initial parasitemia. For example, in Figure 1 Lanes C and D, the majority variant based on phosphorimager intensity (R1) was not detected in the initial parasitemia. These are likely variants that initially existed below the limit of detection of the assay or were sequestered at the time of initial sampling. Another possibility is that these variants existed in a pre-erythrocytic stage at the time of therapy. However, given that failures occurred at Days 19–35 in these six patients it is unlikely. Thus, new technologies and protocols for sampling should continue to be investigated. Technical limitations of HTAs also need to be considered. PCR with agarose gel electrophoresis is technically simple, does not require expensive equipment, and is easily done in the developing world. The major disadvantage of HTAs is that they require the use of radioactive probes, which limits their use in developing countries because there are few facilities with permits to handle radioactive waste disposal. However, probes can potentially be labeled with non-radioactive labels, such as fluorescent tags or biotinylation. More extensive discussions of the advantages and disadvantages of HTAs have been published elsewhere [4, 6].

This study is the first report showing strong evidence for the misclassification of recrudesences as reinfections in antimalarial trials using the currently recommended MMV/WHO protocols. Unfortunately, the rate of misclassification we found was high. Due to this problem, the interpretation of PCR genotyping in its current form should be done with caution. In the future, we need to continue to work on improving the techniques and protocols used for classifying recurrent parasitemias.

Supplementary Material

On-line Supp


Grateful thanks to Paul Trottmanand Jesse Kwiek for valuable assistance. Thanks to Emily Wenink for running the msp-1 HTAs on the samples from Thailand. In addition, thanks to Dr. Socheat Duong for his support.

This project was funded in part by Immtech Pharmaceuticals, Inc., the sponsor of the Phase II clinical trial. Carol Olson is an employee of Immtech Pharmaceuticals, Inc. No other authors on this paper have any conflict of interest to declare.

This project was also funded by Medicines for Malaria Venture (MMV), Infectious Disease Pathogenesis Research Training Grant (DHHS/NIH/NIAID 5 T32 AI07151-29), NIAID 1R21AI076785, the 2007 IDSA ERF/NFID Merle A. Sande/Pfizer Fellowship in International Infectious Diseases, and the World Health Organization (WHO). These agencies had no involvement in the design, collection, analysis, or interpretation of data in this study or in writing this paper or submitting it for publication. The opinions expressed are the private ones of the authors and do not purport to be the official ones of the Department of the Navy. The project described was also supported by Award Number KL2RR025746 from the National Center for Research Resources. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health.


Genotyping data for the samples from Thailand was previously presented at the annual meeting of the American Society of Tropical Medicine and Hygiene, November 4–8, 2007, Philadelphia, PA, Abstract Number 811.


1. World Health Organization. Susceptibility of Plasmodium falciparum to antimalarial drugs: report on global monitoring: 1996–2004. Geneva, Switzerland: World Health Organization; 2005.
2. Jafari S, Le Bras J, Bouchaud O, Durand R. Plasmodium falciparum clonal population dynamics during malaria treatment. J Infect Dis. 2004;189:195–203. [PubMed]
3. Martensson A, Ngasala B, Ursing J, et al. Influence of consecutive-day blood sampling on polymerase chain reaction-adjusted parasitological cure rates in an antimalarial-drug trial conducted in Tanzania. J Infect Dis. 2007;195:597–601. [PubMed]
4. Kwiek JJ, Alker AP, Wenink EC, Chaponda M, Kalilani LV, Meshnick SR. Estimating true antimalarial efficacy by heteroduplex tracking assay in patients with complex Plasmodium falciparum infections. Antimicrob Agents Chemother. 2007;51:521–7. [PMC free article] [PubMed]
5. World Health Organization. ethods and Techniques for Clinical Trials on Antimalarial Drug Efficacy: genotyping to identify parasite populations. Geneva, Switzerland: World Health Organization; 2008.
6. Juliano JJ, Kwiek JJ, Cappell K, Mwapasa V, Meshnick SR. Minority-variant pfcrt K76T mutations and chloroquine resistance, Malawi. Emerg Infect Dis. 2007;13:872–7. [PubMed]
7. Greenhouse B, Dokomajilar C, Hubbard A, Rosenthal PJ, Dorsey G. Impact of transmission intensity on the accuracy of genotyping to distinguish recrudescence from new infection in antimalarial clinical trials. Antimicrob Agents Chemother. 2007;51:3096–103. [PMC free article] [PubMed]
8. Kapoor A, Jones M, Shafer RW, Rhee SY, Kazanjian P, Delwart EL. Sequencing-based detection of low-frequency human immunodeficiency virus type 1 drug-resistant mutants by an RNA/DNA heteroduplex generator-tracking assay. J Virol. 2004;78:7112–23. [PMC free article] [PubMed]
9. Leitner T, Halapi E, Scarlatti G, et al. Analysis of heterogeneous viral populations by direct DNA sequencing. Biotechniques. 1993;15:120–7. [PubMed]
10. Liu S, Mu J, Jiang H, Su XZ. Effects of Plasmodium falciparum mixed infections on in vitro antimalarial drug tests and genotyping. Am J Trop Med Hyg. 2008;79:178–84. [PMC free article] [PubMed]
11. Ngrenngarmlert W, Kwiek JJ, Kamwendo DD, et al. Measuring allelic heterogeneity in Plasmodium falciparum by a heteroduplex tracking assay. Am J Trop Med Hyg. 2005;72:694–701. [PubMed]
12. Ariey F, Chalvet W, Hommel D, et al. Plasmodium falciparum parasites in French Guiana: limited genetic diversity and high selfing rate. Am J Trop Med Hyg. 1999;61:978–85. [PubMed]
13. Resch W, Parkin N, Stuelke EL, Watkins T, Swanstrom R. A multiple-site-specific heteroduplex tracking assay as a tool for the study of viral population dynamics. Proc Natl Acad Sci U S A. 2001;98:176–81. [PubMed]
14. World Health Organization. Guidelines for the treatment of malaria. Geneva, Switzerland: World Health Organization; 2005.
15. Brockman A, Paul RE, Anderson TJ, et al. Application of genetic markers to the identification of recrudescent Plasmodium falciparum infections on the northwestern border of Thailand. Am J Trop Med Hyg. 1999;60:14–21. [PubMed]