Can estimates of antimalarial efficacy from field studies be improved?

Dealing with indeterminate outcomes in antimalarial drug efficacy trials: a comparison between complete case analysis, multiple imputation and inverse probability weighting

BACKGROUND

Antimalarial clinical efficacy studies for uncomplicated Plasmodium falciparum malaria frequently encounter situations in which molecular genotyping is unable to discriminate between parasitic recurrence, either new infection or recrudescence. The current WHO guideline recommends excluding these individuals with indeterminate outcomes in a complete case (CC) analysis. Data from the four artemisinin-based combination (4ABC) trial was used to compare the performance of multiple imputation (MI) and inverse probability weighting (IPW) against the standard CC analysis for dealing with indeterminate recurrences.

METHODS

3369 study participants from the multicentre study (4ABC trial) with molecularly defined parasitic recurrence treated with three artemisinin-based combination therapies were used to represent a complete dataset. A set proportion of recurrent infections (10, 30 and 45%) were reclassified as missing using two mechanisms: a completely random selection (mechanism 1); missingness weakly dependent (mechanism 2a) and strongly dependent (mechanism 2b) on treatment and transmission intensity. The performance of MI, IPW and CC approaches in estimating the Kaplan-Meier (K-M) probability of parasitic recrudescence at day 28 was then compared. In addition, the maximum likelihood estimate of the cured proportion was presented for further comparison (analytical solution). Performance measures (bias, relative bias, standard error and coverage) were reported as an average from 1000 simulation runs.

RESULTS

The CC analyses resulted in absolute underestimation of K-M probability of day 28 recrudescence by up to 1.7% and were associated with reduced precision and poor coverage across all the scenarios studied. Both MI and IPW method performed better (greater consistency and greater efficiency) compared to CC analysis. In the absence of censoring, the analytical solution provided the most consistent and accurate estimate of cured proportion compared to the CC analyses.

CONCLUSIONS

The widely used CC approach underestimates antimalarial failure; IPW and MI procedures provided efficient and consistent estimates and should be considered when reporting the results of antimalarial clinical trials, especially in areas of high transmission, where the proportion of indeterminate outcomes could be large. The analytical solution estimating the cured proportion could provide an alternative approach, in scenarios with minimal censoring due to loss to follow-up or new infections.

Statistical methods to derive efficacy estimates of anti-malarials for uncomplicated Plasmodium falciparum malaria: pitfalls and challenges

The Kaplan-Meier (K-M) method is currently the preferred approach to derive an efficacy estimate from anti-malarial trial data. In this approach event times are assumed to be continuous and estimates are generated on the assumption that there is only one cause of failure. In reality, failures are captured at pre-scheduled time points and patients can fail treatment due to a variety of causes other than the primary endpoint, commonly termed competing risk events. Ignoring these underlying assumptions can potentially distort the derived efficacy estimates and result in misleading conclusions. This review details the evolution of statistical methods used to derive anti-malarial efficacy for uncomplicated Plasmodium falciparum malaria and assesses the limitations of the current practices. Alternative approaches are explored and their implementation is discussed using example data from a large multi-site study.

Where chloroquine still works: the genetic make-up and susceptibility of Plasmodium vivax to chloroquine plus primaquine in Bhutan

BACKGROUND

Bhutan has made substantial progress in reducing malaria incidence. The national guidelines recommend chloroquine (CQ) and primaquine (PQ) for radical cure of uncomplicated Plasmodium vivax, but the local efficacy has not been assessed. The impact of cases imported from India on the genetic make-up of the local vivax populations is currently unknown.

METHODS

Patients over 4 years of age with uncomplicated P. vivax mono-infection were enrolled into a clinical efficacy study and molecular survey. Study participants received a standard dose of CQ (25 mg/kg over 3 days) followed by weekly review until day 28. On day 28 a 14-day regimen of PQ (0.25 mg/kg/day) was commenced under direct observation. After day 42, patients were followed up monthly for a year. The primary and secondary endpoints were risk of treatment failure at day 28 and at 1 year. Parasite genotyping was undertaken at nine tandem repeat markers, and standard population genetic metrics were applied to examine population diversity and structure in infections thought to be acquired inside or outside of Bhutan.

RESULTS

A total of 24 patients were enrolled in the clinical study between April 2013 and October 2015. Eight patients (33.3 %) were lost to follow-up in the first 6 months and another eight patients lost between 6 and 12 months. No (0/24) treatment failures occurred by day 28 and no (0/8) parasitaemia was detected following PQ treatment. Some 95.8 % (23/24) of patients were aparasitaemic by day 2. There were no haemolytic or serious events. Genotyping was undertaken on parasites from 12 autochthonous cases and 16 suspected imported cases. Diversity was high (H E 0.87 and 0.90) in both populations. There was no notable differentiation between the autochthonous and imported populations.

CONCLUSIONS

CQ and PQ remains effective for radical cure of P. vivax in Bhutan. The genetic analyses indicate that imported infections are sustaining the local vivax population, with concomitant risk of introducing drug-resistant strains.

Genetic diversity of Plasmodium vivax in clinical isolates from Bangladesh

Background

Plasmodium vivax is the second most prevalent human malaria parasite in Bangladesh; however, there are no data of its genetic diversity. Several molecular markers are available where Pvcsp, Pvmsp 1 and Pvmsp 3α are most commonly used for P. vivax genotyping studies. The aim of the study was to investigate the population structure of P. vivax in Bangladesh.

Methods

A total of 102 P. vivax-positive blood samples were collected from different malaria-endemic areas in Bangladesh and subsequently analysed for those three genotyping markers. Nested PCR was performed for diagnosis and genotyping analysis followed by PCR–RFLP to detect genetic diversity using Pvcsp, Pvmsp 1 and Pvmsp 3α markers.

Results

Analysis of Pvcsp showed that the VK210 repeat type was highly prevalent (64.7%, 66/102) compared to VK247 (35.3%, 36/102), although the prevalence of VK247 was higher than other Southeast Asian countries. Analysis of these three genes revealed a diverse, circulating population of P. vivax where a total of ten, 56 and 35 distinct genotypes were detected for Pvcsp, Pvmsp 1 and Pvmsp 3α, respectively.

Conclusion

This genotyping observation of P. vivax is the first report from Bangladesh and will provide valuable information for establishing the genotyping methods and circulating genetic variants of these three markers available in Bangladesh.

A better resolution for integrating methods for monitoring Plasmodium falciparum resistance to antimalarial drugs

Effective chemotherapy is the mainstay of malaria control. However, resistance of falciparum malaria to antimalarial drugs compromised the efforts to eliminate the disease and led to the resurgence of malaria epidemics. Three main approaches are used to monitor antimalarial drug efficacy and drug resistance; namely, in vivo trials, in vitro/ex vivo assays and molecular markers of drug resistance. Each approach has its implications of use as well as its advantages and drawbacks. Therefore, there is a need to use an integrated approach that would give the utmost effect to detect resistance as early as its emergence and to track it once spread. Such integration becomes increasingly needed in the era of artemisinin-based combination therapy as a forward action to deter resistance. The existence of regional and global networks for the standardization of methodology, provision of high quality reagents for the assessment of antimalarial drug resistance and dissemination of open-access data would help in approaching an integrated resistance surveillance system on a global scale.

Genetic diversity of Plasmodium falciparum among school-aged children from the Man region, western Côte d'Ivoire

Background

The genetic diversity of Plasmodium falciparum allows the molecular discrimination of otherwise microscopically identical parasites and the identification of individual clones in multiple infections. The study reported here investigated the P. falciparum multiplicity of infection (MOI) and genetic diversity among school-aged children in the Man region, western Côte d’Ivoire.

Methods

Blood samples from 292 children aged seven to 15 years were collected in four nearby villages located at altitudes ranging from 340 to 883 m above sea level. Giemsa-stained thick and thin blood films were prepared and examined under a microscope for P. falciparum prevalence and parasitaemia. MOI and genetic diversity of the parasite populations were investigated using msp2 typing by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP).

Results

Plasmodium falciparum prevalence and parasitaemia were both found to be significantly lower in the highest altitude village. Genotyping of the isolates revealed 25 potentially new msp2 alleles. MOI varied significantly across villages but did not correlate with altitude nor children’s age, and only to a limited extent with parasitaemia. An analysis of molecular variance (AMOVA) indicated that a small, but close to statistical significance (p = 0.07), fraction of variance occurs specifically between villages of low and high altitudes.

Conclusions

Higher altitude was associated with lower prevalence of P. falciparum but not with reduced MOI, suggesting that, in this setting, MOI is not a good proxy for transmission. The evidence for partially parted parasite populations suggests the existence of local geographical barriers that should be taken into account when deploying anti-malarial interventions.

Plasmodium falciparum populations from northeastern Myanmar display high levels of genetic diversity at multiple antigenic loci

Levels of genetic diversity of the malaria parasites and multiclonal infections are correlated with transmission intensity. In order to monitor the effect of strengthened malaria control efforts in recent years at the China-Myanmar border area, we followed the temporal dynamics of genetic diversity of three polymorphic antigenic markers msp1, msp2, and glurp in the Plasmodium falciparum populations. Despite reduced malaria prevalence in the region, parasite populations exhibited high levels of genetic diversity. Genotyping 258 clinical samples collected in four years detected a total of 22 PCR size alleles. Multiclonal infections were detected in 45.7% of the patient samples, giving a minimum multiplicity of infection of 1.41. The majority of alleles experienced significant temporal fluctuations through the years. Haplotype diversity based on the three-locus genotypes ranged from the lowest in 2009 at 0.33 to the highest in 2010 at 0.80. Sequencing of msp1 fragments from 36 random samples of five allele size groups detected 13 different sequences, revealing an additional layer of genetic complexity. This study suggests that despite reduced prevalence of malaria infections in this region, the parasite population size and transmission intensity remained high enough to allow effective genetic recombination of the parasites and continued maintenance of genetic diversity.

Estimating the numbers of malaria infections in blood samples using high-resolution genotyping data

People living in endemic areas often habour several malaria infections at once. High-resolution genotyping can distinguish between infections by detecting the presence of different alleles at a polymorphic locus. However the number of infections may not be accurately counted since parasites from multiple infections may carry the same allele. We use simulation to determine the circumstances under which the number of observed genotypes are likely to be substantially less than the number of infections present and investigate the performance of two methods for estimating the numbers of infections from high-resolution genotyping data. The simulations suggest that the problem is not substantial in most datasets: the disparity between the mean numbers of infections and of observed genotypes was small when there was 20 or more alleles, 20 or more blood samples, a mean number of infections of 6 or less and where the frequency of the most common allele was no greater than 20%. The issue of multiple infections carrying the same allele is unlikely to be a major component of the errors in PCR-based genotyping. Simulations also showed that, with heterogeneity in allele frequencies, the observed frequencies are not a good approximation of the true allele frequencies. The first method that we proposed to estimate the numbers of infections assumes that they are a good approximation and hence did poorly in the presence of heterogeneity. In contrast, the second method by Li et al estimates both the numbers of infections and the true allele frequencies simultaneously and produced accurate estimates of the mean number of infections.

Understanding the population genetics of Plasmodium vivax is essential for malaria control and elimination

Traditionally, infection with Plasmodium vivax was thought to be benign and self-limiting, however, recent evidence has demonstrated that infection with P. vivax can also result in severe illness and death. Research into P. vivax has been relatively neglected and much remains unknown regarding the biology, pathogenesis and epidemiology of this parasite. One of the fundamental factors governing transmission and immunity is parasite diversity. An understanding of parasite population genetic structure is necessary to understand the epidemiology, diversity, distribution and dynamics of natural P. vivax populations. In addition, studying the population structure of genes under immune selection also enables investigation of the dynamic interplay between transmission and immunity, which is crucial for vaccine development. A lack of knowledge regarding the transmission and spread of P. vivax has been particularly highlighted in areas where malaria control and elimination programmes have made progress in reducing the burden of Plasmodium falciparum, yet P. vivax remains as a substantial obstacle. With malaria elimination back on the global agenda, mapping of global and local P. vivax population structure is essential prior to establishing goals for elimination and the roll-out of interventions. A detailed knowledge of the spatial distribution, transmission and clinical burden of P. vivax is required to act as a benchmark against which control targets can be set and measured. This paper presents an overview of what is known and what is yet to be fully understood regarding P. vivax population genetics, as well as the importance and application of P. vivax population genetics studies.

The distribution of Plasmodium falciparum infection durations

OBJECTIVES

The duration of untreated Plasmodium falciparum infections in naturally exposed human populations is of interest for rational planning of malaria control interventions as it is related to the duration of infectivity. The extent of variability in duration is relevant where transmission is seasonal, and for the planning of elimination efforts. Methods for measuring these quantities from genotyping data have been restricted to exponential models of infection survival, as implied by constant clearance rates. Such models have greatly improved the understanding of infection dynamics on a population level but likely misrepresent the within-host dynamics of many pathogens. Conversely, the statistical properties of the distribution of infection durations, and how these are affected by exposure, should contain information on within-host dynamics.

METHODS AND RESULTS

We extended existing methods for the analysis of longitudinal genotyping data on P. falciparum infections. Our method simultaneously estimates force of infection, detectability, and the distribution of infection durations. Infection durations are modeled using parametric survival distributions. The method is validated using simulated data, and applied to data from a cohort study in Navrongo, Northern Ghana. Distribution estimates from exponential, Weibull, lognormal, and gamma models are compared with the distribution of durations in malariatherapy data.

CONCLUSIONS

The Weibull model fitted the data best. It estimated a shorter mean duration than the exponential model, which gave the worst fit. The distribution estimates appeared positively skewed when compared with the distribution of durations in malariatherapy data, suggesting that a significant proportion of infections is cleared shortly after inoculation. We conclude that malariatherapy data, the most important source of information on P. falciparum within-host dynamics, may not be representative of the actual processes in natural populations, and should be used with care. Further, conclusions from transmission models assuming exponential infection survival may be biased.

Uncertain outcomes: adjusting for misclassification in antimalarial efficacy studies

Evaluation of antimalarial efficacy is difficult because recurrent parasitaemia can be due to recrudescence or re-infection. PCR is used to differentiate between recrudescences and re-infections by comparing parasite allelic variants before and after treatment. However, PCR-corrected results are susceptible to misclassification: false positives, due to re-infection by the same variant present in the patient before treatment; and false negatives, due to variants that are present but too infrequent to be detected in the pre-treatment PCR, but are then detectable post-treatment. This paper aimed to explore factors affecting the probability of false positives and proposes a Monte Carlo uncertainty analysis to account for both types of misclassification. Higher levels of transmission intensity, increased multiplicity of infection, and limited allelic variation resulted in more false recrudescences. The uncertainty analysis exploits characteristics of study data to minimize bias in the estimate of efficacy and can be applied to areas of different transmission intensity.

The perils of PCR: can we accurately 'correct' antimalarial trials?

During follow-up in antimalarial drug trials, treated subjects can be newly infected. PCR correction is used to distinguish this re-infection from drug failure (recrudescence) and to adjust final drug efficacy estimates. The epidemiological, biological and technical limitations of PCR correction and how this can lead to misclassification in drug trial outcomes are underappreciated. This article considers these limitations and proposes a framework for reporting, interpreting and improving PCR correction of antimalarial trials.

Comparison of Plasmodium falciparum allelic frequency distribution in different endemic settings by high-resolution genotyping

Background

The diversity of genotyping markers of Plasmodium falciparum depends on transmission intensity. It has been reported that the diversity of the merozoite surface protein 2 (msp2) is greater in areas of high compared to low endemicity, however, results for msp1 were inconsistent. These previous reports relied on low resolution genotyping techniques.

Methods

In the present study, a high-resolution capillary electrophoresis-based technique was applied to genotype samples from areas of different endemicity in Papua New Guinea and Tanzania. For both endemic settings, the diversity of msp1 and msp2 was investigated; the mean multiplicity of infection (MOI) and the F_ST values were determined to investigate whether more accurate sizing generates different results.

Results and Conclusion

The results of the present study confirmed previous reports of a higher mean MOI for both marker genes and increased genetic diversity in areas of higher endemicity as estimated by the total number of distinct alleles for msp2. For msp1 a minor increase in diversity was observed. Measures of between population variance in allele frequencies (F_ST) indicated little genetic differentiation for both marker genes between the two populations from different endemic settings. MOI adjusted for the probability of multiple infections sharing the same allele was estimated by using the msp2 allele frequency distribution and the distribution of observed numbers of concurrent infections. For the high-resolution typing technique applied in this study, this adjustment made little difference to the estimated mean MOI compared to the observed mean MOI.

Revisiting the design of phase III clinical trials of antimalarial drugs for uncomplicated Plasmodium falciparum malaria

Steffen Borrmann and colleagues discuss appropriate endpoints and their measurement during phase III trials of new antimalarial drugs.