Infection dynamics of Plasmodium falciparum in Papua New Guinea

Malaria is one of the leading causes of illness and death in Papua New Guinea (PNG), mainly affecting children under 5 years of age. The current first line treatment for uncomplicated malaria is a combination therapy of sulfadoxine-pyrimethamine and chloroquine (CQ) or amodiaquine, however, frequent treatment failures have been reported shortly after its implementation. Drug resistance has mainly been associated with single nucleotide polymorphisms in five different genes. Different studies have indicated that mutations associated with drug resistance incur fitness costs to the parasite in absence of drug pressure. Among these are reports from different countries where a decreasing prevalence of mutations associated with CQ resistance was observed after CQ has been suspended as first line treatment. As it is experimentally difficult to estimate the fitness of Plasmodium parasites, a surrogate marker is needed to quantify fitness costs associated with drug resistance mutations. We hypothesised that the parasites’ survival within the human host can be used as surrogate marker for parasite fitness. In a pilot study we determined the drug resistance-associated haplotypes of parasite clones at 25 loci on 4 marker genes, and compared their prevalence between newly acquired and chronic infections. A reduced frequency of a 7-fold mutated haplotype and increased frequency of a 5-fold mutated haplotype in long term persistent infections indicated an impaired survival of highly mutated parasites and suggested that the duration of infections is a promising marker for parasite fitness that deserved further investigation.
A further approach to define duration of infections more precisely was undertaken with samples from a longitudinal field survey in PNG. A cohort of 269 1-4.5 years old children was followed over a period of 16 months. This provided consecutive blood samples collected in 2-monthly follow-up visits plus a blood sample from each morbid episode. All samples were genotyped for the polymorphic marker gene merozoite surface protein 2 (msp2) in order to distinguish individual parasite clones within a host. The persistence of genotypes in consecutive blood samples of each child was determined. Unexpected high numbers of antimalarial treatments given in the course of this study led to a high turn-over rate of parasite clones and prevented the establishment of asymptomatic long-term infections. The shortage of long untreated periods in our study participants hampered the determination of a novel molecular parameter termed “duration of infection” for each individual parasite clone. This parameter could have been useful for modelling the transmission success and fitness of drug resistant versus drug sensitive parasite clones. We concluded that this approach requires a cohort of semi-immune individuals where treatment is given rarely, e.g. older children or adults, where P. falciparum infections less frequently cause morbid episodes.
The high incidence of morbidity in our study cohort allowed us to investigate molecular parameters that have an impact on the development of a subsequent clinical episode. We found that in children >3 years, a higher multiplicity of infection (MOI) at baseline reduced the risk of a P. falciparum episode. These results are in agreement with previous reports and support the concept of premunition. Furthermore, we studied the impact of co-infecting Plasmodium species on MOI and found an increased MOI in the presence of a heterologous species. This can be explained by the observation of reduced parasite densities in mixed-species infections, leading to a decreased need for treatment and thus facilitating the accumulation of multiple clones. Our results provide further evidence for interactions among co-infecting P. falciparum clones as well as among co-infecting Plasmodium species.
During the 16 months field survey all morbid episodes were treated with Coartem®. Analysis of msp2 genotyping results revealed an unexpected high number of Coartem® treatment failures. After confirmation of recrudescent parasites with 2 additional marker genes and exclusion of host genetic factors to be responsible for treatment failures, the most likely explanation for the frequently observed Coartem® failures was a combination of poor adherence to the treatment regimen and a lack of fat supplementation which is required for absorption of the drug. Our results disagree with findings from a clinical trial reporting Coartem® to be highly effective in PNG. In contrast, our observations from a non-trial setting highlight potential problems of Coartem® usage in routine clinical practice.
In the course of this thesis, genotyping techniques for merozoite surface proteins (msp) 1 and 2 were optimized and applied. Using these high resolution typing techniques based on capillary electrophoresis, we investigated the effect of transmission intensity on diversity and complexity of msp1 and msp2 in samples from PNG and Tanzania. We observed a greater MOI and a greater number of distinct alleles in samples from Tanzania. Genetic diversity was greater for msp2 than for msp1. In both these areas of different malaria endemicity msp2 was found to be superior for distinguishing individual parasite clones. The probability of two infections carrying by chance the same msp2 allele was lower than this probability calculated for msp1. Based on the frequency distribution of msp2 alleles and on the distribution of observed numbers of infections, we estimated the true MOI adjusted for the probability of multiple infections sharing the same allele. For our high resolution typing technique this adjustment made little difference to the estimated mean MOI compared to the observed mean MOI.
A central aim of this thesis was to measure molecular parameters of infection dynamics. These can be determined from genotyping longitudinal sets of samples. Some of these parameters were successfully determined: MOI, force of infection (FOI), and detectability. Other parameters such as natural elimination rates and persistence of infections could not be determined due to frequent treatments. We estimated detectability of parasite clones based on samples collected 24 hours apart and investigated its impact on the MOI and FOI. Imperfect detection of parasites occurs as a consequence of sequestration or when parasite densities fluctuate around the detection limit. We found that in our study participants detectability was high. This was likely an effect of high parasite densities in children of this age. The benefit of short-term sampling on measures of MOI and FOI was marginal. We concluded that in future studies carried out in this age group, taking repeated samples 24 hours apart has limited benefit and does not justify the additional costs, work load and discomfort for the study participants.
This project contributed to our understanding of the infections dynamics of P. falciparum and the interactions between parasites clones and Plasmodium species. We provided further insights into determinants of malaria episodes and highlighted the potential usefulness of the parameter “duration of infection” as surrogate marker to estimate fitness costs of drug resistance. This thesis provided findings that are relevant for malaria control strategies and treatment guidelines.