Analysing treatment failures by high-resolution capillary electrophoresis genotyping in uncomplicated P. falciparum malaria adapted from patient samples from Bobo-Dioulasso, Burkina Faso

Abstract

In the context of my dissertation I worked for nine months in Bobo-Dioulasso, Burkina Faso, as part of the WANECAM-trial, which compared the antimalarial agents dihydroartemisinin-piperaquine and pyronaridine-artesunate with the comparator drug artemether-lumefantrine. The aim of this pilot study was to implement P. falciparum merozoite surface protein 2 genotyping using high-resolution capillary electrophoresis genotyping with an automated sequencer at the AG Borrmann laboratory at the Institute of Tropical Medicine of Eberhard Karls Universität Tübingen. I analysed if recurrences were caused by new infections or recrudescences. Furthermore, I determined allele frequencies and recrudescence rates of the studied population. In paired filter paper blood samples from patients from Bobo-Dioulasso, I extracted Plasmodium falciparum DNA, amplified it with a primary polymerase chain reaction and subsequently with a secondary polymerase chain reaction. I analysed the amplification products first by gel electrophoresis and second, using capillary electrophoresis with an automated sequencer. The software GeneScan® evaluated the results and presented them as electropherograms, which I interpreted and analysed. Of the 63 paired samples, I was able to extract and analyse 37 successfully. I noticed that 3D7-type alleles appeared in higher polymorphisms than D10-type alleles, but that D10-type alleles were responsible for high allele frequencies. I calculated a recrudescence rate of 21.62 % in total. Genotypes from the D10 family, which were highly represented in the baseline population, were also highly represented in recrudescences. For the 3D7 family, a similar proportionality could not be found. The recrudescence rates for artemether-lumefantrine and pyronaridine-artesunate were high with 25 % each. Further analyses with a bigger sample size as well as equal distribution between treatment arms should be awaitened before drawing further conclusions. The discussion of this study is subdivided into three parts: technical limitations of the methodology, biological aspects of P. falciparum and the comparison of my results with published and unpublished data. The technical limitations concerned mostly the Summary 67 detection limit of P. falciparum when examined by CE genotyping. The biological aspects of the parasite’s life cycle, which cause difficulties with CE genotyping are firstly, that infections might be misclassified as recrudescences due to occurrence of gametocytes on the day of recurrence and secondly, that P. falciparum sequesters into deep tissues and is therefore not always detectable in the blood stream. This can result in certain genotypes not being detected on day 0 even though they have already spread in the human body. A misclassification as a recrudescence may also occur when, by coincidence, patients are re-infected by parasites with the same genotype. Furthermore, I compared my data with published and unpublished data and understood that the results were often consistent with each other: My results as well as the compared results show that firstly, 3D7-type alleles have a higher polymorphism than D10-type alleles, and that secondly each population has certain genotypes that are overrepresented and which tend to belong to D10 genotypes. The results of this study have shown that CE genotyping was well established for samples from Burkina Faso. The high recrudescence rates in this study are not statistically valid due to small sample size and due to unequal number per treatment arm. Further analyses of the WANECAM-trial should be awaited to draw conclusions concerning the efficacy of the examined drugs. The potential dominance of possible of certain genotypes in parasite populations should be monitored in malaria trials to prevent misinterpretation of genotyping results.