Plasmodium vivax molecular diagnostics in community surveys: pitfalls and solutions

Abstract

A distinctive feature of Plasmodium vivax infections is the overall low parasite density in peripheral blood. Thus, identifying asymptomatic infected individuals in endemic communities requires diagnostic tests with high sensitivity. The detection limits of molecular diagnostic tests are primarily defined by the volume of blood analysed and by the copy number of the amplified molecular marker serving as the template for amplification. By using mitochondrial DNA as the multi-copy template, the detection limit can be improved more than tenfold, compared to standard 18S rRNA targets, thereby allowing detection of lower parasite densities. In a very low transmission area in Brazil, application of a mitochondrial DNA-based assay increased prevalence from 4.9 to 6.5%. The usefulness of molecular tests in malaria epidemiological studies is widely recognized, especially when precise prevalence rates are desired. Of concern, however, is the challenge of demonstrating test accuracy and quality control for samples with very low parasite densities. In this case, chance effects in template distribution around the detection limit constrain reproducibility. Rigorous assessment of false positive and false negative test results is, therefore, required to prevent over- or under-estimation of parasite prevalence in epidemiological studies or when monitoring interventions.

Keywords: 18S rRNA transcripts; Gametocytes; LAMP; Mitochondrial DNA; Molecular diagnostics; Plasmodium vivax; Quantification; Surveillance.

Fig. 1

Parasite densities of P. falciparum and P. vivax measured by qPCR (a) and light microscopy (b) in community samples from PNG (5–9 years old children from 6 neighbouring villages) (Data taken from [41])

Fig. 2

Detection of 18S rRNA genomic copies compared to 18S rRNA transcripts. Data used for this blot derived from earlier published work [10]. Dashed line: choice of cut-off (> 10 transcripts per reaction)

Fig. 3

Comparison of Pv-mtCOX1 and Pv18S rRNA assays performed in parallel in 604 community samples from Brazil. a Overlap in positivity by Pv-mtCOX1 and Pv18S rRNA qPCR. b Correlation of log10 template copy numbers detected by Pv-mtCOX1 and Pv18S rRNA qPCR

Fig. 4

Copy numbers of each marker gene detected per sample. Each dot represents one sample, red indicates all samples positive for Pv18S rRNA qPCR, orange indicates samples detected only by Pv-mtCOX1 qPCR. Dashed line: Molecular assays have a theoretical LOD, i.e., at least 1 template copy must be present per PCR reaction

Fig. 5

Correlation between P. vivax parasite density measured by 18S rRNA qPCR and P. vivax gametocyte density determined as pvs25 transcript numbers by qRT-PCR (graph based on data originally published in [2])