A novel, sensitive assay for high-throughput molecular detection of plasmodia for active screening of malaria for elimination

Abstract

Although malaria remains one of the leading infectious diseases in the world, the decline in malaria transmission in some area makes it possible to consider elimination of the disease. As countries approach elimination, malaria diagnosis needs to change from diagnosing ill patients to actively detecting infections in all carriers, including asymptomatic and low-parasite-load patients. However, few of the current diagnostic methods have both the throughput and the sensitivity required. We adopted a sandwich RNA hybridization assay to detect genus Plasmodium 18S rRNA directly from whole-blood samples from Plasmodium falciparum and Plasmodium vivax patients without RNA isolation. We tested the assay with 202 febrile patients from areas where malaria is endemic, using 20 μl of each blood sample in a 96-well plate format with a 2-day enzyme-linked immunosorbent assay (ELISA)-like work flow. The results were compared with diagnoses obtained using microscopy, a rapid diagnostic test (RDT), and genus-specific real-time PCR. Our assay identified all 66 positive samples diagnosed by microscopy, including 49 poorly stored samples that underwent multiple freeze-thaw cycles due to resource limitation. The assay uncovered three false-negative samples by microscopy and four false-negative samples by RDT and agreed completely with real-time PCR diagnosis. There was no negative sample by our assay that would show a positive result when tested with other methods. The detection limit of our assay for P. falciparum was 0.04 parasite/μl. The assay's simple work flow, high throughput, and sensitivity make it suitable for active malaria screening.

Fig 1

Correlation of assay signal with cultured Plasmodium falciparum parasitemia in human erythrocytes. Parasitemia in human erythrocyte culture was determined based on infection rate (number of infected red blood cells [RBCs]/number of infected plus uninfected RBCs [n = 1,000]) and RBC concentration The limit of detection was determined as the minimal amount of cultured plasmodia added to the erythrocytes that gave a net signal above three times the SD of the background erythrocyte control (dashed line). Each dilution was prepared using culture medium. Triplicate samples were used in the assay, and the data are representative of three independent runs. The average limit of detection was 0.04 parasite/μl blood. RLU, relative light units.

Fig 2

Correlation between assay signal and parasitemia for poorly stored blood samples. Stored blood samples from 49 patients, which underwent multiple freeze-and-thaw cycles after collection due to limited resources, were assayed (20 μl each), and the net signals were plotted against parasitemia determined microscopically at admission. Each sample was assayed in duplicate, and the average signal was used. Signal intensities beyond 1 × 10⁷ RLU were considered approaching detection saturation of the photon detector.