A saliva-based rapid test to quantify the infectious subclinical malaria parasite reservoir

Abstract

A large proportion of ongoing malaria parasite transmission is attributed to low-density subclinical infections not readily detected by available rapid diagnostic tests (RDTs) or microscopy. Plasmodium falciparum gametocyte carriage is subclinical, but gametocytemic individuals comprise the parasite reservoir that leads to infection of mosquitoes and local transmission. Effective detection and quantification of these carriers can help advance malaria elimination strategies. However, no point-of-need (PON) RDTs for gametocyte detection exist, much less one that can perform noninvasive sampling of saliva outside a clinical setting. Here, we report on the discovery of 35 parasite markers from which we selected a single candidate for use in a PON RDT. We performed a cross-sectional, multi-omics study of saliva from 364 children with subclinical infection in Cameroon and Zambia and produced a prototype saliva-based PON lateral flow immunoassay test for P. falciparum gametocyte carriers. The test is capable of identifying submicroscopic carriage in both clinical and nonclinical settings and is compatible with archived saliva samples.

Fig. 1

Study design of the cross-sectional studies of subclinical malaria parasite carriage among children in Cameroon and Zambia. Primary schools in the catchment area in Mfou, Cameroon, where screening and sampling performed are listed. School children between the ages of 5 to 16 were enrolled in the study. For Zambia, children <16 years of age who were present in households were sampled. The sampling strategy in Zambia dovetailed on an ongoing Southern Africa International Center of Excellence for Malaria Research (ICEMR) program project in Nchelenge, Zambia, which was conducted in partnership with the Tropical Diseases Research Centre (TDRC) of Zambia. Informed consent for all enrolled children was provided by a parent/guardian. If a child was found to be positive for malaria parasites by blood smear, they were referred for treatment according to the National Malaria Control Program guidelines. *Children who were not approved by their parent or guardian to participate were not included during the screen, and these children were not included in the total number of samples that were used for downstream molecular analyses.

Fig. 2

Saliva and matched blood sample collection, stabilization, and analyses. (A) Photograph and diagram of saliva collection (drool method) and stabilization for transport for two different collection periods from schools (Cameroon) and households (Zambia). The total number of saliva samples is 364. Saliva samples were stabilized for transport and analysis by LC-MS/MS for marker discovery. The prevalence of the marker PSSP17 in the saliva of children was then measured by LC-MRM MS. (B) Table summary of the LC-MRM results and estimated prevalence of marker carriage across all samples collected from both Cameroon and Zambia. (C) Table summary data of samples comparing the LC-MRM results with the reference molecular method of qPCR and PCR for pfs25, 18S rRNA gene in Cameroon, and cytB in Zambia, respectively. (D) Comparison of estimated female gametocytes/µl of blood (as determined by pfs25 mRNA qPCR, left y axis, blue bars) and PSSP17 (PF3D7_1218800) protein expression (as determined by PAR, right y axis, gray bars) for a subset of 93 samples with paired qPCR (pfs25) and PAR data available. Sample case codes are indicated on the x axis. Red arrows refer to examples of discordant results, as described in the main text.

Fig. 3

Prototype LFIA RDT for PSSP17 (PF3D7_1218800). (A) Schematic of the gRAD lateral flow platform and the capture and detection of PSSP17 by EuChelate microparticle-conjugated mAbs. (B) Images of LFIA gRAD strips to estimate the LOD of recombinant PSSP17 in spiked-in assays using naïve human saliva as matrix. (C) Initial analyses of frozen, stabilized samples from Cameroon (n = 12). Samples were randomly selected and, upon unblinding, were found to be either positive or negative by MS (LC-MRM), microscopy, or qPCR. A048 and A048b are two independent saliva samples collected from the same child to show consistency in detection by the LFIA. Positive (+) control: gametocyte lysate spiked into naïve saliva; negative (−) control: irrelevant asexual parasite lysate spiked into naïve saliva. Images on the left side of the figure are gRAD platform strips that were captured by mobile phone camera. (D) A second subset of independent Cameroon samples (n = 10), positive by LC-MRM, was selected and compared across the same categories as in (C). Microscopy was subdivided into gametocyte versus asexual trophozoite positivity/negativity, and qPCR/PCR analyses were subdivided into gametocyte-specific transcript (pfs25) or a transcript present in both trophozoites and gametocytes (18S rRNA). For (B) to (D), 10 µl of saliva containing either recombinant PSSP17 or endogenous marker was used with each test. C, control line; T, test line. Vertical dotted line: Demarcation of the end of the sample loading pad of the gRAD strip.

Fig. 4

Evaluation of the LOD of the prototype PSSP17 LFIA and potential use in clinical settings. (A) Replicate tests using filtered (or unfiltered), axenic supernatant from stage V gametocyte cultures, asexual blood-stage cultures, and complete media (negative control). (B) Estimated LOD of the PSSP17 LFIA based on gametocyte quantification using LM or qPCR for pfs25 transcripts. (C) Schematic of the Sierra Leone study to test the utility of the LFIA in confirmatory diagnosis of malaria parasite infection. Symptomatic individuals (>1 year of age) presenting in the Mercy Hospital in Bo, Sierra Leone, who provided informed consent were enrolled in the study. Matched blood samples were stored either in RNAlater or as DBS on Whatman FTA cards. (D) LFIA test strips showing PSSP17 detection in the saliva of a subset of symptomatic individuals presenting at a clinic in Bo, Sierra Leone. Negative control, uninfected/naive human saliva; positive control, naïve human saliva with lysed P. falciparum gametocytes. (E) Table comparing orthogonal detection approaches, multiplex PCR, LM, and the PSSP17 LFIA for all 34 samples from Sierra Leone. Multiplex PCR score: 0 = negative, 0.5 = weak positive, and 1 = positive. (F) Table estimating the sensitivity of the LFIA in a clinical setting compared to microscopy or multiplex PCR as the reference standard. For all lateral flow tests: C, control line; T, test line.