Field Evaluation of a Molecular Point-Of-Care Test for Detecting Low-Density Malaria Infections: A Community-Based Prospective Diagnostic Accuracy Study in Rural Burkina Faso

Background: Asymptomatic, low-density infections sustain malaria transmission across Africa, but conventional diagnostics routinely miss them, which undermines elimination efforts. We adapted Dragonfly, a cold-chain-free molecular point-of-care (POC) test combining sample preparation with a lyophilised, colorimetric loop-mediated isothermal amplification chemistry, for rapid detection of pan-Plasmodium DNA from finger-prick blood on a portable, battery-powered device. We assessed its field accuracy outside laboratory settings in rural Burkina Faso, against whole-blood quantitative PCR (WB-qPCR) and standard malaria diagnostics.

Methods: We conducted a community-based cross-sectional survey in Nanoro Department, Burkina Faso, at the end of the malaria transmission season. Participants aged ≥6 months were randomly selected from a Health and Demographic Surveillance System. Capillary blood was obtained from each participant and tested using HRP2-based rapid diagnostic test (RDT), expert light microscopy (LM), dried blood spot quantitative PCR (DBS-qPCR), WB-qPCR (reference standard), and the Dragonfly test. Microscopy and qPCR were performed in reference laboratories; RDTs and Dragonfly tests were run in real-time in peripheral facilities and homes.

Findings: Of 601 individuals recruited, 596 (99·2%) were analysed; 472/596 (79·2%) were asymptomatic and 490/596 (82·2%) were WB-qPCR positive, of whom 242/490 (49·4%) had submicroscopic infections (median density 0.9 parasites/µL; IQR 0·2–6·4). The Dragonfly test had the highest sensitivity (446/490; 91·0%, 95% CI 88·1–93·4), significantly exceeding DBS-qPCR (416/490; 84·9%, 95% CI 81·4–88·0), RDTs (229/490; 46·7%, 95% CI 42·2–51·3), and LM (248/490; 50·6%, 95% CI 46·1–55·1). Dragonfly detected 198/242 (81·8%, 95% CI 76·4–86·5) submicroscopic infections, versus 170/242 (70·2%, 95% CI 64·1–75·9) for DBS-qPCR, and 71/242 (29·3%, 95% CI 23·7–35·5) for RDTs. Specificity was high across methods, Dragonfly and LM each achieved 98·1% (104/106; 95% CI 93·4–99·8), DBS-qPCR 93·4% (99/106; 95% CI 86·9–97·3), and RDTs 96·2% (102/106; 95% CI 90·6–99·0), with no significant difference from Dragonfly.

Interpretation: Community deployment of this cold-chain-free molecular POC malaria test is feasible in a resource-limited rural African setting and substantially improves detection of low-density infections, closing a diagnostic gap for asymptomatic and submicroscopic parasitaemias that sustain transmission. Future work should evaluate cost-effectiveness and scalability.