Comparative evaluation of multiplex bead assay, polymerase chain reaction, and rapid diagnostic test for P. falciparum detection in community surveys in Malawi.

BACKGROUND

Accurate detection of malaria infections, particularly asymptomatic and low-density infections, is essential for effective surveillance and control efforts. This study compared the performance of three malaria detection platforms for identifying Plasmodium falciparum infections in a community-based survey setting.

METHODS

Cross-sectional household surveys were conducted in three high-transmission districts in Malawi from February-March in 2022 and 2023. From 2960 participants, samples from 604 children aged 2 months to 14 years were analyzed for P. falciparum infection using three detection methods: WHO-prequalified HRP2-based rapid diagnostic tests (RDTs), a multiplex bead-based assay (MBA) for histidine-rich protein 2 (HRP2) and Plasmodium lactate dehydrogenase (pLDH) antigens, and a real-time quantitative polymerase chain reaction (PCR) targeting P. falciparum 18S rRNA, which was considered the gold standard test.

RESULTS

PCR detected a P. falciparum prevalence of 24.0% (145/604; 95% CI 20.7-27.6%) compared to 29.5% by RDT (178/604; 95% CI 25.9-33.3%). RDT performance was suboptimal compared to PCR, with RDT sensitivity of 48.3% (95% CI 39.9-56.7%) and specificity of 76.5% (95% CI 72.3-80.3%). Seventy-five PCR-positive samples were RDT-negative, possibly representing low-density infections below RDT detection limits, while 108 RDT-positive samples were PCR-negative, consistent with persistent HRP2 antigenemia following parasite clearance. Receiver operating curve analyses demonstrated that the MBA for both antigens showed limited discriminatory power against PCR, HRP2 (AUC = 0.63, 95% CI 0.58-0.69) and pLDH (AUC = 0.61, 95% CI 0.56-0.67). However, MBA demonstrated strong agreement with RDTs, (HRP2: AUC = 0.93 (95% CI 0.90-0.96); pLDH: AUC = 0.88, (95% CI 0.85-0.92), indicating comparable antigen detection capabilities. PCR was estimated to be 41.3% less expensive in start-up costs and 32.2% less in processing costs compared to the MBA.

CONCLUSIONS

In high-transmission, community-based settings, the malaria antigen detection MBA demonstrated a limited ability to detect P. falciparum infections when compared to infection identified by PCR. Real-time PCR is increasingly feasible for use in surveillance where infrastructure permits. These findings underscore the importance of selecting diagnostic tools based on surveillance objectives, local epidemiology, and available laboratory capacity.