Background
The Gambia has successfully reduced malaria transmission. The human reservoir of infection could further decrease if malaria-infected individuals could be identified by highly sensitive, field-based, diagnostic tools and then treated.

Methods
A cross-sectional survey was done at the peak of the 2017 malaria season in 47 Gambian villages. From each village, 100 residents were randomly selected for finger-prick blood samples to detect Plasmodium falciparum infections using highly sensitive rapid diagnostic tests (HS-RDT) and PCR. The sensitivity and specificity of the HS-RDT were estimated (assuming PCR as the gold standard) across varying transmission intensities and in different age groups. A deterministic, age-structured, dynamic model of malaria transmission was used to estimate the impact of mass testing and treatment (MTAT) with HS-RDT in four different scenarios of malaria prevalence by PCR: 5, 15, 30, and 60%, and with seasonal transmission. The impact was compared both to MTAT with conventional RDT and mass drug administration (MDA).

Results
Malaria prevalence by HS-RDT was 15% (570/3798; 95% CI 13.9–16.1). The HS-RDT sensitivity and specificity were 38.4% (191/497, 95% CI 34.2–42.71) and 88.5% (2922/3301; 95% CI 87.4–89.6), respectively. Sensitivity was the highest (50.9%, 95% CI 43.3–58.5%) in high prevalence villages (20–50% by PCR). The model predicted that in very low transmission areas (≤ 5%), three monthly rounds of MTAT with HS-RDT, starting towards the end of the dry season and testing 65 or 85% of the population for 2 consecutive years, would avert 62 or 78% of malaria cases (over 2 years), respectively. The effect of the intervention would be lower in a moderate transmission setting. In all settings, MDA would be superior to MTAT with HS-RDT which would be superior to MTAT with conventional RDT.

Conclusion
The HS-RDT’s field sensitivity was modest and varied by transmission intensity. In low to very low transmission areas, three monthly rounds per year of MTAT with HS-RDT at 85% coverage for 2 consecutive years would reduce malaria prevalence to such low levels that additional strategies may achieve elimination. The model prediction would need to be confirmed by cluster-randomized trials.