A pooled testing strategy for identifying SARS-CoV-2 at low prevalence.

Area of Science:

• Infectious Disease Epidemiology
• Diagnostic Technologies
• Bioinformatics

Background:

• Effective control of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) relies on rapid identification and isolation of infected individuals.
• Current reverse-transcription polymerase chain reaction (RT-PCR) testing is accurate but costly, posing challenges for widespread implementation, especially in low-to-middle-income countries.
• Sample pooling can reduce costs but must balance group size with test sensitivity to avoid false negatives due to sample dilution.

Purpose of the Study:

• To propose a novel algorithm for pooling SARS-CoV-2 subsamples to reduce mass testing costs and time.
• To accurately identify infected individuals using fewer tests and rounds, particularly at low prevalence.
• To investigate the optimal group size and parallel search strategies for efficient mass testing.

Main Methods:

• Development of a sample pooling algorithm based on hypercube geometry.
• Proof-of-concept experiments demonstrating detection of diluted positive subsamples.
• Quantification of sensitivity loss due to dilution and exploration of mitigation strategies like re-testing.

Main Results:

• The proposed hypercube-based pooling algorithm accurately identifies SARS-CoV-2 infected individuals with fewer tests and rounds.
• Proof-of-concept experiments successfully detected positive subsamples diluted up to 100-fold, exceeding previous studies.
• Cost reduction is significant, particularly at low prevalence, with ongoing field trials in Rwanda and South Africa.

Conclusions:

• Massive-scale group testing using the hypercube algorithm offers a promising, cost-effective strategy for controlling COVID-19.
• This approach facilitates continuous population monitoring and rapid isolation of infected individuals.
• The method has the potential to substantially reduce the cost of mass testing for infectious diseases.