Dorfman pooling enhances SARS-CoV-2 large-scale community testing efficiency

Abstract

PCR-based analysis is the gold standard for detection of SARS-CoV-2 and was used broadly throughout the pandemic. However, heightened demand for testing put strain on diagnostic resources and the adequate amount of PCR-based testing required exceeded existing testing capacity. Pooled testing strategies presented an effective method to increase testing capacity by decreasing the number of tests and resources required for laboratory PCR analysis of SARS-CoV-2. We sought to conduct an analysis of SARS-CoV-2 pooling schemes to determine the sensitivity of various sized Dorfman pooling strategies and evaluate the utility of using such pooling strategies in diagnostic laboratory settings. Overall, a trend of decreasing sensitivity with larger pool sizes was observed, with modest sensitivity losses in the largest pools tested, and high sensitivity in all other pools. Efficiency data was then calculated to determine the optimal Dorfman pool sizes based on test positivity rate. This was correlated with current presumptive test positivity to maximize the number of tests saved, thereby increasing testing capacity and resource efficiency in the community setting. Dorfman pooling methods were evaluated and found to offer a high-throughput solution to SARS-CoV-2 clinical testing that improve resource efficiency in low-resource environments.

Fig 1. Number of tests and positivity rate at Cadham provincial laboratory.

Samples were analyzed by PCR analysis, and positive tests are tabulated using results from all samples analyzed per day. Data date range February 19, 2020 to July 28 2022.

Fig 2. Dorfman pooling schematic.

Samples are represented by tubes, and PCR tests are represented by green (negative) or red (positive) rounded rectangles. When a Dorfman pool is positive, deconvolution is required. When a positive pool is detected each sample is individually re-tested. Pool sizes of 4, for example would result in 7 total tests utilized for analysis of 12 samples. In addition, positivity rate, sensitivity, and efficiency gained need to be taken into consideration when choosing an appropriate pool size for large scale testing.

Fig 3. Change in Dorfman Pooled Ct values for each pool size.

Individual unpooled Ct values are represented by solid lines, and pooled Ct values are graphed with dotted lines for each PCR target ORF1/a and E. Ct values are smoothed using an average of the nearest eight neighbours. A default Ct value of 40 was also instituted when pools had a result of “Target not detected” (Supplemental Material).

Fig 4. Test efficiency of Dorfman pooling.

The efficiency of utilizing Dorfman pooling at varying pool sizes adapted to positivity prevalence compared to individual testing. A positive value for tests saved represents a reduction in tests used compared to individual testing, and negative values represent an increase in the number of tests used.