Pooled testing was hailed early in the COVID-19 pandemic as a way to quickly and efficiently increase the number of people tested for the virus. But the approach has fallen short of its promise for a variety of reasons, from supply and labor shortages to high community infection rates. Now, a new testing method – recently approved for use by laboratories in Israel – could help chip away at some of those issues.
The Pooling-Based Efficient SARS-CoV-2 Testing (P-BEST) method is a single-step, group testing approach that uses large pools of samples. The method is novel because it employs an algorithm to distribute samples across multiple pools, allowing researchers to perform only a small number of tests and still pinpoint any positive samples – all in one step.
In contrast, with traditional two-stage pooled testing methods, a re-test must be done of every sample in a pool if that pool tests positive during the initial group testing. That means pooled testing can quickly lose its efficiency as positivity rates climb.
In a new study published in Science Advances, researchers found that if the rate of COVID-19 positivity in a given population is below 1.3%, the P-BEST method offers both an eight-fold improvement in testing efficiency and an eight-fold reduction in test costs over individual testing.
“This really is a game changer,” Tomer Hertz, a co-investigator on the study and associate professor at Israel’s Ben-Gurion University of the Negev (BGU) in the department of microbiology, immunology, and genetics, told BioWorld. “It both allows the lab to test many more individuals using the same equipment and then also the cost per individual is reduced dramatically.”
Hertz explained that like all pooling methods, the P-BEST system becomes less efficient as the population’s positivity rate rises. Usually that tipping point is about 8%, he said.
However, pooling can still be done with subsets of samples that are likely to have a low positivity rate, such as preoperative surgical patients. “Even in places where the carrier rates are much higher, like they are in the U.S. and in many places and in Israel, you can still sort your samples in various ways and come up with subsets of samples that are really low prevalence,” Hertz said.
In August, the Israel Ministry of Health approved the P-BEST method for use in clinical laboratories in the country. Additionally, BGU and the Open University of Israel have formed a new company – Poold Diagnostics Ltd. – to pursue further regulatory approvals outside of Israel. The company will also be seeking to market the integrated software solution that enables clinical laboratories to run the P-BEST method, provided they have a liquid dispensing robot and standard PCR diagnostic capability.
The cost of the system will depend on the individual lab, Hertz explained. The cost structure would be linked to the savings achieved, with labs paying nothing if they fail to achieve savings, he said.
Hertz and his partners at Poold Diagnostics are also actively seeking partners in the U.S. to perform the necessary pilot studies to earn an emergency use authorization from the U.S. FDA.
As of Aug. 27, the FDA has issued emergency use authorizations (EUAs) for four pooled COVID-19 tests since the start of the pandemic, but all use dual-step testing and none use pools with greater than seven samples. Both of the large U.S. testing companies – Laboratory Corp. of America Holdings and Quest Diagnostics – have EUAs for pooled COVID-19 testing, as well as University of California San Diego Health and Poplar Healthcare in Memphis, Tenn.
Additionally, New York Gov. Andrew Cuomo (D) recently announced that the state’s Department of Health had approved the State University of New York (SUNY) to undertake pooled surveillance testing of COVID-19 using saliva samples. These samples would be pooled in batches ranging from 10 to 25 samples. That saliva testing protocol was developed by the SUNY Upstate Medical University and New York-based Quadrant Biosciences Inc.
Accuracy and sensitivity
Hertz and his collaborators at BGU, the National Institute for Biotechnology in the Negev, the Open University of Israel, and Soroka University Medical Center, all in Israel, tested the new P-BEST method among 384 samples that had previously been clinically tested for COVID-19.
The samples were diluted in a lysis buffer and divided into 48 pools. Each pool had a unique set of 48 samples and each sample appeared in six pools. The 48 pools were tested individually using a clinically approved COVID-19 PCR-based diagnostic protocol that included RNA extraction. Overall, researchers found five positive carriers among the 384 samples. The P-BEST method detected one false positive among those five carriers.
In a separate experiment, the researchers used P-BEST to test a group of 1,115 health care workers, more than 90% of whom were asymptomatic. The samples were pooled, and 144 tests were performed, showing that all the pools were negative. Since the infection rate in Israel was so low at the time of the testing, the researchers blindly added a positive sample to another pool, which was correctly identified.
While large pools are efficient, Hertz acknowledged that some sensitivity is lost that way. However, he noted that with pools of 48 samples, researchers were still able to detect positive samples that carried a low viral load.
Romney Humphries, a professor of pathology, microbiology, and immunology at Vanderbilt University Medical Center in Nashville, Tenn., praised the ingenuity of the system, noting that it tackles the significant problem of having to break up the pools for re-testing.
“It’s interesting, for sure,” Humphries told BioWorld. “It’s pretty clever how they’ve set up the testing strategy.”
But she noted that even with its advantages, it continues to suffer from two issues that have plagued pooled testing – a reduction in sensitivity over individual tests and the need to have a sense of the asymptomatic carriage rate in the tested population. For instance, if this method were used to test college students, the lab would need to be able to adjust the size of the pools if they anticipated a spike in positivity due to a party on campus. “That requires some type of upfront management, she said.
Humphries also raised concerns about the use of this method in the U.S., where liquid dispensing robots are typically only available at larger reference laboratories. Even if the more sophisticated technology required was readily available, she said the U.S. is currently experiencing shortages of essential testing supplies, such as pipette tips.
“There are multifaceted challenges,” Humphries said.