Efficiency of Pooled Surveillance Testing in Academic Labs to Detect and Inhibit Covid-19 Outbreaks

Abstract

Robust surveillance testing is a key strategic plan to prevent COVID-19 outbreaks and slow the spread of the SARS-CoV-2 pandemic; however, limited resources, facilities and time often impair the implementation of a widespread surveillance effort. To mitigate these resource limitations, we employed a strategy of pooling samples, reducing reagent cost and processing time. Through utilizing academic faculty and labs, successful pooled surveillance testing was conducted throughout Fall 2020 semester to detect positive SARS-CoV-2 infections in a population of 4400 students. During the semester, over 25,000 individual COVID status evaluations were made by pooling eight individual samples into one quantitative reverse transcription polymerase chain reaction. This pooled surveillance strategy was highly effective at detecting infection and significantly reduced financial burden and cost by $3.6 million.

Phylodynamics of a regional SARS-CoV-2 rapid spreading event in Colorado in late 2020

“…Attenuated strains have the potential to out-compete more severe strains due to the trade-off between virus transmissibility and severity [43]. Studies like this are well suited to capture a snapshot of this kind of variance, as samples were collected from both symptomatic and asymptomatic individuals, per USAFA's randomized testing surveillance protocol [15]. Further, mutations in these strains can create reservoirs of mostly neutral mutations, possibly leading to gradual genetic drift over time [44].…”

“…A large number of the USAFA samples are descended from a series of closely related plausible common ancestors beginning with A12 that are mostly closely related to ancestor A3 (corresponding to NextStrain cluster 20G) but separated by nine mutation events mostly not seen elsewhere in the NextStrain database. This branch of the network appears to have been introduced into USAFA from contact with the local Colorado Springs community [15], and we label it the Colorado Springs variant. The network relationships and identities among PLOS ONE sequences were compared to USAFA contact tracing information (under USAFA IRB FAC20200035E) to examine consistency between our inferred genomic ancestral relationships and what was known about the contact path through the USAFA.…”

“…The CU samples were predominantly derived from clinical samples from Denver and Aurora obtained at the University of Colorado hospital, while the USAFA samples were primarily obtained from asymptomatic, randomly sampled cadets. In the case of the USAFA samples, because the testing and quarantine protocols established at the beginning of the school year resulted in a population free from COVID-19 when classes started in August [15], subsequent infections almost certainly originated from contact with the local Colorado Springs community.…”

“…Although many of the USAFA samples were likely picked up from the local Colorado Springs community based on detailed contact-tracing information and limits on off-base travel implemented by USAFA, a group of samples were also associated with a rapid-spreading event within the USAFA campus starting in late October 2020 [15]. In looking at the distribution of inferred mutations on the Colorado phylogenetic network, we found that most common ancestors [A4, A5, A6, A8, A9] share a shallow network of divergence from the ancestors A1, A2, and A3, with one or two mutations separating each from an earlier ancestor ( ).…”

“…We used replicate sequencing to validate variants across multiple related genomes rather than relying on only a single distant reference sequence to validate variant calls (Fig 1). Having established a credible set of high confidence SARS-CoV-2 genomes, we then describe the phylodynamics of a rapid viral spread event within a relatively controlled environment, estimate its divergence from related samples and compare to its associated epidemiological data [15].…”