Every pandemic is attended by expressions of fear and grief, the imposition of life-altering public health measures, and attempts to blame foreigners or other disfavored peoples. In all these respects, the current COVID-19 pandemic is like its predecessors. But the COVID-19 pandemic is playing out differently in terms of surveillance. COVID-19’s causative agent—the SARS-CoV-2 virus—can be subjected to exceptionally close surveillance because we now possess sophisticated genomic sequencing technology. Sequencing information about the viral genome can be gathered so quickly—almost in real time—that it can power a dynamic public health response.

The information revealed through genomes requires analysis and interpretation on the part of genomic epidemiologists, professionals who rely on a steady flow of new sequences to do their research. Unfortunately, it has become increasingly clear that the United States—despite being home to the firm that generates more than 90% of the world’s sequencing data—is not doing enough sequencing. Not even close.

James Lu, MD
James Lu, MD
Helix

Today, the United States sequences roughly 0.2–0.3% of all positive cases, a fraction compared to the United Kingdom, which is leading the effort with 7%. Until now, the sequencing has been done primarily through state-led efforts, hospitals, and academic institutions. It has been “fairly disjointed” up until now, according to James Lu, MD, PhD, the co-founder and CSO of Helix.

How important is genomic surveillance? The World Health Organization (WHO) recently published a study, “Genomic sequencing of SARS-CoV-2: A guide to implementation for maximum impact on public health,” asserting that sequencing is a top priority. It stated, “The accelerated integration of genome sequencing into the practices of the global health community is a must if we want to be better prepared for the future threats.”

The Centers for Disease Control and Prevention (CDC) seems to agree and is boosting the United States’ sequencing efforts through the National SARS-CoV-2 Strain Surveillance (NS3) program. Part of this program hopes to see state health departments and other public health agencies ramp up their sequencing. Indeed, the program states that its surveillance system has been scaled up to process 750 samples per state per week. In December 2020, the CDC contracted with three commercial diagnostic laboratories to conduct additional sequencing. The CDC has commitments from these laboratories to sequence 6,000 samples per week. Besides Helix, the laboratories include LabCorp and Quest Diagnostics.

Sequence trackers

Some scientists, like Emma Hodcroft, PhD, at the University of Bern, have been analyzing the SARS-CoV-2 genome for the past year, since the first sequences of SARS-CoV-2 were available. In doing so, they have gained insight into different aspects of the pandemic. Hodcroft recently published research on a variant that expanded in Spain and then continued to move throughout Europe. Hodcroft says that this work was “incredibly eye-opening” because it uncovered the role that travel tends to play in the dispersion of a variant.

Emma Hodcroft, PhD
Emma Hodcroft, PhD

In addition to tracking variants through travel, Hodcroft says that genome information can yield information about which changes in the genome are important. “We see a lot of changes in the virus,” she points out, “and it’s not always obvious which ones are the ones we want to be paying attention to.” She adds that when genomic changes can be associated with changes in transmissibility or antigen avoidance, “we can start to piece together a better picture of how the virus works [and] what changes we need to really be paying attention to.”

Indeed, the world is watching the evolution of SARS-CoV-2 as it moves around the globe, with new variants becoming important factors in the state of the pandemic. The B.1.1.7 variant that originated in the United Kingdom has garnered much attention due to preliminary (and unconfirmed) reports of phenotypic changes such as increased transmissibility or virulence.

In South Africa, another variant of SARS-CoV-2 appeared independently of B.1.1.7 that shares some of the same mutations. In Brazil, a variant of SARS-CoV-2 known as P.1 emerged with 17 unique mutations, including several in the spike protein. Although these top the list of known variants at the moment, there is no doubt that many more will follow.

Helix steps up

Helix started working on SARS-CoV-2 sequencing the day before Christmas 2020. That started “a very interesting and busy month,” recalls Lu. Helix had been COVID-19 testing, using the standard PCR method, since July. Known for its population genomics business, the company was a relative latecomer to the COVID-19 testing arena, according to Lu. But Helix quickly ramped up and currently runs half of the samples for the San Diego area.

Helix lab
At Helix, a population genomics company, efforts to sequence SARS-CoV-2 variants have shed light on the transmission of the B.1.1.7 variant. Together with other San Diego–based organizations, Helix published a medRxiv preprint in February reporting data on the prevalence and growth dynamics of this variant in the United States.

Every day in its California facility, Helix runs tens of thousands of samples—from all 50 states. But those are PCR tests, which are designed for detection. Helix has been funded to reach 100,000 samples/day. Lu says that Helix expects to achieve this rate.

Out of those samples, roughly 15–20% are positive. From the positives, roughly 1,000 samples are chosen each week to be driven a mile down the road and run on the NovaSeq instruments at Illumina; the two companies formed a partnership in January. Which samples take the ride? Lu says the partners prioritize samples deemed most likely to give a good geographic representation of the United States. They also prefer to include B.1.1.7 variants in the mix.

How many tests, on a percentage basis, need to be sequenced to detect emerging strains? Lu says that if you sequence about 5% of positives in the United States today, which means sequencing roughly 10,000 samples/day, you can catch an emerging variant that is present in about 0.1% or 1% of samples. At that level of sequencing, an emerging variant that has a circulation figure of 0.1% or 1% would be detectable. This circulation figure is roughly the one currently estimated for the B.1.1.7 variant.

For Lu, this work goes beyond SARS-CoV-2, stretching into a future that will bring more viruses that need to be watched closely. Viral surveillance, he notes, will be part of the pandemic infrastructure over time. But it will take resources. And he wonders where the investment is going to be for readiness in the future.

How much sequencing is best?

If every country in Europe could provide a couple hundred sequences every week, or even every month, Hodcroft says, that would “transform what we could say about different variants in Europe and how they’re spreading.” Hodcroft notes that if sequences were provided quickly—within a week or so of sample collection—it would be possible to “track different variants in real time across Europe.” Such a capability, she emphasizes, would be “game changing.” She adds, “If we could expand this to the world, that would be even better.”

A platform to detect variants—or sequence them

Pathogen detection tests in food are challenging; the test has to be quick, cheap, and robust. Making a genomics-based test with those attributes is no small feat. Started in 2014, Clear Labs built such a test, hoping to modernize how food companies identify pathogens. When COVID-19 hit, the company adapted its platform, ClearDX, for SARS-CoV-2 detection. The ClearDX uses Oxford Nanopore’s gridION sequencing platform to sequence the sample.

The platform can run sequencing-based modes for either detection or whole genome sequencing. The diagnostic mode runs in 11 hours and can process 192 samples at a time; whole genome sequencing takes 18 hours for 32 samples. Clear Labs’ goal is to democratize such testing so that “anyone can have the capabilities to go from sample to answer without the need to go to a specialized lab,” declares Sasan Amini, PhD, the company’s co-founder and CEO.

Sequencing wears many hats

Much emphasis is being placed on sequencing positive COVID-19 samples. More viral surveillance will lead to a deeper understanding of SARS-CoV-2 and may help inform public health decisions. Is it possible, though, for sequencing technology to play a second role in combatting the COVID-19 pandemic? Could it also help bolster the diagnostic testing effort?

Illumina came out early with its sequence-based clinical test. This test, called COVIDSeq, was the first next-generation sequencing (NGS) test granted EUA approval. An Illumina spokesperson declined to provide a detailed list of early adopters without prior permission but said global partners include the Institute of Genomics and Integrative Biology in New Delhi and the Communicable Disease Genomics Network in Australia.

Alex Dickinson, PhD
Alex Dickinson, PhD
ChromaCode

But the merits of using NGS for COVID-19 clinical testing are questioned by Alex Dickinson, PhD, co-founder and executive chairman of ChromaCode. “NGS is great when you don’t know what you’re looking for,” says Dickinson, who spent seven years as a senior vice president at Illumina. “It’s fantastic when you’re doing discovery.” But PCR is amazingly efficient when you do know what you’re looking for. When you know your target, he says, PCR is so simple and inexpensive. That is one of the reasons he thinks that most of the COVID-19 testing is happening, and will continue to happen, in PCR machines.

The biggest issues with NGS assays, according to Lu, concern turnaround time. A PCR run is 45–90 minutes, whereas a sequencer run is 10 hours. And although NGS has prodigious multiplexing ability, given that 25,000 samples can be run on a sequencer, the technique poses a serious operational challenge: loading all the samples at one time.

ChromaCode
ChromaCode asserts that its High-Definition PCR (HDPCR) technology can assess as many as four disease targets for every target that is already part of an assay—and do so in a single reaction. Because the standard SARS-CoV-2 PCR assay includes 4 targets, ChromaCode’s HDPCR SARS-CoV-2 assay could be used to assess 16 targets, facilitating the tracking of emerging SARS-CoV-2 variants.

ChromaCode makes the reagents and computer services to execute PCR tests, including its own COVID-19 test, which has been available for the past eight months. To date, about 3 million COVID-19 tests have been shipped. Unlike traditional PCR tests that can detect 4 targets, ChromaCode’s technology can detect 16 targets with a combination of reagents and software. The extra channels can be used to identify more respiratory diseases or to target other regions of the SARS-CoV-2 viral genome. In the latter case, SARS-CoV-2 variants could be identified.

Birgitte Simen, PhD
Birgitte Simen, PhD,
Ginkgo Bioworks

In May 2020, Ginkgo Bioworks announced a $70 million investment, from Illumina and others, to build up its NGS COVID-19 testing capacity. Birgitte Simen, PhD, Ginkgo’s senior director of genomics and computational biology, says that the company works closely with Illumina and talks to Illumina’s personnel on a regular basis. But Ginkgo doesn’t use Illumina’s test.

Ginkgo has two tests—both are NGS based. One test, Concentric, is a PCR-NGS mashup: it’s not a qPCR test, but it’s not a full genome sequencing test either. It’s a ratiometric test based on the Swab-seq test developed by Octant Bio, but with modifications, explains Simen. The test result comes from the ratio of two quantities: the amount of a “spike in” product (a synthetic RNA sequence that is similar to, but distinguishable from, the one for SARS-CoV-2) and the amount SARS-CoV-2 virus. The samples are barcoded, and the readout is done on an Illumina NovaSeq or NextSeq.

Concentric does not require the full time it takes for a normal sequencing run. The sequencer needs to read only the barcode and enough of the sample to differentiate between the “spike in” and the actual virus.

What is the benefit of Concentric over the standard PCR test or the Illumina COVID-seq? In Simen’s view, the benefit is scale—or the potential to scale (Ginkgo did not reveal how many tests it runs at the moment.) Simen clarifies that when she cites scale, she means that you can run a lot of samples at the same time, driving down turnaround time with a few sequencers. How many samples can be run at one time using Concentric? Ginkgo didn’t say, but it’s safe to say that the number exceeds the capacity of qPCR.

What about surveillance? Ginkgo launched a whole-genome sequencing test in May 2020 that the company runs “at fairly low scale on a weekly basis for a number of partners,” according to Simen. Ginkgo has done 4,000 of those tests total. She says that the company is “looking to scale this up.”

Like too many aspects of the COVID-19 pandemic, the United States is lagging far behind other countries in viral genome surveillance. It’s hard to understand how that can be. A year has passed, and some of the companies involved in surveillance are worth billions of dollars. Time and resource constraints are poor excuses.

SARS-CoV-2 surveillance efforts in the United States
After a slow start, the CDC recently began moving quickly to bolster SARS-CoV-2 surveillance efforts in the United States. These efforts include sequencing and measures to improve the “infrastructure and workflow to ensure efficient sequence data submission to public repositories.” Here, a map shows the cumulative number of SARS-CoV-2 sequences by state that have been published in public repositories from January 2020 to the present (as of February 12). More such data can be found on the CDC’s National Genomic Surveillance Dashboard.

Kári Stefánsson, MD, CEO of deCODE genetics, led an aggressive effort to stamp out COVID-19 in his home country of Iceland—which has current daily case counts in the single digits. In April 2020, he said that doing the same would be “even easier” in the United States due to the country’s talent and resources. He added that it was “pretty sad” that although the methods used by Icelanders to quell the pandemic in their island nation originated in America, the Americans have not been using these methods themselves. According to Stefánsson, it all comes down to having “the will and desire.”