Karen Titus
March 2021—Listening to experts and others make predictions about the pandemic, it’s easy to think they’re obsessed with surfing: How will we deal with the next wave?
The dangers are real. With SARS-CoV-2, the next wave might be swept in by emerging variants, with their uncertain but worrisome impact on transmission, severity of illness, treatments, and vaccine effectiveness. Laboratories are looking to approach these variants of concern in ways that will leave them feeling neither blinkered nor bludgeoned, as happened all too often in the past year.
There’s cause for hope, say those who’ve been doing genomic sequencing or are organizing efforts not only in their own labs, but on the state and regional level as well. Federal funds are beginning to flow; the CDC is, so to speak, back.
Nevertheless, the conversations are regularly punctuated by “Who knows?” and “No one knows.” (Also: “Who’s going to pay for this?”) One year after SARS-CoV-2 tore loose, labs are, in a sense, at a new starting line. They’re racing against a whole new set of concerns, all of them set against a sorrowful marker, as the number of U.S. deaths from COVID-19 surpassed 500,000 in late February.
Meanwhile, new unknowns keep washing ashore. “I guess it’s not really a surprise,” says Eric Konnick, MD, associate director, genetics and solid tumor laboratory, and director, genetics preanalytical services, Department of Laboratory Medicine and Pathology, University of Washington (where he’s also assistant professor). “We did expect variation to happen,” as it will with any RNA virus. “But we thought there were some biological aspects of this virus that would make it less likely to mutate in large ‘jumps,’ if you will.”
The real surprise, he continues, is that some of the variants have a cluster of different alterations, not just a single mutation, that seem to impact the biology of a particular strain. Also somewhat surprising, he says, is that these variants emerged in very specific populations and spread very quickly.
Says Dr. Konnick: “Who knows what the next year is going to look like? We’ve shifted from getting closer to the capacity we need for routine testing, at least in the United States. But now we have this whole different problem, that is quite a different scale.”
“The question is, are the curves going to change for the worse because of the variants?” asks James Crawford, MD, PhD, professor and chair, Department of Pathology and Laboratory Medicine at the Donald and Barbara Zucker School of Medicine at Hofstra/Northwell, and senior vice president of laboratory services, Northwell Health. “No one knows.”
In the meantime, labs are trying to prepare.
[dropcap]T[/dropcap]he need for sequencing seems obvious. Identify variants and stop outbreaks in their tracks. Keep new strains from outpacing the rate of vaccinations. Don’t let year two of the pandemic turn into a high-stakes chase.
Just as form follows function, purpose precedes probes. Public health surveillance and monitoring is a different “ask” than clinically reporting, says Dr. Konnick, who is a member of the recently formed AMP COVID response steering committee.
“You can’t just pour buckets of money on the problem,” says Dr. Konnick (though money does help). “It’s multifactorial. You need to make sure you have everything in place.”
But in place to do what, exactly? Dr. Crawford says he asks the question every time he meets with medical colleagues, public health officials, and government leaders: What is the goal? It is, he says, an intense discussion.
Is it to do surveillance for public health purposes? Variants aren’t stopping at the border. B.1.1.7 (the so-called U.K. variant), B.1.351 (originally identified in South Africa), and P.1 (first linked to travelers from Brazil) are being identified across the United States. Yet another new variant (B.1.427/B.1.429) appeared to be surging in San Francisco and Southern California in late February, while B.1.526 seemed to be infiltrating New York City. Because so little sequencing has been done in the United States, it’s hard to know the true prevalence of variants or what else might be out there. States that aren’t reporting the appearance of variants of concern may truly be unaffected, or they may lack the resources to identify them.
Or is the goal to do sequencing in a CLIA-certified lab, with short turnaround times, to manage patient care? Right now, variants don’t determine treatment. But that could change. There’s also the matter of whether variants can affect the performance of diagnostic assays.
Another question sounds like fodder for a medical thriller—Can vaccines outrace the mutating virus?—though that dramatic query has a basis in reality. Some variants could elude vaccines, and if they become widespread, the vaccines in general will become less effective. “So it’s important to continue to do surveillance,” says Frederick Nolte, PhD, vice chair of laboratory medicine and director of the molecular pathology laboratory, Medical University of South Carolina, where he’s also professor of pathology and laboratory medicine.
As the questions pile up, the lab community continues to try to help the country test its way out of the pandemic—first diagnostically and now genomically. Every step feels like exiting a Victorian sitting room—from day to day, season to season, and now one year to the next, labs find themselves bumping up against any number of obstacles.
[dropcap]W[/dropcap]hen he first started hearing about emerging variants, Dr. Nolte recalls, “It was, ‘Oh, this is interesting, but I don’t have to worry about it because it’s public health.’”
As the new year gained steam, so did the variants, and curiosity gave way to concern. Some appeared to be gaining a toehold, and reports of new variants have surfaced with unnerving regularity, like a Tom Brady Super Bowl appearance. Public health labs were, once again, under pressure to perform tests. “So maybe there are some reasons to be worried,” says Dr. Nolte, given the potential impact on patient care. (South Carolina, he notes, had the dubious distinction of being the first state to report the presence of B.1.351, in late January.)
The B.1.1.7 variant has had some impact on the diagnostic tests, Dr. Nolte notes, particularly those that included a spike protein target. “So one good reason for keeping an eye on the variants is to make sure our diagnostic tests don’t suffer from what I term ‘genotype bias,’” he says.
Dr. Crawford reports hearing from two vendors that the platforms his lab uses detect both the P.1 and B.1.351 strains. “So from a PCR detection standpoint, it does not seem to be an issue.” He pauses. “At this time.”
Another pause. “But—we’ll pay attention.”
“I wish I had a crystal ball,” Dr. Nolte says, naming the one item that no one has successfully sourced this pandemic. “It seems like things are getting better in the U.S.,” he says, speaking in mid-February. “The sky isn’t falling, and the variants probably aren’t going to ruin all of our plans to vaccinate people.
“But who knows?”
The line separating public health and clinical care, never sharp to begin with, has only gotten blurrier in the pandemic. Public health labs have been performing sequencing, but as Dr. Nolte points out, each lab is basically an independent player, with disparate levels of funding and staffing. The CDC has contracted with LabCorp, Quest, and Illumina to perform sequencing of positive samples sent by state public health departments, but funneling testing through public health labs will still result in an uneven picture of which variants are emerging and where. That’s yet another reason clinical laboratories are entering the fray.
Or, in the case of Dr. Konnick’s lab, staying in the fray.
Washington was a first responder of sorts, identifying the first U.S. case of COVID-19 in January 2020. The UW clinical lab oversees clinical testing for the region, which includes screening samples that will be sequenced and reported to the state department of health. A separate UW program for testing and sequencing of community-collected samples as part of a public health initiative called the Seattle Coronavirus Assessment Network, or SCAN, is also contributing to the regional efforts. (Early on, this group used respiratory samples from the Seattle Flu Study to create a SARS-CoV-2 diagnostic assay.)
SCAN has worked closely with Trevor Bedford, PhD, at the Fred Hutchinson Cancer Research Center, to do SARS-CoV-2 sequencing almost from the start. Last spring, when UW sent its EUA applications for diagnostic tests, “All of the samples SCAN submitted to the FDA as our positives had whole genome sequencing at that point. So we’ve been ahead of things,” Dr. Konnick says.
Early sequencing efforts showed particular strains of the virus dying out, as patients who had tested positive quarantined and kept it from spreading. That’s not the case now, however; some observers have predicted B.1.1.7 will become responsible for the majority of new infections in the United States by April.
[dropcap]N[/dropcap]ew York generated its own heavy headlines last spring. Dr. Crawford and others are drawing on those hard experiences to figure out a thoughtful approach to sequencing.
He asks the question writ large: What does the lab community need to be doing?
One answer is, apparently, hop on Zoom.
Like everyone else pushing through the pandemic, Dr. Crawford is in meetings all the time. He’s in regular conversation with colleagues in the New York State SARS-CoV-2 Testing Consortium (he’s the moderator and secretary), a group formed last spring by laboratory leaders from 12 academic medical institutions in the region. He’s also involved with the recently formed New York City Coronavirus Genome Collaborative.
Talking has led to surprisingly quick actions.
Both groups met Feb. 2. The following day, the New York City Department of Health and Mental Hygiene issued a request for applications from regional institutions to provide information on sequencing capacity and platforms—Northwell submitted an application, he says—while the city was charged with developing mechanisms for weekly, aggregate reporting of sequencing data for the city.
Consortium members have been asking three key questions as they ponder their role, Dr. Crawford says:
- What is going to be done differently if sequencing data become more widely available?
- Will such data change clinical management?
- How can labs meet sequencing needs in the region?
Even though he had no answers as of mid-February, Dr. Crawford expects matters to change speedily—in a matter of weeks, not months. “This is evolving rapidly. This story, even at the regional level, let alone the national level, is going to be unfolding quickly,” Dr. Crawford says, adding, “How it all lands remains to be seen.”
These questions aren’t New York-specific, nor are Dr. Crawford and colleagues alone in asking them. “I’m sure there are many, many, many other conversations occurring between the state and laboratories, and obviously the city and laboratories.”
Dr. Crawford urges labs to create a regional collaborative—“even just to talk to one another and share your experiences”—as sequencing ramps up. If last year was an every-lab-for-itself competition, this year is more akin to joining a kibbutz. “Work together for advocacy and communication,” Dr. Crawford says. “Learn from one another. Be a part of that regional discussion with civic authorities, departments of health especially. It’s critically important to open those lines of communication.”
Case in point: Last April, the consortium had a sought-after conversation with the state’s commissioner of public health, Howard Zucker, MD, JD, seeking guidance on diagnostic testing for the region. Dr. Zucker assigned the Greater New York Hospital Association to act as convening authority for academic labs in the state; it was joined by the Healthcare Association of New York State. By working closely with those two regional groups, labs have been closely aligned with the New York State Department of Health and the governor’s office throughout the pandemic.
“It’s a wide-open sharing of ideas, [and] challenges, and the ability to get feedback from the state, and to know our deliberations were helping inform state policy,” Dr. Crawford says. “That was in the heat of the first part of the pandemic, and we’re not done with it. Having that communication with civic leadership has certainly been beneficial to the New York region.”
Could the New York network of labs provide adequate capacity for sequencing? “If I do my own back-of-the-envelope arithmetic for how much capacity is available for the New York region,” Dr. Crawford says, “we could hit 20 percent of positives. But can we harness that? And if so, how? And under what leadership?”
The questions don’t stop there, Dr. Crawford says. “What is enough? What is success? What should you be shooting for?”
[dropcap]A[/dropcap]s Dr. Crawford well knows, no lab is an island (even one, like his, that’s located on Long Island). Figuring out the logistics of sequencing is a national effort.
One island—Great Britain—has been managing well, by most accounts, having launched (and funded) a national sequencing program in March 2020. “But we’re a bigger island,” Dr. Nolte laughs.
Multiple federal leaders and agencies are pushing money into the center of the table to support surveillance. But, says Dr. Nolte, it’s unclear how academic and hospital-based clinical laboratories can take advantage of this.
In February his lab joined a CDC-led group called Spheres, which was formed last April and comprises members from public health, academic, and commercial laboratories to tackle genomic research of SARS-CoV-2. Dr. Nolte said his lab, which had begun validating its SARS-CoV-2 sequencing setup in February, could easily handle 400 to 700 samples a week. While not a massively scaled effort, he says, it would be enough to add useful knowledge about spread of variants in the state.
“But how do we pay for it?” he asks. “We’ve got the instrumentation; we’ve got the people. We don’t have a way of paying for it.”
The CDC, as noted, has partnered with large commercial reference labs to generate sequence analyses, but it’s not clear whether similar partnerships are in the works with academic medical centers, Dr. Nolte says. That’s what many lab associations are pushing for, he says.
“Clearly the federal government is thinking about pouring lots of money into this,” Dr. Nolte continues, with the word “billions” regularly popping up in proposals from legislators and agencies. The key question about such funds, he says, is, “Will they be available to labs like ours?”
Adds Dr. Konnick: “This is not a cheap endeavor.” There’s no straightforward mechanism for reimbursing clinical labs for this work, and the public health/clinical testing divide complicates matters further. “Most laboratories would want to batch things as much as possible to reduce costs,” Dr. Konnick suggests. But again, that push-pull between testing and its purpose will affect what labs do. Batching is cheaper and appropriate for public health surveillance; if results are being used to manage patient care, however, larger (and cheaper) batch sizes won’t work for rapid turnaround times.
[dropcap]A[/dropcap]s the number of labs offering SARS-CoV-2 sequencing grows, so will the data.
Dr. Konnick sees at least one bright spot as he considers the challenges: There’s no shortage of new tools to generate and share information. “I think back to the original SARS-1 virus, in 2003—none of this was even possible.”
Possible doesn’t mean easy, however. The tools and those who use them face a tangle of questions. What will public health reporting look like if labs are required to provide information about variants? How will data be shared? Can an LDT be used? What will the data-sharing requirements and mechanism look like for the raw sequencing data? (“These files are huge,” Dr. Konnick says.)
Dr. Nolte chimes in. “If we start sequencing, what will our public health lab want from us? Are they going to want the raw data files? Are we going to post that to their website? Are we going to post it to one of the sites that are already out there?”
Dr. Nolte’s lab is using the international database GISAID to compare its own isolates with what other labs are reporting. But questions persist. How can labs be certain of the quality of sequence data being posted on various sites? And how much additional information—geographic, patient profiles, etc.—is needed to make sense of the data?
Then there’s the matter of coordinating information across multiple laboratories—dozens? hundreds?—that are eager to join the national effort. “My hats off to the people trying to do that,” Dr. Nolte says.
And if genomic information makes its way into clinical practice, more questions will erupt. As anyone who performs next-generation sequencing for cancer testing knows, there’s nothing simple about the results, even when working with expert medical oncologists. “We find they misinterpret things, or have a different interpretation of what we communicate with them,” Dr. Konnick says, “regardless of how careful and thorough we think we are on the pathology and lab medicine side.”
[dropcap]H[/dropcap]aving fewer labs performing sequencing would make coordination easier, but more players may be better, Dr. Nolte says, particularly if they include academic medical laboratories, most of which are already invested in NGS and could generate, collectively, substantial SARS-CoV-2 sequences without slowing down other clinically important work.
Of course, sequencing might become clinically important at some point, which would force labs to make other adjustments—to TATs, for example, as well as reporting infrastructure.
And who will handle the work?
As labs weigh their options, Dr. Crawford says, “You have to take care of your workforce.” Even if sequencing is outsourced, the lab will still be receiving the specimens and sending them off to a reference lab, he says. “And that’s on top of the still substantial COVID-19 burden.”
Even as he responded to the NYC health department’s request for information on sequencing, Dr. Crawford and his colleagues continue to focus on building capacity for PCR-based diagnostic testing. Unlike labs in other parts of the country, Dr. Crawford says, Northwell and other area institutions continue to see high demand. As of mid-February, Northwell was receiving 14,000 to 15,000 specimens on weekdays for SARS-CoV-2 testing. “It’s [all] a substantial increase in the laboratory effort to deliver health care now, with the same amount of staff.”
Dr. Konnick finds himself worrying about his bench scientists. Those who are trained in oncology biomarker testing have been moonlighting in the virology lab on evenings and weekends, he says. “I’m very thankful they’re willing to do that, but it’s also putting a strain on them.
“I mean, I’m definitely exhausted from the last year,” he acknowledges.
He anticipated that would be the case, he says, recalling that he initially prepared mentally for getting through a year of COVID-19 stressors, after which—he hoped—things would settle down. Emerging variants might topple that notion. “We’re potentially on this trajectory of continuing that very intense, high-level professional work, and putting people at risk for burnout if it continues another six months to a year.”
All of which prompts Dr. Crawford to return to one of the persistent questions about sequencing: Why do it? “If we’re already maxed out with PCR-based testing and some serology antibody testing thrown in, should we also be aspiring to do sequencing? Currently, that’s a public health activity, not a medical activity. And until there’s affirmative evidence that sequencing impacts medical care, we should not divert our workforce from medical diagnostics. That is the dilemma.”
[dropcap]D[/dropcap]eciding to do SARS-CoV-2 sequencing is complex. So are the subsequent steps.
Consensus needs to emerge on quality issues—sequencing depth, percent of genome covered, etc. Says Dr. Nolte: “If you’re doing amplicon sequencing, which I think most people are, the primer set you use is important. You’d like to see that be consistent.” As with any type of sequencing, this will require the proper bioinformatics pipeline, as well as clustering software.
In an early run of about 500 samples, Dr. Nolte’s lab noticed an approximate 15 percent failure rate. They didn’t pay much attention, simply wanting to see what would happen with a representative group of samples. But it’s clear that sequencing won’t perform well on samples with a low viral burden—hardly a surprise, he says, but important to know.
Will this be a zero-sum game, with SARS-CoV-2 pulling focus from, for example, oncology applications? It depends, Dr. Konnick says. A lab may have the equipment and expertise, but depending on its organization and structure, competition may ensue for resources, space, computational capacity, and all the rest, he says. “In the near term, it’s probably not going to be a huge shift of resources, but it’s certainly a concern.”
On the other hand, SARS-CoV-2 testing has pushed labs to increase redundancy, which may spill over to sequencing. “I think like any other area of the laboratory, we’ve all learned you can’t rely on a single instrument,” Dr. Konnick says. For now, “We have enough instrumentation that this isn’t an issue.” If variant sequencing becomes widespread, personnel, robotics, and automation could become choke points within a system. “But we’re not there yet.”
What about supply issues, which haunted and taunted labs for nearly a year on the diagnostics side? “In our oncology lab, we have seen multiple disruptions in the supply chain for our general-purpose plastics and reagents for sequencing applications that are now impacting care,” he said in late February. “While some supply chains have become more stable, we are still seeing stochastic disruptions across the board.”
Dr. Nolte reports a calming of the waters in his region. Though his lab still performs a high volume of SARS-CoV-2 diagnostic tests, “It’s not that insane demand we were dealing with when things were really bad. So we have the ability to now shift staff around in response to different initiatives.”
Using CARES dollars, his laboratory purchased two NGS instruments that also offered EUA diagnostic testing for SARS-CoV-2 (the lab’s original need).
He’s also had conversations with Abbott, which maintains a global viral surveillance program. The company has offered to help validate MUSC’s sequencing platform and, moving forward, sequence a portion of the lab’s isolates. “They’ll try to culture any unusual variants and make those available,” he says. “So there are resources out there. We are trying to leverage them all.”
At UW, Dr. Konnick and colleagues are talking through the different potential clinical uses of sequencing variants.
One might be to screen patients who are eligible for antibody treatment, if there are data to suggest differential benefit to patients with or without variants. Current TATs would make that approach challenging—usually patients need to be identified, and medications delivered, fairly quickly—although larger centers that run continuous sequencing might be able to do so.
More targeted assays that allow faster screening—reflexively testing positive cases using real-time PCR, for example—might work, although the ideal use would be when there’s only one main variant of concern. If variants proliferate, and are located in different regions of the genome, “that could get complex very, very quickly,” pitting speed against comprehensiveness.
Testing archival samples presents its own difficulties. Resampling patients who’ve undergone antibody treatment could be challenging, Dr. Konnick says, because many sequencing techniques require a higher viral titer for detecting nucleic acid.
And then there’s the storage issue. A lab that can perform high volumes and has tested a large number of positive cases is a plus, but doing so requires the resources and system for storing and retrieving them. And samples can degrade, depending on the extraction method. Dr. Konnick reports seeing degradation of samples that undergo freeze-thaw, like Chicago roads in winter, which may result in insufficient nucleic acid, especially if a sample was borderline to begin with.
[dropcap]O[/dropcap]ne year out, labs are seeing improvements as well as possible parallels with the past.
Dr. Crawford takes a fairly rosy view. “Here we are a year later,” he says. “We have a world community that is now educated in a way we were not educated a year ago about this new pathogen. We also have a different administration, which is putting science first.”
The previous dynamic is now being replaced, he says, by “thoughtful consideration of the federal response”—a combination of federal funding combined with CDC and FDA actions, and coordination with the laboratory industry, all of whom have a year’s experience.
“It’s been an extraordinary year of knitting together these communities,” Dr. Crawford says. “One of the shining aspects of this brutal year is this coming together, to the best of our ability. That’s so different from a year ago, when we were just beginning to find our way.”
Dr. Crawford sees another change as well: an extraordinary explosion of knowledge, as evidenced by the six-figure results that come up on PubMed searches related to SARS-CoV-2 and COVID-19. “It’s been the journey of a lifetime in terms of knowledge,” he says. “It’s a reason for tremendous hope.”
He adds: “’We’re talking about the lab effort differently. We are changed. We are not the same people we were a year ago.”
Dr. Nolte sounds a little more wary. “Not only are we behind the rest of the world in terms of sequence analysis, but it feels like a redo of the diagnostic thing.” First comes the turn to the CDC and public health labs, which are soon overwhelmed. Next, testing gets sent to major commercial reference labs, whose capacity to handle large volumes falls short. “If this really ramps up, I imagine they’ll be overwhelmed,” Dr. Nolte says. “It does feel a little like we’re doing the same thing with variant analysis that we did with diagnostics.”
At least on the surface. The advantage this time around is that more thought has gone into this, he says—the difference between reacting and responding. “I just hope labs like ours will find a way to contribute,” Dr. Nolte says. “I think we could make a significant impact.”
Karen Titus is CAP TODAY contributing editor and co-managing editor.