NEW YORK – While COVID-19 held the world in its grip throughout much of 2020, the urgent need for reliable, accurate, and fast diagnostics had much of the medical community largely turning to PCR technology. But as the urgent rush for tests has abated, diagnostics developers are exploring how they can use more complex technologies like next-generation sequencing or CRISPR to create ever more accurate or informative assays.
While both technologies individually have unique advantages and functionalities, scientists from the Translational Genomics Research Institute, or TGen, and genome technology company Jumpcode Genomics have teamed up to develop a potentially more powerful test using both platforms.
PCR remains the gold standard for infectious disease testing in general. But in the case of SARS-CoV-2, the advent of multiple vaccines as well as the emergence of several dangerous viral variants has prompted researchers to begin exploring diagnostic technologies that provide more information than a simple infection result.
CRISPR and NGS, in particular, have emerged as possible challengers to PCR's crown. Several research groups published work in 2020 that showed CRISPR's diagnostic utility including assays that improved on the speed or ease of use of previous tests, could be performed in one step, worked with smartphones, or could be combined with isothermal amplification methods or microfluidics. These assays also highlighted CRISPR's adaptability and potential for multiplexed viral testing.
Some companies, such as Sherlock Biosciences and Mammoth Biosciences, have even received Emergency Use Authorization from the US Food and Drug Administration for CRISPR-based SARS-CoV-2 diagnostics.
On the NGS front, meanwhile, a case study last May from the pediatric intensive care unit at Children's Hospital Los Angeles, or CHLA, clearly illustrated the strengths of NGS for SARS-CoV-2 diagnostics.
While PCR excels at giving a yes-or-no answer for infection, it cannot do anything to help with contact tracing, epidemiological analyses, and genetic studies of viral mutations. As the CHLA case study showed, NGS testing allowed the clinicians there to monitor a small outbreak at the hospital, trace the transmission pattern, and confirm the safety of the hospital — all of which would have been impossible with PCR-based testing only.
Then in June, Illumina received FDA EUA for its COVIDSeq Test, the first such authorization for an NGS-based SARS-CoV-2 test. This was followed by an EUA in October for UCLA's NGS-based SwabSeq COVID-19 Diagnostic Platform and one last month for Twist Bioscience and Biotia's SARS-CoV-2 Next-Generation Sequencing assay.
According to Nicholas Schork, deputy director and distinguished professor of quantitative medicine at TGen, the institute will soon add to that list with an NGS-based SARS-CoV-2 diagnostics of its own. But TGen's test will have an added bonus: a CRISPR-based sample prep kit, courtesy of Jumpcode Genomics.
Jumpcode, which was founded in 2016, has developed a negative selection technology called CRISPRclean, which works by depleting unwanted nucleic acid sequences from NGS libraries to leave behind only targets of interest. Unlike PCR, which seeks to amplify targets of interest, CRISPRclean uses CRISPR-Cas guide RNAs, or gRNAs, to attach to DNA or RNA sequences that aren't of interest and deplete them from a given sample.
The platform uses a catalytically active Cas9 locks on to the targeted region in the sample, and then cuts the targeted sequence so that it cannot be amplified during NGS library construction. The nucleases could theoretically be programmed to remove either DNA or RNA-derived double-stranded cDNA targets. Users are only left with the sequence or sequences that they want to study, or they're able to search for unknown signals that had been previously obscured by all the background noise.
Jumpcode had previously said that its technology could be packaged into clinical kits for infectious disease, where it could be used to remove human DNA for clearer signals of viral, bacterial, or fungal pathogen DNA. And that's exactly what Schork and his colleagues at TGen are aiming for.
"The problem [with NGS] is that a lot of the stuff that will be in those samples is not what you're looking for — it's more noise. And most of it will just be human DNA, because it's taken from a human," he said. "What one would want to do is find the stuff that's not human, that is… the DNA sequence reflecting the pathogen, and somehow get rid of all the noise. And that's what Jumpcode's technology can do."
By using Jumpcode's CRISPR-based depletion reagents, he added, the user is left with a much higher signal-to-noise ratio, and hence a confident and reliable identification of the pathogen.
The way the test would work, Schork said, is that each sample would be subjected to the Jumpcode depletion reagents, so that all that's left are the targets of interest, whether that's RNA for specific viral variants, RNA for all possible variants, or even genes that could help signal human host responses to infection. Once that's done, the samples would be run through a sequencer, and then the results would be returned to doctors and patients.
There are multiple ways that a sample could be collected and returned for testing, he added. TGen is looking to start with in-person collection at facilities that would be capable of carrying out the sequencing, or perhaps sample collection at point-of-care locations, with the samples being sent to the lab for processing.
The researchers have also been testing all kinds of input sample types to determine which one would be best, whether that's nasal swabs, saliva, blood, or another type of biospecimen that's already been validated for PCR-based diagnostics. They've determined that the Jumpcode reagents perform basically the same way across the different sample types.
"So, we're not seeing that the sequencing-based test with the Jumpcode reagents only works for the nasal swabs as opposed to saliva. It looks like it's working pretty well no matter what the starting material," Schork said. Once other studies have determined what sample type contains the most viable source of viral RNA, he added, that's what the test will be designed to use.
Aside from the sample type, the Jumpcode and TGen teams are also still working on the optimal mix of gRNAs to include in the depletion kit.
"Jumpcode can make different kits depending on what it is that you want to deplete from a sample, so you don't want to run into situations where you deplete everything," Schork said. "The depletion constructs [that Jumpcode has made] are reliable and work. It's just, in what context are they going to apply them and then what are they going to apply them to, to deplete? That might be context-specific."
That context for depletion could be what sets this test apart from other commercial tests. While the researchers are fairly certain about what they don't need to see in the samples, there are a few targets that they're considering keeping, such as markers for more than one SARS-CoV-2 variant, genes that would indicate human immune response to the virus itself, or even markers that would indicate the presence of other viral infections.
Knowing the presence or absence of a virus in someone without knowing the context is helpful only to a certain extent. "For example, someone could have been infected a long time ago and then had a remnant of the pathogen that was detected from a standard PCR test. But that wouldn't necessarily tell you what the sequence was of the pathogen, which could be very important for determining whether or not someone's been infected with one of the super spreader variants or not," Schork said. "That [also] wouldn't put into context whether or not they're likely to still be infective or whether their body is actually responding to the infection at the time of the diagnosis."
Having a test that not only indicates the presence of the pathogen but also the host's response to it is important to understanding whether the patient is still infective or is actually at the end of their infection period. "Are there genes that are starting to be expressed that give an indication that the body is fighting off an infection? That would incredibly important," Schork said. "And, if through the sequencing, you could not only see that the person is infected with SARS-CoV-2, but what strain or what variant might be in the genome of the virus they've been infected with, that could help with contact tracing and put it into context how well they're going to do in the ICU if they start to develop symptoms."
That process of picking and choosing the gRNAs is still ongoing, and the exact mix can be tailored to create a more reliable and comprehensive test, he added.
Longer term, Schork and his colleagues are envisioning creating a multiplex test that would search for more than one infection at the same time. If a patient walks into a doctor's office with symptoms like headache and sore throat, "you don't know if it's SARS or the flu," he said, and having to run multiple tests to determine a diagnosis is wasteful or might be error-prone.
"But if you could just sequence everything in the sample … that might give people a greater insight into what's going on than only having a negative result on one test and then not being able to know what the real issue is," Schork said.
The next step for the TGen and Jumpcode teams is to secure a funding grant to finish the research and development of the test. They've talked to a few agencies, including the Biomedical Advanced Research and Development Authority, which has funded the development of new diagnostic assays that could be deployed at scale. Schork is hopeful that the team will have more clarity on funding before the fall for additional work. "That would include not just refining what goes on in the laboratory, but actually doing a prospective study comparing the gold-standard PCR test against the sequencing-based test," he said.
That kind of study would also help the researchers generate the kind of data they'd need to pursue regulatory clearance for their test. Right now, they're documenting the CRISPR-based depletion protocols carefully in terms of reliability, reproducibility, and safety, Schork said. Once they have the necessary data from a wider prospective study, they plan to apply for FDA EUA.
Importantly, he noted, while this test has potential for use during the COVID-19 pandemic, it's not something that has to be confined to testing for SARS-CoV-2. "If we used the COVID crisis as the test case to show the utility of sequencing-based approaches to diagnosis coupled with depletion reagents and a couple of other tricks on the downstream part, it's an easy thing to show that'll work in other contexts," he said. "And then, we won't have a COVID test. We won't have a flu test. We won't have a MRSA test. We'll have one test that looks at all these things, and that's where we want to go."