NEW YORK – A startup company called Caspr Biotech is aiming to prove the utility of nucleases from extremophile organisms in CRISPR-based diagnostics by developing a testing platform that would be within reach of countries with limited resources.
The ambitious vision of the Buenos Aires- and San Francisco-based firm collided at the beginning of 2020 with the COVID-19 pandemic, giving Cofounder and CEO Franco Goytia and his colleagues a chance to put a product on the market much sooner than they were anticipating. With it, Caspr looks to compete with CRISPR companies like Sherlock Biosciences and Mammoth Biosciences while making the next generation of precision molecular diagnostics accessible to more people worldwide.
Goytia and two collaborators founded the company at the beginning of 2019 in Argentina to use CRISPR for epigenetic reprogramming applications, but eventually shifted their focus to CRISPR-based molecular diagnostics.
The firm raised pre-seed funding from an investor in Argentina, and also received investments from IndieBio, Grid Exponential, and Paul McEwan, the former cofounder and CSO of Kapa Biosystems. The nearly $3 million in total funding allowed Caspr to establish an office in San Francisco and start fleshing out its vision out into an actual product line.
A new kind of Cas12
But before the company could start developing a CRISPR-based diagnostic, it had to decide on which Cas enzyme to use — and that was the first problem.
"We initially tried to approach the foundational universities or institutions — UC Berkeley, the Broad Institute — for licensing possibilities and had no success, even though we tried to narrow the scope of a territory or a particular application," Goytia said. "They were quite restrictive in terms of the access for the technology."
Because the freedom to operate as they wanted was important to Goytia and his colleagues, they decided on a strategy of biodiscovery to find their own novel nucleases for which they could file patents and get intellectual property protection. That strategy began by researching publicly available metagenomics databases like the NCBI's. This helped the researchers generate a hypothesis that they could find novel microorganisms within extreme environments that might contain novel CRISPR systems with differential activities that could be useful for diagnostics.
Since 2019, they've been collaborating with research groups working in extreme environments, accessing unpublished metagenomic data from, for instance, Antarctica or the South American Puna grassland and salt desert ecoregion, Goytia said, even taking expeditions to La Puna, staying with the native communities, and immersing themselves in the environment.
This immersion and respect for the native communities is an important part of the company's bioprospecting strategy, Goytia said. The firm has signed onto the Nagoya Protocol on Access to Genetic Resources and the Fair and Equitable Sharing of Benefits Arising from their Utilization to the Convention on Biological Diversity, an international agreement which aims to fairly share the benefits from the utilization of genetic resources. More than 100 countries have signed the protocol, though the US isn't one of them.
The protocol aids in the "fair prospecting and commercialization of components which are based out of genetic material of a particular environment," Goytia said. "So, from these communities, we take just a scoop of a lake or a scoop of soil or a scoop from a particular environment, and we can generate a source of income. And on that source of income, that can be something that goes back to the original communities, in part."
The biodiscovery strategy has uncovered several new nucleases that the firm's researchers are now characterizing for possible use in future diagnostics. For example, they've found enzymes that were expected to only cut DNA but also had nonspecific kinetic activity for RNA, Goytia said. They've also found enzymes that are similar to Cas9, Cas12, Cas13, and Cas14, but have less than 50 percent — in some cases, less than 30 percent — sequence identity with those more familiar nucleases.
"Also, we characterize novel functionality or differential functionality when compared to the previously disclosed or discovered enzymes from other [research] groups," he added.
When they do find enzymes that aren't the best fit for diagnostics purposes — enzymes that might better be used for therapeutics, for example — Caspr reaches out to other firms to see if there are possibilities for partnerships to make the best use of those nucleases.
Importantly, Caspr's La Puna expeditions yielded what Goytia and his colleagues are hoping will be the nuclease that puts the company on the map — a Cas12 enzyme that they've integrated into the COVID-19 assay that will likely be the company's first product to market, the Caspr Lyo-CRISPR SARS-CoV-2 Kit.
For now, Goytia isn't saying much about how this Cas12 works or how it's different from current commercially available Cas12 enzymes that would fall under Broad or UC Berkeley IP protection, except that it has about a 50 percent sequence identity with more familiar Cas12 enzymes. The nuclease also has differential activity at room temperature, allowing for Lyo-CRISPR kit to be transported and performed at room temperature. The enzyme can also be lyophilized into beads.
The company is preparing three scientific papers for publication in the next couple of months that will contain more details about the enzyme itself. Caspr has also applied for Emergency Use Authorization from the US Food and Drug Authorization and has submitted parallel regulatory applications in other regions and countries for the Lyo-CRISPR kit using its novel nuclease, and expects to receive approval in the coming weeks.
A platform approach
In the meantime, though, the company is also refining its platform technology, which is what it was developing before the COVID pandemic grabbed the world's attention.
"Since the very origin of Caspr, we've had this platform approach to molecular detection. That includes the applications within infectious diseases in healthcare, but also doing validations beyond healthcare for applications in agriculture or farming, for example," Goytia said. "This, in part, comes from the ease of reprograming or reconfiguration that CRISPR brings by tailoring the RNA guide of our system."
Throughout 2019, the company validated its technology on 12 to 15 different targets, including tropical viruses like Dengue and Zika, endemic viruses like hantavirus, so-called bacterial superbugs, and antimicrobial resistance genes. The work was done in what Caspr thought was preparation for the commercialization and production of a variety of diagnostics.
In fact, Caspr researcher Carla Gimenez, along with collaborators in Argentina, published a study in Emerging Microbes and Infections in June showing the validations they did in 2019 using the firm's platform for detection of emerging infectious diseases.
In that paper, they used a commercially available Cas12a (Lba Cas12a). Taking advantage of Cas12a's reported unspecific single-stranded DNAse activity upon target recognition, they coupled the enzyme with an ssDNA reporter, generating a fast, accurate, and ultrasensitive molecular detection method. Further, they demonstrated that Cas12a was able to detect DNA target sequences corresponding to carbapenemase resistance genes such as KPC, NDM, and OXA. And with the addition of a reverse-transcription step, the researchers were able to detect viral RNA sequences from dengue, Zika, and hantavirus genomes.
In all cases, the assay run time was less than two hours, at attomolar levels of detection. The resulting use of Cas12a to detect both DNA and RNA targets made it an appropriate and convenient tool to detect all types of pathogens, they wrote.
As COVID emerged, the researchers quickly performed initial validations and set a more aggressive timeline to fully develop a prototype assay that they could take to the FDA for EUA, and also manufacture at scale.
Gimenez and two Argentinian colleagues published a proof-of-concept paper on the BioRxiv preprint server showing a similar approach for the COVID assay as they had for the overall infectious disease platform.
They again used a commercially available Cas12a enzyme (LbCas12a), but showed that their test was characterized by a limit of detection lower than the minimum levels needed to detect the virus in clinical samples for most commercially available tests. They also pointed to the CRISPR diagnostic's portability and low cost as advantages compared to many PCR-based tests.
"As of July 2020, we're on the track to [achieving] our initial objectives in the sense that we have developed what we call the Lyo-CRISPR SARS-CoV-2 Kit, which is based out of a lyophilized format of isothermal amplification combined with our CRISPR detection through our Cas12 enzyme," Goytia added. "And given this lyophilized presentation and the accessibility of CRISPR, it allows for detection with minimal external equipment — a heat block, a fluorescent plate reader."
Making it accessible
The fact that the equipment can be transported at room temperature is one of the assay's most important features, Goytia said, given that the company is looking to deploy its diagnostics in resource-limited regions.
Even though the firm hasn't yet received EUA for the Lyo-CRISPR kit, it's already working with several early-access partners to validate the assay, both on the clinical side and on the workflow side, to compare the CRISPR-based process to qPCR workflows.
"Those results that we've been getting from those partners in the US, in Canada, in South America, in Africa, they have been very, very good," Goytia said. "From our initial clinical validations of 30 positive and 30 negative [samples] of extracted RNA, we had 100 percent sensitivity and specificity. And now having processed more than 300 samples from our external partners on their own, that degree of sensitivity and specificity remains at that very, very high standard of over 99 percent for those two variables."
He added that those validations were done on RNA extracted from nasopharyngeal or nasal swabs from COVID patients.
Importantly, the assay's simplified workflow allows for results to be obtained in less than 60 minutes, and Goytia estimated that the test will cost about $10 to $15 per patient. The improvement of both time and cost as compared to PCR-based tests is why Caspr believes its tests will be more accessible than those of its competitors.
"We're trying to take our cost of production and our price point lower and lower," Goytia added. "We believe that within the next one year, as we shift a lot of the production of many these components internally, our price point on a per-patient basis could be even much lower." Between the possibility of a lower price point and the fact that real-world validations have proven robust with a variety of RNA extraction workflows and diverse lab operations, the company is hoping to "turn this into millions of tests in the second half of this year," he said.
The CRISPR diagnostic landscape shapes up
Assuming the Lyo-CRISPR kit receives FDA EUA, it would have to be performed in laboratories rather than at the point of care, because it requires an RNA extraction step. However, it can be used in low-complexity labs or facilities, with minimal capital expenditures.
Caspr is following a familiar pattern for CRISPR-based diagnostics companies this year: it's at least the third firm in the past couple of months that's used COVID to develop a smaller, single-use version of a larger, multiplex, or more complex future diagnostic platform. Caspr's Lyo-CRISPR kit fits neatly into the competitive landscape, alongside Sherlock Bio's test for use in CLIA labs (until now, the only CRISPR-based test to receive FDA EUA), its planned point-of-care test in partnership with Binx Health, and Mammoth Bio's planned CRISPR-based SARS-CoV-2 diagnostic test, which it's developing in partnership with the consumer healthcare arm of GlaxoSmithKline for use by consumers at home and in clinics.
But the COVID tests are only the beginning of what Sherlock Bio and Mammoth Bio have planned. Sherlock Bio and Binx are already in talks to develop a multiplex assay that includes several viruses, such as SARS-CoV-2 and influenza. The company is also continuing to develop its own at-home SARS-CoV-2 test that would likely work like an at-home pregnancy test, and is even considering whether there's value in integrating its two foundational technologies — SHERLOCK and INSPECTR — into one platform for viral detection.
Mammoth Bio is also working on different version of its test with GSK. While it ultimately plans to develop an at-home COVID test, it also plans to create a version for healthcare facilities. Importantly, the company also has the potential to develop many more similar tests based on its own CRISPR technology, such as an at-home disposable test for the flu, and has been in talks with GSK about possibly developing additional tests.
So, it comes as no surprise that Caspr is also planning to create additional formats for its tests. The company will be developing a direct-from-sample application for the COVID test, which could be applied to decentralized settings such as schools, warehouses, airports, or even at home, Goytia said. This system would utilize a fully automated, reusable device and single-use cartridges for single-patient disease detection.
The core biology would still utilize the same Lyo-CRISPR technology as the initial assay, but it would be integrated into small, disposable cartridges that would be fed into the reusable device. This device could then have its own optical readout and connectivity systems, or it could be connected to a smartphone through an app, which would receive the results.
Caspr is planning to deploy the cartridge and device solution by the end of 2020, Goytia said.
"The two major developments that we're trying to do out of our technology and our enzyme is the simplified multiplex capability through CRISPR, and also to avoid the need for the isothermal amplification step," he added. "So, when you take into account that we have an enzyme that has this differential performance at room temperature, that can be lyophilized into various formats, and used directly for detection at room temperature, we envision that we could have a solution that doesn't require any kind of heating and is very accessible for molecular detection in this decentralized manner in less than ten or fifteen minutes. We're working to achieve that vision."