NEW YORK (GenomeWeb) – At the Clinical Virology Symposium this week in Savannah, Georgia, Arc Bio introduced its first metagenomics next-generation sequencing kit — a research-use-only, end-to-end solution that detects and quantifies 10 viral species commonly associated with transplants — that it dubbed Galileo Pathogen Solution – Transplant.
Preliminary data in a well-attended company-sponsored workshop suggested the kit results were comparable to quantitative PCR, with a caveat of low specificity for one of the targets in cases where patients had high titer infections with another virus.
In an interview prior to the conference, Arc Bio CEO Todd Dickinson said that the company's mission is to "fundamentally change how we approach diagnosing and treating infectious disease." This led the company to name the product line Galileo.
As previously reported, the product pulls together "a bunch of exciting technologies" in sequencing, Dickinson said.
The Galileo platform is an integrated solution around an NGS platform, he said. It consists of reagent kits that have integrated sample prep — DNA extraction, library prep, as well as proprietary host-removal technology — as well as internal and external controls, both for the reagent kit and the software.
"The overall system is intended to be bullet-proof and easy for anyone to run," Dickinson said.
The calibration controls are used to generate calibration curves that then enable calibration of the viruses in the sample.
On the back end, the firm has developed sophisticated analytics that make sense of the data coming out of the NGS instrument, Dickinson said. A cloud-based software solution is used to generate a report that users can interface within a simple web app.
Arc Bio said the workflow takes 48 hours to run from start to finish.
The GPS kits essentially perform unbiased metagenomic sequencing, with all non-human DNA sequenced and the results compared to the database.
The company plans to take a tiered approach to assay development, targeting the first assay to a narrow list of the 10 viruses that most commonly cause problems to patients who are immunosuppressed after receiving solid organ or stem cell transplants.
Specifically, the 10 viruses that the test reports on are adenovirus, cytomegalovirus, Epstein-Barr virus, human herpes viruses 6A and 6B, herpes simplex viruses 1 and 2, varicella zoster virus, JC virus, and BK virus.
However, among these 10 viruses, the technology can detect and identify more than 375 different strains.
Promising to unlock the power of NGS without the learning curve, the Arc Bio transplant kit is the first in what is expected to be a series of metagenomic solutions. It can potentially be validated by a clinical virology or other lab and used as a laboratory-developed test to diagnose viral infections in patients who have had solid organ or stem cell transplants.
The firm chose transplant as the first application for GPS because there is a "dire need for better approaches," Dickinson said. "A full 30 percent of transplant patients will acquire a viral infection," he said, adding that the way doctors are currently testing is essentially by guessing which test to order.
The main target market for this kit is the 270 transplant centers in the US, Dickinson said. "We’re not limiting ourselves to just the US," he added, noting that the technology is available to any transplant center moving in the direction of sequencing-based solutions to viral detection.
"We're finding that, as in any market or field, there are early adopters, and those are the folks that we'll largely be going after," he said, including institutions that are embracing metagenomic analysis or have sequencers in their labs. The firm will work with them to make the technology available on an RUO basis, and they'll be free to validate the test in their own labs.
Patients who are immunosuppressed after transplant can have reactivation of prior infections, or acquire them newly, and because their defenses are lowered, the infections can be dangerous. Patients can also acquire infections from the donor tissue they receive.
For example, a CMV infection acquired after transplantation can cause pneumonia, gastrointestinal tract disease, hepatitis, encephalitis, or retinitis. If the donor was known to be CMV positive, the recipient can be put on prophylactic antiviral treatment, or the recipient can be treated — typically with gold-standard intravenous ganciclovir — if they develop a CMV infection after transplantation. Infection is usually diagnosed with PCR-based testing, and viral load is monitored to determine when to stop treatment.
Greg Storch, a pediatric infectious disease specialist at Washington University School of Medicine in St. Louis, attended the CVS workshop and said in a subsequent interview that the Arc Bio transplant panel could theoretically have clinical utility.
"I think it is a well-selected panel, especially for patients who are undergoing organ transplantation, and no such panel exists right now," Storch said, adding that NGS is a quite interesting approach, although it still remains to be seen whether it is the best approach.
NGS-based clinical diagnostics have been on the horizon for some time, and Storch said he thinks laboratorians are very interested in applications of the technology. Agnostic approaches are especially appealing, but also present unusual challenges, particularly from a regulatory standpoint. A narrower approach, such as the one currently being used by Arc Bio, may be more practical, at least for getting started, Storch said.
Dickinson also agreed that agnostic detection is a higher goal. "Ultimately, that's where these technologies are headed, but we felt it was important to step into this in a rigorous way … and expand from there."
And, although there may not be many virology labs that have the equipment in house to do sequencing, ones in major academic medical centers frequently have access to a sequencing core, Storch noted.
The kit's ability to detect rare genetic variations, particularly ones that confer drug resistance, would be good to address, he added, because this is an important issue in transplantation and something for which NGS would be well suited.
Preliminary data
In the workshop, Benjamin Pinsky, the medical director of clinical virology at Stanford Health Care, presented a retrospective analysis of patient samples using the panel, while Julianne Brown, a clinical scientist at Great Ormond Street Hospital for Children, described her lab's overall use of metagenomics for encephalitis cases and showed beta testing of the GPS kit using pediatric samples.
Pinsky pointed out that there is only one assay cleared by the US Food and Drug Administration for viral detection in transplant scenarios, specifically a qPCR-based test for CMV in solid organ and hematopoietic stem cell transplant. A CMV test from Qiagen was cleared last year. Otherwise, viruses are typically quantified using single-target real-time PCR assays, Pinsky said.
The Stanford labs' evaluation was a retrospective study done as a research contract with Arc Bio. Pinsky's lab compared the Galileo panel to a laboratory-developed workflow involving extraction on Qiagen instruments followed by qPCR testing with reagents from companies like Altona Diagnostics and Artus as well as two in-house laboratory-developed tests.
The initial study set included 49 plasma samples with a median sequencing depth of 55 million and a median of about 2 million non-human reads sequenced.
Total percent agreement with qPCR-based tests was 89 percent, Pinsky reported, with a negative percent agreement of 91 percent and positive percent agreement of 85 percent.
Most of the qPCR-positive, Galileo-negative samples appeared to have low levels of virus. Using an alternative, previously published bioinformatics pipeline, the researchers found the same samples were also negative, "suggesting that the Galileo pipeline is not less sensitive than at least one other pipeline," Pinsky said.
Pinsky and his team also tried to tease apart the samples that were negative by qPCR but positive by Galileo. Only 30 percent of these were positive by the alternative analysis pipeline, he said.
One problem child of the panel appeared to be the BK virus, which has a very small genome of 5.1kb, Pinsky said, and so is "more difficult to detect at low levels by sequencing." The pipeline also called seven samples as containing JC virus that was not detected by qPCR, and these were in samples that were highly positive for BK virus.
Importantly, there was "good quantitative agreement" between samples quantified by qPCR and Galileo, which Pinsky found to be "the most impressive part."
One interesting factor was also the level of co-infection with a particular virus known as torque tino virus, or TTV. "It is very common – virtually everyone is infected with TTV," Storch explained.
In his CVS presentation, Pinsky said nearly two-thirds of the clinical samples his lab tested in the Arc Bio evaluation had extremely high concentrations of TTV, on the order of 2 million copies per milliliter. "This did not interfere with the detection of any of the other viruses that we expected to detect," Pinsky said.
Clinical virologists are interested in the virus because quantifying it could potentially serve as a biomarker of a patient's immune status, for example, in kidney transplant patients, providing insight into whether they are immunosuppressed enough to prevent graft rejection but not so immunosuppressed that ubiquitous non-pathogenic viruses like TTV can run rampant. Indeed, incorporating TTV quantification might be especially useful in a transplantation-related viral NGS panel in the future, Storch said.
Brown, meanwhile, reported on her lab's metagenomics testing in encephalitis cases — using a pipeline called MetaMix that the lab has recently uploaded onto BlueBee — which she said can help identify pathogens in pediatric cerebrospinal fluid samples where the lab could likely run out of sample before it would be able to run all the PCR tests needed.
Her retrospective beta study of the Galileo solution used 23 samples from stem cell transplant patients. For the vast majority, there was concordance between the methods, she said. "The kit provides all necessary internal and external controls, and this makes it easy to adhere to accreditation standards," Brown also said. Her group, as well, saw a handful of samples with high CT values that were undetected by sequencing, as well as some cross-reactivity with JC virus and high viral load BK virus.
The potential clinical utility of detecting these particular 10 viruses remains to be seen. Storch said that there is a catalog of antivirals and other treatment strategies for targets in the panel.
In Brown's case, she noted that identifying pathogens with metagenomics in encephalitis cases is important because non-infectious encephalitis can be treated with immunosuppression, a treatment that would make the illness much worse if it were actually caused by an infectious agent. The drawback, however, is that sequencing can be much more expensive and time-consuming than PCR.
Competing technologies — like ones from Karius, IDbyDNA, and others — offer sequencing for viral ID as a service, but Dickinson said deploying the technology into a customer's own lab is a very different approach, with the upside of reduced turnaround time.
In addition, "The power of metagenomic sequencing is that it has the potential to be truly agnostic and hypothesis-free," Dickinson said. The firm is already starting to work on the step in its tiered approach to targeted metagenomics assay development, but for now it will focus on getting the RUO transplant test into key peoples' hands and establishing centers of excellence around the world, he said.