Abstract
Clinical next-generation sequencing (NGS) can mean different things to different people, so when trying to understand the current landscape of clinical NGS, it is important to appreciate that it covers a wide range of applications—from single-gene testing to whole-genome sequencing.
Also known as high-throughput sequencing—owing to its more rapid resolution of DNA samples than the original Sanger sequencing methods—NGS was introduced at the start of the 21st century with the launch of Lynx Therapeutics's massively parallel signature sequencing platform.
Lynx Therapeutics was acquired in 2005 by Solexa, who was then acquired by Illumina in 2007. Illumina has grown to control around 80% of the global NGS market since its formation in 1998. The company offers several different sequencing platforms, of which two—the MiSeq Dx and the NextSeq 550Dx—currently have regulatory approval for in vitro diagnostic use. They both work on short-read technology, meaning that the DNA is cut into short fragments, up to 300 bp in Illumina's case, before the individual fragments are sequenced simultaneously.
Advantages of this method over long-read sequencing include lower cost and higher accuracy.
Clinical applications of NGS
The major clinical applications for NGS are currently in the diagnosis of rare diseases, the analysis of tumor samples for mutations that might guide treatment, and in prenatal and neonatal screening.
Bruce Korf, associate dean for genomic medicine at the University of Alabama at Birmingham Heersink School of Medicine, explained that, currently, clinical genetic testing can be broadly grouped into three distinct approaches. The first involves examining a single gene for a pathogenic variant. This is often done in conditions with well-defined characteristics that generally point to a likely clinical diagnosis but where finding a mutation may give valuable additional information.
Bruce Korf
associate dean for genomic medicine, University of Alabama, Birmingham Heersink School of Medicine
The second approach looks at a panel of multiple genes, any of which might explain a particular condition. Then finally there is whole-genome sequencing (WGS) or whole-exome sequencing (WES), which may be done when the list of possible genetic causes of a condition is too long to focus on any one gene or if the clinical manifestations do not point towards any specific well-defined diagnosis.
Korf says that over time these three approaches will “probably converge because as sequencing technology gets more and more powerful, it may become less expensive and more efficient to sequence the whole genome than to focus on just one gene.” He adds that if the technology “reaches a point where it's able to detect all types of pathogenic variants then you might end up sequencing the genome in preference to looking at a single gene because a major advantage to sequencing is that the technology is gene agnostic.”
Having said that, Korf stresses that we are not at that point yet but believes “I believe that's where things are headed.” In the United States, commercial laboratories such as Novogene and Quest play a major role in providing NGS solutions. The majority carry out testing using Illumina's technology but a fast-growing number also offer services that incorporate other systems such as Thermo Fisher Scientific's Ion Torrent Technology.
Many academic laboratories such as the Broad Institute's Genomics Platform also carry out clinical NGS, often using laboratory developed tests (LDTs) in conjunction with FDA-cleared instruments to overcome regulatory hurdles, but the advantages of outsourcing NGS can include lower costs and faster turnaround times, particularly for smaller labs that may have to wait for enough samples to fill a test run.
However, the need for large test runs may become less of an issue with new technologies that are lowering the sample threshold needed to make in-house testing more efficient and faster than an outsourcing-based approach.
Even so, Korf says, “Being at an institution that does not have a lab that does sequencing does not serve as an impediment to getting access to it.”
Current obstacles to widespread adoption of clinical NGS
What does serve as an impediment, however, is reimbursement. Heidi Rehm, chief genomics officer at Massachusetts General Hospital says that although there is momentum to make clinical NGS more widely available “the biggest barrier is the lack of insurance coverage.”
A 2021 report1 by the U.S. Department of Health and Human Services found that Medicare payments for genetic tests quadrupled between 2016 and 2019. In addition, the number of laboratories that received more than $1 million in Medicare payments per year for genetic tests almost tripled, and the number of providers ordering genetic tests for beneficiaries more than doubled. Yet a 2020 study2 by Kathryn Phillips, director of the Center for Translational and Policy Research on Precision Medicine (TRANSPERS) at the University of California, San Francisco, and colleagues showed that just 39% of Medicaid enrollees have cover for WGS or WES. That proportion increases to 63% for insured individuals. Coverage for tumor sequencing was 80% for insured individuals and 56% for those with Medicaid, while a respective 97% and 90% had coverage for non-invasive prenatal testing (NIPT).
TRANSPERS program manager Michael Douglas said that data from November 2021 show that 17% of commercial policies covered WGS and 69% of commercial policies covered WES. For WGS, about 65% of policies only provide cover for rapid WGS in the inpatient (typically neonatal intensive care unit [NICU] patients).
Michael Douglas
program manager, TRANSPERS
“In general, there has been a slow increase in coverage for WES. Coverage for WGS hasn't increased by much and is limited to very specific settings,” he remarked.
Douglas added that “reimbursement remains a barrier to wide implementation of WGS and WES (as well as other NGS tests), as coverage is very dependent on payer's evaluation of clinical utility for these tests.”
The definition of clinical utility varies depending on who you ask —a payer may have a different view to a researcher or clinician. Therefore, “if one wants to move the needle on reimbursement, the key is conducting studies that generate data that payers need or use to make informed decisions,” Douglas said.
Education gaps need to be filled
Another factor that can be a barrier to NGS uptake is education. “Many clinicians who are practicing today didn't have easy access, or had no access at all, to NGS technology when they were in medical school or early clinical practice. As a result, they may be unfamiliar with its benefits and unaware about when it can be used to inform decisions in caring for a patient,” said Luca Quagliata, global head of medical affairs for clinical next-generation sequencing and oncology at Thermo Fisher Scientific. In addition, clinicians may be unprepared to explain the results of an NGS test to a patient when they do not fully understand it themselves.
Luca Quagliata,
global head of medical affairs for clinical next-generation sequencing and oncology, Thermo Fisher
Laboratory technician loads samples into the Ion Torrent Genexus Dx System from Thermo Fisher Scientific.
Indeed, a recent survey of 46 pediatrics residents by Ryan Gates and colleagues from Stanford University School of Medicine3 found that 28% felt they had insufficient knowledge of the basic concepts of genetics and 35% were insufficiently confident in discussing these basic concepts with patients. More than half (57%) said they were not confident about the indications for genetic testing and almost two-thirds (63%) felt they had insufficient knowledge of genetic testing limitations.
Of note, confidence was higher among residents who were in the earlier years of their training than among their more senior counterparts. This could be because they are closer to the training they received in medical school and thus feel more knowledgeable, the researchers suggest.
Quagliata says that “medical associations are now putting a lot of attention on filling the educational gap that exists toward the understanding of molecular profiling and NGS.” Commercial providers are also playing a part. For example, Thermo Fisher Scientific is working with patient advocacy groups like LUNGevity to help educate clinicians while incorporating a patient centric view on the benefits of molecular testing.
For the clinicians who are comfortable ordering genetic tests, there is the administrative burden to overcome. Tests and results are often not well-integrated into electronic health systems and physicians need to take time to obtain the additional consent required for genetic tests.
However, companies like Concert Genetics, who provide digital infrastructure for the management of genetic testing and precision medicine, are working hard to make life easier for clinicians. Last year they announced the release of their comprehensive genetic test identification system, which they say “uniquely identifies all 160,000 genetic testing products, organizes them into a multi-level taxonomy, and incorporates new tests as they enter the market, which is currently happening at a rate of more than 30 per day.”
The system reduces administration time because “a test can be selected, authorized, ordered, resulted, billed, adjudicated, and correctly paid” because all of the systems use the same test identification code.
Another factor that compounds the lack of education among clinicians is insufficient reimbursement for genetic counseling services. “Clinical departments are having trouble funding genetic counselors to serve their clinics. And so, even though genetic counselors are out there, healthcare providers can't pay for them and then they don't have access to those services,” said Rehm.
However, researchers are now looking at ways to overcome the limited access to genetic counsellors by making adjustments to traditional service delivery models. A team at the HudsonAlpha Institute for Biotechnology in Huntsville, Alabama, has investigated whether genome sequencing results can be effectively communicated to NICU patient families by non-genetics providers who underwent training in genome sequencing technology and psychosocial considerations as part of the SouthSeq study4. Following the training, participants reported significantly increased confidence in reading and interpreting genome sequencing results and in managing a patient's care based on genome sequencing results. Kelly East and co-investigators said they now plan to analyze data on the frequency and nature of errors in results disclosed by the non-genetics providers and compared outcomes with results disclosed by genetic counselors.
Can anyone disrupt Illumina's monopoly?
Illumina has held the monopoly in the NGS market for a long time now, but with patents about to expire, things could soon change.
The Chinese sequencing giant, MGI, is already selling its CoolMPS (massively parallel sequencing, another term for NGS) system to the U.S. market, with their StandardMPS system expected to be available in January. MGI uses similar technology to Illumina but is currently cheaper, which may make the company more appealing to some users.
The cost of NGS is often cited as a barrier to NGS implementation. And while costs have declined substantially since 2009 when Illumina announced the launch of its own personal full-genome sequencing service for $48,000, the median price health plans paid for clinical WES, which includes pre- and post-sequencing steps, was still $8,800 in 20215.
Bruce Korf pointed out the cost of a magnetic resonance imaging scan is in the few thousand dollar range, so clinical NGS is now in the same ballpark. Illumina is also believed to have plenty of scope to reduce their reagent costs, which will lower overall costs further.
Although perhaps not as low as those claimed by Ultima at the Advances in Genome Biology and Technology (AGBT) meeting in Florida in June, where they announced the launch of their UG 100 ultra-high-throughput platform that will deliver a $100 genome ($1/Gb) using short-read technology that works with 300-bp fragments. Not only is it cheaper, the massively parallel novel sequencing platform is also faster, with a 20-hour turnaround time compared with 48 hours on Illumina's NextSeq technology.
Critics may argue, however, that the $100 price tag is unrealistic because it only includes the cost of reagents and not labor or and pre- and post-sequencing steps such as data analysis. It also doesn't account for the instrument amortization, which is rumored to cost as much as $2 million.
Another newcomer to the market, Element Biosciences, launched their large benchtop sequencer in March. The AVITI system is competitively priced ($5–7/Gb) compared with Illumina's NextSeq ($20/Gb) due to its reduced reagent use and it creates a higher signal-to-noise ratio, which improves quality. The technology also has the ability to produce synthetic long reads (many short reads joined together) as large as 10 kb, which sets it apart from the competition.
The third potential disrupter that has launched in the past 12 months is Singular Genomics with their benchtop G4 sequencer. Another short-read system that uses sequencing by synthesis, the USP for Singular is a platform that they claim will deliver up to three times the data output per hour than any other benchtop instrument currently available, with a flexible run capacity that doesn't require high throughput for cost savings. The price per Gb ranges from below $10 up to $45.
Shawn Baker, who through his company SanDiegOmics advises start-ups and investors on all aspects of genomics, says that the announcements by Ultima, Element, and Singular mean that “for the first time in many many years Illumina's going to have competition.”
He believes the competition will help to bring prices down and will not be easy for Illumina. “But it's not exactly rosy for any of the new entrants either because Illumina is so entrenched,” he remarked.
At present, it is unclear whether any of these companies will seek regulatory approval for their technology as in vitro diagnostic devices and therefore where they will fit into the clinical NGS marketplace. However, Rehm notes that “most clinical labs do not use the IVD versions of the instrument and instead use the research instruments and validate the instrument and test themselves as an LDT.”
The recently launched CE-IVD (IVDD) Oncomine Dx Express Test is used with the Ion Torrent Genexus Dx Integrated Sequencer from Thermo Fisher Scientific in clinical labs.
One company that already has FDA approved tests, including the Oncomine Dx Target test, which in 2017 was the first FDA-approved NGS test for non-small-cell lung cancer (NSCLC) on market, is Thermo Fisher Scientific. The Oncomine Dx Target test, carried out on the short read Ion Torrent Genexus System, screens for mutations associated with NSCLC that can be used to guide targeted therapy. In the U.S. the test is approved for seven NSCLC targeted therapies, with regulatory approval available for 15 targeted therapies in 17 countries worldwide.
PacBio's Sequel IIe long-read whole-genome sequencing system.
Last year, Thermo Fisher announced that it will be collaborating with AstraZeneca to develop companion diagnostics for targeted therapeutics in AstraZeneca's precision medicine portfolio. The tests will be co-developed for use on the Ion Torrent Genexus System.
“By continuing to work with our pharma partners on companion diagnostic solutions, we are developing assays that ensure newly approved targeted therapies are readily available for patients. Solutions along with the Ion Torrent Genexus Dx System will enable ultra-fast clinical NGS results and report delivery,” said Quagliata.
What about long reads?
All of the machines mentioned so far work on short-read technology, but there are two companies in the field—Pacific Biosciences (PacBio) and Oxford Nanopore Technologies—that use long reads in their systems. Long reads, as the name suggests, are sequences of sections of DNA that are thousands of bases long. The methods are more expensive and can be less accurate than short reads, but on the other hand they can interrogate regions of the genome that are inaccessible with short reads.
It is unclear where long-read technology might fit into the clinical landscape, but both companies have announced collaborations that may improve their standing. PacBio will be working with Berry Genomics, a leading company in clinical genomics and life science in China, to develop desktop systems based on their HiFi technology for the Chinese clinical market. And Oxford Nanopore has just announced that they will provide sequencing services to Netherlands-based Genome Diagnostics for NGS-Turbo, a high-resolution HLA-typing assay with a turnaround time of less than four hours that can help match transplant patients with organs.
Where is clinical NGS headed?
Regardless of which test is used, it is likely NGS will be at the backbone of healthcare in future. In March this year, the Telomere-to-Telomere (T2T) Consortium announced that they had assembled the first fully complete human genome, which offers endless possibilities for future clinical research.
An area that is currently generating a lot of interest is creating polygenic risk scores (PRS) for common diseases like diabetes, cancer, dementia, and heart disease. These scores can be created using genome sequencing and will identify people at greatest risk for developing certain conditions.
The Ion Torrent Genexus System from Thermo Fisher Scientific requires minimal hands-on time and delivers NGS results in as little as a single day.
PRS are developed using data from population-based screening studies such as the 100,000 Genomes project, Our Future Health Initiative in the U.K., and the All of US research program in the U.S.
Our Future Health6 aims to recruit up to five million adults from across the U.K. who will all have their DNA genotyped, using an array produced by Illumina, and will provide detailed information about their lifestyle and medical history.
The data generated will be used by researchers to not only create PRS for participants but also assess the impact that disease risk information being returned to participants will have on the future health of the U.K. population.
Baker said its “pretty clear that at some point every newborn will get sequenced” because it will be inexpensive and the knowledge generated will ultimately reduce overall health care costs. “More questionable is when does that happen? Is that two years? 10 years? 20 years? You'll have different people argue different things,” he added.
So, is clinical NGS on the cusp of a revolution? Not quite yet—but watch this space.