By Monica Heger
This article was originally published March 21.
Clinical sequencing is certain to impact diagnostics, therapeutics, and patient care, according to discussions at a recent conference, though opinions vary as to the specific short- and long-term implications of the approach.
Researchers, clinicians, industry leaders, and payors who took part in last week's X-Gen Congress meeting in San Diego all had slightly different thoughts as to the immediate and future applications of clinical sequencing, but nearly everyone agreed that it is poised to have a huge impact on health care, despite looming questions and hurdles.
"Next-gen sequencing is going to turn diagnostics completely upside down and on its head," Richard Resnick, CEO of GenomeQuest, said at a panel discussion on next-gen sequencing-based diagnostics during the conference.
"It's going to be pervasive and inform the future," agreed François Ferré, CEO of Althea Dx.
Many predicted that whole-genome sequencing will have the most immediate use in guiding cancer treatment and that other applications will follow. Targeted sequencing for disease diagnostics has already begun, with several companies and academic labs launching panels for diseases such as X-linked mental retardation, cardiomyopathies, and others, and most thought this trend would continue.
However, most seem to believe that applications such as whole-genome sequencing for disease risk prediction and diagnosing more complex diseases will lag farther behind due to the complexities of interpreting whole-genome sequencing as well as regulatory hurdles.
Sequencing's Sweet Spot
The advances of next-gen sequencing have been the most obvious in cancer research. While at last year's X-Gen Congress, researchers were still questioning the clinical utility of sequencing tumor genomes (IS 3/23/2010), this year most were in agreement that not only is the approach clinically useful but that next-gen sequencing will likely make its biggest initial impact in cancer diagnostics and in guiding treatment.
Several groups have already begun using sequencing to guide therapy. For example, researchers at the Mayo Clinic and the Translational Genomics Research Institute demonstrated last month that whole-genome sequencing of a tumor could be used to make treatment decisions (IS 2/22/2011).
"Cancer is definitely the sweet spot for this technology," said Gregory Heath, senior vice president and managing director of diagnostics at Illumina.
Heath said that at Illumina's CLIA lab, sequencing whole tumor genomes has been particularly useful in helping to diagnose rare cancer subtypes and determining the best course of treatment.
"Cancer genetics is no question the biggest and most popular way to go," said Mike Moradian, director of operations in Kaiser Permanente's genetics division. He suggested that it would be useful to use whole-genome sequencing to create an assay of all the known cancer oncogenes, tumor suppressor genes, and transcription factors.
Althea Dx's Ferré added that whole-genome sequencing of cancer tumors is starting to attract the attention of pharmaceutical companies. "Pharma finally realizes that cancer is a genetic disease," he said, adding that every pharmaceutical company now has a team focused on finding new biomarkers for cancer that can be used for diagnostics and therapeutics. Next-gen sequencing is "truly changing the way we're treating patients," he said.
Close on the heels of cancer are other applications, such as determining how viruses like HIV acquire drug resistance and in creating drug sensitivity profiles for patients, said experts at the meeting.
Illumina's Heath said that aside from cancer, next-gen sequencing would begin making an impact in the clinic in diagnosing infectious disease and determining viral resistance.
For fungal diseases, which can be difficult to culture, and other bacterial and viral diseases that may be hard to diagnose, sequencing will be particularly useful, he said. In those cases, he foresees sequencing being done in hospitals.
At a roundtable discussion on the matter, Tudor Constantin, a scientist at sequencing services firm Eureka Genomics, agreed with Heath, adding that the most common question he has received is concerning sequencing HIV to determine drug resistance and study mutational patterns.
[ pagebreak ]
Additionally, Kaiser's Moradian said that one immediate application of sequencing would be creating panels of cytochrome oxidases — germline mutations that would tell a clinician the types of drugs that are effective for a given patient. "It will be easy and accessible for physicians to interpret those," he said.
Targeted vs. Whole Genome
While whole-genome sequencing is important in cancer diagnostics and therapy, it still does not always make sense as a diagnostic for other types of diseases, Eliot Shearer, an MD/PhD candidate at the University of Iowa who is developing a sequencing-based test for deafness (CSN 3/17/2011), said at the panel discussion.
He said his team is taking a targeted sequencing approach because they have studied the genetics of hearing loss, have an expertise in the field, and feel comfortable interpreting sequence information from those genes. Sequencing a patient's whole genome to diagnose deafness would not make sense because the sequencing would inevitably turn up mutations of unknown significance as well as variants that are pathogenic for other diseases.
"That's why we're focused on a targeted test. …We're an expert on those genes and can develop a good curated database to handle that information," he said.
Targeted sequencing is "directly translatable to the patient. We're not interested in uncovering variants of unknown significance or variants that we don't know what to do with," he added.
Kaiser's Moradian agreed: "How to train personnel and how to translate results to clinicians so they can use them in the proper way" are major hurdles toward bringing sequencing to the clinic, he said.
Using whole-genome sequencing for diagnostics will also pose regulatory challenges, said Tony Marble, a senior scientist who assesses new technology at Abbott Molecular.
The US Food and Drug Administration is "not comfortable with reporting alternate indications," Marble said. So, if a clinician sequences a patient's genome to diagnose, for instance, deafness, but uncovers pathogenic mutations for a different disease, it is unclear if and how the clinician would be able to report that mutation under FDA guidelines.
A diagnostic test would have to demonstrate reproducibility and specificity, which may be difficult criteria to reach if using a whole-genome sequencing method. Nevertheless, the price of sequencing will soon reach the point where it makes more sense to do whole-genome sequencing rather than targeted sequencing, said Illumina's Heath.
Genomes in the Cloud
Further into the future is the idea that everyone will have their genome sequenced at birth, and clinicians will be able to do a disease-risk profile for patients as well as call up each patient's genome when making therapeutic decisions. Experts at the meeting did not want to pin a timeline on this scenario, though most agreed that it will not occur in the near term.
"Paying for whole-genome sequencing for every person to base diagnostics and therapy on, I don't see in the near future," said Moradian. However, while widespread adoption of next-gen sequencing in the clinic has not yet happened, it is fast approaching, and has the potential to change patient care, he added. "We will make savings by early diagnostics and cutting hospitalization time," he said.
Others noted that specific applications of sequencing would be further down the road, and that several scenarios could emerge.
Heath suggested that whole-genome sequencing could either become so cheap that it would be akin to being blood-typed and there would be no need to store genomic information, or instead of storing people's entire genomes, only relevant components of the genome would be stored in the cloud, such as pharmacogenomic components or individual variants.
"We're just seeing the tip of the iceberg right now," Heath said.
Have topics you'd like to see covered in Clinical Sequencing News? Contact the editor at mheger [at] genomeweb [.] com.