Caitlin Smith has a B.A. in biology from Reed College, a Ph.D. in neuroscience from Yale University, and completed postdoctoral work at the Vollum Institute.
Clinical scientists are on the cusp of a power surge. Their diagnostic toolboxes are swelling with the incorporation of so-called “next-generation” DNA sequencing (NGS). NGS, with its greater parallelization, lower price tags and higher throughput than its predecessor, Sanger sequencing, already is increasing clinicians’ diagnostic power, especially with genetic diseases. Medical fields as disparate as reproductive health, oncology, organ transplantation and infectious disease also are using NGS-based diagnostic tests. Here’s a look at how NGS diagnostics are driving new developments in medicine.
Advantages of NGS diagnostics
The advantages of using NGS to diagnose diseases, especially genetic disorders, are manifold. One NGS-based genetic test yields the same or more information as multiple Sanger-based tests, but faster and at a significantly lower price. “NGS can interrogate many targets within one test, up to all of the individual nucleotides in the genome,” says Daniel Grosu, chief medical officer at Illumina. “Using tagging [or ‘barcoding’] technology, NGS also allows users to multiplex samples from many patients in a single run.”
That’s good news for scientists and patients alike. According to Matt Ferber, assistant professor in laboratory medicine and pathology at the Mayo Clinic, “genes that were once thought to be too rarely involved in disease to justify the cost of clinical-laboratory test development can now very easily be added to the diagnostic test without significantly increasing the cost of the test to the patient.”
Another advantage of NGS is its ability to gather multiple types of diagnostically useful information. According to Grosu, NGS “can detect disease signatures based on different types of abnormalities in DNA sequence”—such as single-nucleotide polymorphisms, “indels,” copy-number variants and translocations—as well as RNA transcripts, microRNA and DNA methylation.
NGS platforms are available from Illumina, Life Technologies, Roche/454 and Pacific Biosciences. Mostly those are research instruments. But companies such as Illumina are developing NGS platforms for clinical labs, too. Many clinical labs use Illumina’s high-end HiSeq and benchtop MiSeq platforms after internal validation, for instance, and a new platform joined the family in July.
Illumina’s MiSeqDx, currently under review by the U.S. Food and Drug Administration (FDA), is a desktop NGS sequencer specifically designed for diagnostic testing. The company’s MiSeqDx Cystic Fibrosis System (currently available in Europe and pending FDA clearance in the United States) incorporates a clinical assay for the cystic-fibrosis gene CFTR and is the first NGS-based diagnostic system on the market.
As with any tool, leveraging the power of NGS diagnostics requires knowing when and how to use it. “Knowing when a single gene test is most appropriate, vs. a panel [of tests], vs. a whole genome [test] is the key to using the technology appropriately,” Ferber says. His group typically uses one of two approaches for genetic testing with NGS. If the patient’s symptoms suggest only a few specific genes, they use a panel of tests for those genes. In contrast, they use NGS-based exome-wide analysis when they suspect many genes could be involved—or perhaps an as-of-yet undiscovered gene.
Disadvantages of NGS diagnostics
Of course, as with any technology there also are downsides to NGS diagnostics. For instance, says Ferber, although the technology excels at identifying rare mutations, it is less adept at finding insertions and deletions that are more than 50 base pairs long. “It's important to know the technical limitations of the diagnostic test being used, as they all have their Achilles’ heel, and [to know] how that relates to the analytic sensitivity for each potential clinical application,” he says.
There also are significant informatics challenges in NGS, in terms of data storage, manipulation and interpretation. For instance, Sami Amr, assistant director of Harvard University’s Lab for Molecular Medicine, which is housed within the affiliated Partners Healthcare Center for Personalized Genetic Medicine (PCPGM), says the sheer volumes of genetic variations revealed by NGS methods are overwhelming currently, but that obviously will improve with time. “The major drawbacks [of NGS] are the analysis and interpretation bottlenecks due to the generation of large number of variants, which incur an impact on both cost and turnaround time,” says Amr, who is also director of the PCPGM research core. “This drawback can be addressed by increasing communication and data sharing between laboratories through public variant databases such as ClinVar.”
Another disadvantage of NGS diagnostics is researchers’ ignorance of the extent of normal genetic variation, Ferber says. That leads to confusion as to whether a given variant is likely to be important, information that can confuse genetic counselors, physicians and patients. “As we sequence more genomic real estate, the higher the likelihood is for encountering indeterminate variants that can lead to patient confusion,” Ferber says. In the short run, sequencing the genes of more individuals and sharing the information can help but, “each individual harbors so many unique variants that no matter how many genomes we have in our shared databases, we will always be faced with new and perplexing variants.”
New clinical developments
Despite these difficulties, advances in genetic testing are already directly impacting people’s lives. “The most striking example of the power of NGS to revolutionize an entire field of medicine has been the rapid adoption, since late 2011, of noninvasive prenatal testing for detection of fetal aneuploidies in high-risk pregnancies,” says Grosu.
Another high-profile development is a genetic test called STAT-Seq [1], from Stephen Kingsmore’s group at the Center for Pediatric Genomic Medicine at Children’s Mercy Hospital and Clinics. Kingsmore, the Center’s director, was recently named one of Medscape’s best physicians of the year. One reason is that the lives of children have already been saved by STAT-Seq. Before STAT-Seq, the families of infants born with genetic diseases waited 4 to 6 weeks for the chance to receive a diagnosis. Furthermore, with more than 3,500 genetic diseases (and tests available for only a few hundred), some families never knew what ailed their newborn, even after testing was complete. For the lucky ones that did ultimately receive a diagnosis, it’s possible that the genetic disease had already claimed the life of their newborn in the meantime.
This is where STAT-Seq comes in. It cuts the diagnosis time from weeks to two days using high-speed, whole-genome sequencing and an automated bioinformatic pipeline. In two days, there is a much better chance to save a newborn diagnosed with a treatable disease or to prepare the family if the diagnosed disease is untreatable. Underscoring the technique’s impact, Time magazine named STAT-Seq one of its Top 10 Medical Breakthroughs of 2012.
Another approach called whole-exome sequencing (which only sequences the genes that will become translated into proteins) offers a powerful alternative for identifying the genes underlying rare genetic disorders. Exome sequencing is especially valuable when a patient’s symptoms suggest numerous possible genetic causes. In this case, it can be easier, faster and less expensive to sequence the entire exome at once rather than design panel after panel of tests for possibly irrelevant diseases. The UCLA Clinical Genomics Center offers whole-exome sequencing, which it runs on Illumina’s HiSeq 2000 (after using either Illumina's TruSeq exome capture kit or Agilent Technologies's SureSelect kit).
Last year, a group of researchers led by Eric Vilain, a professor of human genetics at the David Geffen School of Medicine at UCLA, used whole-exome sequencing to discover that two different mutations in one gene cause two distinct, but related, genetic diseases. One mutation causes IMAGe syndrome, in which stunted growth leads to small, underdeveloped infants. The other mutation causes Beckwith-Wiedemann syndrome, which is characterized by very rapid growth and unusually large children. In publishing their findings, Vilain and colleagues suggested that the two mutations appear to manifest as opposite functions of the same gene [2]. Other groups offering whole-exome sequencing include the Medical College of Wisconsin, Washington University’s School of Medicine and the company Ambry Genetics.
Kingsmore’s group recently developed a new test called TaGSCAN (Targeted Gene Sequencing and Custom Analysis). “It is an NGS test for 768 childhood genetic diseases,” he explains, including many that are otherwise unavailable in the United States. TaGSCAN works by zeroing in on potentially relevant regions of the genome according to the symptoms presented by the patient.
But there’s a good chance researchers will soon develop additional tests to complement or extend tools like TaGSCAN. “NGS diagnostics is still a very young field,” says Kingsmore. And a dynamic one, too. As clinicians sequence more and more individuals, once indeterminate results should increasingly come into focus. Says Ferber, “The genetics report of the future is a living document, not a static and concrete book of record.”
References
[1] Saunders, CJ, Miller, NA, Soden, SE, Dinwiddie, DL, Noll, A, Alnadi, NA, Andraws, N, Patterson, ML, Krivohlavek, LA, Fellis, J, Humphray, S, Saffrey, P, Kingsbury, Z, Weir, JC, Betley, J, Grocock, RJ, Margulies, EH, Farrow, EG, Artman, M, Safina, NP, Petrikin, JE, Hall, KP, Kingsmore, SF, “Rapid whole-genome sequencing for genetic disease diagnosis in neonatal intensive care units,” Sci Transl Med, 4(154):154ra135, 2012. [PubMed]
[2] Arboleda, VA, Lee, H, Parnaik, R, Fleming, A, Banerjee, A, Ferraz-de-Souza, B, Délot, EC, Rodriguez-Fernandez, IA, Braslavsky, D, Bergadá, I, Dell'Angelica, EC, Nelson, SF, Martinez-Agosto, JA, Achermann, JC, Vilain, E, “Mutations in the PCNA-binding domain of CDKN1C cause IMAGe syndrome,” Nat Genet, 44(7):788-792, 2012. [PubMed]
Image: The Ion 318™ Chip for Life Technologies' Ion Personal Genome Machine (PGM).