this article was originally published March 18.
Emory Genetics Laboratory is now offering a clinical exome sequencing test with enhanced coverage of around 4,600 disease-associated genes that the lab hopes will bump up the diagnostic yield of exome testing, currently hovering between 25 and 30 percent.
The design of the so-called medical exome has been a work in progress collaboration led by Emory Genetics Laboratory's Madhuri Hegde; Avni Santani, scientific director of the Molecular Genetics Laboratory at the Children's Hospital of Philadelphia; and Birgit Funke, director of clinical research and development at Harvard University's Laboratory of Molecular Medicine.
While clinical exome sequencing has had a number of high-profile successes in diagnosing children's unexplained disease and in some cases leading to dramatic clinical improvement in their conditions, the diagnostic rate has stalled at around 25 percent, in part because current exome capture kits do not cover the entire exome.
Off-the-shelf kits cover up to around 92 percent of the exome, but between 11 percent and 20 percent of the known clinically relevant exons are either poorly covered or not covered at all.
At the Future of Genomic Medicine conference in San Diego, Calif., earlier this month, a number of clinical laboratory directors spoke of the problem of exome sequencing missing important disease-related genes, including even in protein-coding regions. The Medical College of Wisconsin's Howard Jacob said that of 25 patients subjected to both exome and whole-genome sequencing, exome sequencing missed 212 actionable variants, while whole-genome sequencing missed three.
Due to these issues, EGL, CHOP, and LMM have had an ongoing collaboration to identify and curate a list of medically relevant genes, define precisely which of those exons were being missed by commercial capture kits and why, and develop approaches to target those areas.
EGL has now launched its version of the medical exome, dubbed the Medical EmExome, which it will offer as a diagnostic service to ordering physicians.
Hegde, executive director of EGL, told Clinical Sequencing News that the laboratory began offering the Medical EmExome internally in January and is now expanding the service broadly. In addition, it is reducing the price of clinical exome testing to $6,000 from $9,000. Trio sequencing, which Hegde said the lab prefers to do because it facilitates analysis, will cost $6,700.
In addition, she said, EGL has re-priced all 80-plus of its gene panels at $3,000, half the price of an exome test. The goal of the pricing system is to make it easier for physicians to order tests by having clear, uniform pricing, she said.
To design the medical exome, the three groups tested exome capture kits from various vendors, and ultimately decided to use Agilent's version 5 exome capture kit as a starting point. The groups had already identified the 4,600 genes they wanted to ensure coverage of, and they then tested the kit for coverage of those genes, Hegde said.
There are two main reasons for no or low coverage of specific exons, Hegde said. Either the kits have not been designed to capture those exons in the first place, or there is lost coverage due to sequence complexity even though the capture kit attempts to target them.
The first step was to "identify the exons without probes. Those were the easy ones," she said. "Then we identified the regions that still drop out, even with probes and boosted those regions."
Working in collaboration with Agilent, the three labs gave the firm feedback about their findings, and Agilent researchers then made adjustments to the probe design — changing the probe position, for instance, to try to cover the troublesome area.
According to Hegde, the final product still does not provide 100 percent coverage; however, the three labs and Agilent have an ongoing collaboration to continue improving on the design, with the goal of eventually getting to 100 percent coverage of all 4,600 genes.
"We know now which exons still don't work," she said. In general, they tend to be for pseudogenes, or areas that are extremely GC rich.
For those areas, Emory supplements coverage with spike-in probes from Integrated DNA Technologies, or by doing Sanger sequencing to fill in those areas.
Another unique facet of Emory's diagnostic exome, Hegde said, is that physicians can order a "boost" at no extra charge. The boost is essentially a targeted gene panel of a phenotypically defined region. For instance, said Hegde, in the case of a suspected neuromuscular disorder, a physician can order the Medical EmExome with boosted coverage of the neuromuscular-related genes.
The panels can be ordered separately and are run on Illumina's MiSeq system. The EmExome is run on the HiSeq 2500.
In a separate collaboration, EGL is working with Oxford Gene Technologies to design a deletion/duplication array to accompany the Medical EmExome. The array will assay the same 4,600 genes for copy number variants and Hegde said she expects that to be finalized by the end of April.
One library, many tests
EGL has also switched from using RainDance's target enrichment technology for its 70 different gene panels to IDT Lockdown probes, Hegde said, because the IDT probes provided the lab with more flexibility, in terms of adding probes, and spiking in coverage on specific exons with exome sequencing.
In an abstract that Hegde will present at the upcoming American College of Medical Genetics and Genomics meeting in Nashville, Tenn., she detailed how the laboratory was able to design one library using IDT Lockdown probes that targeted all exons in HGMD listed genes from which it could pull out specific gene panel tests.
The single library approach decreased cost compared to individual panels by 30 percent and decreased technologist hands-on time by 20 percent, according to the abstract. Exon coverage of each individual panel increased to around 99 percent from 95 percent, with all known variants detected at an average coverage of 75x.
Hegde anticipates that gene panels will remain popular. Panel tests tend to have a higher diagnostic rate, she said, both because coverage is deeper and more complete and also because the patient usually has a more specific phenotype. Additionally, each panel is complemented with a gene array to detect deletions and duplications at no extra cost, in order to ensure coverage of the entire mutational spectrum.
EGL's neuromuscular panel has a diagnostic rate of 63 percent, and the eye disorders panel has a diagnostic rate of 71 percent, Hegde said. "The more specific your phenotype," she said, "then the clinical yield from the gene panels is higher."
The goal with the Medical EmExome is to increase the diagnostic rate of exome testing to the levels of the gene panels.