Whole-Genome Sequencing Solving Medical Mysteries

Ron Zimmerman

March 17, 2014

SAN DIEGO — Whole-genome sequencing can determine the cause of undiagnosed diseases and can be done on a small scale, Howard Jacob, PhD, from the Medical College of Wisconsin in Milwaukee, said here at the Future of Genomic Medicine VII.

His hospital champions the use of whole-genome sequencing, not merely exome sequencing. And some patients owe their lives to it.

A recent study showed that whole-genome sequencing conducted on 2 different machines matches up only one-third of the time (JAMA. 2014;311:1035-1045). But Dr. Jacob said he'll take those odds if that's the only way to diagnose an otherwise fatal disease.

He described the case of Nicholas Volker, a poster child for whole-genome sequencing. "At age 4, Nic was in the process of dying," Dr. Jacobs told Medscape Medical News.

Volker's disease remained undiagnosed until Dr. Jacob and his team developed analytic tools to evaluate approximately 18,000 genetic variants. "We came up with a diagnosis of an XIAP mutation that was likely pathogenic. No one had ever seen this before," Dr. Jacob explained.

"The defect in the XIAP protein caused an immune dysregulation syndrome resulting in early-onset, rapidly progressing, and catastrophic inflammatory bowel disease," he said.

We need to be looking across the genome or there are a lot of things we'll miss.

"We ended up doing a core blood transplant; 42 days later he was eating. He had been on total parenteral nutrition for 9 months; after treatment, he enjoyed steak. Nic is now 9. It's coming up on 4 years since his transplant and his gut disease completely resolved. It's a great story."

In fact, the Milwaukee Journal Sentinel won the Pulitzer Prize in 2011 for its series on this case, entitled One in a Billion: A Boy's Life, a Medical Mystery.

Not only is Volker's story a great example of the power of genomic sequencing, but Dr. Jacob's clinic has been praised as an outstanding example of what can be done in genomic medicine, even on a small scale. "We've built an entire genetic testing clinic around 3 geneticists and 2 genetic counselors. We do both exomes and whole genomes," Dr. Jacob explained.

However, he makes an impassioned argument for doing whole-genome sequencing rather than the simpler whole-exome sequencing. For some phenotypes, Dr. Jacob pointed out, whole-exome sequencing works just fine, but for others, it isn't a good test because it will not reveal causal variants.

"Do we really need whole-genome sequencing when all the information is in the exomes?" he asked.

Whole-exome sequencing is cheaper. "From a research perspective, it makes sense," he said. "But from the rare disease perspective, there is not a lot of advantage to exome sequencing, because there's a bias in it toward poor coverage. We need to be looking across the genome or there are a lot of things we'll miss."

Dr. Jacob described a recent study of 100 patients in which whole-genome sequencing was more accurate. "With the exome, there are 2700 actionable variations, but it's still missing 212 actionable variants. In your exome reports, it's often not reported what variants they didn't cover. With whole-genome sequencing, we only missed 1 variant that is unique to the whole genome and we only missed 3 in common. Whole-genome sequencing is just fundamentally more accurate — even at 100-times coverage."

To manage their data and phenotypes and variants from other databases, the team at the Wisconsin clinic developed the ClinMiner data integration tool. The tool helps to accurately place variants into 1 of 5 reporting categories.

Table. Reporting Categories in the ClinMiner Data Integration Tool

Category Characteristics
Pathogenic Previously reported variants that are a recognized cause of the disorder
Likely pathogenic Previously unreported variants that are of the type expected to cause the disorder
Variants of uncertain significance Previously unreported variants that are of the type that might be causative of the disorder
Likely benign Previously unreported variants that are probably not causative of the disorder
Benign Previously reported variants that are recognized as neutral

The variants of uncertain significance "are the ones we all talk about," said Dr. Jacob. Benign or likely benign variants are rarely reported. "But just because it hasn't been reported as pathogenic doesn't make it benign. These are the ultimate negative data. We have to know what genes don't do. All we report is what they do."

He presented the case of a 3-year-old girl with delayed development and seizures. Many standard tests were performed over a period of 6 months, but the cause remained unclear. Finally, genetic testing revealed an intracellular cobalamin metabolism disorder. Treatment halted progression of the disease, but the existing neurologic abnormalities were irreversible and permanent.

Dr. Jacob said he believes that whole-genome sequencing would have found the mutation earlier, and earlier treatment would have resulted in normal neurologic development instead of lifelong mental impairment.

Genomics at a Community Hospital

Dr. Jacob's clinic services a small 200-bed hospital and a mid-sized 400-bed hospital. Genomic medicine can be practiced successfully at a community hospital level, even when dealing with rare diseases, he explained.

In fact, "23% of all pediatric admissions to hospital are for 'rare' disorders," Dr. Jacob reported. "And 20 million Americans suffer from so-called rare diseases. We need to come up with a new name. We can't be calling something that is this common rare."

Dr. Jacob said he hopes his clinic's success will help other medical groups initiate their own programs. Since the Volker case, his team has been successful in diagnosing about 27% of their cases. "We're making the diagnosis 25% more with whole-genome sequencing than with whole-exome sequencing. We believe this will change the way you practice medicine."

When the Medical College of Wisconsin sequenced Nicholas Volker, it took 12 physicians and scientists 6 months to interpret the data manually, noted Ali Torkamani, PhD, director of genome informatics and drug discovery at the Scripps Translational Science Institute in La Jolla, California.

We believe this will change the way you practice medicine.

Today, in the Scripps Idiopathic Diseases of Man project, which uses whole-genome sequencing, a debilitating neuromuscular disease takes 1 scientist less than 1 day to find a causational pathogenic mutation. Dr. Torkamani said he thinks that by 2017, driven by the decrease in sequencing cost, clinical diagnoses will be even faster, and will be done at the point of care by a single physician.

"We want to make these techniques available to everyone," Dr. Torkamani said. "We should think about applying them to everyone with a rare disease. Because rare is not rare; the cumulative total is 5% to 10% of all diseases, which means there are more people living today with rare diseases than are living with cancer."

Currently, the highest volume of exome sequencing is "in developmental delays, autism, neurologic disorders," said Heidi Rehm, PhD, associate professor of pathology at Harvard Medical School in Boston.

"That is a very complex spectrum of phenotypes that is difficult to tease out and know specifically what to look at," she said. In such cases, it first has to be determined whether there is a fundamental basis for a developmental disorder.

For example, you can largely diagnose Marfan syndrome on the basis of clinical features, so the utility of genetic testing is debatable, she explained. However, "when you're talking about the spectrum of neurologic disorders, it's very difficult to sort out what's going on. That's an area where genetics can help physicians better see their treatment options and eventual outcomes."

Dr. Jacob said he recommends the immediate adoption of whole-genome sequencing.

And the cost? "With these rare diseases, the cost of whole-genome sequencing is insignificant compared with hospitalization and treatment costs. Think of it that way," he said.

Dr. Jacob and Dr. Rehm have disclosed no relevant financial relationships. Dr. Torkamani is a founder and stockholder in Cypher Genomics.

Future of Genomic Medicine (FoGM) VII. Presented March 7, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: