Skip to main content
Premium Trial:

Request an Annual Quote

UNC Team to Use NHGRI Grant for Study of Exome Sequencing in Clinical Care


By Molika Ashford

This is the fourth in a series of profiles of centers awarded grants under the NHGRI's new Clinical Sequencing Exploratory Research Project program. We previously profiled centers at the University of Washington, Brigham and Women's Hospital, and the Children's Hospital of Philadelphia.

The University of North Carolina will embark next month on a National Institutes of Health-funded study exploring how best to use genome sequencing in the practice of medicine.

The study — dubbed "North Carolina Clinical Genomic Evaluation by NextGen Exome Sequencing" — will enroll at least 750 subjects from several patient groups including cancer sufferers, children with dysmorphic birth defects, and adults and children with neurodegenerative diseases.

The UNC team plans to sequence patients' whole exomes and randomize the study participants to two groups — one that will be able to choose to receive a range of incidental findings and one that will only receive medically actionable findings from their sequence data. The researchers hope to evaluate how well sequencing performs as a diagnostic tool, as well as the impact of both diagnostic and incidental information on patients and families.

In December, the UNC project received a four-year, $6.4 million grant for the project from the National Human Genome Research Institute under its new Clinical Sequencing Exploratory Research Project program. Four other centers — the Children's Hospital of Philadelphia, Baylor College of Medicine, Brigham and Women’s Hospital, and the University of Washington — also received funding under the NHGRI program (CSN 12/6/2011).

The UNC team expects to receive approval from its institutional review board in the next week and the researchers plan to begin enrolling patients in the NCGENES effort starting in February, according to James Evans, a genetics professor at the UNC School of Medicine and the principal investigator of the project.

Evans told Clinical Sequencing News that the UNC group has four major goals for the effort. First, they hope to assess the diagnostic potential of clinical sequencing in the patients they plan to sequence.

"These are people who have not been able to be diagnosed by standard techniques," he said. "Yet they have some indication in their history that a primarily genetic ideology is at work, so they are promising for this kind of a study."

The group plans to roughly divide its 750 total subjects between individuals with cancer that either develops at a young age or those with a strong family history of cancer, children with developmental disorders, patients with cardiac abnormalities, and both children and adults with neurodegenerative diseases.

"We are [hoping to] be begin to get some handle on the types of patient characteristics that translate into a promising result by applying this technology," Evans said.

Evans said UNC will be sequencing patients' DNA in house at the university's High Throughput Sequencing Facility using Illumina machines. Any variants the group plans to report to patients will also be independently verified in a CLIA-approved lab, he added. "We won't report anything that hasn't been confirmed."

According to Evans, the group chose whole-exome sequencing mostly due to its lower cost. "It remains significantly cheaper to do whole-exome sequencing and we still know so little about so much of the genome that I don't think we are losing any appreciable clinical information," he said.

"We decided in the end that [whole-genome information] would give us little enough added value that we'd be far better off [being] able to sequence more people with whole-exome [sequencing], than sequencing fewer whole genomes."

Another overarching goal for the study will be to design and optimize the best strategy for sorting through sequencing data in a way that allows information to be used in the care of patients.

"You do whole-exome sequencing and you get hundreds of thousands of variants," Evans said. "As we go forward with this, it's utterly unrealistic to think someone can sit in a lab and go through all this if you are going to integrate this technology into medical care."

The UNC team is creating an algorithm to "thin" a whole exome into "clinically relevant categories," Evans said.

"By allowing a machine to take a first pass through the genome and do the bulk of the analysis, it distills those findings that are most likely to have actual clinical significance and leaves the human interpretation for the most critical sliver of clinically important data."

As part of a third main focus, the researchers plan to randomize the cohort into two groups, which will experience different levels of access to their genomic information. The group plans to evaluate how two different strategies for sharing incidental data affect patients. One half of the study subjects will be randomized to receive only clinically actionable incidental findings and nothing else. The other half will be able to choose what types of information they want to see.

"The amount of counseling and education that goes with the release of those results will be calibrated to the potential for [them] to cause problems," Evans explained. "If someone is randomized to the choice group and says they want their pharmacogenomic profile, their cytochrome P450 genes, that’s low-risk stuff and we'll give that to them without overtaxing educational efforts."

"But if someone says they want to know if [they] have Huntington's, we certainly feel like it’s a patient's right, but we have an added responsibility to do more counseling," he said.

While patients randomized to the no-choice group will not be able to see any non-actionable incidental results during the course of the study, after the study is complete they will be able to go back and access their full genomic data.

Evans said the researchers will follow up with patients using questionnaires to measure what kind of information people tend to want, how they use it, and how they react to it. "We're still in very uncharted territory here so we're hoping to generate the kind of data that will really allow the field to know, as we go forward, what people really want," he said.

Finally, the group considers it a fourth main goal to fully represent UNC's diverse patient population, according to Evans. "An important component of our effort is to include African American, Hispanic, and Native American [subjects] who make up a decent component of our patient population [at UNC]," he said.

"One of the problems with genetics in particular and any new advances is that sometimes access to those things is quite stratified. So given the genetic differences between different ancestral populations we're excited about applying this in a very broad way."

Like the other projects funded by the NHGRI, the UNC effort is divided in a three-part structure.

Evans and Jonathan Berg, a UNC geneticist, head the clinical project and are responsible for the overall study and patient selection.

Karen Weck, a UNC professor of pathology, and Kirk Wilhelmsen, a professor of genetics and neurology, lead the informatics arm of the effort and will be involved in performing the sequencing and designing the informatics process, Evans said.

Gail Henderson, chair of the department of social medicine, will head the section of the project dealing with return of incidental findings.

While the NHGRI funding is for four years, Evans said his hope is for the project to be a foundation for a much longer study. "There is enough in the human genome to keep us busy for a long time," he said.

"My fervent hope is that we'll be able to secure funding beyond [four years] because we're still just at the early stages of understanding how best to use this information in the clinic."

Evans said the project could also switch to whole-genome sequencing within or after the period of the NHGRI grant, depending on how quickly WGS costs continue to fall.

Have topics you'd like to see covered in Clinical Sequencing News? Contact the editor at mashford [at] genomeweb [.] com.