By Monica Heger
A group led by Joris Veltman at the Radboud University Nijmegen Medical Center in the Netherlands is conducting a 500-person pilot program designed to demonstrate that reimbursing for exome sequencing is more cost-effective than single-gene tests, particularly in cases where the diagnosis is uncertain.
The study is based on a diagnostic exome sequencing test that the group began offering last month for a range of disorders, including deafness, blindness, movement disorders, mitochondrial disorders, and intellectual disability — diseases where the team has "expertise," Veltman, an associate professor of genomic disorders at the Nijmegen Centre for Molecular Life Sciences, told Clinical Sequencing News last week at the combined International Congress of Human Genetics/American Society of Human Genetics meeting in Montreal.
Veltman said that the major Dutch insurance companies are currently reimbursing the test for €1,500 ($2,071) per patient as part of the pilot project.
While the reimbursement does not cover all costs for the sequencing-based test — which include consumables, equipment, data analysis, personnel, and counseling — it is twice the amount typically reimbursed in the Netherlands for genetic tests, said Veltman.
The remainder of the costs, which Veltman said vary between patients, are covered by funding that the laboratory receives from the university and other grants.
The 500-person pilot should be completed by the end of the year, and then the team will renegotiate with insurance companies about reimbursement.
"We are front runners for this," said Veltman, and therefore more willing to accept the initial added costs to help drive sequencing into the clinic. He added that it's clear that next-generation sequencing "should go into the clinic, but carefully."
The group is conducting the test in an accredited laboratory with a diagnostic license, using Life Technologies' SOLiD 5500 with Agilent's SureSelect enrichment technology. Turnaround time is currently three to four months, but Veltman aims to reduce that to four weeks, putting it on par with its Sanger sequencing-based tests.
Patients are referred for testing by a physician and a clinical geneticist. The goal is to "select cases where exome sequencing will be beneficial," said Veltman.
In cases where the physician is trying to determine the cause of intellectual disability in a child, the team will sequence both the patient and the parents, said Veltman, in order to also look for de novo mutations. For the other diseases, only the patient is sequenced.
Following the sequencing, a committee that includes physicians with expertise in the disease area, ethicists, and legal experts meets to decide on which results should be returned. Veltman said that the team returns only results with clinical utility that have been validated by Sanger sequencing.
"We don't want to increase uncertainty," said Veltman.
Patients all receive counseling before and after the test. Prior to receiving the test they are counseled on the possibility that variants unrelated to their disease may be found, Veltman said. Patients cannot opt out of receiving unrelated results, but Veltman said if a significant number of patients decide not to have the test because of that requirement, the team may change its policy.
Additionally, in the Netherlands, minors cannot be informed of a genetic predisposition to adult disease, so if variants related to risk of early-onset Alzheimer's disease or cancer risk are discovered in a child, those results will not be returned, Veltman said. Once a minor turns 18, he or she would have to go through a consent process to receive the results of the test.
Veltman said the team is pretty conservative in the results it returns and typically returns just a handful of variants related to the patient's condition.
The Dutch group is not the first to offer clinical whole-exome sequencing services. Recently both 23andMe and Ambry Genetics launched exome sequencing services (CSN 10/5/2011), however neither service is for the diagnosis of specific diseases.
While Ambry's service will only be offered through a doctor's prescription and for patients with undiagnosed disease, 23andMe's service is geared toward consumers and does not offer interpretation.
Exome vs. Targeted
Veltman said the team decided to go with exome sequencing over a targeted approach for a variety of reasons. For one thing, he noted that exome sequencing allows for the test to be comprehensive and adaptable, so that all patients receive the same test, regardless of their condition, and as new variants are discovered they can easily be added to the interpretation.
For example, Veltman said that in his experience with doing targeted sequencing, "by the time you finish a study, new genes have been published [that are causative] and you don't have the flexibility to update your test."
Last year, Veltman's group published an exome sequencing study identifying de novo mutations as causative of intellectual disability (IS 11/16/2010). Then this year, a separate team, also using exome sequencing, identified an additional 50 genes causative of intellectual disability (CSN 9/28/2011). Had Veltman designed a targeted test based on his original findings, he would not be able to easily incorporate those new findings in the test, he said.
Also, the Radboud University researchers wanted to design a test for multiple diseases. Using a targeted approach, they would have had to design separate tests for each disease, as opposed to one test with different analytical frameworks.
Other universities and labs have launched disease-specific targeted sequencing tests — such as the University of Iowa's deafness test (CSN 3/22/2011) and Emory University's tests for congenital disorders (IS 5/25/2010) — in part because the interpretation of a panel of genes is thought to be easier than an entire exome or whole genome. Veltman argued, however, that in some cases interpreting an entire exome can actually be more straightforward.
When sequencing a handful of genes, it is not always clear whether a variant is disease-causing or not, he said. But looking at the entire exome makes it easier to rule out variants that are more likely to be benign, even if they are in genes suspected of being associated with the disease. When looking at a few genes, researchers have only the variants in those genes to focus on. Unless there is an obvious deleterious variant, determining whether a variant is disease causing can be difficult since researchers won't know what potentially deleterious variants are in other genes. But looking at the whole exome can make ruling out variants of unknown significance easier, if there is a more obvious candidate. Finding a "smoking gun" can be easier, he said.
During a panel discussion about the use of next-gen sequencing in medicine at the ICHG/ASHG meeting last week, a number of experts expressed concern about incidental findings that would inevitably turn up when doing whole-exome or whole-genome sequencing of patients.
Patients who are not educated about genetics or who do not have a family history of genetic disease could have a negative response to sequencing results, particularly results not related to their current disease, said Les Biesecker, a senior investigator at the National Human Genome Research Institute.
Biesecker heads the NHGRI's ClinSeq program, which enrolls both healthy and diseased patients for exome sequencing to genomically characterize both novel phenotypes and disease phenotypes. Patients' reactions to unexpected results "include utter disorientation and puzzlement," he said. "This is a challenge we are going to face on a regular basis."
"It's a risk factor," Veltman acknowledged, noting that the patients in his group's study "have all been counseled that this could happen."
He stressed that the study is not designed as a fishing expedition but rather to look for variants that are likely to cause specific diseases, and the analysis is all done by computer. For instance, in a patient with unexplained blindness, all the known and suspected blindness genes will be interrogated first, he said.
"Even if a person has a breast cancer risk mutation, we won't always find it," he said. Nevertheless, "it can happen," which is why the patient receives genetic counseling both before and after the test and why the team has established an external, independent committee to discuss these unrelated findings..
"It's important to do it carefully," he said.
Have topics you'd like to see covered by Clinical Sequencing News? Contact the editor at mheger [at] genomeweb [.] com.