By Monica Heger
Incidental findings uncovered in whole-genome and whole-exome sequencing studies have posed a contentious issue in the move to bring next-generation sequencing into the clinic, as the community has debated when and whether to return such results to patients' physicians.
In the absence of consensus on the issue, academic labs and companies that have begun offering clinical whole-genome and whole-exome sequencing tests have so far come up with their own standards and protocols for returning incidental findings.
Now, the results of a survey published this month in Genetics in Medicine indicate that physicians generally agree that those findings should be returned to adult patients, especially when medical intervention is possible. They differ, however, in their opinions on whether the findings should be returned to pediatric patients, and on findings for which medical intervention is not possible.
The study authors hope it will lay the groundwork for establishing uniform guidelines on the return of incidental findings.
"This is the start of a discussion," Howard Jacob, director of the human and molecular genetics center at the Medical College of Wisconsin, told Clinical Sequencing News. "As more labs get involved with this type of work and as more physicians order [clinical sequencing tests] and there's feedback from the patients who are getting the data back, we're going to be able to make some policy around those experiences," he said.
Jacob co-led the study with Robert Green, associate director of the Partners Center for Personalized Genetic Medicine at Brigham and Women's Hospital and Harvard Medical School. The researchers surveyed 16 genetics specialists on their recommendations for returning genetic information from clinical whole-genome sequencing to the patients' physicians.
The survey, which covered 99 genetic conditions and disease predisposition genes, questioned specialists on whether they would return known pathogenic mutations, mutations that were likely to be pathogenic, and variants to known disease genes but with unknown consequences.
The specialists were asked to assume that the sequencing was accurate, that no family history was available for the patient, and that the patient had no clinical features consistent with the incidental finding.
Around 80 percent of the specialists agreed on the return of known pathogenic mutations for 64 diseases and all of them agreed on the return of results for 21 conditions such as cancer risk and other diseases where there is the potential for medical intervention.
There was less concordance, however, about returning results about children or results for which there was no possible medical intervention. For children, the 16 specialists agreed on four conditions for which known pathogenic mutations should be disclosed if found incidentally.
Additionally, concordance diverged as results became less certain — such as in the case of missense variants in disease genes that were computationally predicted to be deleterious.
Jim Evans, who directs clinical cancer genetics services at the University of North Carolina and who was one of the specialists surveyed in the study, told CSN that the results indicate it should be possible to build a consensus on dealing with incidental findings.
The study was a "first pass" to see what the thinking was among experts. "There was no effort to get together and decide" prior to the survey, he said, suggesting that it will be possible to reach a consensus on what types of incidental findings to return.
He said the next steps would be to convene a group of stakeholders to come up with a list of genes and variants that should be returned to the ordering physician.
Children vs. Adults
Such a list of reported genes may end up differing between children and adults. The experts that were surveyed showed more disagreement over what to return to the ordering physician when the patient was a child.
For instance, while all the participants agreed that known variants conferring risk for hereditary breast and ovarian cancer should be returned to adults, only 75 percent thought that information should be shared with children.
There was even more divergence when the disease was something for which there is no medical intervention, such as Huntington's disease or Alzheimer's.
Around 60 percent of the specialists said that known mutations in the genes for Huntington's disease should be returned to adults, but only 31 percent thought those findings should be returned to children. Half thought that mutations to APOE, which suggests a predisposition to Alzheimer's, should be returned to adults, while 25 percent said that the finding should be returned to children.
"As an adult, you have the right to decide what you want," said Jacob. But, when dealing with pediatric patients, the issue is trickier, because the physician has to "balance the child's rights" with the parents' rights.
"As a parent, you have the right to decide a lot of things about your kids," Jacob said. "But, in this case, you're also talking about having data that could impact them as an adult, and you're making the decision for them."
He said he wasn't sure whether there would end up being separate guidelines for returning incidental findings for adults and children. "There needs to be careful thought about this. I can argue both directions," he said.
Currently, within the MCW and Children's Hospital of Wisconsin clinical whole-genome sequencing program, parents have a range of options and can elect to receive no incidental findings, non-actionable incidental findings about childhood diseases, actionable findings related to adult-onset diseases, or non-actionable findings related to adult-onset diseases.
At the other end of the spectrum is the approach adopted by the University of California, Los Angeles, which recently launched a clinical exome sequencing test. The UCLA team does not plan to return any incidental findings that are not relevant for the patient's disease (CSN 3/7/2012).
Moving forward, Evans said it would be necessary to get the opinions of all stakeholders, including patients themselves.
Then, "there should be a transparent process for feedback," he said. Additionally the list "should not be written in stone" and could be revised periodically.
Additionally, he thinks an organization like the American College of Medical Genetics could spearhead the creation of guidelines and best practices around the issue.
"The community needs a generally respected and reputable organization to coordinate the arrival at a list of genes and the types of variants in those genes that would be reported," he said.
Aside from the ACMG, Jacob said that the College of American Pathologists, the US Food and Drug Administration, as well as patient advocacy groups should also be involved.
"There needs to be dialog between all of them," Jacob said. They each have "different expertise and different views."
"Fundamentally, it's still going to come down to making sure we practice medicine appropriately," he said.
Have topics you'd like to see covered by Clinical Sequencing News? Contact the editor at mheger [at] genomeweb [.] com.