Controversy over incidental findings in genetic testing
Clinicians ordering whole genome sequencing for a patient with cancer or another indication for sequencing are likely to generate incidental findings that were not being looked for, such as a cancer geneticist finding a gene variant associated with heart disease.
"Targeted genetic testing that focuses on one or more specific genetic variants will probably not generate much in the way of incidental findings," says Karen J. Maschke, PhD, a research scholar at The Hastings Center in Garrison, NY. It is when many genetic variants are being interrogated, such as with whole genome sequencing and whole exome sequencing, that incidental findings are more likely, perhaps occurring in 1% to 3% of patients.1
In April 2013, the American College of Medical Genetics and Genomics (ACMG) recommended that clinical labs doing exomic or genomic analysis be required to analyze 56 genes that increase the likelihood of diseases for which there is an intervention, with the results reported back to the ordering clinician, regardless of the specific disease the clinician intended to investigate when he or she ordered the test, and regardless of patient preference or age.2
"This is a huge controversy, and it crosses several domains," says Susan M. Wolf, JD, McKnight Presidential Professor of Law, Medicine & Public Policy and faculty member at the Center for Bioethics at University of Minnesota in Minneapolis. There are some limited indications in which sequencing is already moving into clinical application, notes Wolf. At times, individual patients are sequenced for a primary indication, in order to help clinicians determine a treatment strategy for their cancer, for example.
In 2012, the ACMG recommended that individual patients must be allowed to opt out of additional analysis to preserve patient autonomy.3 "We know that genetic information can pack a wallop psychologically, and also can have significant social consequences," Wolf says. "To the surprise of a great many, the 2013 statement urged that whenever an individual is sequenced for any primary indication, [that the additional analysis] be done without the patient’s specific consent." If the individual doesn’t want the extra analyses, she explains, then the person’s only choice is to decline the sequencing altogether, even if they need it medically.
Consensus has stood since at least 1995 that children should not undergo genetic testing for adult-onset disorders, adds Wolf, but should decide for themselves when they reach adulthood. "The reaction to the [ACMG’s] rejection of autonomy and the long history of choice in genetics was explosive, and even more so because they said it applied to kids," she says. "What they said was that even if we are talking about variants for adult-onset disorders in kids, the 56 extra genes should be analyzed whenever clinical sequencing is done in children."
While the ACMG recommendations do not extend to research participants, the current controversy over incidental findings in clinical sequencing grows out of a longstanding debate over how to handle incidental findings in research. "There has been great debate over which research results are eligible for return, how to offer the results back to people, and how to address the possibility of incidental findings when informed consent is obtained at the beginning of the research process," says Wolf.
Wolf has served as principal investigator of projects funded by the National Institutes of Health (NIH) beginning in 2005, which has issued consensus recommendations about when researchers should offer genetic and genomic research results and incidental findings to research participants.4,5
"There is a tremendous amount of research going on now to answer all of these questions," says Wolf. "NIH has really stepped up to the plate, and has formed a consortium of investigators to speed progress."
Actionable or not?
The two primary ethical considerations are whether the incidental findings are "actionable" to allow a better preventative or treatment outcome, and whether the patient wants the information, according to Gail Jarvik, MD, PhD, professor of medicine and genome sciences at University of Washington in Seattle. "Maybe searching is reasonable if the patients want the information. If they do not, then it is not," she says.
Jarvik says that if the lab is aware of an actionable finding and the patient wants the information, it should be shared; but if it is not actionable, sharing the information may have more negatives than positives. Notably, the ACMG recommendations extend only to "actionable" genes.
"Labs searching for actionable incidental findings is different than what we do in other parts of medicine. This is presented by some as a simple thing that falls out of a genome, but it uses extra resources," says Jarvik. "It is more akin to adding a glucose test to every blood sample, than noticing an extra spot on an X-ray."
Of the highly contentious issue of whether adult-onset findings in children should be returned to parents, Jarvik says this is reasonable if the parents want the information. "It may appear to differ from our general policy of not testing children for adult-onset disease, but is very different, in that the genetic variant would not be known in the family," she explains. "Thus, its detection may benefit the transmitting parent."
Both in research and in clinical care, there is debate about how much information should be offered back to individuals. Some argue that only information with clinical significance should be given to research participants and patients who are not in research. "Others argue that patients and research participants have a right to get access to their full set of genomic information," says Maschke.
Maschke says these are the central ethical considerations:
- individuals’ autonomy to decide for themselves what they want;
- researchers’ obligations to inform research participants to satisfy the principle of individual autonomy/decision-making;
- the potential harm of returning information to research participants and patients that has unknown clinical significance. For instance, an individual might obtain risk information about genetic variants associated with heart disease and misinterpret the information in a way that leads her to engage in lifestyle behaviors that are harmful to her health.
"Yet, survey data from some research participants and patients suggest that many people want broad access to their genomic information when their biospecimens are examined in the research context, even if there is not evidence of its clinical significance," says Maschke.
There are still many significant questions about how to integrate the sequencing of the entire genome into clinical care. "That’s why this debate is relevant to ethicists who consult with clinicians," says Wolf. "Also, the whole domain of exome and genome sequencing, right now, straddles the clinical and research worlds."
Most major medical institutions that are conducting studies using whole genome sequencing have some type of bioethics program at their institution. "So there is ongoing education and discussion at these institutions about the ethical issues involved with incidental genomic information," says Maschke.
In addition, many medical institutions are developing programs to offer whole genome and whole exome sequencing to selected patients. "These institutions are working with genetic specialists, clinicians, and bioethics experts at their institutions to develop appropriate educational materials about the issues surrounding genomic incidental findings," says Maschke.
The debate over incidental findings is "enormously important" for ethicists to understand, says Wolf, adding that any protocol that is submitted involving human subjects research needs to consider incidental findings. "Learning how to work with researchers and clinicians to ethically manage incidental findings is essential, and a big challenge," she says. "This has become a crucial part of the ethicist’s knowledge base and tool kit."
- Dorschner MO, Amendola LM, Turner EH. Actionable, pathogenic incidental findings in 1,000 participants’ exomes.
Am J Hum Genet 2013;93(4):631-640.
- Green RC, Berg JS, Grody WW, et al. ACMG recommendations for reporting of incidental findings in clinical exome and genome sequencing. Genet Med 2013;15:565574.
- Board of Directors, American College of Medical Genetics and Genomics. Points to consider in the clinical application of genomic sequencing (2012). Available at: http://www.acmg.net/StaticContent/PPG/Clinical_Application_of_Genomic_Sequencing.pdf.
- Wolf SM, Lawrenz FP, Nelson CA, et al. Managing incidental findings in human subjects research: Analysis and recommendations. J Law Med Ethics 2008;36:219-248.
- Wolf SM, Crock BN, Van Ness B, et al. Managing research results and incidental findings in genomic research involving biobanks and archived datasets. Genet Med 2012;14(4):361-384.
- Gail Jarvik, MD, PhD, Professor of Medicine and Genome Sciences, University of Washington, Seattle. Phone: (206) 221-3974. E-mail: firstname.lastname@example.org.
- Karen J. Maschke, PhD, Research Scholar, The Hastings Center, Garrison, NY. Phone: (845) 424-4040, ext. 223. E-mail: email@example.com.
- Susan M. Wolf, JD, McKnight Presidential Professor of Law, Medicine & Public Policy, University of Minnesota, Minneapolis. Phone: (612) 625-3406. E-mail: firstname.lastname@example.org.