Cancer Gene Sequencing Raises New Medical Ethics Issues

Janis C. Kelly

September 06, 2013

Cancer researchers are routinely analyzing hundreds of individual cancer genomes in the quest for better, more personalized treatment. As a byproduct of this work, new genetic sequencing methods are identifying medically important genomic variants not related to the cancer under study (the "incidentalome"). This has created an unanticipated set of problems for researchers, clinicians, and patients.

Researchers are wrestling with what to look for in genomic studies, how to handle the resulting information, and whether to contact patients for additional consent before genomic studies. Clinicians want to know more about what constitutes a "medically actionable" gene variant, how to respect patient autonomy without shirking ethical responsibility, and what to do about potential genetic discrimination and the threats to privacy posed by electronic medical records.

Patients may not be sure how much of this information they want to know and may not have access to adequate genetic counseling, which is not usually covered by insurance.

The Rise of the Incidentalome

Cancer genome studies are typically done by sequencing both a tumor cell and a normal cell from the same patient. Subtracting the normal (germline) sequence from the tumor cell sequence reveals the tumor genome.

"However, the process of decoding the genome of a patient's tumor may incidentally reveal information about inherited predispositions to cancer and other diseases.... There is a need to establish approaches to decision making with respect to the return of these incidental results," Yvonne Bombard, PhD, from the clinical genetics service at Memorial Sloan-Kettering Cancer Center in New York City, and colleagues write in a viewpoint essay published in the August 28 issue of JAMA.

The genomic information of concern involves mutations linked the risk for cancer or other diseases and mutations that can affect either the efficacy or toxicity of a treatment drug.

"The problem with genomics is that it is information. Once I sequence your genome, I can throw the test tube of your blood away and put it on a computer chip. Genomic information immediately becomes electronic information.... Just as people can hack into any system and take information, there are problems of privacy protecting," Robert Klitzman, MD, told Medscape Medical News." Dr. Klitzman, professor of clinical psychiatry and director of the masters of bioethics program at Columbia University in New York City, is coauthor of a viewpoint essay published in the July 24/31 issue of JAMA (2013;310:369-370) that addressed the implications of balancing secondary genomic findings with patient autonomy. In addition, he is the author of the book, Am I My Genes?: Confronting Fate and Family Secrets in the Age of Genetic Testing (Oxford University Press, 2012).

Researchers Puzzle Over What to Look For

Genomics researchers are debating whether to set limits on what they look for to avoid the ethical and logistical complications associated with having "too much information" about an individual patient. This represents something of a pushback against the 2013 American College of Medical Genetics and Genomics (ACMG) recommendation on incidental findings (Genet Med. 2013;15:565-574).

The ACMG recommended that laboratories "seek and report" 57 specific genetic variants in all clinical germline exome and genome sequencing, "including the 'normal' of tumor-normal subtractive analyses in all subjects, irrespective of age, but excluding fetal samples." Twenty-three of those mutations pertain to cancer risk. Other mutations are associated with Marfan syndrome, Loeys–Dietz syndrome, familial thoracic aortic aneurysms and dissections, hypertrophic or dilated cardiomyopathy, a number of heart rhythm defects, familial hypercholesterolemia, malignant hyperthermia susceptibility, and the vascular type of Ehlers–Danlos syndrome.

In their viewpoint essay, Dr. Bombard, Mark Robson, MD, and Kenneth Offit, MD, MPH, recommend that in retrospective studies, germline variants should not be explicitly defined at the time of tumor genetic analysis, as a way of striking a balance between research goals and protecting patient privacy.

Dr. Robson, also from the clinical genetics service at Memorial Sloan-Kettering Cancer Center, told Medscape Medical News that this could be a way to ethically deal with samples when the subjects' preferences regarding incidental results are not known and additional consent would be difficult or impossible to obtain.

However, Dr. Klitzman argues that avoiding troublesome information might not always be an ethical approach.

"If there are actionable findings, such as data indicating that the patient has a mutation for a serious illness for which there is definitive treatment, there's a serious question of whether the researcher has some ethical responsibility toward the patient," Dr. Klitzman explained. "One could argue that the researcher should at least try to contact the patient. It may not be possible to obtain additional consent, but there may also be people whom you can contact."

"Optimal interpretation of the cancer genome requires a comparison with the inherited genome, but it is possible to avoid explicit listing of inherited variants. This strategy is justifiable in retrospective genomic research, but is less supportable in the prospective setting," Dr. Bombard and colleagues write.

For prospective studies, they recommend discussing disclosure of incidental findings as part of the informed-consent discussion.

Who Knows? Who Wants to Know? And What Is There to Know?

Clinicians puzzle over what constitutes a "medically actionable" genetic variant, what and how to tell patients, and how to address genetic privacy issues such as the risk for genetic discrimination. Dr. Bombard and colleagues point out that although the ACMG position is that information about all "medically actionable variants" should be given to the patient, there is no consensus on what constitutes medically actionable.

The 2 main types of medically actionable gene variations are those associated with increased risk for cancer or other diseases and those with a pharmacogenomic impact that increases or decreases drug efficacy or adverse effects. Dr. Klitzman explained that more data are needed on the pathogenicity of rare variants and on prevalence. "We don't know the prevalence of those mutations in the healthy population; it may be much higher than we assume," he said.

"And what if there is information that is not medically actionable but is personally actionable? There's nothing you can do about Huntington's disease, but there are patients who have told me, 'I want to know. If I have it, I'm not going to save all my money up and raise kids. I'm going to move to Tahiti and be an artist,'" Dr. Klitzman said.

"We certainly need more empiric information regarding what kind of information patients really want to know and how best to inform their choices. Many people, at first, say they want to learn 'everything', but when one digs a bit further, that is often qualified. We need more research to address these questions," Dr. Robson said.

Steering Through a Tsunami of Genetic Information

"This is a challenge for everyone involved in medicine. We are going to get this tsunami of genetic information in the next few years. Doctors don't understand it. Patients don't understand it. There is a huge need for education of everyone about it. There is a huge shortage of genetic counselors. The technology is driving the ship here. We are going to have all this information and not know how to process it, how to interpret it, what to give to patients, or how to explain it to them," Dr. Klitzman said.

This is a challenge for everyone involved in medicine. Dr. Robert Klitzman

Developing personalized medicine depends on researchers being able to access many people's genomes. "Genomes are going to be passed around. Privacy is under siege; it is already somewhat of an illusion," Dr. Klitzman said.

The Genetic Information Nondiscrimination Act covers health insurance but does not cover life insurance, disability insurance, or long-term care insurance.

"If you apply for life insurance, the company is within its right to ask, 'Have you ever undergone a genetic test? Has anyone in your family ever undergone a genetic test? What was the result?' If it included the BRCA1 breast cancer gene, the company can refuse to insure you or can charge a much higher rate. Genetic discrimination is very real and is a problem," Dr. Klitzman said.

Dr. Bombard and colleagues conclude that "for prospective research and clinical translation, the path forward depends on approaches that better define actionable variants and the development of the evidence base and clinical infrastructure to support preference-based disclosure of incidental findings. This will require the creation and evaluation of decision tools and the clinical capacity to provide genomic counseling for which no standards of care exist."

Dr. Bombard, Dr. Robson, Dr. Offit, and Dr. Klitzman have disclosed no relevant financial relationships.

JAMA. 2013;310: 795-796. Abstract



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.