BOSTON -- Should patients undergoing broad DNA testing for a specific ailment be told of unexpected findings that signal risk of cancer or other serious diseases, even if they don't request the information?
The question is at the core of a battle brewing among doctors and ethicists amid growing use of gene sequencing for clinical use and the plethora of information that results from such tests. Writing Thursday in the journal Science, a team of ethicists said patients should decide how much they want to know and how deeply scientists should look into their genome.
The disagreement is with an April recommendation by the American College of Medical Genetics and Genomics. The group said labs testing large swaths of patients' DNA should routinely seek and report back on 57 gene abnormalities that signal treatable or preventable illnesses. Patients should be given the option not to learn about those risks, the ethicists led by Susan Wolf of the University of Minnesota, said in the Science article.
"If doctors are going to analyze all that information, they should get consent from the patient," said Sherman Elias, a geneticist at Northwestern University in Chicago who cowrote the essay criticizing the guidelines. "This is a question of whether patients have a right to decide what testing is done on them and for their children."
A separate contribution to the journal supports the guidelines that were issued by the American College of Medical Genetics and Genomics.
The European Society of Human Genetics also released recommendations today that call for limits on the use of broad DNA sequencing. Doctors should target genetic sequencing, when possible, to avoid deciphering the content of genes that may contain information patients might not want to know, the European society said today in a statement.
"Analysis should be limited to genome regions linked to the clinical problem for which the analysis is being undertaken," according to the Vienna-based society's statement. "We are opposed to the type of opportunistic screening that throws up large numbers of incidental results."
Faster, cheaper DNA sequencing with machines made by Illumina Inc. and Life Technologies Corp. allows doctors to quickly sift through the 6 billion letters, or bases, of a patient's DNA code for possible causes of serious disease. While the technique sometimes finds a genetic cause for rare or unexplained conditions, the genome is still largely unexplored, and many cases go unsolved.
To make sure they investigate all possible genetic sources of illness, doctors read either the entire genome or just the portion that contains protein-coding genes, called the exome.
The American society's guidelines were written to help doctors who fear that some patients might not be getting potentially beneficial information from the test that might help avoid known diseases, said James Evans, a geneticist at the University of North Carolina, Chapel Hill.
A patient being evaluated for a broken rib should be informed if imaging indicates lung cancer, whether or not they have asked to be told, said Evans, who didn't contribute to the guidelines or to either of today's two journal essays. Likewise, if a broad-based gene test for an unexplained condition shows that a patient has a high risk of colon cancer, the information should be relayed along, he said.
"It's incumbent on you to do a systematic analysis of the data that might reveal something highly important to your patient," he said.
The lung imaging analogy doesn't apply in this situation, because laboratories have to make an effort to seek out the 57 disease-related genes that the guidelines say should be analyzed, Elias said.
"This isn't something that you just trip over," he said. "We're saying, let the patient know and decide whether to proceed with this or not."
The recommendations also don't give much guidance about how to counsel patients about the mutations that appear on the list of 57 abnormalities, said Arthur Caplan, director of the division of medical ethics at New York University's Langone Medical Center.
"Most doctors wouldn't know much about half of these diseases to begin with," he said. "They should be given concrete information about how to do the communication and how to do it well."
Children are particularly vulnerable when information becomes known about their risk of disease later in life, Caplan said. Even the most well-meaning parents might treat their child differently because of supposed vulnerability to disease, and a very high standard must be met for giving that information to them, he said.
The panel's list of mutations that should always be investigated when broad sequencing is done includes the BRCA1 and BRCA2 genes that raise the risk of breast cancer in women, and Lynch syndrome genes that increase the likelihood of colon and rectal cancer. It also includes genes linked to hypertrophic cardiomyopathy, a heart condition that can kill young athletes unexpectedly.
The recommendations, including the list of mutations and diseases that should be reported, is likely to change as more understanding is gained, said scientists led by Amy McGuire of the Baylor College of Medicine's Center for Medical Ethics and Health Policy in Houston, who wrote an editorial in the journal defending the U.S. guidelines.
The U.S. society "acknowledged that the list will evolve over time and is developing a mechanism for community input," McGuire and her colleagues said in the essay.