When a Medical Test Reveals an Unwelcome Surprise

In a split second someone can be transformed into a patient-in-waiting.

A genetic scan ordered to test for one suspected condition picks up an elevated risk for a different, unexpected disease. Or a brain scan for a concussion detects a suspicious shadow. There’s no guarantee that those abnormalities may develop into something. There may not even be a treatment or therapy for the resulting condition. The pressing question then is: Should the patient be told about the results of such incidental findings? And the answer is more complicated than it seems.

As the cost of sequencing an entire genome drops, the day will soon arrive when documenting a full record of a patient’s DNA may become as routine as a cholesterol test. The results will inevitably lead to unexpected findings, and in many cases the medical establishment is ill-prepared to counsel patients about some uncertain surprises. No federal or state statutes directly address a clinician’s duty to return such findings to patients. And apart from legal or administrative considerations, there is no clear answer about what to do when faced with these difficult dilemmas in the lab, hospital or company boardroom.

Encountering these challenges is a “growing certainty” that will affect everybody, says Amy Gutmann, the chair of the Presidential Commission for the Study of Bioethical Issues. And such decisions may also have ripple effects for family members who may be carriers of the same genetic mutations.

The commission took a stab at these complicated questions, issuing new analysis and recommendations for how to manage these increasingly complex issues in medicine, research and with direct-to-consumer tests that allow a patient to send in a cheek swab and get health information (like the personalized genome services kit offered by 23andMe, that was recently blocked by the U.S. Food and Drug Administration). The recommendations do not prescribe specific directions. Rather, they are a call to action that presses professional organizations to draw up their own protocols for everything from abdominal scans to brain imaging. Each medical test may have its own range of findings that could pertain to the immediate condition for which a patient is being treated, but alternatively could lead to uncertainty and distress when an unexpected result emerges. In short, no simple answer exists, notes the commission report.

The document emphasized that more information is not always better for every patient. And an incidental finding of a gene mutation or other abnormality may not always lead to a particular disease. “For example, each of us has scores of deleterious mutations in our genes, and these will be picked up every time a whole genome sequence is obtained,” says committee member Stephen Hauser, chair of the Department of Neurology at the University of California, San Francisco. “The vast majority of these mutations do not lead to disease because we have backup systems in place, and we do not yet fully understand these backup systems,” he says. When each surgery or therapy is a balancing act, what if an exploratory surgery motivated by an incidental finding leads to serious complications?” the report asked.

Some 32 professional organizations and working groups have begun to address the issue ahead of the report, half of them in the U.S. Their plans in different areas helped informed the commission’s findings, alongside public comments and expert presentations to the panel. Take the American College of Medical Genetics and Genomics (ACMG). This year it developed guidelines for what to do when tests give rise to more information than the patient (and clinician) bargained for and how to manage the data. The ACMG advised that laboratories always look for and return findings for 24 genetic conditions that arise from large-scale genetic sequencing regardless of individual patient preferences because those two dozen conditions, gene variants or mutations all have existing medical interventions. As clinical sequencing becomes more common, pretest counseling will become more impractical and “lack of standardization and its application to patients of all circumstances might result in deeply varying levels of truly informed preference setting,” the group reasoned. In deciding what to list, the ACMG said the condition “needed to be fairly well understood and that only variants in the associated genes that met a standard of relatively high likelihood of being disease-causing are to be reported.”