Plan ahead for imaging IFs, experts say
Consider informed consent, expertise issues
Advancements in imaging technology mean a greater chance of detecting incidental findings in imaging research subjects. As IRBs turn their attentions to figuring out how to handle and manage IFs, some may not know where to begin or how to get their plans to the implementation stage.
"[IRBs] are very aware of the issue, but are not sure how to approach it," says Judith Carrithers, assistant dean for Human Research Protection and director, Human Research Protection Program at Johns Hopkins University School of Medicine in Baltimore. "Many IRBs do have plans in places, and others are working on plans but are not quite ready to implement them."
The Johns Hopkins Medicine IRBs, the Wake Forest School of Medicine IRB, and the Dornsife Cognitive Neuroscience Imaging Center all shared their strategies for creating IF management plans with IRB Advisor.
To further assist IRBs that may be facing uncertainty, the Presidential Commission for the Study of Bioethical Issues (Bioethics Commission) released the IRB Primer: Incidental and Secondary Findings in April. The primer was created to assist IRBs in understanding and implementing the commission’s recommendations for ethically managing IFs. Among the recommendations in the primer are suggestions for informed consent, researcher expertise, participant preferences, and researcher responsibilities. (The primer can be found at http://bioethics.gov/sites/default/files/IRB%20Primer%20-%20Incidental%20and%20Secondary%20Findings.pdf.)
Responsibilities of researchers
The IRB Primer suggests that researchers’ IF management plans include what the responsibilities of the research team are if an IF is discovered.
The Johns Hopkins Medicine (JHM) IRBs require that investigators plan for incidental findings when conducting imaging research. The IRB’s electronic application process includes 10 questions that investigators must answer to help develop the plan. (For more details, see box on page 68.)
"The IRB reviews the protocol and how they will manage the findings, and if they think the investigator has to rethink how to handle the findings, the IRB work with them to revise their plan," Carrithers says.
Informed consent, participant preference
Researchers should communicate the plan for IFs to study subjects in the informed consent, including how, when, and if IFs will be reported to them, in order for subjects to decide whether they are comfortable with proceeding, the IRB Primer suggests. The investigators may also choose to allow subjects to opt out of receiving findings. It also recommends that investigators decide how a subject’s opt-out preference could affect study participation, and advises that IRBs should be prepared to answer questions if a researcher approaches them about disclosing a life-saving IF to a subject who opted out.
"In some very limited cases, investigators offer subjects the opportunity to follow up with them if they wish to know about non-serious and non-actionable findings, rather than simply reporting them to the subject," says Joseph Andrews, PhD, director of the Human Research Protection Program and IRB at Wake Forest School of Medicine in Winston-Salem, NC. "The reason is that in cases where the effect on a subject’s health is negligible and no prevention or treatment is available, subjects may prefer not to be told about findings because their inability to lessen or stop the non-serious health issue could be a source of frustration. Thus, no subject is denied access to the information if they want it, but those who would rather not be told of this type of finding have the opportunity to decline the information."
At Wake Forest, the consent form sample states plainly that unexpected results will be shared with the subject and, with permission, the subject’s physician. It also states that the images will be saved in the radiology study library and copies available upon request.
"Typically, if there are scans or some sort of diagnostic that is being performed for research purposes, the board will ask how the findings will be handled," Andrews says. "We want to make sure that the plan is set in an appropriate way. We look at the particular protocol, and if there are diagnostic or genetic procedures going on for research purposes, we will have the investigator speak to that in the consent form."
So far, Andrews has not seen situations in which an investigator has not reported relevant incidental findings to research subjects. "The only circumstance I can think of when relevant incidental findings would not be made available to a subject is if the samples or data were de-identified. In such a case, subjects are told in the consent form that the samples will be de-identified and once that occurs there will be no way to withdraw the samples or to receive individually specific research results," he says. "The circumstances we see where identifiable subjects are not provided research test results are when the tests themselves are investigational, and therefore it is not clear if these direct findings are valid and reliable as predictive tools."
The JHM IRB consent forms must inform subjects of possible IFs with IRB-approved language. A sample consent form explains incidental findings and tells the subject that a qualified expert will read the study image as if it were a part of routine medical care. The form indicates that the investigators will contact the subject by phone or mail if an IF is detected, or go to the person’s home if it is an emergency. The sample form also indicates that a member of the study team will contact the subject’s physician if he or she wishes, or will make a physician referral. It explains that the discovery of an IF may cause the subject anxiety, and explains potential costs and implications for health insurance. The template indicates that subjects do not have the option to decline IF information. "I don’t think declining has ever come up as an issue," Carrithers says.
Expertise
Not all researchers will have the expertise to read and interpret research images from a clinical standpoint. In these cases, researchers may rely on outside experts or facilities to read images for incidental findings.
"We’ve seen where researchers are very transparent saying that the scan is for research only and not diagnostic and will not be read by a physician — in that case nothing will be reported because it’s not reliable," Andrews says. "Some investigators do high-resolution imaging that could be used for that. We also determine who will read it and the time frame in which it will be read."
At imaging research facility Dana & David Dornsife Cognitive Neuroscience Imaging Center (DNI), part of the University of Southern California in Los Angeles, investigators do not seek or inform subjects of incidental findings.
All brain imaging scans at DNI are performed for research purposes only, and no staff are allowed to interpret scans as normal or abnormal, the consent form states. Rather, all scans on normal patients are sent to an outside neuroradiologist for review. The neuroradiologist does not know the identity of the subjects. If an IF is detected, the radiologist will contact the center for the subject’s information so that the subject and/or his or her physician can be alerted. DNI is never told what the IF is, or of any subsequent diagnosis. "The procedures we adopted guarantee that subjects are alerted of possible findings but that the DNI, not being a health providing institution, has no access to the subject’s potential health issues," says Hanna Damasio, MD, university professor, Dana Dornsife Professor of Neuroscience and director of the Dornsife Center.
Scans from patients with pre-existing conditions are not sent to the neuroradiologist. "We cannot expect a neuroradiologist to render an opinion on a subject with pre-existing conditions, or a new finding in someone with a pre-existing condition, without having clinical information about the subject. For that reason, scans obtained in patients with known brain conditions are not sent for review. Such scans are sent to the attending physician of the subject, so that they can either review the scan or consult with a neuroradiologist informed of all pertinent facts," Damasio says.
"We thought long and deep about it before we came to the solution we developed. In our view it was the best and most ethical we could come up with," Damasio adds.
Ten questions for JHM investigators
Questions form basis for IF plans
Johns Hopkins Medical Center’s electronic IRB application process prompts investigators to answer 10 questions on managing incidental findings when conducting imaging research. This gives investigators a foundation on which to build an IF management plan for the study. The IRB’s questions include:
• Type of imaging procedure.
• Will the research imaging be of clinical quality?
• Will the images be read at Hopkins?
• Description of plan for external reading center, including credentialing, time frame of reading, and method of IF notification.
• Method of contacting participant (at least two methods must be maintained).
• Will the participant’s personal physician be notified?
• Has the participant given permission for his or her physician to be notified?
• Describe the process for documenting incidental findings and follow-up action in the medical record.
• Will Hopkins serve as the reading center for scans produced elsewhere?
• Describe the plan for notifying the local research group of incidental findings.
