Changes to Common Rule could affect biobanks

For years, federal regulations have deemed the research use of deidentified blood and tissue samples collected in clinical procedures to be non-human subjects research, and therefore, they have not required informed consent from the patients from whom they were taken.

Now the Department of Health and Human Services (HHS) proposes to change that rule with a revision that could potentially affect the operations of some institutions' biobanks. In its advance notice of proposed rulemaking (ANPRM), the department proposes requiring written consent for any research use of biospecimens collected for clinical purposes, as well as strengthened data protections to minimize the risk of re-identification. In its discussion of the proposed change, HHS notes that advances in technology put even deidentified samples at risk of being used in ways that could allow re-identification.

They also point to surveys that show people want to be able to decide whether their samples are used for research. Public opinion in this area is clear, says Jeffrey R. Botkin, MD, MPH, associate vice president for research integrity at the University of Utah in Salt Lake City. "What we've found, what others have found, is that the majority of people are willing to have their samples used, they're willing to have their children's samples used," Botkin says. "But they want some element of choice in there. These proposed changes try to honor those desires and so are quite good and appropriate in terms of what we understand the public's request to be in this regard."

Effects on "opt out"

The changes as proposed would apply only to specimens collected after the effective date of the final rule. However biobanks that take in leftover deidentified samples from clinical procedures might be required to change the way they do business going forward.

The degree of those changes would depend on the operations of an individual biobank. Vanderbilt University in Nashville, TN, for example, collects leftover blood samples from clinical use for its BioVU bank. But everyone who receives care at Vanderbilt signs a form stating that they understand specimens taken from them might be used for research, says Todd Rice, MD, medical director of the Vanderbilt Human Research Protection Program. People who do not want this research use of their specimens may check a box to opt out of the program.

"The [ANPRM] says that you can do a quick, short-form consent, and our BioVU patients actually do that," Rice says. "We think in the overall scheme of things, it's not going to affect us that much, although we may have to tweak that form some."

However, he says, the hospital has long collected and archived tissue samples that have been used in research. "In some of those cases, there's no consent for the use of those. They didn't sign the BioVU form," Rice says. Depending on whether the existing samples are grandfathered in, their use in research might be curtailed.

Parental consent for leftover blood

The new regulations would affect another potential source of biobank samples: leftover blood samples from the required state screening that babies receive at birth.

They are considered a potentially valuable research resource, since the samples come from all babies born in a state. However, their use without parental permission has been controversial and prompted lawsuits in three states: Texas, Michigan, and Minnesota. (For more information about pediatric biospecimens, see story, below.)

At least one state, Michigan, has created a program to seek parents' consent for the use of leftover newborn blood spots. However, Botkin says many other states still are considering how to handle this tricky issue. "The proposed changes would provide a more substantive set of guidelines about what would be consistent with the regulations," Botkin says. "On the other hand, this may mean that folks will sit back to a certain extent and wait to see what happens at the federal level before they change their policy."

He says that while the change requiring explicit consent for research use of biospecimens is a good one, the proposal would require only a brief consent form generally agreeing to permit future research. In the questions section of the ANPRM, there is some discussion of giving those who donate samples an opportunity to check off potential research uses for which they don't want their sample used.

"What they've proposed is a very brief consent process that's open-ended in nature," Botkin says. "And therefore it's at substantial risk, in my opinion, of becoming a perfunctory process. It's just another form to sign when you come into the hospital or clinic. You've officially gotten their written consent, but you've not really had any sort of meaningful education or engagement about the issues."

He's working on a program to educate parents about the research use of newborn bloodspots, which could be delivered during prenatal visits. (For more information about the education process, see story, below.) Botkin says this new requirement for consent should prompt institutions to beef up their education to patients about the benefits and risks of biobank research.

"We've found that people in the general public oftentimes have a very limited understanding of what research is all about," Botkin says. "They have a difficult time conceptualizing what the benefits and risks are of research."

Education about active consent

An education process has been under way for nearly a decade at the Marshfield Clinic in Marshfield, WI, which operates an active consent biobank called the Personalized Medicine Research Project (PMRP).

Marshfield uses a community advisory board, focus groups, newsletters and other strategies to educate the community about the importance of the biobank and how samples would be used. Jonathan C. Reeser, MD, chairman of Marshfield's review board, says the institution's efforts helped establish a bond of trust between the clinic and the community.

"It seems to me that without that, we would not have been as successful as we were," he says. "Right now, our biobank is beyond 20,000 enrollees or samples, and to obtain that in a community of our size is pretty remarkable." He says that he doesn't foresee any major changes in the PMRP's operation as a result of the proposed revisions to the Common Rule.

Two issues that weren't addressed in detail in the ANPRM have some potential impact on biobanks. One is the notion of group harm: whether review of a research project should consider the risks to a group from which samples are taken, says Jeffrey R. Botkin, MD, MPH, associate vice president for research integrity at the University of Utah in Salt Lake City.

The other is return of research results to participants. Botkin notes that the revised regulations would not allow a study to be exempt from review if the researcher plans to return results to participants. What this doesn't address, he says, is the possibility that serious results could come up in the course of a study prompting a decision to seek out the participant.

"For example, if in the conduct of your research, you come across a chromosomal rearrangement or a BRCA1 mutation [indicating an increased risk of cancer], is it appropriate simply to say, 'We told people we weren't returning results, so we're not going to return those results'?" Botkin says. "Some have argued, once you get that level of information, then you ought to feel an ethical obligation to get back to those people irrespective of what the original consent said. I think that's an interesting and important debate that still has to go on here."

Pediatric biospecimens lead to ethical concerns

When Vanderbilt University's BioVU biobank considered whether to add pediatric biospecimens to the adult samples already being gathered, officials faced a range of ethical questions.

Kyle Brothers, MD, a pediatrician in Nashville, TN, and adjunct professor of law at Vanderbilt in Nashville, says the leaders of the biobank examined those questions in the framework of BioVU's opt-out procedure, in which all patients who arrive at the hospital or its related clinics sign a form stating that they know their leftover biospecimens could be used for research. Those who don't wish such use check a box to opt out of the program.

Brothers says biobank officials wanted to know how parents felt about the idea of collecting these types of samples from their children during pediatric visits. At that point, officials weren't sure whether to extend the same opt-out procedure to the pediatric samples or to require a more stringent consent process.

"I originally became involved because I'm a pediatrician and an ethicist," he says. "I went into clinics, and while parents were waiting to talk to their children's doctor, I took advantage of that time to tell them about BioVU, to show them the opt-out forms, and get their opinion about the whole thing."

Parental support

Brothers says he met with more than 60 parents in a variety of settings and found overwhelming support for the idea.

"Parents recognized that research in children was important, and they really liked that this was a way to do that research without having to give children an extra needle stick," he says.

Although Brothers intentionally avoided settings where children were extremely sick, he did visit clinics to talk to parents whose children had chronic illnesses. He says those parents were particularly supportive of the opt-out plan. Based on these results, the decision was made to keep the opt-out process essentially the same for children, he says.

The process is set up to make opting out easy, regardless of whether the decision is made by the parent, by the child if he or she is able, or even by another caregiver who brings the child in for a doctor's visit. "We don't require for this that you have to have legal authority to consent [in order to opt out]; it's the opposite," he says. "You're opting out, and we don't want any unnecessary barriers to opting out. "So if Grandma brings the kids to the clinic and reads that and says, 'I don't want my baby to be involved in that,' she can check that box. We won't go back to her and say, "I'm sorry you're not the parent, you can't legally do that.'"

Likewise, he says, a teen-ager can read the form, check the box, and samples will not be taken.

Re-identification risk

Brothers says one key ethical element to consider in biobanking is the risk of re-identification of participants, even those whose data has been stripped of the common HIPAA identifiers. As an example, he points to a county near Nashville, which has high-income residents and is much more homogenous than the city itself.

"So a person of a certain ethnic group, of a certain age, of a certain gender in that county could be identifiable to one person," Brothers says.

Vanderbilt has addressed this issue through an innovative approach. For any proposal to share BioVU's information outside the hospital — for example, to the Database of Genotypes and Phenotypes (dbGaP) — the biobank uses an informatics tool that closely examines the data being shared.

In effect, Brothers says, it attempts a sort of reidentification of those records, by comparing the data to other publicly available databases. If this comparison turns up records at high risk for re-identification, those records have more data removed. "What the software does is it gives us the opportunity to reduce the amount of information in that particular record, so that it becomes less identifiable," Brothers says.

The process is not necessary for research conducted in-house, he says, because all of Vanderbilt's researchers have signed data use agreements that forbid them from attempting to re-identify participants. Brothers serves on a committee, separate from the institutional review board (IRB), which reviews all proposals for use of the biorepository. That committee assesses the scientific value of the research and operational questions, as well as the risk of re-identifiability in the study's design.

"I have not talked to a lot of people who report that they are the ethics reviewer for a biorepository," Brothers says of his role. "I think that generally there's a compliance question that the IRB helps answer. But these are relatively unique questions about the re-identification risk."


Brothers K. Biobanking in pediatrics: The human nonsubjects approach. Per Med 2011; 8:79.