An overview of stakeholder comments on the redefinition of “biospecimen” as a “human subject” and the consent requirements for secondary use of these biospecimens.
On September 8, 2015, the U.S. Department of Health and Human Services (HHS) and 15 other federal agencies issued a notice of proposed rulemaking (NPRM) to update the federal policy for the Protection of Human Subjects, more commonly known as theCommon Rule. More than 2,000 comments were submitted in response to the NPRM. This update focuses on the discussion surrounding the redefinition of “biospecimen” as a “human subject” and the consent requirements for secondary use of these biospecimens. In sum, stakeholders’ general response to these proposed changes was unenthusiastic and evidenced a concern for the future of cost-efficient research. What the relevant agencies do with these comments is likely years away.
The Common Rule defines “human subject” as (1) a living individual about whom an investigator conducting research obtains data through intervention or interaction with the individual or (2) individually identifiable private information that may readily lead to the ascertainment of the individual’s identity. Currently, researchers may use biospecimens that have been stripped of all identifiers without obtaining the individual’s consent because the data and specimens do not meet the definition of “human subject.”
The NPRM expressed concern that technologic advances and whole-genome sequencing could result in the re-identification of biospecimens or data that have been stripped of identifiers. As a result, the NPRM proposed to change the definition of “human subject” to include all biospecimens, regardless of identifiability. Researchers would then be required to obtain consent from patients to use these biospecimens for unspecified secondary research purposes different from the original purpose for which they were collected. The NPRM proposed doing so by means of a broad consent.
Concerns and Criticisms
While stakeholders universally complimented the agencies’ attempt to update and modernize the Common Rule, they nevertheless offered stark differences in their views of some of the major proposed changes. For example, the American Medical Association (AMA) stated that it “strongly support[s] the development of guidance specific to research involving biospecimens or identifiable private information, including the secondary use of specimens or information initially collected for non-research purposes.” Others were less complimentary. Representing five independent institutional review boards, the WIRB-Copernicus Group commented that the distinctions between biospecimens and other types of data were “confusing and not warranted,” and it encouraged HHS to treat biospecimens and data the same under the regulations, even if biospecimens are theoretically identifiable.
Acknowledging that informed consent to use identifiable biospecimens remained crucial for patient autonomy, stakeholders representing the medical device, pharmaceutical and biotechnology manufacturing industries expressed concerns over the future of research if surplus biospecimens lacking individual identifiers also required consent before use. Because each biospecimen’s owner would need to be contacted and consented, the Advanced Medical Technology Association (AdvaMed) estimated that costs to conduct research would raise 10- to 50-fold and studies that normally take days or weeks would require months and years. This, AdvaMed stated, “would be starkly inconsistent with FDA’s statutory and regulatory mission of promoting public health and ensuring that medical devices provide a reasonable assurance of safety and effectiveness.”
Pharmaceutical Research and Manufacturers of America (PhRMA) similarly observed that researchers and hospital administrators overseeing biorepositories would need to develop tracking systems to match biospecimens in storage with the consent status of the person who provided the specimen. The AMA questioned whether the biorepository or the requesting researcher would be responsible for contacting and consenting patients whose specimens were not originally donated for research. This administrative burden and consideration of non-research patients’ privacy may lead hospitals to forgo making samples available to researchers.
AdvaMed, PhRMA and St. Jude Children’s Research Hospital all noted the push toward precision medicine and translational research. These cutting-edge studies require convenient access to vast stores of samples with multiple biomarkers for cancer and other diseases. Identifying and contacting thousands of patients certainly is not practicable, and may be unethical, according to AdvaMed, as it would require researchers to contact terminally ill patients possessing unique biomarkers to request the use of their already available biospecimens. Combined, the stakeholders expressed misgivings that this could undermine the intent of the Common Rule, delay important breakthroughs in drug and device development, and compromise the principles of respect, beneficence and justice described in the Belmont Report.
Stakeholder comments revealed similar concerns about patient privacy. For example, both AdvaMed and the Association of Clinical Research Organizations (ACRO) cautioned that researchers would need to track previously unidentifiable biospecimens with a unique patient identifier to determine whether the patient granted consent for unspecified future use. Additionally, researchers would need to monitor whether the patient withdrew consent or whether such consent expired, given the NPRM’s proposed 10-year duration for broad consent – an “arbitrary” and “difficult to track and enforce” period, according to St. Jude Children’s Research Hospital.
Interestingly, ACRO hypothesized that redefining “biospecimen” as “human subject” may have unintended political and ethical consequences, co-opted by groups citing a federal definition of “personhood” as precedence for what “human life” means. Such a consequence would be entirely unrelated to the Common Rule’s intent.
Along with criticisms, stakeholders had a number of thoughtful suggestions. Many advocated for an “opt-out” broad consent option rather than an “opt-in” option. In an opt-out option, patients would assent to the use of non-identifiable biospecimens for unspecified future research unless they affirmatively said no. As ACRO advocated, patient education at the point of specimen collection regarding de-identification of biospecimens and the risk of re-identification would increase patient autonomy by giving patients the opportunity to discuss privacy risks rather than distracting them with a misleading consent alluding to unspecified future research. AdvaMed further argued that an opt-in approach would likely become routine rather than being a meaningful discussion about the likelihood of risk, thereby undermining patient autonomy. PhRMA agreed, stating that an opt-out approach would maximize the opportunities for sample donations and enhance scientific research without increasing risks to subjects.
Regarding the potential future risk of re-identification through technologic advances, AdvaMed reminded the agencies of their regulatory authority to “encourag[e] HHS under its existing statutory authorities … to impose administrative, regulatory, civil, and even criminal penalties on … the unauthorized re-identification of biospecimens that are not individually identifiable.” ACRO also cited the success of data use agreements under HIPAA with regard to a limited data set. These data use agreements allow for sharing of data while specifically prohibiting re-identification of de-identified data through penalties and other restrictions. A similar application could be used for shared, unidentifiable biospecimens. These regulatory and contractual considerations would protect patient privacy without compromising scientific research.