|Regulations & Standards|
Originally Published March 2000
For IVD companies, access to human tissue samples is essential. Tissue specimens are required for a variety of purposes. They are especially useful during the development of new diagnostic tests and when companies are generating the data needed to support submissions for premarket clearance or approval from FDA.
In seeking to gain possession of such tissue samples, IVD manufacturers have always had to pay attention to relevant legal and regulatory concerns. However, recent events will compel IVD companies to pay even greater attention to such issues relating to sample acquisition. Two forces in particular are driving the need for companies to reexamine their policies on tissue sample acquisition: the dramatic increase in public concern over patient privacy, and the ongoing Human Genome Project and related spin-off areas.
The current crop of legal and regulatory issues affects both prospective studies and retrospective studies. However, the changing environment of public opinion is having the greatest impact on the way that companies conduct studies using banked samples.
Prospective clinical trials to support IVD marketing applications have long been subject to FDA's institutional review board (IRB) and informed consent regulations. Such regulation generally requires that an IRB exercise oversight over the study's protocol and subject consent forms, and that subjects who participate in the trials consent in advance to that participation. A standard consent form typically advises the subject that patient privacy will be maintained to the extent permitted by law, subject to inspection by FDA and the sponsor.
In recent years, however, the rapid growth of research in the field of human genetics has focused attention on an important question: What is the scope of a subject's consent? In the past, the consent forms used for most studies referred explicitly to use of the tissues in connection with a specific diagnostic test. With increasing frequency, however, sponsors are seeking to use donated tissues for some additional, unrelated purpose, such as testing for the presence of a particular gene or as part of a pharmacogenomic research project.
FDA regulations do not directly address the propriety of such actions. One could argue that the use of tissue in this fashion is outside the scope of FDA's informed consent regulation, which pertains to research involving human beings rather than specimens. But in any case, the regulation was certainly not drafted with this particular situation in mind.1
This situation is considered in the August 1999 report issued by the National Bioethics Advisory Commission (NBAC), Research Involving Human Biological Materials: Ethical Issues and Policy Guidance.2 Among the wide range of genomic research issues covered by this report is the scope of informed consent forms used in prospective studies.
Although generation of the NBAC report was triggered by concerns related to the Human Genome Project, the logic underlying the report's recommendations is not limited to application in genomic research. NBAC also considered questions that can arise with research using more-conventional methodologies. Thus, the informed consent issues addressed by NBAC—and the commission's recommendations—transcend genomic research.
Like an earlier report by the American Society for Human Genetics, the NBAC report recommends a tiered approach to informed consent.3 The most restrictive level listed by NBAC is the use of tissue solely for the clinical trial described in the consent form. A less restrictive option allows consent for the clinical trial described in the form, but permits further contact with the donor for the purpose of requesting permission to conduct additional research.4
At the other end of the spectrum, the NBAC report recommends that patients be permitted to consent to coded use of their biological materials for any kind of future study. This recommendation was controversial. Of the 23 recommendations adopted by NBAC, this was the only one with dissenting views. One of the dissenters argued that neither IRB review nor informed consent is possible at the time when subjects are asked to provide consent for the future use of stored materials, because those studies "encompass unknown risks and benefits."5 A fall-back position, which was criticized by three NBAC members, is to formulate the consent agreement to cover coded samples "for any study relating to the condition for which the sample was originally collected, with further contact allowed to seek permission for other types of studies." In the end, a majority of the commissioners endorsed the recommendation of a consent agreement that would authorize the use of specimens in future studies.
In accord with the NBAC recommendations, IVD manufacturers that are conducting prospective trials would be well advised to consider the potential future uses of acquired specimens when they begin their study. If the samples are likely to be used in other research projects, the informed consent form should reflect this potential broader application. The consent form must be reviewed by one or more IRBs. If the reviewing board holds to the view that open-ended consent is inherently not informed consent, then the manufacturer may be required to scale back the scope of its consent form, thereby limiting the uses that can be made of the tissue.
With release of the NBAC report, sponsors should now be aware that when tissue samples are obtained for a prospective study, the future use of those samples will depend on the scope of the subject's consent.
According to the NBAC report, repositories in the United States possess a vast library of banked tissues—all together numbering approximately 300 million stored samples. Many such samples are potentially valuable to academic and corporate researchers. Indeed, a flourishing market in the sale of remnant samples has emerged.
Although it may be easier for a manufacturer to obtain banked samples than to acquire new samples for a prospective trial, the legal, regulatory, and ethical issues stemming from the use of banked samples can be far trickier. FDA officials have occasionally suggested that manufacturers should obtain informed consent for all banked specimens used in their research, but implementing such a recommendation is often impracticable, if not impossible.
Historically, the use of banked tissues in retrospective studies has not been considered a significant regulatory issue—particularly if the study sponsor did not know the patient's identity. In recent years, however, the public view of such matters has changed dramatically because of the rapid strides that have been made toward decoding the human genome and the consequent growth of public apprehension over the potential for loss of privacy. And, while genomic research using banked samples is the greatest single cause of such public concern, equivalent fears have also spilled over into other types of tissue research.
In cases where samples were obtained from a subject who consented to research, addressing the legal, regulatory, and ethical issues related to the use of banked specimens can be relatively straightforward. If the study sponsor believes that the original consent agreement is broad enough to encompass its proposed new research, and an IRB agrees, the study can proceed without hindrance. (Testing of tissue for the same disease or condition as before, but with a new analytical tool, would not require a new IRB approval.) The IRB review may be achieved through expedited review without consideration by the full IRB.
In most cases, however, banked specimens are not accompanied by written consent, or the original consent is not broad enough to embrace the new proposed test. In this event, the manufacturer may need to consider whether such tissues may still be used for its study. According to the NBAC report, the key factor for answering this question is the degree to which the samples are truly anonymous.
In cases where samples are completely anonymous, the NBAC report unequivocally affirms that they can be used in a new study. Since no one knows the individuals who provided the anonymous samples, NBAC reasons that there should be no restriction placed on using such specimens in research. Unfortunately, completely anonymous samples are relatively uncommon.
The more typical case is that the laboratory or some other entity has some record that identifies the subject who provided the sample. By removing all identifiers, however, it is possible to make a known sample into an "anonymized" one. Once this step has been performed, neither the laboratory nor the IVD manufacturer can identify the person whose tissue was tested. FDA regulations do not address this practice, which is addressed at length in the NBAC report.
In analyzing this issue, NBAC focused especially on the requirements specified in the Department of Health and Human Services IRB regulation.6 Although this regulation does not govern studies designed to support marketing applications or test development—and is therefore not legally binding on IVD manufacturers who are conducting such research—it nevertheless provides a helpful analytical framework.
NBAC carefully critiqued the IRB regulation and concluded that the practice of using anonymized specimens represented "minimal risk to the subject." For IVD manufacturers, the practical significance of this recommendation is great. When accompanied by IRB consent, it allows companies to use remnant samples from which all identifiers have been irreversibly deleted. The NBAC acknowledges that even anonymized samples may be accompanied by clinical and demographic data about the patient. Such samples may still be used, so long as the data are not so extensive as to permit identification of the subject.
Moreover, because such a study is considered a minimal risk, IRB consent can take the form of clearance by a single authorized IRB member. The member's review should focus primarily on ensuring that anonymity can be maintained; that is, that the test results cannot be linked back to a specific patient.
IVD manufacturers seeking to make use of NBAC recommendations in this area should note that the report's conclusions may be subject to several important conditions. First, the NBAC recommendations are not legally binding; if FDA were to issue a contrary interpretation requiring informed consent in all cases, that interpretation would supersede the NBAC recommendations. Considering the fact that it is almost never possible to obtain such informed consent, however, there is no question that NBAC's interpretation is both preferable to and more practical than the alternatives. As the NBAC report says, its approach allows "important research to proceed while also considering potential harms to the subject."7
Second, IVD firms should be aware that state governments have also begun to regulate genetic testing. A statute adopted by New York, for example, provides that except within the context of a very limited exemption "no person shall perform a genetic test on a biological sample taken from an individual without the prior written consent of such individual."8 IVD companies must ensure that their activities do not run afoul of such emerging state laws.
Finally, the NBAC report does not endorse anonymization as an approach for all studies involving newly collected tissue. Rather, the commission considers this practice a stopgap measure to allow the use of previously collected samples. NBAC believes that, in light of the legal and ethical issues associated with research in this field, prospective consent should be obtained for all future studies.
Securing written informed consent in which the patient authorizes the use of tissue for research will remove many of the legal, regulatory, and ethical uncertainties concerning the use of the samples. However, a physician who is in the midst of treating a patient will not typically be thinking about obtaining consent that would enable a tissue sample to be tested by an IVD company with whom he has no contact. Thus, while prior consent could resolve many issues, obtaining such consent is still not part of the routine in most clinical settings.
Provided that IRB consent is obtained and patient confidentiality truly protected, anonymization of banked samples would still seem to be the best interim solution for meeting the needs of IVD manufacturers while protecting patients.
Manufacturers should not underestimate the potential impact of public concern over privacy and human genomics on their research efforts. As illustrated by the recent public relations debacle related to the use of genetically modified organisms in food, public rejection of a new technology can have devastating consequences.
IVD companies should ensure that they develop procedures that adequately protect donor subjects—even those whose identity they do not know. In the current climate of public opinion, failure to be attentive to the rights of study subjects—including patients and others whose tissue is used in genetic research—could have profoundly negative consequences.
1. 21 Code of Federal Regulations (CFR), sect. 50.20.
2. National Bioethics Advisory Commission (NBAC), Research Involving Human Biological Materials: Ethical Issues and Policy Guidance [NBAC Report], (Rockville, MD: NBAC, 1999).
3. American Society of Human Genetics, "Statement on Informed Consent for Genetic Research," American Journal of Human Genetics 59 (1996): 471–474.
4. NBAC Report, p 65.
5. NBAC Report, p 65.
6. 45 CFR 46.
7. NBAC Report, p 70.
8. New York Consolidated Law Services, Civil Rights Section 79-1(2)(a) (1999).
Jeffrey N. Gibbs is a partner in the law firm of Hyman, Phelps & McNamara (Washington, DC).