Even after a trial is underway FDA has the authority to place it on clinical hold in order to ensure the safety of human subjects.
In contrast, through RAC, NIH's relationship is with the Principal Investigator, not the sponsor. Investigators must submit to OBA a copy of the proposed research protocol and must comply with the policies and procedures for the conduct of human gene transfer clinical research set forth in the NIH Guidelines. As mentioned previously, a single sponsor may be collaborating with several Principal Investigators under a single IND. Thus, for the same scope of research activity, FDA may have on record one sponsor and one IND, whereas NIH may have on record multiple Principal Investigators. Unlike FDA, NIH does not have statutory authority to either put a protocol on hold or allow it to proceed, rather NIH is responsible for the NIH Guidelines and for convening RAC, which on a selective basis conducts public review and discussion of the scientific, safety, and ethical issues raised by specific human gene transfer research protocols.
In addition to registering with OBA, filing an IND with FDA, and following the NIH Guidelines, investigators and/or sponsors also must obtain all applicable local institutional and federal regulatory approvals-this includes approval by their institution's IRB (as required by 45 CFR 46 or the parallel FDA regulations for the protection of human subjects in research (21 CFR, Part 50 and 56), and IBC (as required by the NIH Guidelines). The IRB reviews human subjects research supported with federal funds or conducted at institutions that receive federal funds. The IBC is governed by the NIH Guidelines and is responsible for and authorized by the research institution to review and approve potentially biohazardous lines of research. It is charged with conducting an independent assessment of the containment levels required by the NIH Guidelines for the proposed research; assessing the facilities, procedures, practices, and training and expertise of personnel involved in recombinant DNA research; and ensuring compliance with all surveillance, data reporting, and adverse event reporting requirements required by the NIH Guidelines.
It is important to note that in order to initiate a gene transfer trial, an investigator must obtain authorization from FDA, and IRB and IBC approvals. Investigators governed by the NIH Guidelines are required to submit their proposals to OBA/RAC, but neither this submission nor the RAC discussion are necessary prior to the initiation of a trial.
Evolution of RAC
NIH established RAC in 1975. In that year it held its first meeting to create appropriate biological and physical containment practices and procedures for recombinant DNA research. These were later developed into a set of guidelines for the safe conduct of recombinant DNA research, the NIH Guidelines. NIH has continually refined its oversight of recombinant DNA research as the field has developed. From 1979 to 1983, several major revisions were made to the NIH Guidelines as the understanding of the actual levels of risks became clear, that is, they were lower than had been expected. In 1991, NIH ceded oversight of environmental release of genetically modified organisms to the U.S. Department of Agriculture and the Environmental Protection Agency when putative risks to the public of such releases did not materialize.
In 1990 the NIH Guidelines were amended to include guidance on clinical gene transfer research. In the earliest stage of this research each protocol invariably set a new precedent, thus it was critical for RAC to conduct a case-by-case review. At the same time FDA was developing its guidance in this area of research. In 1991 FDA issued a document "Points to Consider in Human Somatic Cell Therapy and Gene Therapy," which discussed FDA regulatory requirements.
By 1995, RAC had reviewed and recommended to NIH the approval of 113 gene transfer protocols. RAC, the scientific community, and the public had a substantial base of information regarding the use and safety of many of the vectors employed in, and target diseases addressed by, human gene therapy. Subsequent analyses revealed that the human health and environmental safety concerns expressed at the inception of gene transfer clinical trials had not materialized.
As the number of protocol submissions burgeoned, in 1995 NIH Director, Harold Varmus, appointed an ad hoc review committee (the "Verma Committee") to assess the activities of RAC and to make recommendations about its role in oversight and facilitation of gene transfer research. The Verma committee recommended that in view of the reasonably safe experience with clinical gene transfer research, RAC should no longer conduct case-by-case review of all protocols in that this presented duplication of effort with FDA. However, the committee recommended that RAC continue to review novel protocols in an open forum (i.e., protocols that involved new disease indications, vulnerable populations, and new classes of viral vectors). Also in 1995, NIH convened a committee (referred to as the "Orkin/Motulsky Committee") to assess the status of gene therapy's efficacy and to make recommendations about NIH's investment in the field. (These reports are available at www.nih.gov/oba/documents)
Simultaneously, absent evidence of substantial safety concerns for gene transfer protocols that had been previously tested, on March 6, 1995, RAC voted to recommend approval of amendments to the NIH Guidelines that would eliminate RAC review and approval of human gene transfer experiments not considered to be novel. Under this mechanism, all protocols determined not to represent a novel gene transfer delivery strategy, or that did not pose unusual safety concerns were considered exempt from RAC review and approval (although IRB and IBC review were still required). This streamlined process, which became known as the NIH and FDA "Consolidated Review," was believed to eliminate unnecessary and time-consuming duplication of effort by NIH and FDA. On April 17, 1995, NIH Director Varmus approved these amendments to the NIH Guidelines.
In December 1996, the RAC review process was modified to require a rapid initial analysis of every protocol to determine which protocols presented significant novel scientific, safety, ethical, legal or social issues. Only those protocols deemed to present novel issues would receive full RAC review and public discussion.
In 1997, after public notices and the receipt of public comment, NIH no longer required RAC approval of gene transfer protocols, novel or otherwise. These changes did not alter the ongoing requirement for all investigators conducting gene transfer trials supported by NIH funding or at institutions receiving NIH funds to register their protocols with OBA, adhere to the NIH Guidelines, receive IRB and IBC approvals, and participate in public discussion as determined by RAC.
Reporting Adverse Events
All gene transfer clinical trials are subject to FDA regulations, found in Title 21 of the Code of Federal Regulations (CFR), including specific requirements at 21 CFR 312.32 related to adverse events. FDA regulations require that unexpected fatal or life-threatening adverse events associated with the use of the product are to be reported as soon as possible, but in no event later than seven calendar days after the sponsor's initial receipt of the information. Serious and unexpected events associated with the use of the investigational product are to be reported as soon as possible, but in no event any later than 15 calendar days after the sponsor's initial receipt of the information. Sponsors also must submit a comprehensive report of all adverse events to FDA in annual reports and other regulatory documents.
The NIH Guidelines currently require the immediate reporting to OBA of all serious adverse events associated with human gene transfer clinical research. The Guidelines do not define serious adverse event; however, on a practical basis NIH uses the FDA definition of adverse event. If a reported adverse event is judged by NIH/OBA to be serious and related to a gene transfer protocol, OBA confers with FDA and the RAC Chair regarding follow-up evaluation of the event and any need for action. As appropriate, the following groups are notified: RAC, OHRP, IBCs, and all Principal Investigators engaged in related clinical research.
Although NIH and FDA both require expeditious reports of serious adverse events in human gene transfer, the agencies initiate different, but complementary, processes in response to this information. FDA reviews the information provided in the adverse event reports in the context of other information from that study and from other studies of the same or related products. Actions that FDA might take to assure the safety of study participants include:
FDA also conducts an analysis of the event and relevant data and, if necessary, places the study, and others like it, on clinical hold until the safety issues have been adequately addressed. FDA is required by law to maintain the confidentiality of all information connected with an INDa. However, FDA does contact other sponsors if adverse events might affect the health and safety of individuals in other clinical trials. The FDA Commissioner may disclose a summary of selected portions of safety and effectiveness as are appropriate for public consideration of a specific pending issue (601.51(d)), including important safety issues.
In contrast to FDA, NIH does not have statutory authority either to authorize the initiation of a clinical trial or place a clinical trial on hold. But, the reporting to NIH/OBA of serious adverse events initiates, if warranted, a notification process and public discussion of the event. Serious adverse event reporting by investigators allows OBA to notify RAC rapidly and, as appropriate, other IBCs, IRBs, and Principal Investigators in the field. Expeditious reporting also provides a mechanism for early recognition of trends regarding the occurrence of serious adverse events that may raise significant implications for the safety of individuals enrolled in similar human gene transfer studies. When deemed appropriate, OBA initiates additional data collection for a comprehensive and public review by the RAC and ad hoc experts. This process fosters broad public awareness of issues and developments in human gene transfer research.
Exactly what must be reported when a serious adverse event occurs is the topic of much discussion, in part because there are different reporting requirements not only for FDA and NIH, but also for OHRP. For example, FDA requires that the sponsor notify FDA of any adverse experience associated with the use of the drug that is related and both serious and unexpected. The NIH Guidelines require that the principal investigators must report immediately any serious adverse event to the local IRB, IBC, OHRP (if applicable), NIH/OBA, and FDA. The Common Rule (with oversight by OHRP) requires that institutions (through the Multiple Project Assurance mechanism) have in place written procedures for ensuring prompt reporting to the IRB, appropriate institution officials, and the department or agency head of any unanticipated problems involving risks to subjects or others.
Activities following the death of Jessie Gelsinger are illustrative of the process. When notified of Gelsinger's death, FDA immediately placed the trial on clinical hold and informed the sponsor that presentation at the next meeting of the RAC would likely be required. FDA reviewers then identified other gene transfer protocols in which adenoviral vectors were being administered intravascularly, contacted these sponsors to alert them of the death, requested that they add appropriate language to the informed consent, and inquired as to whether other adverse events were observed in their INDs, particularly thrombocytopenia, DIC, and/or abnormal liver function tests. In October 1999, FDA placed two other protocols that used a high dose adenoviral vector via intrahepatic inoculation for cancer on clinical hold.
On September 21, 1999, the day after OBA received notification of the Gelsinger death, OBA sent a notification letter to RAC members, IBC chairs, FDA, and OPRR informing them of the death and describing plans for RAC review and discussion of issues associated with adenoviral vectors. Also in October, NIH sent a letter to all Principal Investigators (approximately 70) who were using the same or similar method of gene delivery, requesting safety and toxicity data, including adverse events.
Subsequently, a working group of RAC was formed to conduct an in-depth analysis of the data and, if necessary, develop guidance regarding the use of adenoviral vectors in gene transfer studies. In addition, a scientific symposium was held at the December 1999 RAC meeting to provide comprehensive scientific review and public discussion of the data.
As a result of these enhanced oversight activities, it has become clear that many investigators and sponsors either did not submit serious adverse event reports to NIH as they occurred or labeled them as confidential. In September 1999, NIH requested that RAC consider the need to clarify and strengthen the NIH Guidelines for adverse event reporting, specifically, what NIH should do to ensure that adverse events reports required by NIH not be classified as confidential, trade secrets, or proprietary information. During its September 2-3, 1999 meeting, RAC developed the following consensus statement regarding serious adverse event reporting:
"Adverse event reports shall not be designated as confidential, either in whole or in part. Adverse event reports are essential to decision making by IBCs, IRBs, and potential subjects of gene transfer research in humans. The public disclosure of adverse events [in human gene transfer research] is also essential to public understanding and evaluation of gene transfer in humans. Adverse event reports must be made available for public discussion [by the RAC] without the inclusion of proprietary or trade secret information."
RAC subsequently directed NIH to prepare a proposed amendment to the NIH Guidelines that states, "adverse event reports should be devoid of individually identifiable information and should not be marked as confidential."
Proposed Changes to the NIH Guidelines
Several proposals have been made in the past six months for modification of the NIH Guidelines.
A November 22, 1999 proposed action published in the Federal Register intended to strengthen compliance with the NIH Guidelines (November 22, 1999, Federal Register notice) would add three provisions to the Guidelines: 1) a definition of serious adverse events and a stipulation of the time frame in which serious adverse events are to be reported in writing; 2) a mandate that serious adverse event reports must not contain any trade secret or commercial or financial information that is privileged or confidential and that all information submitted in accordance with the Guidelines will be considered public unless NIH determines there to be exceptional circumstances; and 3) a directive that serious adverse event reports submitted to NIH be stripped of individually identifiable information in order to ensure confidentiality. These changes are intended to clarify the NIH requirements for reporting adverse events, as well as ensure that all pertinent information regarding the safe and ethical conduct of human gene transfer trials is provided in a timely fashion to RAC for review and analysis.
Also in 1999, in another effort to optimize and streamline this process, NIH considered modifying the requirements for protocol submission by allowing proposals to be submitted for RAC review before they have been approved by the local IBC and IRB. Pursuant to this revised process, clinical trial investigations would not be initiated (that is, no human subjects enrolled) until the RAC review process had been completed, and IBC and IRB approvals and applicable regulatory authorization(s) have been obtained. The proposed changes would allow investigators to receive RAC input at an earlier stage of protocol development, allow multiple levels of protocol review to occur simultaneously, and reduce the delays in initiating clinical protocols. A human gene transfer protocol that meets the requirements set forth in Appendix M-I of the NIH Guidelines would be submitted for RAC review prior to receiving final IBC and IRB approval. For the purposes of this proposed action, "enrollment" is defined as the process of obtaining consent from a potential research subject, or a designated legal guardian, to undergo any test or procedure associated with the gene transfer experiment.
NIH accepted RAC's recommendations but has delayed implementation pending related reviews of RAC activities by one of RAC's internal working groups and by this Advisory Committee to the Director Ad Hoc Working Group.
Also of relevance, RAC formed a working group in July 1999 to redefine the scope of the NIH Guidelines. The working group developed a proposal that was discussed and met with broad acceptance by the RAC at the September 2-3,1999 meeting. The proposal broadened the scope of the Guidelines by acknowledging the development of many new technologies, and expanded, accordingly, the definition of encompassed research to include research that is based on the aim/intent of modifying the human genome, as opposed to research that utilizes a particular, currently recognized, genome-modifying technology. RAC has delayed final action on the recommendations of that working group pending the outcome of this and other reviews.
RECOMMENDED PROCESS FOR RAC REVIEW OF PROTOCOLS
1. Concurrent registration by the investigator of the original protocol with OBA and submission of the protocol for review to the relevant IRB and IBC. The IBC cannot grant final approval of a protocol until the RAC process has been completed.
2. In addition to the protocol, the submission to OBA should include documentation of the status of IRB and IBC submission and review, and the status of the IND authorization if an application to FDA has been made. RAC should work with OPRR to encourage IRBs to consider their reviews provisional pending RAC review.
3. Upon receipt of a protocol, OBA staff determines whether the application is complete and, if so, initiates the process to determine whether the protocol is novel. NIH should ask investigators to encourage sponsors to withhold application for an IND until either: 1) the proposal is deemed non-novel, or 2) the protocol has undergone the RAC review process.
4. Protocols deemed non-novel are subject to the standard scrutiny and review now required by FDA for IND authorization. If RAC deems a protocol non-novel, notification is sent to the IRB, IBC, FDA, and the investigator, informing them of this determination. The IBC can now grant approval. Once an IND is authorized, the investigator should register the final protocol with OBA.
5. Protocols deemed novel are scheduled for public RAC review and discussion, the results of which are reported to the investigator, relevant IRB and IBC, and FDA. Following receipt of these comments, and investigator response to the RAC, the IBC can now make a decision regarding approval. IRBs should work with investigators to ensure that the results of RAC review are incorporated into the informed consent process.
6. The investigator is responsible for communicating with the IRB, IBC, and the sponsor and FDA, as appropriate, to prepare a publicly available response to RAC. It is assumed that the investigator will notify the sponsor, and that any FDA requirements will be followed. The investigator may submit an interim response outlining a plan that may include, for example, more pre-clinical work that may alter the protocol design enough to require a new submission at a later date. If the response is final, it must include the final IRB and IBC approval, the final consent document, the final IND (once received), and a letter describing any actions taken in response to the RAC review.
7. If no actions are to be taken, the response must contain a detailed explanation in support of the investigator's rationale for electing not to comply with RAC's recommendations. The investigator should copy the IRB, IBC, and FDA on the final response.
8. As with all gene therapy protocols, as stated in Appendix M-VII-A of the NIH Guidelines: Upon receipt of notification of permission [from FDA] to proceed with an Investigational New Drug Application for a human gene transfer protocol, the Principal Investigator(s) shall submit a written report [to NIH OBA] that includes the following information: (1) how the investigator(s) responded to RAC's recommendations on the protocol (if applicable), and (2) any modifications to the protocols as required by the FDA. Once the IND is authorized, the protocol should be registered with OBA.
9. Subject enrollment begins and the protocol is entered into the NIH Human Gene Transfer database.
The Working Group was appointed because something is seriously wrong with the conduct of gene transfer research in human subjects. While the catalyst was the death of Jesse Gelsinger, subsequent investigations revealed additional violations of regulations at other research sites. Perhaps the most disturbing aspect is that the failures are not just of one man or one institution but extend far beyond that.
Gene transfer stands on a cusp: It is still basic research and not quite yet medicine. It has the potential to be the next big wave of innovation in biomedical research. The public has been encouraged, perhaps at times unreasonably so, to see it as relief from some of life's most devastating and elusive illnesses.
If it is to fulfill these promises, gene transfer research must lead not only with the high tech of cutting edge science, but also with the high touch of human interactions that value and empower patients as full partners in the research process.
Presently, the vast majority of American men, women, and children who participate in gene transfer enter the earliest phases of clinical trials where the goal is to identify toxicities and a reasonable dosing regimen in light of those toxicities. Therapeutic improvement is a long-term hope rather than a likely immediate benefit for those who enter the trials.
Protection of participants in clinical trials derives from two critical and independent sources that complement each other. One is through creation of an institutional system that instills within each member of the research community the highest ethical value for human life, and demands that behavior reflects those values. Accountability and enforcement is crucial.
The recommendations made in the majority report will enhance institutional protection of human subjects and we support all of them.
The other, less apparent, route of protection of patients is through decisions made by patients themselves. It is the basis for informed consent. Empowerment of patients for self-protection is achieved through information and through creation of an attitude of equality within both researchers and patients. Thus participants become more than passive "subjects," they become truly equal partners in the research endeavor.
But consent is only "informed" if it is predicated upon both full knowledge of the proposed trial and an understanding of medical options other than those available through trial participation. This points to the need for greater public transparency of information on all clinical trials for the prospective patient to be able to make truly informed consent.
Recommendation: One of the principal goals of gene transfer oversight should be:
To enhance the base of public information so that patients and their independent advisers can make better informed decisions on whether or not to enter or continue in a clinical trial protocol.
Adverse Events and Reporting
The Working Group faced its most difficult challenges in examining what constitutes adverse events, which ones need to be reported, when do they need to be reported, and the extent to which those reports shall be open to the public.
The HHS Office of Inspector General recommended increased use of data safety monitoring boards (DSMBs) in "high risk" clinical trials in the 1998 report Protecting Human Research Subjects. In an April 2000 Status of Recommendations it lamented that little progress had been made in this area.
Gene transfer epitomizes "high risk" clinical trials. Factors leading to that conclusion include: The transforming nature of the science; the basic questions surrounding it that remain unanswered; the small patient populations and data sets involved with virtually every trial; and the fact that no gene transfer mechanism has moved from basic research completely through the approval process to become a prescribed therapy. Of the hundreds of clinical trials initiated, only three have advanced as far as Phase III trials.
It is precisely because of its nascent state and its potential as the next revolution in medicine that gene transfer should become the model paradigm of how expanded patient protection, public participation, and information-based research is conducted. It is a once in a generation opportunity to improve the standards for biomedical discovery.
The NIH, FDA, and OPRR, now reshaped as the Office of Human Research Protections (OHRP), regulate gene transfer. Each has overlapping but differing audiences, responsibilities, and information requirements. This fragmentation of oversight has resulted in some confusion and duplication of activity. There are calls for consolidation and harmonization of reporting from all corners. These efforts are worthy of support.
However, in areas involving high risk research, such as gene transfer and likely others, it has become too tempting to allow the goal of harmonization to blur our focus on the public's need for and entitlement to more comprehensive, timely reporting of serious adverse events.
Gene transfer is not the Nth version of a "me too" drug but an area of biomedicine where much remains unknown. There are valid issues of science and public perception that justify treating gene transfer differently.
Harmonization should be in the direction of greater transparency and greater availability of information, not less. If we are to err with oversight of this new science, let it be in the direction of protecting patients "too much."
Recommendation: All serious adverse events (SAEs) both anticipated and unanticipated, whether believed related or unrelated to the trial, should be reported electronically, in real time, to a centralized agent, using a standardized reporting mechanism and nomenclature.
The agent would immediately retransmit relevant sections of the report to FDA, NIH, OHRP and others as appropriate. The agent may be housed within one of those organizations or elsewhere.
Patient privacy should be respected to the greatest extent possible. Data on SAEs, stripped of patient identifiers, should be made publicly available as soon as is reasonably possible. NIH should make public its analysis of data sets prior to or simultaneous with public release of that data.
Historically, the NIH has led the way in expanding patient protection and in opening up the research process to the American public. But the scope of its jurisdiction is limited. The May 23 initiatives of President Clinton and Secretary Shalala promise to further explore enhancements to patient protections in clinical trials within a context that is broader than NIH. We heartily support those efforts.
Christine K. Cassel, M.D., Co-Chair
The Mount Sinai Medical Center
New York, NY
Victor J. Dzau, M.D.
Debra Lappin, Esq.
Ruth Macklin, Ph.D.
Claudia Mickelson, Ph.D.
Thomas Murray, Ph.D.
Stuart H. Orkin, M.D., Co-Chair
Harvard Medical School
David R. Parkinson, M.D.
Robertson Parkman, M.D.
Kathryn Zoon, Ph.D.