| Comment Record|
Dr. Robert Larsen ||
2002-05-07 11:55:25 |
University of Southern California |
| Comments for FDA General |
1. General Comments
RE: Docket No. 01N-0322 “Institutional Review Boards: Requiring Sponsors and Investigators to Inform IRBs of Any Prior IRB Reviews”
My response represents my personal views and do not reflect the institution I serve. I hold the position of the Vice-Chair of the Institutional Review Board for the University of Southern California on the Health Sciences Campus. This IRB is responsible for all human studies research conducted on the Health Science Campus.
The proposed regulatory change is misguided, burdensome, and is not likely to enhance subject safety unless it is limited in scope, and carefully crafted to achieve a limited goal. Institutional Review Boards are independent and thus do come to differing opinions about the value and safety of research for any given project. Their charter is to minimize the risk of conducting human research in the facilities for which they are given that authority. What may be safe research to conduct in one location may not be safe in another location. Learning that there are unsafe settings and/or locations will not benefit a local IRB in making this determination.
The timing of IRB reviews can be very problematic. Many clinical studies conducted by the FDA are conducted at multiple sites, including international locations. It is not uncommon for sites to “come on board” many months following initial IRB review and approval. The added burden to review the decision of each IRB will expand the already substantial review conducted, and may lead to requests for clarification regarding findings that may or may not have bearing on the local approval and significantly delay orderly accrual of subjects to a multi-center study while these issues are resolved. Furthermore, clear guidance will be required about which issues must be referred for full board review and which can undergo an expedited review process for those sites that have previously approved the research activity. Additionally, will IRBs be required to translate documents from international sites, or will that be a responsibility of the sponsor? How long will the interval be between an IRB action at one site and the mandated reporting at all other sites? What will be the action if a project is approved, but not given final approval and no subjects are enrolled?
The more expedient process is for all IRBs to be provided documentation from sponsors of IRB correspondence indicating disapproval for any version of the protocol. Commonly, clinical protocols undergo evolution. This is particularly true if the original application is denied. It is of value to understand the process by which a protocol is improved to satisfy the requirements maximizing safety while maintaining the original scientific objectives. Furthermore, the sponsor should be well able to address the changes in a clinical protocol in response to disapprovals.
The regulatory requirements for obtaining IRB approvals are the investigators and sponsor’s responsibility. The FDA has the authority to demand of the sponsor that all IRB correspondence from all sites proposing to conduct the specific research project be submitted, including those that did not receive final approval. Upon discovery of significant divergence in IRB approval, e.g. “IRB shopping”, the FDA has the authority to report such divergence to OHRP who could, if deemed appropriate, conduct an investigation regarding the “apparent” IRB misbehavior. The IRBs that are not in substantial compliance with regulations can then be appropriately sanctioned or decertified by the FDA and/or OHRP.
To respond to the questions as outlined in your “Proposed Rules” I offer the following:
1. How significant is the problem of IRB shopping? The IRB at the USC Health Science Campus is the only IRB governing human research at our site. No human clinical research can be conducted without our approval and as such it is not possible to shop for IRB approval at another site and conduct research on the Health Science Campus. Our investigators are not permitted to conduct research at other sites, even if governed by another IRB, without the approval of the IRB at our site. There is no “IRB shopping” permitted under these rules.
2. Who should make these disclosures? The sponsor should be the responsible party for reporting all IRB correspondence, including sites that attempted to initiate the clinical protocol but did not get final approval or for which approval was denied. Failure to make the appropriate disclosure should then lead to regulatory action by the FDA. Investigators should not hold this responsibility. The IRB making the unfavorable decision “disapproval” should be required to report that decision to the site principal investigator, the study sponsor and the FDA.
3. Who should receive the disclosure? In consideration of rule making, it may serve the purpose of the FDA to require IRBs to report “disapprovals” to the sponsor and the FDA, if the project is an FDA regulated activity. The reporting IRB could then supply the FDA with the sponsoring agency, the IND and/or IDE number, and the sponsor’s protocol number. The FDA, in its monitoring role may then seek comment from the sponsor, if necessary. There are sufficiently few such events that the additional burden to IRBs would be minimized.
4. What information should be disclosed? It is sufficient to disclose unfavorable “disapproved” IRB decisions to all IRBs involved in the project, although it is our current practice to notify only the on-site principal investigator. Included in the notification of the unfavorable IRB decision, all comments shared with the on-site principal investigator (at the site of the unfavorable review) should be included for review. However, the sponsor’s response to those unfavorable IRB decisions would be valuable and appropriate. Thus, a reasonable sponsor response time is needed, if there is to be a value added component to guide the deliberations of other IRBs. Upon receiving an unfavorable IRB review, the sponsor should determine if the trial: 1) should temporarily suspend enrollment at all sites while preparing their response to an unfavorable action by an independent IRB, and 2) if the unfavorable IRB decision should be reviewed by the full board or can undergo expedited review because of the sponsor’s response and modification of the protocol detail (if required). While the judgment of the sponsor is required for these issues, it is in the sponsor’s best interest to address the issues of an unfavorable IRB review promptly and completely.
5. If a proposal would not require disclosure of all prior IRB decision, what information should be disclosed? It is universal that there are comments on all initially approved research activities. It would provide almost no value to read volumes of IRB correspondence asking for changes in the informed consent document and local non-significant issues, e.g., budget discussions, laboratory usage forms, etc. Thus, only unfavorable reviews which meet the threshold of “disapproved” are of sufficient value to warrant review. Further, all IRBs, including those that gave prior approval, should be apprised on any “disapproved” IRB decision.
6. To permit a subsequent IRB to assess the value of a prior IRB decision, should the information about the basis of the prior decision be disclosed? Yes, the full correspondence from the IRB with the “disapproved” decision should be provided, including any prior correspondence preceding the adverse decision.
7. How should the FDA enforce the requirement? Two elements are necessary to enforce notification of unfavorable IRB decisions. 1) The IRB that makes an unfavorable decision of an FDA regulated clinical trial, e.g., “disapproved”, should be required to provide the FDA with that decision, the name of the sponsor, the title of the protocol, the sponsors protocol number, and all correspondence with the site’s principal investigator regarding the IRB deliberations. 2) The Sponsor should include in their regulatory submission all IRB approvals (initial and continuing) and all adverse IRB decisions. FDA review of IRB activities which reveal potential IRB misbehavior could prompt an FDA audit “for cause” of IRB compliance with appropriate regulations and the findings dealt with in the usual fashion. These finding should also be forwarded to OHRP.
8. Are there other ways to deal with IRB shopping other than disclosure of prior IRB reviews? Possibly not. However, local IRB reporting to the sponsor and FDA requiring disclosure of all adverse IRB actions (disapproval in addition to the notifications already required) would address this issue without unduly burdening the IRB and does not requir modification of existing rules. The FDA in its regulatory capacity could then independently indicate which disapprovals appear to have raised important safety concerns. The FDA in its regulatory capacity should at that point require the sponsor to address those issues of concern which may or may not lead to changes in protocol detail or informed consent template document(s).
Robert A. Larsen, M.D.
Associate Professor of Medicine
University of Southern California
Health Sciences Campus