| Comment Record|
Ms. Cassandra Kennedy ||
2002-06-04 15:22:42 |
Quintiles, Inc. |
| Comments for FDA General |
1. General Comments
5927 South Miami Boulevard
Morrisville, NC 27560
4 June 2002
Dockets Management Branch (HFA-305)
Food and Drug Administration
5600 Fishers Lane
Rockville, MD 20857
Subject: Docket No. 01N-0322: Institutional Review Boards: Requiring Sponsors and Investigators to Inform IRBs of Any Prior IRB Reviews
The above Advanced Notice of Proposed Rulemaking (ANPRM), published in the 6 March 2002 Federal Register (Vol. 67, No. 44), has been reviewed and discussed by representatives of Quintiles Quality Assurance, Regulatory, and Clinical Operations units. Our comments to these proposed rules are summarized below.
The commendable goal of this proposed rule, in response to the OIG report dated 14 June 1998, will be to prevent individuals from “IRB Shopping”. IRB shopping is defined as receiving a negative review from one IRB and subsequently going to another IRB for review/approval without the new IRB being aware of the other’s prior involvement. Prevention of such IRB shopping would be a positive addition to the current regulations governing clinical research. The ANPRM outlines numerous mechanisms to accomplish this fundamental goal and also proposes several additional requirements. However, Quintiles has concerns that the reporting and enforcement of the proposed requirements would not achieve its purpose and, moreover, would be confusing and burdensome to sponsors, investigators, and IRBs in the conduct of clinical trials. Below, we make several recommendations that, we believe, would accomplish FDA’s admirable purpose without imposing unnecessary burdens on the clinical trial process and, accordingly, would not delay the approval of important new therapies for patients.
From a Clinical Research Organization perspective, IRB shopping has not been observed as a frequent occurrence, as approval occurs in an overwhelming majority of submissions to an IRB, both local and contract IRBs. However, during the IRB review process, Quintiles does encounter questions and requests for additional information regarding legal issues, budgetary questions, and the informed consent form; but rarely regarding the research study itself.
In reviewing the process of obtaining IRB review, it appears that the possibility of IRB shopping occurs with the use of non-local, contract IRBs. This would be attributed to the requirement that investigators must utilize their local hospital/university IRB, as applicable. Accordingly, those investigators would not have the ability to utilize another IRB if the local IRB did not approve the submitted study, thereby reducing the overall frequency of IRB shopping.
In discussions at Quintiles, it was our opinion that the most efficient way of stopping IRB shopping would be to place the responsibility on IRBs to request information for all submissions as to whether the protocol had a prior negative review from any other IRBs. Indeed, FDA has addressed this in the current Information Sheets: Guidance for Institutional Review Boards and Clinical Investigators; 1998 Update: Question and Answer section.
Question #26 reads:
“If an IRB disapproves a study submitted to it, and it is subsequently sent to another IRB for review, should the second IRB be told of the disapproval?”
The response is:
“Yes. When an IRB disapproves a study, it must provide a written statement of the reasons for its decision to the investigator and the institution [21 CFR 56.109(e)]. If the study is submitted to a second IRB, a copy of this written statement should be included with the study documentation so that it can make an informed decision about the study. 21 CFR 56.109(a) requires an IRB to ... review ... all research activities [emphasis added] ....
The response goes on to state that “ FDA regulations do not prohibit submission of a study to another IRB following disapproval.” This is appropriate, as an IRB may return a negative decision due to a number of different reasons, such as a lack of scientific expertise to perform a sound review. Providing the reporting requirements to allow a sponsor or investigator to take the research study to another IRB for review is seen as a positive addition to the current regulations.
The ANPRM contains several proposals regarding what should be disclosed to a reviewing IRB regarding prior IRB submissions --BOTH positive and negative reviews -- and by whom (sponsor and/or investigator).
Rather, Quintiles believes that this proposed rule should be restricted to controlling IRB shopping when a negative not a positive review is obtained. Reporting of all positive reviews for the same protocol or similar protocols would be confusing and burdensome to sponsors and investigators and would not target the offending IRB shopping. The ANPRM proposed several advantages of reporting positive reviews to a reviewing IRB, such as instances where a larger, more established IRB may have scientific experts, whose review would be beneficial to another IRB. The concern with that approach is that many of the larger IRBs have multiple review panels. Therefore, it could not be assumed that a positive review by a particular IRB would have included a review by that scientific expert. To address this issue, the ANPRM further proposed that in order for an IRB to place the initial IRB rejection/acceptance in proper perspective, the sponsor/investigator would submit the initial IRB’s basis for that decision to the new reviewing IRB, including a disclosure of the reviewing members. Again, it is believed that the expansion of the initial concept of this proposed rule will add layers of additional reporting requirements by investigators and sponsors, who may not have such information. The impact to IRBs would also increase as their reporting requirements would be increased. As the principles for approval/rejection of research are outlined in 21 CFR 56 and are also required to be documented in the IRB’s written procedures, it is believed that IRBs would gravitate toward offering broad statements as their decision basis, offering another IRB very little additional insight/assistance.
One obvious concern with the approach above, as identified in the ANPRM, was that a favorable decision by a prior IRB could be of little or no value to a new IRB and may actually lead to a defensive acceptance by a smaller, less experienced IRB. Current FDA regulations (21 CFR 56.107) require that IRBs contain sensitivity to community attitudes. Therefore, regardless of another IRB’s review status, the proposed research study may or may not be appropriate in the community for which it is being reviewed. It is imperative that each IRB possess the scientific and ethical membership to enable it to make a sound decision, independent of other IRB reviews.
The ANPRM questioned the appropriate timing to report prior IRB decisions in question #3. Specifically, the ANPRM proposed reporting of subsequent review information to the reviewing IRB after that IRB had rendered a decision. The limited benefit of such information would be greatly outweighed by the difficulty in obtaining and reporting the information.
Question #7 discussed the issue of enforcement and possible sanctions for failure to comply with disclosure requirements. Enforcement for any new disclosure requirement should be consistent with current regulations for clinical research, but sanctions for failure to follow this reporting requirement do not appear to be consistent with the current regulations, which do not carry a regulation-specific sanction for non-compliance.
In summary, Quintiles agrees that the practice of IRB shopping, as described in the OIG report, should be regulated. An IRB should know of prior rejections of a particular research study being submitted by an investigator or sponsor, which will be accomplished by incorporating the statements in FDA’s Information Sheet for IRBs and Clinical Investigators into 21 CFR 56 and 312. In our opinion, this would efficiently accomplish the agency’s goal without slowing down the equally important goal of approval of new therapies for patients.
Cassandra S. Kennedy
Executive Director, Clinical Quality Assurance