Skip to : [Content] [Navigation]
 

Originally Published MX July/August 2002

GOVERNMENTAL & LEGAL AFFAIRS

IRB Compliance: The Clinical Study Sponsor's Obligations

A noncompliant oversight board at a study site could jeopardize a manufacturer's new-product introduction.

Anna J. Baldwin-DeMarinis

Clinical trials of medical and in vitro diagnostic (IVD) devices can take years to complete and cost a million dollars or more. One major imperative of such trials is the protection of the human subjects who are enrolled, an FDA requirement. To satisfy FDA in this regard, an institutional review board (IRB) must review the clinical research, and trial participants must give their informed consent to be subjects.

IRB reviews and the form for obtaining informed consent must both comply with federal regulations. Touching on informed consent only incidentally, this article focuses on how the manufacturer of an investigational device or diagnostic product can determine the FDA compliance status of IRBs at the sites where the firm intends to conduct—or has already conducted—clinical trials. A trial sponsor faces risks if an IRB's performance is not compliant. Strategies outlined here for assessing IRB compliance can minimize those risks.

Clinical Trial Oversight

The protection of human beings subjected to medical research is a relatively recent concept, not widely implemented in the United States until several disturbing reports were publicized. Exposure of the notorious Tuskegee Syphilis Study led to Congressional hearings and ultimately the creation of the current federal framework for human-subject protection. The publication of the Belmont Report quickly led to the development of harmonized regulations that define human-subject protection today (see Table I).1–5

Health and Human Services Department (HHS)

45 CFR Part 46: Protection of Human Subjects.

  • Subpart A: Basic HHS policy for protection of human research subjects (the “Common Rule”).
  • Subpart B: Additional protections for pregnant women and fetuses.
  • Subpart C: Additional protections for prisoners.
  • Subpart D: Additional protections for children (adopted by
    FDA April 30, 2001, as an interim rule).
FDA

21 CFR Part 50: Protection of Human Subjects.

21 CFR Part 56: Institutional Review Boards.

Table II. Common findings of FDA’s BIMO Program and OHRP in reviewing IRB operations.

FDA, through its Bioresearch Monitoring (BIMO) Program, audits and inspects clinical research and research data. BIMO investigators also review the records and operating procedures of IRBs responsible for subject-protection activities at clinical trial sites. In addition, the IRBs of institutions conducting clinical research using federal funding come under the scrutiny of the Department of Health and Human Services (HHS) Office for Human Research Protections (OHRP). The common findings of both agencies are similar (see Table II).

Compliance with both HHS (45 CFR) and FDA (21 CFR) regulations is required when clinical studies are federally funded. IRB review of research is a central element of each set of rules.6 A set of internationally accepted guidelines also addresses IRB functions and interactions.7

Human-subject protection is a work in progress. The United States Congress is considering improvements now (see sidebar, page 58).

IRB Compliance Failures

Clinical trial sponsors have a vested interest in confirming the compliance status of the IRB at any institution where a trial is planned or under way. Deleterious publicity about the suspension of clinical research at prestigious institutions—including Duke University, the Johns Hopkins University, the University of Pennsylvania, and the Fred Hutchinson Cancer Center—has proliferated. The following recent examples of IRB noncompliance have been gleaned from a review of FDA Form 483 inspectional findings posted on the Web site of the agency's Office of Regulatory Affairs (ORA), and from warning letters available at the EFOI reading room.

FDA Bioresearch Monitoring (BIMO) Program
  • Inadequate written procedures.
  • Inadequate documentation of activities.
  • Deficiency in continuing review timeliness and adequacy.
  • Inadequate informed-consent documents.
  • Inadequate informed-consent processes.
HHS Office for Human Research Protections (OHRP)
  • Inadequate conduct of reviews, both initial and continuing,
    including conflicts of interest and failure to document the
    existence of quorums.
  • Inappropriate use of expedited reviews.
  • Deficient informed-consent documents.
  • Inadequate records and minutes.
  • Lack of policies and procedures.
Table II. Common findings of FDA’s BIMO Program and OHRP in reviewing IRB operations.

The June 2001 report of the death of a healthy volunteer in an asthma study conducted at the Johns Hopkins University (Baltimore) prompted an immediate three-month FDA inspection of the university's IRB. A Form 483 was issued, citing the IRB's failure to maintain adequate minutes of meetings and conflict-of-interest disclosures, failure to follow its own procedures, failure to conduct a proper review of research, and failure to ensure that informed-consent documents were appropriately written.

An inspection of the IRB of Bayview Medical Center, a Hopkins-affiliated institution where the study was actually conducted, drew similar citations. In addition to the violations noted above, the Bayview IRB was cited for failing to document the existence of a quorum at its convened meetings. Press reports indicated that regulatory sanctions imposed on Hopkins by OHRP reflected these same failures.

The IRB of ImmunoGenetics (Birmingham, AL) received a warning letter in August 2001 from FDA's Center for Biologics Evaluation and Research (CBER). The 12-page letter cited the board for lacking written procedures, for failing to conduct proper review, for having members deficient in the necessary expertise, for conflicts of interest, and for numerous other regulatory violations.

Bayshore Community Hospital (Holmdel, NJ) received a warning letter in June 2001 from FDA's Center for Devices and Radiological Health (CDRH) for IRB violations involving inadequacies in written procedures, research review, and documentation. CDRH also sent St. Mary's Hospital (San Francisco) a warning letter in March 2001. That hospital's IRB violations included failing to follow its own procedures and maintaining inadequate documentation. Finding an inspected IRB in violation of its regulations, FDA has the authority to impose sanctions on the board and its parent institution. The agency can halt the enrollment of new subjects in a particular clinical trial or in all trials going on at the institution. It can also refuse to permit new clinical trials there until further notice. Findings of violations and any subsequent sanctions are public information readily available via FDA's Web site.

Adverse inspectional findings do not simply create terrible publicity for the institution. They also put at risk all industry-sponsored research undertaken at that institution. The device or IVD manufacturer choosing a research facility as a site for clinical study of its new product is responsible for protecting its own project and reputation. The only real option for a clinical trial sponsor is to itself assess the IRB's capacity to comply with federal regulations. A variety of strategies can be employed.

How to Assess IRB Compliance

A clinical study sponsor should be able to trust that an institution's review board is in compliance. Nevertheless, it should verify compliance. The assessment should be conducted when the sponsor qualifies the investigative site. But even if the clinical trial has already begun or has been completed, it is not too late to assess compliance as long as the manufacturer addresses any adverse findings promptly.

The critical first stage in any IRB compliance assessment is the gathering of appropriate technical resources, principally the FDA compliance program manual that the agency investigators use to conduct an IRB inspection.8 This document is available through the ORA Web site. FDA's on-line information sheets for IRBs and investigators are also very helpful.6 At the end of that resource is a link to a very valuable tool, FDA's IRB self-evaluation checklist (Appendix H). This detailed checklist was developed to enable any IRB to conduct an operational self-assessment. It specifies necessary policies and practices and gives appropriate regulatory references.

FDA Center Inspecting
Clinical Trial Party Inspected
Total
Clinical
Investigator
IRB
Sponsor/Monitor
CDER
399
146
23
568
CDRH
126
57
62
245
CBER
143
15
10
168
Combined
668
218
95
981

Sponsors should periodically check FDA warning letters and OHRP compliance determination letters to see whether the IRB at an institution under consideration as a site has a problematic regulatory history.9,10 They also can consult the OHRP IRB guidebook for more-detailed federal guidance on IRB operational requirements.11

Human resources are as important in IRB compliance assessment as documentary ones. The sponsor should select a quality assurance or regulatory affairs staff member with compliance auditing skills to conduct the assessment, which principally involves an on-site document review (unless the IRB is willing to provide copies of requested documents to the manufacturer). Alternatively, the sponsor can outsource the assessment, hiring a consultant with IRB experience to do the review. All the requirements by which FDA reviews an IRB are public information to which any IRB auditor has ready access for guidance in formulating a custom auditing plan and checklist.

In assessing an IRB's compliance with regulations, the sponsor should review a number of documents and records. The IRB's response to requests for such information can range from "not a problem" to "not a chance." If the board is reluctant to be audited, asking the principal investigator to convince the IRB to participate may be helpful. Academic research institutions sometimes have an office of compliance or office of research development that can get the institutional review board to accommodate an audit. Investigators and research facilities have a vested interest in the compliance of their IRB, which affects their ongoing ability to attract industry-sponsored research. If an IRB is unwilling to provide access to documents relevant to an assessment of its compliance status, the sponsor would be wise to reconsider that institution as a clinical trial site.

Compliance Assessment Criteria

IRB Member Roster. The membership of the IRB is a major compliance criterion. An audit includes an assessment of both regular and alternate IRB members' qualifications and their contributions to the deliberations. The assessment should cover their degrees, medical specialties, licensures and certifications, and professional affiliations, and should specify nonaffiliated and lay members. Racial, gender, and cultural diversity of the membership can be difficult to gauge on paper, but this factor contributes to compliance.12,13 Acquiring a list of board members from the IRB administrator, either with actual names and credentials or with credentials and areas of expertise for anonymous individuals, should pose no problem.

Failure to achieve a quorum is noted frequently in IRB inspections. The more extensive the member roster, the harder it might be to achieve a quorum consistently when meetings are called. This phenomenon of committee dynamics has particular importance when it affects regulated IRBs.

Policies and Procedures Manual. A common regulatory finding is that IRB policies and procedures are absent or inadequate. Since regulations require each IRB to have written policies and procedures, it is unacceptable for an IRB to be derelict in this area. Some IRB policies and procedures manuals are actually posted on the Internet, so these manuals clearly are not considered confidential information. If an IRB demonstrates hesitancy in providing access to its manual, that could be an acknowledgement that the manual is not fully compliant. Whether a copy of the IRB's manual can be delivered to the auditor's location or the auditor has to travel to the study site to review it, the review is important and should be conducted. The FDA resources mentioned above serve to direct a desk audit of the content of an IRB policies and procedures manual.

Investigator Submission Package. A well-run IRB should have a comprehensive list of instructions to investigators regarding the documentation needed for the initial or continuing review of research, including copies of forms to be used in the submission process. An auditor should review this package in conjunction with the policies and procedures manual to determine whether submission due dates are clearly specified and whether the IRB asks for information appropriate for performing a complete review of the research. This information ought to be publicly available, even from the Internet. No IRB should be reluctant to provide it to an auditor.

IRB Resources. A number of IRBs have been cited for having inadequate resources to conduct the reviews submitted. Assessing compliance in this area may require asking some sensitive but legitimate questions. However, getting good answers is important, given the disastrous potential outcome for the trial of an inspection citing inadequate resources in a finding of IRB noncompliance.

Informed Consent Review. FDA often cites IRBs for violations of the informed consent requirements. Therefore, informed consent is a critical area for the auditor to scrutinize. Focus should be on whether the IRB has templates for standard informed consent, short-form consent, and the assent of children, and whether the IRB staff uses an informed consent checklist to assess the adequacy of each submitted document. Such procedural tools can support a determination that the board has established systems by which to review and revise the informed consent documents it receives. The auditor should request copies of all such documents to review their adequacy, and should receive no resistance from the IRB in doing so. In addition, the sponsor would be wise to conduct a review of the investigator's informed consent document before submission to the IRB, and of the IRB's suggested editing of that document.14

Redacted Minutes. FDA can read IRB meeting minutes routinely, but sponsors cannot, and this significantly affects a sponsor's ability to assess IRB compliance. The sponsor would find it very useful even to obtain a set of redacted minutes in order to assess the way an actual IRB meeting was conducted. Many FDA inspections have revealed IRB decisions being made without evidence of a quorum, and minutes failing to record the voting, abstentions, and conflict-of-interest exclusions adequately.

Ideally, an auditor would be able to examine a few sets of IRB meeting minutes, with the confidential information appropriately redacted, and so determine how key meeting activities are carried out and documented. An alternative more likely to be provided by the IRB is a copy of the redacted minutes of a meeting during which the sponsor's own research was discussed and acted upon.

FDA Inspection History. FDA typically inspects IRBs on a five-year cycle, a schedule the agency cannot always maintain given limited available resources and the competition from other high-priority FDA inspection programs. Of more than 2000 IRBs in the United States, only 218 were inspected in FY 2000 (see Table III).

An IRB should be able to provide an auditor with the dates of all its past FDA inspections, though the auditor would be prudent to examine FDA inspection-related correspondence that confirms the inspection results. Favorable inspection correspondence will probably be readily available. The IRB may hesitate to supply supporting correspondence for an unfavorable inspection. The auditor may wish to ask about viewing any Form 483 observations resulting from an inspection. A clinical study sponsor might well consider using a site where an IRB with an unfavorable FDA inspection history has implemented improvements that satisfy FDA as documented in agency correspondence.

Study Files. If a sponsor's trial at a particular institution is already finished, its auditor should review the completed IRB file on the research. This can help in identifying issues before FDA does, and it can enable the sponsor to determine whether to use that institution for a future study. Certainly, a request to review the sponsor's own study files should not pose a problem for any IRB.

Conclusion

IRBs are increasingly likely to fall under the regulatory microscope, particularly at sites where high-profile research is conducted. The result of that heightened scrutiny will inevitably trickle down to all research conducted at the site, including clinical trials for medical and diagnostic devices. The increasing emphasis being placed on informed consent, even for repository samples, also portends a closer relationship between the device manufacturer and the IRB. The compliance status of every IRB therefore becomes more critical. Device manufacturers must undertake to assess that status as part of their investment in an investigational project.


References

1. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, Ethical Principles and Guidelines for the Protection of Human Subjects of Research (Belmont Report) (Washington, DC: Department of Health, Education, and Welfare, 1979).

2. Code of Federal Regulations, 45 CFR 46: Protection of Human Subjects.

3. Code of Federal Regulations, 21 CFR 50: Protection of Human Subjects.

4. Code of Federal Regulations, 21 CFR 56: Institutional Review Boards.

5. Federal Register, 66 FR:20589–20600, April 24, 2001.

6. FDA, Guidance for Institutional Review Boards and Clinical Investigators, FDA Information Sheets, Update (Rockville, MD: FDA, 1998); available from Internet: http://www.fda.gov/oc/ohrt/irbs/default.htm.

7. International Conference on Harmonization (ICH), Good Clinical Practice: Consolidated Guideline, ICH E6 (Geneva: ICH, 1997).

8. FDA, Compliance Program Guidance Manual 7348.809, Institutional Review Boards (Rockville, MD: FDA, n.d.); available from Internet: http://www.fda.gov/ora/compliance_ref/bimo/default.htm.

9. IRB warning letters available from Internet: http://www.accessdata.fda.gov/scripts/wlcf/subject.cfm?FL=I.

10. Compliance determination letters available from Internet: http://ohrp.osophs.dhhs.gov/compovr.htm.

11. IRB guidebook available from Internet: http://ohrp.osophs.dhhs.gov/educmat.htm.

12. Code of Federal Regulations, 21 CFR 56.107.

13. Code of Federal Regulations, 45 CFR 46.107.

14. J N Gibbs, "Informed Consent for IVD Studies: The Manufacturer's Role," IVD Technology 7, no. 9 (2001): 22–25.

Anna J. Baldwin-DeMarinis is principal of the DeMarinis Group (North Attleborough, MA) and a regulatory consultant to the New England IRB (Wellesley, MA).

Photo illustration by Don Farrall/photodsic

Copyright ©2002 MX