FDA IRB inspections: What, why, and how
Friday, January 1, 2016
Has your institutional review board (IRB) been inspected by the FDA for compliance with IRB regulations? These inspections focus on IRB operations and responsibilities. If your IRB reviews and approves FDA-regulated clinical trials, this article will provide you a guide for what to expect during an IRB inspection.
FDA’s regulation of IRBs
Because new people enter the clinical trial community almost every day, we have many colleagues who are unfamiliar with FDA regulations and inspections of clinical trial-related activities. This is an update of an article that appeared more than 3 years ago, to serve as an introduction to our colleagues who are new to the field and as a refresher for those who are more experienced, but who have never undergone an FDA inspection.
FDA regulations that govern clinical trials establish specific responsibilities for clinical investigators (CI, also the principal investigator or PI), sponsors, and IRBs. These regulations ensure the proper conduct of clinical trials whose data are intended for submission to FDA and the protection of the rights and welfare of subjects enrolled in those trials. The November–December 2012 issue of Research Practitioner focused on good clinical practice (GCP) inspections and responsibilities of the IRB. This article will update that discussion. An IRB inspection is a type of FDA bioresearch monitoring (BIMO) inspection.
As noted previously, these GCP responsibilities are defined in FDA regulations found at Title 21 of the United States Code of Federal Regulations (abbreviated 21 CFR). The regulations directly applicable to IRBs are found in 21 CFR Part 56 (Institutional Review Boards). IRBs also are responsible for reviewing and approving informed consent documents. Those regulations, which must be observed by the IRB, are found in 21 CFR Part 50 (Protection of Human Subjects). A related IRB review function is to examine any financial conflict of interest (COI) held by the PI, and the IRB must consider FDA’s regulations on that topic, found at 21 CFR 54 (Financial Disclosure by Clinical Investigators). Finally, IRB scrutiny also is applied to the PI’s professional COI. Readers should review the relevant regulations along with this article to complete the inspection “picture.” While no FDA regulations specifically address the PI’s professional COI, there is substantial guidance from government agencies.
Guides for conducting IRB BIMO inspections
GCP inspections are the responsibility of FDA’s BIMO program staff. IRB inspections typically are conducted by FDA field staff (from FDA’s Office of Regulatory Affairs or ORA). In addition to many general resources for conducting inspections, FDA investigators have access to several documents that direct the conduct of a specific BIMO inspection. These document are called Compliance Program Guidance Manuals, which are internal policy/procedure manuals. The five FDA Compliance Programs (CP) related to clinical trials are listed in Table 1. FDA inspections of IRBs are conducted according to the principles and requirements stipulated in CP 7348.809. The contents of this CP are presented below. Before conducting such an inspection, an FDA investigator will prepare, in part, by reading this compliance program. Other preparatory activities include reading the IRB’s FDA file (if one exists), which contains prior inspection reports and other documents or correspondence related to those prior inspections. FDA also has a useful document for IRBs to review in preparation for an inspection. This self-evaluation checklist can be used to assess an IRB’s standard operating procedures (SOPs) and is available at http://www.fda.gov/ScienceResearch/SpecialTopics/RunningClinicalTrials/GuidancesInformationSheetsandNotices/ucm118063.htm.
Starting an IRB BIMO inspection
For IRB inspections, FDA will determine if the IRB is in compliance with its own SOPs and with FDA regulations. FDA inspections are either surveillance (periodic, routinely scheduled every 5 years) or directed (unscheduled, following a complaint, PI misconduct, or safety issues pertaining to a study or a site). FDA usually (but not always) will make an appointment for a suitable date to start the inspection, although an unannounced inspection is possible (typically for cause, such as a complaint). If the date proposed by FDA is not suitable due to extenuating circumstances, such as business or vacation travel by the IRB chairman or other principals, FDA may allow some flexibility in the start date, but it is not a good idea to delay the inspection start date by more than a few days.
As described previously, certain preliminary steps are followed for all inspections. The FDA investigator will ask to see the person in charge, typically the chairman or a top executive. When that person arrives, he or she will be given a Form FDA 482, Notice of Inspection. The person who receives the Form 482 should be sure to see the FDA investigator’s credentials and photo identification card. This starts the inspection.
If the IRB refuses access to records or copying records, or if the IRB’s actions appear to be a partial refusal, the FDA investigator will call attention to FDA’s authority to inspect the facility, found in the Federal Food, Drug and Cosmetic Act. If this does not resolve the issue, the FDA investigator will proceed with the inspection and notify the BIMO program contact person at the assigning FDA review center for further guidance. These centers are named in Table 2.
Inspections typically will include interviews, a review of IRB policies and SOPs, evaluation of the IRB’s performance through inspection of selected records, and facility tours (particularly with regard to IRB record storage). FDA will determine an IRB’s compliance with regulations based on these activities.
Documents that are inspected typically include policies and SOPs; records of IRB membership; IRB reports; IRB approvals and corresponding documentation; study folders; meeting minutes; significant risk (SR) vs non-significant risk (NSR) device determinations and correspondence to/from sites, sponsors, and FDA; organizational charts; emergency use approvals; complaints from subjects; training/qualification records; and IRB study databases.
The FDA investigator takes handwritten or electronic notes and is permitted to make copies of documents. FDA personnel may copy relevant IRB records as needed to submit to the BIMO reviewer. In some cases, the FDA investigator may wish to capture someone’s statement on an affidavit (Form FDA 463a), which is essentially a sworn statement, but it is acceptable to have a policy that prohibits IRB members and staff from signing an FDA affidavit. An IRB should have a detailed SOP on managing an FDA inspection.
Overview of the IRB Inspection
The current version of CP 7348.809, which was implemented on November 28, 2011, is based on compliance with FDA regulations on IRB oversight of FDA-regulated clinical trials. Its principal elements are presented below.
In 1981, Congress gave IRBs a legal mandate to oversee human subject research that is conducted using FDA-regulated products. FDA published its IRB regulations in the Federal Register on January 27, 1981. The regulations for informed consent also were published on the same date. This FDA inspection program provides for the inspection of domestic IRBs that review and approve investigational studies involving human subjects and FDA-regulated articles (e.g., drugs including biologics, food or color additives, and medical devices). FDA has no authority to inspect foreign ethics committees.
FDA must periodically inspect every IRB that reviews FDA-regulated research. These inspections are typically “routine” inspections. FDA consults the IRB registration database maintained by the Office for Human Research Protections (OHRP) for the name and address of the institution, and when available, the name of an IRB contact person. The assigning center may provide a specific protocol to review during the inspection, when available. IRBs without significant violations will usually be assigned for re-inspection every 5 years.
Alternatively, FDA may assign a “directed” or “for cause” inspection, which occurs when the assigning center receives information that calls an IRB’s practices into question (e.g., a complaint). A directed inspection may be limited to one area of concern or assigned to cover the entire IRB BIMO compliance program. IRBs found to have major deficiencies (Official Action Indicated or OAI classification) will usually be assigned for re-inspection within 1 year to confirm that adequate corrections have been made.
FDA’s criteria for selecting IRBs for inspection include:
- IRBs with a previous OAI inspection (these IRBs are likely to be re-inspected within 1 year);
- IRBs that have never been inspected;
- IRBs that were recently established or have limited experience reviewing FDA-regulated research (for example, an IRB that previously only reviewed social behavioral research begins to review FDA-regulated clinical trials).
IRB inspections involve an evaluation of the IRB’s written SOPs and records to determine the IRB’s compliance with 21 CFR Parts 50 and 56. In general, assignments are issued from an FDA review center and may identify several studies (usually up to three) for inspection in order to establish that FDA-regulated research has been reviewed by the IRB. The FDA investigator will evaluate these studies during the inspection unless otherwise directed by the assignment.
The chosen studies should reflect current IRB practices, preferably within the previous 3 years, and should be ongoing. If possible, at least one study that has been through a continuing review cycle will be examined. In addition to “typical” clinical trials, target studies might include:
- studies employing novel regulatory mechanisms involving cutting edge technologies (e.g., cell, gene, and tissue-based therapies);
- studies involving vulnerable populations (e.g., pediatric studies, studies involving an exception from informed consent);
- device studies that involve a significant risk determination; or
- studies with privacy/confidentiality protections (e.g., HIV studies, etc.).
Communications between FDA field and center personnel before the inspection include confirming inspection dates, obtaining updates to the IRB’s information as appropriate, and providing guidance on complicated issues. A BIMO reviewer and/or a backup will be available at the center for consultation during the inspection. After an inspection, the BIMO reviewer will receive a completed Establishment Inspection Report (EIR), copies of all documents/records that were taken during the inspection, a copy of any Form FDA 483 (Inspectional Observations) that was issued at the conclusion of the inspection, and the IRB’s written response (as applicable). Written responses to a Form FDA 483 should be addressed to the District Office indicated on the Form FDA 483. After reviewing this information, the BIMO reviewer will issue its final classification.
FDA’s selection of studies reviewed during the past 3 years is intended to ensure that the IRB’s records for those studies (e.g., meeting minutes, membership rosters, continuing review records) are available and that the inspection covers the IRB’s “current” review SOPs and practices. However, if an IRB’s SOPs have been revised within the past 3 years, only studies reviewed by the IRB since the date the SOPs were revised should be inspected. This change in SOPs should be pointed out to the FDA investigator. This will prevent citing the IRB for SOPs that are no longer applicable. As the inspection proceeds, the FDA investigator will likely collect copies of the following documents:
- IRB-written SOPs;
- IRB membership rosters for the time period covered by the inspection;
- Copies of IRB minutes that show recent practices, SOPs that violate the regulations, and approval and follow-up on selected studies; and
- Records of selected studies such as the protocol and investigator brochure (but only if there are observations involving these two documents), consent form(s), correspondence between the IRB and the PI, and correspondence between the IRB, FDA, and appropriate institutional officials that report any unanticipated problems or serious or continuing noncompliance with regulations.
Elements of an IRB inspection
IRB registration. Since September 14, 2009, every IRB that reviews FDA-regulated research has been required to register on the OHRP website; that registration must be renewed every 3 years. During an inspection, the IRB’s registration status will be verified.
IRB membership. FDA will determine whether the IRB membership has the representation specifically outlined by 21 CFR 56.107. If an IRB regularly reviews research that involves a vulnerable category of subjects, the inspection will determine if at least one of the members is knowledgeable about and experienced in working with those subjects. FDA will confirm that no IRB member participates in deliberations or voting during the initial or continuing review of any study in which that member has a COI, and that except for expedited reviews, the IRB reviewed the research at convened meetings with a majority of members present. FDA also will examine whether the use of alternate members, when they occur, was conducted appropriately. For example, does the alternate member have comparable qualifications and expertise as the member he replaced? Did the alternate member receive the same materials for review as the regular board?
Meetings. FDA will confirm that, except for expedited reviews, the IRB reviewed research at convened meetings with a majority of members present. A majority (quorum) is calculated as “half-plus-one” (rounded up for IRBs with an odd number of members). FDA will confirm that a majority was maintained at all times throughout the meeting, even after the recusal of members with a COI or of late arrivals and early departures. FDA will inspect whether the IRB’s written procedures address how a majority is calculated, whether the IRB’s meeting minutes document that a majority of voting IRB members were present at each meeting, and that the majority was maintained throughout the meeting for each vote taken on FDA-regulated studies. FDA also will determine whether the IRB’s meeting minutes indicate whether any IRB member was counted toward the majority on projects for which the member had a COI.
Written procedures. FDA will determine whether the IRB has adequate written SOPs. Those SOPs include:
- Conducting initial and continuing review;
- Reporting findings and actions to the PI and institution;
- Deciding which projects require more often than annual review and which projects need independent verification (through IRB site visits);
- Ensuring prompt reporting to the IRB of research changes;
- Ensuring the site makes no changes in approved research during the approval period or not initiate any changes without IRB review and approval except where necessary to eliminate apparent immediate hazards to human subjects;
- Ensuring prompt reporting of unanticipated problems, any serious or continuing noncompliance, and any suspension or termination of IRB approval;
- Reviewing humanitarian use devices (HUDs) and humanitarian device exemptions (HDEs);
- Making SR and NSR device determinations.
Initial IRB review of research. FDA will assess the IRB’s authority to approve, modify, or disapprove proposed studies, and to modify or suspend or terminate approval of ongoing studies. It will determine how the IRB receives and distributes materials submitted by the PI and how it assesses whether the PI has the necessary experience, research staff, and facility to conduct the investigation. FDA will want to confirm that all IRB members have access to copies of the complete IRB submission. At convened meetings, what was reviewed, who was present, and who voted will be examined (Was there a quorum? Did the majority approve?). FDA will be interested in how the IRB ensures that PIs make all required modifications prior to enrollment of subjects.
FDA will determine whether the SOPs describe IRB actions for deciding when approved research requires review more than annually (the standard requirement). That decision should be recorded in meeting minutes and IRB approval correspondence to the PI.
Minutes also will be examined to see if they show the IRB’s determination that risks to subjects were minimized in relation to anticipated benefit. The basis for that decision (preclinical testing, any other information provided by the sponsor/PI) will be determined. Readers should note that the required contents of meeting minutes are fully described in FDA regulations at 21 CFR 56.115(a)(2). If an IRB assigns a primary reviewer to review the research, FDA will note that. Additional items to be inspected regarding the initial review and approval process include the IRB’s assessment of the adequacy of the PI’s site and the PI’s qualifications. In the November 5, 2015, issue of the Federal Register, OHRP (with FDA concurrence) issued a draft guidance document on IRB meeting minutes, with a call for public comments through February 3, 2016. This guidance document is described in the Regulatory Update section of this issue.
Continuing IRB review of research. FDA will examine if the PI promptly submits periodic progress reports according to the IRB’s instruction; if each IRB member has access to progress reports; if at least one member is assigned responsibility for review of each progress report; and how progress reports are evaluated to determine if the study should be amended, terminated, or allowed to continue as originally approved. In addition, FDA will determine whether the protocols and consent forms approved by the IRB are actually used by the PI; whether unanticipated problems involving risk to subjects or others are promptly submitted by the PI and reviewed by the IRB; whether the IRB files confirm that continuing review procedures were completed; and whether the IRB followed the criteria for approval of research as outlined in 21 CFR 56.111. For selected studies, FDA will determine if the review for ongoing studies is performed by the IRB at the assigned frequency and before the approval expiration date.
Adverse event (AE) reporting. FDA will determine if the IRB has SOPs in place for investigator reporting of AEs to the IRB, including how AEs that are designated unanticipated problems are reported to the IRB, appropriate institutional officials, and FDA. This includes reporting serious AEs observed during bioavailability or bioequivalence studies in humans. FDA also will confirm the IRB has SOPs for notifying subjects of changes in a protocol or informed consent triggered by an AE. The corresponding requirements for medical device reporting (unanticipated adverse device effect), which are slightly different from drug/biologic reporting, will be confirmed.
IRB reporting to the PI and the institution. FDA will determine whether and how the IRB notifies PIs and their institutions in writing of the IRB’s initial and continuing review decisions and the expectations for PI follow-up to those decisions, including reports to the IRB of proposed changes to the research, unanticipated problems, any serious or continuing noncompliance with applicable regulations, and any suspension or termination of IRB approval. FDA will confirm whether IRB files contain copies of the IRB notifications to and subsequent responses from the PI. FDA also will determine whether the IRB has records to document that both the IRB and the PI have met their responsibilities. Relevant SOPs will be examined to confirm these processes.
Expedited review. FDA will determine whether the IRB’s use of expedited review procedures meets the requirements of 21 CFR 56.110, and will verify that these actions are documented in the IRB records. In particular, assurance will be sought that expedited review procedures are not used for circumventing the convened meeting requirements. Examples of such misuse may include interim approval granted by the chairman pending review of the proposed study at a later convened meeting, or approval granted for a one-patient non-emergency use in a protocol that does not meet 21 CFR 56.110 requirements. FDA also will ensure that the IRB has a method for keeping all members advised of research proposals that have been approved via expedited review.
Exception from informed consent. FDA will determine whether all uses of FDA-regulated test articles exempted from prior IRB review on the basis of “emergency use” or “emergency research” meet FDA requirements, as well as review IRB procedures for that exception.
Informed consent. FDA will determine whether the consent form contains the required and additional elements relevant to the study. IRB procedures for approval of consent forms will be confirmed throughout the study, including consent for non-English speaking people and significant new findings and waiver of informed consent, as well as minutes that support those approvals. The latter is important because FDA and the HHS Common Rule differ somewhat on waiver of informed consent. During an inspection, if the FDA investigator asks to see a study where informed consent was waived, the IRB should provide an FDA-regulated study, not one that is just an HHS-funded study. FDA also will confirm whether consent language for applicable clinical trials include the mandatory ClinicalTrials.gov statement.
Pediatric studies. If a study involves children as subjects, the FDA investigator will determine if the IRB has documented that the research study complies with 21 CFR Part 50 Subpart D — Additional Safeguards for Children in Clinical Investigations. Subpart D requires that the IRB review each investigation involving children and approve only those that meet the criteria with respect to the level of risk posed by the study (see Subpart D for details). That review and the extent of parental consent that is required (one or both parents) and assent by the child must be documented in meeting minutes.
If an IRB regularly reviews research that involves pediatric studies, FDA will review its written procedures for that review, including how to conduct that review, and having one or more pediatric representatives on the IRB when pediatric studies are reviewed. If the IRB membership does not include such individuals, FDA will determine if the IRB used consultants or invited individuals with appropriate pediatric expertise to assist the IRB deliberations. The meeting minutes will be inspected and should document the risk determination and its rationale as described in Subpart D.
Electronic records and electronic signatures. Computerized systems are commonly used by IRBs to collect and preserve records. Regardless of the type of computerized system used by the IRB, FDA believes the regulatory requirements for adequate documentation of IRB activities do not change whether the documentation is captured on paper, electronically, or a mix of the two. FDA’s regulations at 21 CFR Part 11 (Electronic Records; Electronic Signatures) describe the technical and procedural requirements that must be met if an IRB chooses to maintain records electronically and/or use electronic signatures. Furthermore, FDA issued a useful guidance document (Computerized Systems Used in Clinical Investigations, May 2007) that IRBs can reference.
FDA will determine the scope of inspection regarding computerized systems, but IRBs should be prepared to discuss the elements of their systems and to show they are in compliance. The FDA investigator will collect some amount of information about the system and how it is used to report to the BIMO reviewers at FDA headquarters. For example, are electronic records available for inspection and retained for the required period of time (at least 3 years after completion of the research)? Do the electronic records and documentation meet the requirements applicable to paper IRB records? How are changes to the information in the system made? Are changes made with write-protected audit trails to include the name, date, and reason for change? Interested readers should consult the regulations and the guidance document for a complete understanding of the implication of Part 11 requirements on IRB records.
Central or independent IRBs. The same inspection process is followed, whether the IRB is a local one affiliated with a medical institution or a central or independent IRB. However, if a central/independent IRB is used by a PI whose institution has its own local IRB (but the PI chooses or is required for some reason by his institution to use a central/independent IRB), FDA will determine if a written agreement exists between the institution and its local IRB and the central/independent IRB on how review responsibilities will be apportioned between the two (partial or complete transfer of jurisdiction). FDA will collect a copy of such an agreement. IRB SOPs should address such agreements. Another item for assessment during the inspection of a central/independent IRB is how that IRB evaluates geographically remote sites for participation in the study (e.g., whether the site has medical services appropriate for the complexity of the study).
Conclusion of IRB inspection
An IRB inspection where no significant violations are found might take about a week (up to 5 business days). At its conclusion, FDA will present a verbal summary of the inspection and if some minor violations were observed, also present those verbally to the IRB principals. When significant violations are found, the inspection may take considerably more time. If significant violations are encountered during the inspection, the assigning FDA center will be notified and will provide guidance to the FDA investigator on continuing the inspection. Inspections where many significant violations are observed could last for weeks.
At the conclusion of the inspection, if significant violations are observed, the FDA investigator will issue a Form FDA 483, Inspectional Observations, a written summary of the deviations from regulations that were observed. Deviations from guidance documents should not be listed on the Form FDA 483, but they will be discussed with FDA management and documented in the final inspection report, the EIR. The information in the EIR may be used in support of further enforcement actions. This may indirectly affect the status of any clinical trials that are reviewed by that IRB and any marketing submissions that include clinical trial data that were overseen by that IRB.
The EIR must document all findings that could significantly affect FDA’s decision-making process and must include sufficient information and documentation to support an inspection classification. The classification will be decided by BIMO staff at the FDA reviewing center. The classifications are defined below.
No action indicated (NAI). No objectionable conditions or practices were found during the inspection, or the significance of the objectionable conditions found does not justify further FDA action. Any post-inspectional correspondence will acknowledge the IRB’s basic compliance with pertinent regulations.
Voluntary action indicated (VAI). Objectionable conditions were found and documented, but FDA is not prepared to take or recommend any further regulatory (administrative or judicial) action because the objectionable conditions do not meet the threshold for regulatory action. Regulatory violations uncovered during the inspection are few and do not seriously affect subject safety or data integrity. Post-inspectional correspondence may identify the issues and, when needed, state that FDA expects prompt, voluntary corrective action by the IRB. This would be considered an “untitled letter” (in contrast to the “titled letter” e.g., a warning letter).
Official action indicated (OAI). An OAI recommendation is appropriate when regulatory violation(s) uncovered is/are significant or serious and/or numerous, and the scope, severity, or pattern of violations(s) support a finding that:
- Subjects participating in studies approved by the IRB would be or have been exposed to an unreasonable and significant risk of illness or injury; or
- Subjects’ rights would be or have been seriously compromised; or
- Data integrity or reliability is or has been compromised.
When applying the classification criteria, FDA will generally evaluate the effect of the IRB’s actions (number, scope, and severity of the regulatory violations) on subjects’ rights, safety, and welfare. There are gradations in the severity of each example, and the specific observation(s) should support the seriousness of the violation(s) and the effect(s) on subjects’ safety and welfare and/or the reliability and acceptability of data for FDA decision-making purposes. FDA will also consider whether it has cited the IRB for the same or similar violations during a previous inspection.
FDA may take several actions against IRBs that are found in serious violation of the applicable regulations. Administrative actions include untitled letters, warning letters, re-inspection to verify corrective actions, meetings with FDA, withholding the approval of new studies that are conducted at the institution or reviewed by the IRB under scrutiny, directing that no new subjects may be recruited for ongoing studies, terminating ongoing studies, or disqualification of an IRB or institution.
If an OAI decision is reached, additional information (e.g., previous inspections or other information about the IRB) assists FDA in determining the type of post-inspectional correspondence that is appropriate (untitled or warning letter). If FDA chooses to issue a warning letter, the IRB should submit a detailed corrective action plan. If the IRB fails to respond, or fails to develop an adequate corrective action plan, or is found during a subsequent inspection to have failed to comply with a corrective action plan, IRB disqualification proceedings may ensue.
Disqualification may occur if the IRB has refused or repeatedly failed to comply with any of the regulations, and the noncompliance adversely affects the rights or welfare of the human subjects in the clinical trial. FDA sends a notice of disqualification to the IRB or institution. FDA will not approve an application for a research permit (e.g., an investigational new drug or IND application) for a clinical trial that is to be under the review of a disqualified IRB or that is to be conducted at a disqualified institution. FDA also may refuse to consider any data that are submitted in support of a marketing permit when the data are from a clinical trial that was reviewed by a disqualified IRB. Common IRB inspectional observations include:
- Inadequate meeting minutes
- Inadequate/not following written procedures
- Failure to have a majority of members present during convened meetings
- Inappropriate use of expedited review
- Failure to conduct continuing review
- Failure to have nonscientific member during IRB meetings
- Failure to maintain IRB member rosters
- Failure to make SR/NSR determinations
- Failure to make prompt reports to FDA
- Failure to correct recent IRB inspectional observations.
For IRBs inspections conducted during fiscal year 2014, these metrics can be found at: http://www.fda.gov/ScienceResearch/SpecialTopics/RunningClinicalTrials/ucm261409.htm (see also Tables 2 and 3 and Figure 1).
Code of Federal Regulations. Title 21, Part 11, Electronic Records; Electronic Signatures.
Code of Federal Regulations. Title 21, Part 50, Protection of Human Subjects.
Code of Federal Regulations. Title 21, Part 54, Financial Disclosure by Clinical Investigators.
Code of Federal Regulations. Title 21, Part 56, Institutional Review Boards.
By Anna J. DeMarinis, MA, CQA (ASQ), MT(ASCP)SBB
This article was reprinted from Research Practitioner, Volume 17, Number 1, January-February 2016.