On September 28, 2007, the Office of the Inspector General for the Department of Health and Human Services (OIG) released a report entitled "The Food and Drug Administration’s Oversight of Clinical Trials" (the Report). The OIG issued the Report pursuant to a congressional request to review oversight of clinical trials by the U.S. Food and Drug Administration (FDA) after a series of news articles detailed vulnerabilities in enforcement and monitoring of clinical trials in humans (For example, see David Evans et al., Drug Industry Human Testing Masks Death, Injury, Compliant FDA, Bloomberg News, November 2, 2005). A copy of the Report is available here.

The Report concluded that "[w]eaknesses in Food and Drug Administration’s (FDA) information systems and management processes hinder the agency’s ability to oversee clinical trial inspections," according to a September 28, 2007 OIG press release, entitled "OIG Releases Report of FDA’s Oversight of Clinical Trials, Concludes Improvement of Information Systems and Processes is Needed." The Report’s findings and conclusions provide insight into current and future changes to the FDA’s processes and regulations for conducting investigations and taking corresponding enforcement actions.

Recommendations by the OIG

The Report recommended five steps the FDA should take to improve its monitoring of clinical trials:

1. Develop a comprehensive clinical trial database that constitutes an inventory of all clinical trials regulated by the FDA

2. Create an Institutional Review Board (IRB) registry that constitutes an inventory of all existing IRB

3. Create a cross-center database to track onsite bioresearch monitoring (BiMo) inspections between and among the Center for Drug Evaluation and Research (CDER), the Center for Biologics Evaluation and Research (CBER) and the Center for Devices and Radiological Health (CDRH) (collectively, the Centers)

4. Establish an internal feedback mechanism to promote uniformity of reporting by BiMo investigators

5. Seek additional legal authority to monitor more effectively current clinical trial practices, such as delegation to subordinates and use of foreign clinical trial sites

Findings

To support these five conclusions, the OIG made three principal findings.

Recommendations 1 & 2: Developing a Comprehensive Clinical Trial Database and IRB Registry. The OIG found that data management limitations inhibit the FDA’s ability to efficiently and effectively manage the BiMo program, particularly the lack of a comprehensive clinical trial registry, the absence of an IRB registry and the inconsistency and incomplete nature of the BiMo inspections databases in use at the Centers. The OIG observed that the lack of coordination of data within the FDA hindered its ability to conduct a comprehensive review of BiMo inspections and take follow-up actions.

Recommendations 3 & 4: Creating a Cross-Center Database of BiMo Investigations and Establishing an Internal Feedback Mechanism for BiMo Investigators. The OIG found that several other factors contributed to the FDA’s inability to effectively manage the BiMo program:

  • Inconsistent classifications of official action indicated (OAI) and no action indicated (NAI) inspections by the Centers prevented effective inspections and enforcement action. For example, CDER revised 68 percent of the OAI recommendations made by the FDA’s Office of Regulatory Affairs (ORA) and CDRH revised 23 percent of ORA’s OAI recommendations, highlighting the lack of consistent policies and standards.
  • The FDA’s reliance on voluntary compliance to correct violations, with only 3 percent of OAI or voluntary action indicated (VAI) classifications resulting in follow-up by CDRH during fiscal years (FYs) 2000-05. The Report also highlighted the absence of linkages between the violative inspections and any resulting follow-up visits.
  • The FDA’s failure to coordinate BiMo inspections and the Centers’ inability to predict how many applications will be received, leading to uncertainty in budgeting for appropriate personnel and inefficiencies in inspection processes.
  • The FDA has fallen behind industry practices, such as delegation to colleagues or subordinates of direct care to human subjects and the use of trials outside the United States. This limitation reduces the FDA’s ability to initiate enforcement action for any violations it discovers.

Recommendation 5: Seeking Additional Legal Authority to Monitor Current Clinical Trial Practices.

The OIG found that the FDA inspected only 1 percent of clinical trial sites for the Investigational New Drug Applications and Investigational Device Exceptions received during FYs 2000-05. The Report observed that 75 percent of BiMo inspections were surveillance inspections, involving analysis of already-completed trials for data verification purposes. Because these surveillance inspections are conducted after the clinical trial is completed, such inspections are less effective in ensuring necessary protection of human subjects during the course of a study. In addition, the Report observed that the FDA inspected less than 40 percent of all IRBs each year during FYs 2000-05. The Report stated that increased inspection of IRBs may be a more effective monitoring methodology, reaching a larger population of trial subjects per inspection.

Methodology

In preparing the Report, the OIG utilized data from seven data sources for FYs 2000-05. These data sources included BiMo inspections data, file reviews of BiMo inspections classified as OAI, an e-mail survey of BiMo investigators, interviews with FDA officials, observations of BiMo inspections, analysis of the National Institutes of Health (NIH) clinical trial registry and reviews of FDA documents. Review was limited to BiMo investigations of clinical trials and subsequent actions taken as a result of onsite investigations, and only considered this component of the FDA’s overall oversight authority. Notably, the OIG was required to use data sources outside the FDA due to its own incomplete databases and lack of coordination of information; in some instances, the OIG extrapolated data points from existing data due to an inability to track certain numbers.

Conclusions

In its Agency Comments, the FDA concurred with four of the five recommendations made by the Report, but failed to address the recommendation to create a feedback process for BiMo investigators. The FDA has addressed or has under consideration several initiatives that respond to the OIG’s recommendations, including the following:

On September 27, 2007, the 2007 Food and Drug Administration Amendments Act of 2007, Pub. L. No. 110-85, § 801, 121 Stat. 823, was signed into law. Section 801 of this act establishes the Clinical Trial Registry Data Bank, which requires both device and drug clinical trials to submit data in a coordinated manner to a single data repository.

In 2004, the FDA and the Office for Human Research Protections (OHRP) issued a proposed rule for the establishment of an IRB registry, which is currently under revision, pending issuance of a final rule

The FDA has recently promulgated draft guidance to begin to address the gaps in legal authority to mirror current clinical trial practices and has begun consideration of rulemaking in this and other related areas. For example, see Food and Drug Administration, Draft Guidance for Industry: Protecting the Rights, Safety, and Welfare of Study Subjects – Supervisory Responsibilities of Investigators, May 2007.

Additional changes are to be expected in the future revision of FDA policies, procedures and regulations to further reflect some of the recommendations of the Report, including more detailed inspections and a greater number of enforcement actions and other follow-up activities.

Implications

Clinical trial sites and IRBs should expect increased scrutiny in BiMo investigations of the efficacy of drugs, devices and biologicals, particularly adverse reactions that may have been evident during clinical trials. Any OAI or VAI classifications, particularly those that result in issuance of a warning letter, should encounter more comprehensive follow-up, as the FDA revises its policies and data management practices. Finally, a general trend towards greater efficiency, coordination and consistency of BiMo investigations likely will result in a net increase in the overall level of FDA monitoring of clinical trial sites and the level of FDA enforcement and compliance activity.

Entities engaging in clinical research should consider modifications to their policies, procedures and practices to address likely changes foreshadowed by the Report and the FDA’s response to the Report, especially those entities that have received warning letters as a result of BiMo investigations. They would also be well-served by monitoring closely other changes to FDA regulations and policies as they may evolve.