Healthy Skepticism Library item: 1126

Warning: This library includes all items relevant to health product marketing that we are aware of regardless of quality. Often we do not agree with all or part of the contents.

Publication type: report

Office of Inspector General, Dept of Health and Human Services.
FDA’s Review Process for New Drug Applications: A Management Review
2003 Mar
http://oig.hhs.gov/oei/reports/oei-01-01-00590.pdf

Abstract:

PURPOSE

To assess how well the Food and Drug Administration manages its new drug application review process.

BACKGROUND

The Food and Drug Administration (FDA) receives new drug applications (NDAs) from sponsors, typically pharmaceutical companies, and reviews these applications for scientific evidence pertaining to the safety and efficacy of drugs. Based on its assessments, the FDA determines whether drugs can be marketed in the United States.

The Prescription Drug User Fee Act (PDUFA), enacted in 1992, authorized FDA to collect user fees from sponsors to help speed up the review of NDAs. It also established time goals for FDA’s reviews. In 1997, the FDA Modernization Act reauthorized user fees for another 5 years. It shortened the time goals and called for FDA to work more collaboratively with sponsors. In June 2002, the Public Health Security and Bioterrorism Preparedness Act of 2002 once again reauthorized user fees. The part of this Act addressing user fees is referred to as PDUFA III.

This inquiry focuses on FDA’s Center for Drug Evaluation and Research (CDER), which reviews NDAs. This inquiry does not assess the scientific merit of the decisions that FDA has made. Instead, it examines how well FDA carries out its NDA review process. This report draws heavily on the opinions of CDER officials. We surveyed CDER reviewers, receiving an estimated 47 percent response rate (N=401) and interviewed about 80 CDER officials, including managers. In addition, we surveyed sponsors, receiving a 60 percent response rate (N=72), reviewed files for all 15 new molecular entities approved by CDER in fiscal year (FY) 2001, analyzed CDER data regarding the number of advisory committees, observed 17 CDER meetings, interviewed 20 stakeholders, and reviewed relevant FDA policies and procedures.

We also drew on data from an internal survey conducted by CDER of a random sample of 188 reviewers that had a 72 percent response rate.
We conducted this inquiry prior to the implementation of PDUFA III. Where appropriate, we indicate the potential impact of PDUFA III on our findings.

FINDINGS

FDA’s new drug application review process has several strengths that contribute
significantly to its effectiveness.

Both FDA reviewers and sponsors have confidence in the decisions FDA makes. Our
review underscored that FDA’s NDA review process is science-based and comprehensive. This is supported by the comments of both FDA reviewers and sponsors. Seventy-eight percent of FDA respondents and 86 percent of sponsors indicated in our surveys that they were confident in the decisions FDA makes with regard to a drug’s efficacy.

FDA is highly responsive to the time goals required under the Prescription Drug User Fee Act and the FDA Modernization Act. In 1993, median total approval time for CDER was 27 months for standard NDAs classified as new molecular entities; in 2001, it was 19 months. The reduction in approval times helps to ensure timely access to new medications that can benefit the public health.

FDA works collaboratively with sponsors. In FY 2001, CDER conducted 1,021 formal meetings with sponsors. In these meetings, FDA provides valuable advice to sponsors that can help speed up the drug development process.

FDA has taken numerous steps to improve efficiency and consistency. In 2000, CDER issued about 40 guidance documents, most of which it directed to sponsors. Between 1996 and 2001, CDER issued about 140 policies to help guide reviewers. It also now accepts applications electronically.

FDA relies on expert scientific reviewers. Both sponsors and reviewers agreed that FDA’s in-house expertise is a key asset of the review process. Funds from user fees have allowed FDA to increase the number of employees for drug reviews by about 700 employees over the past 10 years.

But workload pressures increasingly challenge the effectiveness of the review
process.

Reviewers are under constant pressure to meet time goals. They not only review NDAs, but also other key documents submitted by sponsors, some of which also have time goals attached.
At the same time, reviewers must provide advice to sponsors and stay abreast of the latest scientific advances in their fields. Below, we present the consequences of these workload pressures.

Reviewer concerns about time pressures. Forty percent of FDA survey respondents who had been at FDA at least 5 years indicated that the review process had worsened during their tenure in terms of allowing for in-depth, science-based reviews. Respondents cited lack of time as the main reason. According to 58 percent of FDA respondents, the allotted 6 months for a priority review is inadequate. This is considerably higher than the 25 percent of respondents who indicated that the allotted 10 months for a standard review is inadequate.

Reviewer concerns about time constraints do not necessarily mean that there is a threat to public health. We have no evidence of a public health concern nor did we seek to obtain such evidence. Reviewers commented in interviews that they did not believe that they were ignoring key information or data contained in the applications in order to meet time goals. The FDA has also received the 4th highest composite score out of the 13 operating divisions within the Department of Health and Human Services on the 2002 Secretary’s Quality of Work Life Survey on Organizational Climate, which indicates a positive work environment. However, our survey data do indicate a significant management issue warranting attention.

The PDUFA III should help to address reviewers’ concerns about time pressures, as CDER estimates hiring close to 300 additional employees over the next 5 years with funds from user fees.

Less use of advisory committees. Advisory committees are comprised of independent scientific experts who provide advice to FDA during the review process. The number of advisory committee meetings CDER held for NDAs decreased from 40 in 1998 to 23 in 2001. Although the declining number of NDAs submitted by sponsors has contributed in part to this decline, FDA managers also pointed out that they have little time to hold these meetings.

Insufficient time for raising scientific disputes. Pressure to meet time goals may inhibit the raising of disputes. Reviewers may be reluctant to raise disputes due to concerns about slowing down the process. Twenty-one percent of FDA respondents indicated that the work environment allowed for the expression of differing scientific opinions to a small or no extent.

Contributing to staff turnover. The FDA data show that medical officers and
pharmacologists had the highest attrition rates within CDER for FY 2001, 8.4 percent and 6.9 percent respectively, compared to the overall rate of 5.5 percent. On an internal CDER survey, 50 percent of reviewers who responded indicated that their workloads are influential reasons to consider leaving FDA.

Less time for reviewers to participate in professional development and to conduct research to improve drug development. The FDA has policies and programs in place to encourage professional development, yet 59 percent of FDA respondents indicated that they have little time to participate in professional development activities. Similarly, reviewers have little time to conduct research on drug development using the clinical trial databases FDA has
obtained from sponsors.

Several factors have contributed to the workload pressures.

Time goals have been beneficial, but at the same time they have created pressure on reviewers to work quickly. The FDA has little flexibility in reassigning staff to handle increased workloads. The FDA’s dual roles as advisor and reviewer demand substantial time and resources; the CDER held over 1,000 meetings with sponsors in FY 2001. Incomplete and disorganized applications can cause delays. The 15 new molecular entities we reviewed
contained, on average, 38 amendments to the original application. Inefficiencies in the process also contribute to workload pressures.

As we have already indicated, PDUFA III will provide FDA additional resources to hire more staff that should help address these workload pressures. It also calls for FDA to conduct several studies aimed at improving the efficiency of the process.

Other factors also challenge the effectiveness of the review process.

Rush to finalize drug labels at the end of the process. Although labeling negotiations must occur toward the end of the process, we found that negotiations were considerably compressed. Eighty-two percent of FDA respondents indicated that labeling negotiations contribute to delays. Twenty-seven percent of labeling amendments for the 15 new molecular entities we reviewed were submitted in the last 14 days of the review process. The rush to
finalize labels at the end of the review process can be caused by the lengthy discussions that often occur between FDA and the sponsor regarding the information to include on the label.
The FDA has numerous activities underway to help address this issue.

Reviewers’ uncertainty about postmarketing commitments. Postmarketing commitments are made by the sponsor at the time of approval and can include additional studies to further define the safety and effectiveness of the drug. Reviewers indicated that they were often uncertain about what types of postmarketing commitments to request of sponsors. The PDUFA III calls for FDA to issue several guidance documents regarding risk management after the drug is approved. These documents should help to clarify the use of postmarketing
commitments.

Limited public disclosure of FDA’s rationale for decisions. We reviewed the information on CDER’s website for 15 new molecular entities, and in no case did FDA provide a summary document that explained the overall basis for approval. The FDA does not routinely provide summary information for approved drugs, nor is it required to do so. We found it took 7.6 months, on average, for FDA to post the technical information it does disclose on its website
after a drug is approved. For drugs that FDA reviewed but did not approve, FDA disclosed almost no information regarding the basis for its decisions. The FDA’s regulations limit such disclosure.

CONCLUSION

FDA’s NDA review process has several strengths. However, reviewers face
workload pressures that increasingly challenge the effectiveness of the process.

Beyond these pressures, three other factors threaten the effectiveness of the process: the rushed review of drug labels that takes place toward the end of the review process, the limited guidance available to reviewers in determining the extent and type of postmarketing commitments to request of sponsors, and the limited information that FDA makes available to the public on the basis for its decisions concerning NDAs. Overall, these findings present a
significant warning signal, one, that if not fully addressed, could jeopardize the gains that FDA has made in recent years.

The enactment of PDUFA III presents significant opportunities to address many of the findings in this report.

We recognize that FDA has already identified many of the concerns presented in this report and has numerous efforts underway to address them. In particular, the enactment of PDUFA III, which FDA played a critical role in developing with sponsors, presents significant opportunities to address many of our findings. It calls for an increase in user fees that CDER estimates will allow it to hire close to 300 additional employees over the next 5 years. Over
time, this could help considerably in relieving the workload pressures that we have emphasized.
In addition to resources, PDUFA III calls for FDA to conduct several activities aimed at improving the process. These activities will help to address many of the findings in this report, including efficiency, labeling negotiations, and the use of advisory committees. In addition, PDUFA III calls for significant attention to be placed on postmarketing commitments. For the
first time, funds from user fees can be used to monitor drugs after they are on the market. It also calls for FDA to develop several guidance documents on risk management.

Our first recommendation offers additional steps that FDA can take as it implements PDUFA III to ensure that it takes full advantage of these opportunities to address our findings. We also make four other recommendations to FDA to improve the NDA review process that are not addressed in PDUFA III. We direct all our recommendations to CDER.

RECOMMENDATIONS

1. Take full advantage of the opportunities in PDUFA III.

< Conduct a retrospective examination of recent NDA reviews to determine the capacity of reviewers to conduct in-depth, science-based reviews.

< Evaluate the adequacy of current staffing levels and the workload distribution among the 15 review divisions within CDER, and implement a system that in real time would indicate the status of an NDA and the time spent in reviewing its specific parts.

< Assess how amendments to the original application, internal processing delays, and labeling negotiations affect FDA’s capacity to make timely, first-cycle review decisions.

< Offer further guidance on the best way to handle scientific disputes that occur among reviewers, and how to balance the role of reviewing NDAs and the role of advising sponsors throughout the drug development process.

< Include case studies of past reviews as part of training programs for reviewers to illustrate good review principles and foster consistency among divisions.

< Provide a list of the various postmarketing commitments that FDA reviewers can request of sponsors and suggestions for when each could be considered.

2. Determine whether the significant workload pressures discussed in this report justify any exceptions to the current time goals regarding new drug applications to allow for more in-depth reviews.

As we have indicated, the Prescription Drug User Fee Act and the FDA Modernization Act have been positive forces for the review process. However, it is important that an appropriate balance exists between timeliness and the ability to conduct a comprehensive review. Our data show that reviewers face significant workload pressures. Therefore, FDA could examine if it would be beneficial to extend the review clock by 1 or 2 months when it chooses to use an advisory committee and to consider modifying the current 10-month time goal for standard NDAs by 1 or 2 months.

3. Reject applications that are incomplete and of poor quality that can create
delays in the new drug application review process.
A timely review process depends not only on FDA, but also on sponsors submitting complete and well organized NDAs. Toward that end, FDA could reexamine its policies for refusal-tofile and its guidance to sponsors on the content and format of applications to ensure that they make explicit FDA’s requirements.

4. Provide the public with a clear and timely explanation of decisions on new drug applications.
The FDA could provide on its website a succinct explanation of its rationale for approving an NDA. It could also provide the same explanation when it decides not to approve an NDA. Disclosing such information could help convey to the public the independent role that FDA plays in the review process and that FDA does not approve all drugs. Further, this could help sponsors gain a better understanding of the criteria FDA uses in its review process and could
lead to improved NDAs in the future.

5. Conduct or support research that takes greater advantage of its vast clinical trial databases to identify ways to improve drug development.
The results of this research, over time, could be highly cost-effective, contributing to better clinical trial designs and more efficient drug development.

AGENCY COMMENTS

The FDA reviewed a draft of this report, and overall, it concurred with our conclusions and recommendations. In its comments, FDA outlined numerous activities it has underway or planned to address our recommendations. Specifically, FDA indicated that it is reviewing its workload distribution and has studies underway to examine delays in the review process. The
full text of FDA’s comments can be found in Appendix A.