FDA Performance Goals for Approving Drugs and Biologics

The omnibus spending bill for fiscal year (FY) 2009 that President Obama signed on March 11 contains a milestone for the US Food and Drug Administration (FDA): it is the first time Congress has appropriated $1 billion for the FDA to regulate human drugs and biologics.¹ The total amount Congress appropriated for the FDA in FY 2009, including user fees from industry, is $2.62 billion, which is an increase of $224.3 million, or 9%, from the amount Congress appropriated for the FDA in FY 2008 (including a special supplemental appropriation of $150 million in June 2008). The proposed budget for FY 2010 that President Obama sent to Congress on May 7 would provide the FDA with an additional $511 million (including user fees) for a total budget of $3.2 billion—an increase of 19% from FY 2009. If enacted, this would be the largest annual budget increase in the FDA's history.²

The budget increases in recent FYs have been substantial for the components of the FDA that are primarily responsible for regulating human drugs and biologics: the Center for Drug Evaluation and Research (CDER) and the Center for Biologics Evaluation and Research (CBER). In FY 2006, Congress appropriated $699.3 million for CDER and CBER, including industry user fees. That amount increased in FY 2007 to $776.8 million, an increase of 11%. In FY 2008, the appropriation (including the June 2008 supplemental appropriation) for CDER and CBER increased to $960.5 million, an increase of 24%. The appropriation for CDER and CBER in FY 2009 is $1.05 billion, and President Obama has proposed increasing this amount to $1.2 billion in FY 2010. These budget increases for the FDA as a whole, and especially for CDER and CBER, have been a welcome change for this small public health agency that has been chronically starved for money.

Almost half the money Congress has appropriated for CDER and CBER is in the form of user fees paid by the pharmaceutical industry under the Prescription Drug User Fee Act (PDUFA).³ President Obama's proposed budget for FY 2010 contemplates increasing substantially the PDUFA user fees and adding new user fees. There is nothing problematic, at least in principle, with user fees paid by industry. For example, the nuclear industry pays 90% of the budget of the US Nuclear Regulatory Commission in the form of user fees. In the United Kingdom, the Medicines and Healthcare Products Regulatory Agency is supported largely by user fees, as are its counterparts in the European Union and in Japan. Unlike user fees paid to other agencies, however, the PDUFA user fees are implicitly tied to detailed “performance goals” for the FDA's review of drugs and biologics.⁴ It is the performance goals, not the user fees, that are problematic for the public health.

The PDUFA was first enacted in 1992 and since then has been reauthorized every 5 years, most recently in October 2007. Congress provided that user fees under the PDUFA “will be dedicated toward expediting the drug development process and the process for the review of human drug applications, including post market drug safety activities.”⁵ Under the PDUFA, each pharmaceutical company pays the FDA for submitting a New Drug Application (NDA) or Biologics License Application (BLA). For FY 2009, the fee for filing an NDA or a BLA that includes clinical data is $1.25 million. In addition, each pharmaceutical company pays an annual fee for each approved drug or biologic and for each manufacturing facility. The FDA calculates the amount of the total fees for each FY using a complex formula specified in PDUFA, and Congress then appropriates this total amount. For FY 2009, the total estimated fees to be paid to the FDA under PDUFA are expected to total $510 665 000.⁶ This amounts to 20% of the FDA's total FY 2009 budget and 49% of the total budget for CDER and CBER. Congress permits the FDA to spend PDUFA user fees only on specified activities related to expediting review of NDAs and BLAs and some limited postmarketing safety activities.

While Congress requires the FDA to negotiate performance goals with the pharmaceutical industry, with limited public comment,⁷ the performance goals are not enacted into law, and there is no explicit legal requirement that the FDA meet the performance goals. Instead, there is an implicit, and inappropriate, political bargain that the industry will accept user fees only in exchange for the FDA's commitment to meet the performance goals. When then Secretary of Health and Human Services Michael Leavitt sent the current set of performance goals to Congress in 2007, he stated in his transmittal letter that the goals outlined “the agreements between the Agency [ie, the FDA] and the industries with regard to the application approval timeframes . . . and other activities to be supported by user fees,” and that the goals “represent the commitment of the Department and the FDA to carry out the goals under mutual agreement with the industries.”⁸ There is no pretense that the PDUFA performance goals are anything other than the “mutual agreement” between the regulator and the industry it regulates.

The current PDUFA performance goals take up more than 20 single-spaced pages of text broken down into 14 separate sections—only 1 of which deals explicitly with drug safety. Most of the other sections focus on timetables for FDA action on industry applications and requests for meetings. For example, the FDA is obligated to make a decision on 90% of “standard” NDAs within 10 months of “receipt” of the application. (A “standard” NDA is one for a drug that is not a significant therapeutic improvement compared with other available drugs or therapies.) The 10-month time frame or “review cycle” for an FDA decision includes the 2 months that the FDA requires to determine that an NDA received is sufficiently complete to be accepted for filing.⁹ The time frames for FDA action are shorter for “priority” drugs that reflect a therapeutic advance.

In 2008, Carpenter et al¹⁰ reported (as corrected) that drugs approved within 2 months before the performance goal deadline for that drug were significantly more likely to have postapproval safety issues, with odds ratios ranging from 2.1 to 3.6. The authors concluded that “it appears to be the deadline, not the speed of approval, that explains the difference in the risk of such problems.”¹⁰ The observation that the deadline is the problem is consistent with the FDA's own PDUFA guidance, in which the FDA notes the potential for “errors associated with crisis-style management dealing with unresolved issues at the end of the review cycle.”¹¹ Crisis-style management of drugs and biologics, driven by the performance goals, is not consistent with public health.

Another questionable performance goal deals with Special Protocol Assessments (SPAs). An SPA is an agreement by the FDA that if a drug company conducts a clinical trial according to a protocol that has been reviewed and approved in advance by the FDA, and if the results of the trial are positive, then the FDA will permit the trial to be used to support the approval of the drug. The SPA performance goal is based on the FDA Modernization Act of 1997, which provided that if the FDA agrees in writing to a protocol for a clinical trial, it cannot later change its position unless it determines that “a substantial scientific issue essential to determining the safety or effectiveness of the drug” was identified only after the trial began.¹² The FDA Modernization Act of 1997, however, did not obligate the FDA to make any such agreement or to explain its reasons for not agreeing to do so. In contrast, the user fee performance goal obligates the FDA, on request, to either enter into an SPA or provide the drug company with a written statement of its reasons for not doing so—and to provide that statement in 90% of cases within 45 days after the FDA's receipt of the protocol. (The FDA estimates that in the current FY, industry will request 485 SPAs.) Having agreed to the protocol, the FDA may renege, according to the performance goal, only if there are “public health concerns unrecognized at the time of protocol assessment.”¹³

There is a significant difference between the FDA Modernization Act of 1997 statutory standard for a change in the FDA's position on a clinical trial—“a substantial scientific issue” essential to safety and effectiveness—and the negotiated performance goal standard for an FDA change in position—“public health concerns unrecognized at the time of protocol assessment.” The question here is less whether these 2 formulations would lead to different outcomes in practice (although the performance goal of “public health concerns” would appear to be a far higher standard for the FDA to satisfy). Rather, the contrasting formulations are important because they reflect the lack of congressional or public oversight of the essentially private process of negotiating the performance goals between the FDA and the pharmaceutical industry. While pharmaceutical and biotechnology companies sometimes tout the existence of SPAs to give their investors confidence that, if their trials are positive, FDA approval of their products will surely follow, SPAs are confidential unless the company makes them public. This provision insulates SPAs from public review.

The PDUFA linkage between user fees and the performance goals is not only unique in the federal government, it is unnecessary. On June 22, President Obama signed into law the Family Smoking Prevention and Tobacco Control Act, which grants the FDA sweeping authority to regulate tobacco products.¹⁴ Under this important statute, the cost of regulating tobacco by the FDA, which will be $235 million in FY 2010, is to be paid entirely by user fees from the tobacco industry. Yet nothing in this statute contemplates that the FDA will privately negotiate performance goals for tobacco regulation with the tobacco industry. This legislation shows that performance goals negotiated with industry are not essential to a user fee system.

The FDA's performance goals, as currently structured, seem inappropriate and appear to have a deleterious effect on public health. Congress should revise the PDUFA either by eliminating the performance goals altogether or, at least, by eliminating the implicit linkage between the performance goals and the FDA's funding. If performance goals are retained, they should be established by Congress in the normal, open legislative process. In addition, the performance goals, if retained, should address the FDA's entire range of public health responsibilities in regulating drugs and biologics instead of focusing merely on expediting the FDA's review of NDAs and BLAs. Finally, Congress should take action as soon as possible and not delay until the reauthorization of the PDUFA in 2012.