Safety versus Speed in Drug Development

The unstated tradeoff for drugmakers' increased investment in postapproval drug-safety monitoring is supposed to be greater US Food and Drug Administration flexibility in approving new drugs for market. In reality, the agency's burden of implementing a host of new drug-safety policies and programs has overtaxed staffers and slowed the new-drug review process. Outside criticism and internal dissent have heightened reviewers' fears about approving new products that raise safety concerns, thus exacerbating FDA's struggle to fulfill its many new statutory requirements and to meet review goals set by the Prescription Drug User Fee Act (PDUFA).

Jill Wechsler

Janet Woodcock, director of the Center for Drug Evaluation and Research (CDER), acknowledges these challenges. Drugs are getting approved, but more slowly, she explained at the FDA Regulatory Symposium in September 2009. The problem is that the FDA Amendments Act of 2007 (FDAAA) added new requirements for assessing postmarketing studies and determining the need for Risk Evaluation and Mitigation Strategies (REMS) at the time of approval, but with no extension in review times. CDER has hired more staff and is implementing new procedures to make the review process more efficient. Electronic submission and review systems for drug applications would help, Woodcock noted. But these innovations remain far from reality, and many applications take much more than one review cycle to gain market approval.

As a result, the US may be falling behind Europe in approving innovative therapies for patients. At a meeting in September 2009 sponsored by the Institute of Medicine (IOM) to evaluate safety initiatives established by FDAAA, Peter Honig, executive vice-president of Merck (Whitehouse Station, NJ), noted that first-cycle approvals are down and that user-fee approval dates are "routinely missed" because of increased scrutiny of safety issues. FDA is "clearly struggling" with postmarket safety demands, he said, adding that "drug lag" may be rearing up once more as European regulators approve some new drugs for market faster than US regulators.

The approval slowdown is noticeably affecting drugs to which the agency has granted priority-review status, a designation traditionally reserved for the most innovative and important new therapies. In the past, about 70% of priority-review applications gained first-cycle approval, but this proportion dropped to 50% in 2008, according to a recent report from Parexel Consulting. User-fee approval targets are 10 months for new drug applications (NDAs) and six months for priority applications. It is particularly difficult for reviewers to accelerate the latter review goal.

The good news is that signs indicate a surge in NDAs filed with FDA and a more robust pipeline for new drugs. Parexel cited 147 NDAs pending at FDA at the beginning of 2009, a notable increase from the 86 under review one year earlier. The number of applications for new molecular entities (NMEs) appears to be holding steady, but the percentage enjoying first-cycle reviews has been dropping slightly, but steadily.

FDA to revise approval process for combination products and medical devices

Safety first

The main culprit in the NDA-review slowdown is pressure on FDA to implement FDAAA postmarketing safety requirements, which is an enormous task for the agency and a growing burden on manufacturers. CDER's Safety First initiative aims to meet the challenge by integrating drug-safety activities and establishing a broader approach to ensuring appropriate drug use throughout the product life cycle.

The REMS program is the most visible new assignment. CDER has approved 63 new REMS in the past two years, 47 of which only require pharmacists to hand Medication Guides to patients filling new prescriptions. Ten of the REMS have additional communication plans that usually involve letters to healthcare professionals about product risks and how to prevent them, and six REMS include Elements to Assure Safe Use, which may impose restrictions on prescribers or patients, limit drug distribution, or require special monitoring.

CDER's Office of Surveillance and Epidemiology has expanded its staff to assess REMS plans, among its other assignments, but the backlog of REMS evaluations is growing, and more work is on the way as requests for periodic assessments emerge. The process of devising, proposing, and negotiating a REMS with FDA is complex and time-consuming, as seen in the long-awaited draft guidance on REMS content and format for drugs and biologics, which FDA issued on Sept. 30, 2009. Even a Medication Guide-only REMS requires a manufacturer to explain why a moderate postmarketing strategy is sufficient to ensure safe product use, and to allow FDA to determine that patient information or more detailed directions for use could prevent serious adverse events.

The need for a restricted REMS program, which requires considerable analysis to develop and implement, will be determined based on the size of the patient population, the seriousness of the disease, the drug's benefits, the duration of treatment, and the known adverse-event profile. The REMS rules for generic drugs are even more complex and will be addressed in further guidance.

The draft guidance describes how manufacturers should submit a REMS proposal to FDA to appropriately explain the risks addressed, program goals and elements, materials involved, and how and when the plan will be implemented. A REMS supporting document should provide a thorough explanation of the rationale for the program and give details about how the elements or tools in the REMS are expected to mitigate risks without disrupting established drug distribution and dispensing systems.

An important REMS component is a timetable for assessing the program after 18 months, three years, and seven years, or more often if warranted. The policy spells out detailed procedures for modifying a REMS after it is approved and adopted if goals are not met or circumstances change. FDA also may unilaterally modify a REMS if new safety or effectiveness information emerges.

A major challenge for sponsors and for FDA is to determine how best to measure whether a REMS is effective. FDA suggests that sponsors conduct surveys, collect prescriber information, or establish active systems to assess these programs. Although it's fairly straightforward to verify that pieces of paper have been handed out as part of a Medication Guide distribution program, it's much trickier to determine whether risk information influences prescribers' and patients' behavior. And manufacturers cannot do much to compel practitioners to report adverse events or to limit inappropriate prescribing if physicians ignore advisories and restrictions.

The pages of the detailed guidance on REMS submissions plainly show why FDA's review of these plans takes so much time. Manufacturers report that information requests and negotiations with FDA on REMS can go on for months. If FDA determines late in the review process that a product requires a more comprehensive REMS program than that initially expected by the sponsor, it's unlikely that the application will be approved in one review cycle, explained John Jenkins, director of CDER's Office of New Drugs, at the IOM workshop.

An important decision for manufacturers is whether it's better to propose a REMS voluntarily before filing an NDA, or to wait and see whether FDA reviewers determine that such a program is necessary. No one wants to implement a REMS if it's not required, because the process consumes considerable resources and elevates the risk associated with the product. But FDA seems to be requiring REMS for most NMEs, and developing a REMS during the application review period will slow the approval process. More drugmakers are talking about bringing up REMS issues at end-of-Phase-II meetings with FDA to avoid surprises later on.

Whatever tack a company takes, it's important to get the details of a REMS right and to establish reasonable and practical goals and timetables. Manufacturers that violate a REMS requirement face fines as high as $10 million and the possibility that FDA will pull the product off the market.

Timely postmarketing studies

Periodic REMS assessments also include updated information about postapproval studies or clinical trials undertaken by the sponsor, including study status, expected completion date, and whether difficulties have been encountered in meeting set goals. Similarly, this information is provided in annual reports on postapproval studies, as required by FDAAA to address complaints that pharmaceutical companies have failed to complete agreed-on postmarketing studies in a timely manner. Congress authorized FDA to mandate postapproval studies that achieve specific objectives and to levy fines on companies that miss agreed-on deadlines.

The result is that FDA reviewers now have to determine the scope and timing of postmarketing studies as part of the complex application-review process. Instead of merely vetting manufacturers' study proposals, agency staffers must examine what scientific data can best assess known and unknown serious risks and set appropriate timeframes for the completion of postmarket studies and trials.

This scrutiny of postmarketing studies seems to be having a notable effect. FDA's latest annual report on the program showed considerable industry progress in achieving "on schedule" status for fulfilling postmarketing study obligations. About 20% of studies are ongoing (i.e., were started on schedule) or submitted to FDA, and most are "pending" (i.e., not yet started, but not delayed). This category includes many studies involving pediatric populations, which have to wait until all other safety information is submitted and reviewed.

FDA should improve its ability to track and review postmarketing studies following Booz Allen Hamilton's analysis of these obligations. The contracted analysts cleaned up FDA's confusing database of about 1531 postmarketing studies, identifying many that have been submitted to FDA or that the agency no longer considers necessary, and determining that more than 80% are proceeding according to established timelines.

FDAAA also authorizes FDA to require additional postmarket studies for approved drugs and to require changes in labeling when new safety information emerges. The agency sent 14 letters requiring additional studies from mid-2008 to mid-2009 and issued 18 safety-labeling notification letters during that period. These additional tasks tap the agency's resources.

Because some postmarketing studies are likely to involve thousands of patients, CDER plans to ask advisory committees to weigh in on acceptable trial designs, numbers of patients, and endpoints. These committee sessions will address study designs for classes of drugs—rather than for each individual product, which would only extend approval times even more. And broader public buy-in for innovative studies may help FDA parry second-guessing later on if research is delayed or changed.

Streamlining reviews

Woodcock hopes that CDER's 21st Century Review initiative will deal with these new challenges by better managing the application-review process. The aim is to compress review times up front to provide more leeway at the end of the review period to address safety issues and resolve internal and external disputes. But streamlining reviews is a considerable challenge, Woodcock explained at the symposium, because the average NDA consists of 10 gigabytes of material. The data may no longer come in truckloads, but CDER reviewers still have to digest a massive amount of information.

CDER conducted a pilot program last year that reviewed 17 applications for NMEs under the new review process. This approach now is being extended to most NMEs and new biologics license applications and is slated to affect all applications and supplements by 2012. The streamlined system involves meeting in advance with manufacturers to identify key issues and ensure that applications are complete when submitted. Earlier internal deadlines have been set for establishing multidisciplinary review teams, for identifying application deficiencies, and for reviewing product labeling. FDA will assess up front whether an advisory committee meeting will be needed, what potential postmarket safety requirements might apply, and other possible "showstoppers" that could delay the approval process.

The agency and industry are optimistic that these efforts, along with staff expansion and training, will get CDER back on track with review goals and deadlines. But any changes must support FDA's capability for assessing and preventing drug-safety problems and ensuring safe drug use throughout the product life cycle.

Jill Wechsler is Pharmaceutical Technology's Washington editor, 7715 Rocton Ave., Chevy Chase, MD 20815, tel. 301.656.4634, jwechsler@advanstar.com