FDA Seeks New Strategies to Improve Clinical Research

New technologies and research methods aim to reduce study failures and spur drug development.

The Food and Drug Administration unveiled a report in March that aims to stimulate new thinking about how to convert new biomedical discoveries into safe and effective therapies more quickly and more efficiently. A main objective is to reverse the recent decline in new drug applications (NDAs) filed with the agency, especially for innovative treatments. Former FDA commissioner Mark McClellan frequently lamented the lag between gains in genomic and other cutting-edge biomedical research and the development of new medicines able to improve public health.

This report, entitled Innovation or Stagnation?Challenge and Opportunity on the Critical Path to New Medical Products,¹ aims to address this conundrum by identifying obstacles and possible solutions to reducing the high failure rate of potential new medical products during the development process. At the same time, FDAs Center for Drug Evaluation and Research (CDER) is examining ways to make its review process more flexible and efficient (see Changes at CDER sidebar).

Downstream focus
The report notes growing interest in translational research at the National Institutes of Health (NIH), nonprofit research organizations, and European institutions that aims to accelerate the transfer of basic biomedical research into the clinic. As the next step, FDA is looking to provide new tools to improve downstream product development related to preclinical safety testing, clinical efficacy studies, and manufacturing processes. This involves developing assays, standards, computer modeling techniques, biomarkers, and clinical endpoints better able to demonstrate safety and effectiveness.

Such tools would allow earlier identification of those products that do not hold promise, thus reducing wasted time and resources on unpromising R&D. The vast majority of investigational products that enter clinical trials fail, the report notes, pointing the need to move away from cumbersome assessment methods that make the path to market for even successful drug candidates long, costly and inefficient. If a drug does not show efficacy in Phase III studies, asks Janet Woodcock, currently FDA acting deputy commissioner and author of the critical pathways report, how did it get that far?

FDA wants to work with industry to help identify problems common to similar products, tapping into the agencys internal information resources on failed drug development programs. FDA holds the only broad, cross-cutting knowledge about how certain investigational products fail, why certain therapeutic areas remain underdeveloped, and when certain development hurdles persist despite advances in technology that could mitigate them, the report explains. The agency seeks more publicprivate collaboration on genomics, proteomics, bioinformatics systems, and new imaging technologies to help detect safety problems early, identify patients likely to respond to therapy, and address product design, characterization, and manufacturing issues.

One obstacle to efficient drug development is that many in vitro screening techniques and animal models often are unable to identify drug candidates with a high probability of effectiveness. FDA believes that greater understanding of the nature of disease will lead to a better understanding of how humans respond to treatments. Such knowledge can permit a shift away from reliance on highly empirical large-scale clinical trials that are unable to control for sources of variability.

Making a list
FDA is launching this initiative by creating a critical path opportunities list. Agency staffers are examining opportunities for developing standards and policies most likely to remove or overcome specific obstacles to drug development. FDA also is seeking additional suggestions from manufacturers, academics, and NIH researchers involved in its roadmap initiative to improve clinical trials. These activities were discussed at the April meeting of FDAs Science Board, along with proposals to devote a portion of the agencys $135 million research budget to critical path projects.

One model lies in FDAs Orphan Products grant program, which provides modest funding ($150,00$300,000 a year) to support development of treatments for small patient populations. Since 1989, the program has yielded 36 novel medical products to treat rare diseases, contributing significantly to medical progress.

Gaining information earlier
Another strategy cited in the report is to use more proof-of-concept trials to confirm drug activity in humans before launching full-scale product development. This may involve development of new statistical models of drug efficacy and safety able to determine the relationship between drug dose, plasma concentration, pharmacokinetics, and pharamacodynamics.

Similarly, Robert Temple, associate director for medical policy at CDER, proposes a range of design tricks for Phase II studies that may give sponsors important information on a drugs potential efficacy and appropriate dosing. Some of these approaches include:

• Enriched patient selection. Temple considers it acceptable to try to select study participants most likely to comply with and respond to treatment. Better understanding of the biological and genetic factors underlying a disease can help identify individuals more likely to benefitor less likely to be harmedfrom a therapy and thus improve chances of a study showing efficacy. Information on potential patient response can allow sponsors to conduct small studies of nonresponders and of those intolerant to standard therapy, which may be useful in showing a particular value of a new agent. There may be debate later about the generalizability of the study data, Temple acknowledges, but at least the sponsor gains assurance of some efficacy, which, he notes, is what Phase II is for.
• Randomized withdrawal studies. Instead of initially assigning individuals to either a test or a placebo study arm, participants enter an open treatment study, and responders then are randomized to control or treatment. This approach is useful to assess long-term effectiveness and also is attractive for pediatric trials because it permits a shorter period without treatment. It also tends to be enriched with responders, again useful if the aim is to document an effect before proceeding with larger studies.
• More dose-finding studies. Temple advises sponsors to examine a broad range of doses in early studies and to continue dose/response studies in larger Phase III trials that are better able to detect small differences in effect and provide useful information on safety. Early dose/response studies should be carried out in known responders to increase study sensitivity or in nonresponders to see if much higher doses have an effect on this population. It also would be useful to learn whether the typical dose-response curve represents increasing response in all patients, or if some patients have a full response to a low dose while others respond more to larger doses. These studies also can examine maintenance doses for long half-life drugs.

In its search for new tools and strategies to improve clinical trials, FDA supports efforts to identify new biomarkers and surrogate endpoints that can be linked with clinical outcomes. New imaging technologies may provide insight into the distribution, binding, and other biological effects of pharmaceuticals, which may lead to important biomarkers. FDA seeks to identify clinical trials under development that could address uncertainties and to explore further how pharmacogenomics and proteomics can help target responders, monitor clinical response, and serve as biomarkers of drug effectiveness.

References
1. See www.fda.gov/bbs/topics/news/2004/NEW01035.html.

SIDEBAR: Changes at CDER
To better handle these new challenges, FDA officials are exploring ways to reorganize staff and procedures. John Jenkins, director of CDERs Office of New Drugs (OND), is contemplating a reorganization of his six review offices.

The current 10-year-old structure is based on clinical indications and product categories that no longer make sense in many cases, Jenkins notes. Some offices, such as the division of neuropharmacological drug products, are overloaded with applications, while others receive only a handful. The oncology office receives half of CDERs INDs (investigational new drug applications), but few NDAs. Jenkins expects to retain an indication-based structure, but to improve the logical groupings of the office to better balance the workload.

The changes at OND will involve integrating reviewers of applications for biotech therapies into this indication-based review structure. Staffers transferred to CDER last year from the Center for Biologics Evaluation and Review (CBER) currently remain in a separate new drug evaluation office. The plan is to add the oncology reviewers from CBER to the larger CDER oncology division and to disperse the others among CDER review groups.

CDERs Office of Pharmaceutical Science (OPS) also plans to reorganize how its Office of New Drug Chemistry (ONDC) reviews manufacturing data in applications. ONDC currently assigns review chemists to work directly with certain OND clinical review groups. With the emergence of new products that require special expertise to assess CMC data and manufacturing systems, OPS Director Helen Winkle seeks to establish more flexible review teams better able to adjust to changing work demands. ONDCs three chemistry review divisions now operate in silos, Winkle comments, and may lack specific expertise to address issues related to a new product. However, Winkle expects to leave intact OPS Office of Biotechnology Products, which houses former CBER staffers who conduct research and review the manufacturing portions of biotech therapy applications. While the clinical portion of biotech applications is fairly similar to that for drugs, the manufacturing sections are quite different.

These changes are likely to occur after much of CDERs staff moves to the newWhite Oak FDA office complex next year, a major change that will bring most drug, clinical, and chemistry reviewers under one roof for the first time in decades. Driving all these changes at CDER is pressure to maintain timely review processes despite increasingly tight resources.JW