| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
COMMENTARIES |
From The Critical Path Institute, Tucson, Arizona.
Address for reprints: Address correspondence to: Raymond L. Woosley, MD, PhD, The Critical Path Institute, 4280 N. Campbell Ave, #214, Tucson, Arizona 85718; e-mail: RWoosley{at}c-path.org.
Key Words: FDA • modeling • simulation • clinical trial
The paper by Wang et al in this issue of the Journal carries a message from the scientists at the Food and Drug Administration (FDA) that is both important and timely because of the unacceptably high and rising failure rate in drug development. In this paper, the authors encourage early and frequent communication with the FDA during new drug development and the use of modern pharmacometric methods. The experience described in this report is an important outcome of a pilot "end of phase 2A" program in the FDA's Critical Path Initiative (CPI). The case studies included in this report provide those working in new drug development concrete examples of the potential benefit of applying modeling and simulation in drug development and the value of undertaking discussions with the FDA early in drug development.
The FDA has been a leading force in advancing the applied sciences in drug development. In the 1980s and '90s FDA scientists such as Carl Peck, Larry Lesko, and Tom Ludden, with their former colleague Lewis Sheiner, demonstrated the value of pharmacometric methods to bring a higher level of quantification and prediction to the fields of pharmacokinetics, pharmacodynamics, and pharmacogenomics.1-4 The use of population pharmacokinetics, mixed-effect modeling, and Bayesian statistical approaches germinated at the FDA and have become standard tools in drug development. Through the encouragement and initial exploration of Peck, his colleagues, and students, the field of modeling and simulation has been applied to clinical trial design and drug development.5 Many large pharmaceutical companies and commercial service organizations have developed teams of scientists with the skills to employ modeling and simulation in the design of clinical trials. The paper by Wang et al is further demonstration of the general acceptance of this exciting and powerful quantitative tool.
The best examples of the value of modeling and simulation are those in which the number of variables to be managed exceed the ability of the human mind to comprehend, and the computer enables scientists to better predict the outcomes and the relative importance of the many factors influencing those outcomes. Case study 2 in this report demonstrates how the modeling and simulation process can incorporate pharmacogenetic variation and biomarkers to improve the predictive accuracy in drug development and begin to accommodate the needs of personalized medicine.
However, the accuracy of predictions from modeling and simulation is dependent upon the quality and availability of the data that populates the model. Wang et al were able to gather data from multiple previously submitted new drug applications in order to perform the simulations described in this report. Yet, because of the proprietary nature of drug development, there are limits on how information and data available in a new drug application can be shared with other companies and the scientific community. However, this can be inefficient, and it can be dangerous if companies unknowingly pursue paths that have previously failed for other companies. While the FDA must protect trade secrets, it is extremely valuable when they can use their knowledge and observations to advance the process of medical product development. Likewise, because any given company's experience in a therapeutic area is often relatively limited, the opportunity for the FDA to share experience and data from the development of other agents can be extremely useful and serve the public health. In the examples given by Wang et al, the FDA was able to take data from previously submitted new drug applications to create models enabling simulation of clinical trials. For example, they used disease progression rate, placebo effect, patient dropout rates, and the relation of biomarkers to primary disease endpoints. This information is extremely valuable and is rarely available in the published literature. When it is available, it is rarely current or in the detail needed to aid in designing clinical trials.
For the FDA to be able to continue advancing the science of drug development and to offer the kind of expertise exemplified by Wang et al, it will need additional resources. Although the CPI was launched in March of 2004, at the time of this writing, the FDA has yet to receive Congressional appropriations to support the initiative. Legislation under consideration at this time and the newly authorized "user fee" legislation includes new positions and new but relatively modest funding to support the CPI. Congress, for the first time, is recognizing that in order for the FDA to be an effective regulator of the industry, it must conduct research in applied sciences. This research is quite different from the basic science research supported by the National Institutes of Health or the National Science Foundation.
Most successful businesses invest some fraction of their revenues to improve the process by which the make new products. Because the pharmaceutical industry is regulated by the FDA, any investment in "process improvement" should best include the active participation of the FDA or at least a willingness to consider a change. Industries such as the semiconductor (computer chip) industry have developed an organization like Sematech in which the companies can work together to develop new methods and establish the standards needed to advance the field. The food industry joined with the FDA to create the National Center for Food Safety Technology where industry scientists work with scientists from the FDA and academia to develop better methods to assure the safety of foods. The CPI and the further development of tools such as modeling and simulation to advance drug development will require the combined efforts of scientists from the FDA, the industry, and scientists from a broad range of disciplines. The pharmaceutical industry is being well served by the CPI activities and are encouraged to actively support and participate in the sharing of "precompetitive" data from their development programs. The Predictive Safety Testing Consortium was one of the earliest and has been one of the most successful projects in the CPI that focuses on the cross-qualification of more sensitive and specific laboratory tests of the safety of new drugs.6
To improve the clinical testing of drugs, the data from the control or comparator arms of clinical trials could be pooled and made available to the FDA and the public without having to employ the "freedom of information" tactic. Continued progress on the critical path initiative will require that the talented scientists at the FDA be encouraged and incentivized to participate in projects such as the one described by Wang et al. This will require the recruitment of additional scientist to the FDA and novel modes of collaboration.
ACKNOWLEDGEMENTS
Financial disclosure: None declared.
REFERENCES
1. Peck CC, Barr WH, Benet LZ, et al. Opportunities for integration of pharmacokinetics, pharmacodynamics, and toxicokinetics in rational drug development. Clin Pharmacol Ther. 1992; 51(4): 465-473.[Web of Science][Medline] [Order article via Infotrieve]
2. Pfister M, Labbe L, Hammer SM, et al. Population pharmacokinetics and pharmacodynamics of efavirenz, nelfinavir, and indinavir: Adult AIDS Clinical Trial Group Study 398. Antimicrob Agents Chemother. 2003; 47(1): 130-137.
3. Lesko LJ. Paving the critical path: how can clinical pharmacology help achieve the vision? Clin Pharmacol Ther. 2007;81(2): 170-177.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
4. Kraiczi H, Jang T, Ludden T, Peck CC. Randomized concentration-controlled trials: motivations, use, and limitations. Clin Pharmacol Ther. 2003;74(3): 203-214.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
5. Lee H, Kimko HC, Rogge M, Wang D, Nestorov I, Peck CC. Population pharmacokinetic and pharmacodynamic modeling of etanercept using logistic regression analysis. Clin Pharmacol Ther. 2003;73(4): 348-365.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
6. Woosley RL, Cossman J. Drug development and the FDA's Critical Path Initiative. Clin Pharmacol Ther. 2007;81(1): 129-133.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
CiteULike 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
