Looks like they just left Bio 2003. Here is an article from that conf. that has a quote from GNSC.
June 22 - 25
BIO 2003 Washington D.C.
Pharmacogenomics a Powerful Tool Requiring New Regulations
by George Miller
bio.org
Among many other uses, genomic technology is now helping identify patients in whom a given drug effect is more precisely predictable than in patients at large.
Heralded as a revolutionary advance five years ago, the application -- pharmacogenomics -- is already influencing drug development and medical practice, while posing new challenges for FDA regulators, according to speakers at BIO 2003.
For biotechnology companies struggling to find the best indications and patients for their medicines, pharmacogenomics can be a boon.
At Millennium Pharmaceuticals, researchers used the technology in the development of Velcade, a newly approved treatment for multiple myeloma, according to Jeffrey S. Ross, MD, Vice President for Molecular Medicine and Head of Molecular Pathology at the Cambridge, Mass., company. Millennium used genome-wide profiling in a Phase II clinical trial to help predict the likelihood of a patient's response to the drug, which eventually enjoyed an accelerated approval from the FDA.
An added benefit of the profiling, according to Dr. Ross, is that it advanced researchers' understanding of less-understood mechanisms at play in the drug's action. Dr. Ross was one of several speakers discussing pharmacogenomics and cancer Tuesday at BIO 2003.
Regulators are as enthusiastic about the technology as drug developers. According to Janet Woodcock, Director of the FDA's Center for Drug Evaluation and Research, pharmacogenomics is leading to nothing less than "a move from the current empirical process [of drug development and approval] to a mechanism-based process that is hypothesis driven, resulting in lower-cost, faster, more effective, less toxic treatments for smaller populations.
But pharmacogenomics also necessitates changes in the regulatory approval process, and some companies are reluctant to share such information for fear of slowing down down the approval for their product.
"We need free and open exchange of results between FDA and industry to ensure appropriate development of policies," said Dr. Woodcock. Concern about how regulators might use the data has stifled this exchange.
There's a lot of variability in human response to drugs, said Dr. Woodcock. Yet each drug has a consistent pattern of side effects over placebo rates. Current medical application, which operates at the level of organs, is not able to predict this pattern.
"There is an inherited [genetic] component to this variability," said Dr. Woodcock. Efficacy response is determined by genetic diversity and variable drug metabolism, making the optimal dose difficult -- if not impossible -- to predict for individual patients. "Metabolisms handle drugs wildly differently," said Dr. Woodcock.
Current drug development provides a "satisfactory determination of efficacy," she said, but on a population basis. Pharmacogenomics promises to bring that down to the individual level.
Potential uses of pharmacogenomics in drug development, she said, are to improve candidate drug selection; to develop new sets of biomarkers for toxic responses; to eventually minimize animal studies; to predict who will respond and who will have serious side effects; and to rationalize drug dosing.
"The challenge we face is how to integrate pharmacogenomics smoothly into the existing regulatory process. Existing statute (the Food, Drug and Cosmetic Act) requires well-controlled trials and the test of all methods."
Dr. Woodcock said the FDA has a working proposal for how to do that, although it's "not ready for prime time yet. We will establish policies for what types of data are required to be submitted--what types of data are appropriate for regulatory decisionmaking."
Within CDER is the Office of Clinical Pharmocology and Biopharmaceutics, directed by Larry J. Lesko, PhD. Dr. Lesko said that OCPB is planning workshops to facilitate the needed exchange of ideas with industry. The office has also developed a "safe harbor" provision, called Genomic Data Submission, to help "strike a balance between openness and uncertainty. We have to work within the regulations."
A draft guidance is being written for August. A workshop is planned for Washington, D.C., in November , by which time Dr. Lesko expects comments will have come back on the draft guidance and been reviewed. Goals for the November meeting are "to increase awareness of the Genomic Data Submission proposal, to see if it fits, and to define and format the submission process."
Balancing Risk, Benefit and Cost
Drs. Lesko and Woodcock, both from the FDA, see the use of pharmacogenomics in drug development and approval from a traditional regulatory viewpoint: risk vs. benefit. The manufacturer point of view, however, is different. "Stakeholders look at risk/benefit vs. cost," said Paul Radensky, M.D., J.D., a partner at the law firm McDermott Will & Emery in Miami.
He described how drug companies, with this different perspective, may decide that some efforts just aren't worth the payback. How much clinical benefit is provided? At what cost?
"This is where pharmacogenomics may very well help us think through regulatory matters in a different way," said Dr. Radensky. "Genomic markers will help with difficult public policy questions, but will not likely provide complete answers."
This lack of complete answers is in some ways a reminder that for all its potential, pharmacogenomics represents an important advance in medicine, but not a magic wand.
"It's important to remember that it's probabilistic, not definitive," said Gualberto Ruano, M.D., Ph.D., Vice Chairman and CSO of Genaissance Pharmaceuticals, a New Haven, Conn., biotech company.
"Genetics is not totally predictive," agreed Mark Watson, Clinical Genomics/Clinical Pharmacology at Whitehouse Station, N.J.-based Merck. "In studies, we've found that 10 percent to 15 percent of cloned mice in the same environment don't get the disease that they were predicted to." |
