The Food and Drug Administration's (FDA) proposal to reform the drug development process have been met with criticism from the Association of American Medical Colleges (AAMC), but with praise from other groups, including the Pharmaceutical Research and Manufacturers Association (PhRMA).

The AAMC charges that the FDA plans begin with a faulty premise: that the federal biomedical investment should translate directly to new treatments. David Korn, senior vice president of Biomedical and Health Science Research at AAMC, also warned that any attempt to involve the National Institutes of Health (NIH) in drug-related research could dilute funding for "curiosity driven" investigations that have, in the end, contributed to new therapies.

"The vast majority of NIH money is in fundamental research, not product development," Korn told The Scientist. "To equate the success of the NIH investment to whether there are 25, 40, or 60 products [submitted to] the FDA is misleading. It's a dangerous way to look at it. People who don't understand the purpose [of NIH] could be misled."

Korn's comments reflect the AAMC's response to the FDA "Critical Path" initiative, launched in March, which outlines obstacles to drug development. The AAMC was among a number of groups that submitted comments on the initiative. The comment period ended July 30, but most suggestions are being publicly released this week.

The agency's March white paper is the first step in what the agency said would be a reform in the way that new drugs are discovered, tested and regulated. By the end of the year, the FDA plans to bring new scientific methods into the development cycle. Eventually, the agency hopes that diagnostic and genomic tools will reduce the need for repeated clinical trials because such tools can cut drug discovery failures.

But observers inside and outside the agency caution that it will take time before such benefits can be realized. "The priority will be to do everything possible to reverse the fall in the number of new molecular entities, [submitted to the FDA for approval]," said Chris Webster, director of regulatory strategy and intelligence at Millennium Pharmaceuticals, based in Cambridge, Mass. "It's not going to be this year or next year, but over the next 3 to 6 years."

The initiative is prompted by the FDA's perception that "the applied sciences needed for medical product development have not kept pace with the tremendous advances in the basic sciences," according to the agency's white paper. Despite a growth in government and private investment, the number of new drugs with novel chemical structures has fallen from roughly 70 in 1993 to less than 30 in 2003, the paper said. New biologics applications have fallen from just under 30 to just under 20 during the same time period.

Despite the AAMC's concern that linking research investment to improving these numbers could threaten investment in basic science, Janet Woodcock, director of the FDA's Center for Drug Evaluation and Research, said the initiative should not affect NIH funding or programs. "We are not saying that money should be diverted from the research establishment," she said. The opportunities list is not considered a direct request for additional money from the US Congress.

Nevertheless, the agency lacks money for the tasks already identified as important, such as developing new biomarkers for disease, Woodcock said. "We do not have the resources," she told The Scientist. "We can identify needs, and we can start on some of the projects. [With the Critical Path initiative], we hope we will attract collaborators."

The FDA has already launched a project, called Voluntary Genomic Data Submission, through which the industry provides pharmacogenomic data confidentially to the FDA. The agency plans to eventually use this data in regulating drugs and biologics. Wyeth was the first company to submit materials, although other companies have since contributed, Woodcock said.

These databases could eventually reduce failures in clinical trials. "You might find, from building a database, that a compound may not be effective," said Gail Cassell, chair of the American Society for Microbiology Public and Scientific Affairs Board, and vice president of Scientific Affairs for Indianapolis-based Eli Lilly & Co. The database could prevent continuing work on a certain class of compounds if "it's been tried many times and failed," she said.

PhRMA also endorsed the FDA plan for compiling company data and using it to define disease biomarkers. In the early 1990s, it was the pooling of company data on biomarkers in the treatment of HIV that led to the development of reduction in viral RNA copy number as a surrogate endpoint in the accelerated approval of new drugs for this condition, Millenium's Webster wrote in an E-mail. Companies are required to validate the efficacy of their anti-HIV drugs against a clinical endpoint after approval, he said.

For PhRMA, the AIDS experience provides a model of future FDA–industry collaboration. "PhRMA would be interested to explore creative ways to do this," its FDA submission said. The trade association also requested that the agency analyze the cost and benefits of building pooled databases and requested that the agency translate the new information into regulatory changes, which, Woodcock said, will come only after the agency learns more about new scientific tools.

Even with the use of databases, consumers should be involved in any definition of biomarkers, said Diane Dorman, vice president for public policy of the National Organization on Rare Disorders.

But Webster said that only technically trained staff should develop biomarkers. Still, "patient groups and interested members of the public would certainly have an opportunity to know what was going on," he wrote, "and perhaps to indicate their views as to, for example, the need for development of different biomarkers or their preferences in the use of particular biomarkers in decision making—how would they feel about particular risk–benefit trade-offs."