Washington Editor

WASHINGTON - At last, the FDA this week issued guidelines to the drug industry on pharmacogenomics, providing a framework for data submissions to supplement marketing applications and regulatory review, while also outlining a new mechanism for the voluntary submission of research data.

In the long run, many think this final guidance, which is the result of several years of public and industry input, will help usher in better therapeutic options for patients and renew focus on the science underlying pharmacogenomics-based drug development. At present, the agency's language points to a ringing endorsement of such research.

"I think that the FDA believes, based on data they've seen to date, that pharmacogenomics does hold the promise to deliver on personalized medicine," said Carol Reed, vice president of medical affairs at Genaissance Pharmaceuticals Inc. "It's been touted for a long time, and people have been disappointed, but I think with this guidance and accompanying comments, they're saying We do think it's a potential solution on both the efficacy and safety sides and we want to encourage you, drug developers, to do more of it.'"

That sentiment was evident in the agency's statements that accompanied its published guidelines, titled "Guidance for Industry Pharmacogenomics Data Submissions."

"This new technology will allow medicines to be uniquely crafted to maximize their therapeutic benefits," said Janet Woodcock, the FDA's acting deputy commissioner for operations, "and minimize their potential risks for each patient."

Pharmacogenomics research, especially in certain disease areas such as oncology, central nervous system disorders and inflammatory indications, is nothing new. Genaissance, of New Haven, Conn., has long based its drug development activities on pharmacogenomics, and Reed noted that the agency's action represents an invitation for drug companies to extend such research.

"I really believe that the main message here," she told BioWorld Today, "is that the FDA wants to work with industry on this. They believe we can help each other in bringing these tools to drug development, and in my mind this is a real opening of the door by the FDA."

The guidelines suggest a number of procedures for drug companies to follow throughout the development cycle. Such a "manual of policy and procedures," said Chris Webster, director of regulatory strategy and intelligence at Millennium Pharmaceuticals Inc., of Cambridge, Mass., "was something the industry asked for."

For investigational new drug applications, pharmacogenomic requirements vary. Per existing rules, data related to animal or in vitro studies must be submitted if a sponsor intends to use the findings to make a scientific case or when the pharmacogenomic test is a known valid biomarker. Sponsors also must submit human data of known relevance such as known valid pharmacogenomic biomarkers, and the FDA can request pharmacogenomic information it considers relevant.

But submissions are not required if data stem from exploratory studies or from general gene expression analyses in cells, animals or humans, or single nucleotide polymorphism analysis of trial participants, or if the information consists of results from test systems in which the validity of the biomarker is not established. Although such data are not required under the regulations, the FDA said it would welcome voluntary submissions, a procedure Genaissance already has followed.

"Millennium [also] has already made a voluntary submission under this process," Webster told BioWorld Today. "We look at the guidance as allowing us to go have a conversation at an early stage of development about real data, so you're not talking in the abstract. On the other hand, those data will not be used in the FDA's decision making, so they will not carry the normal regulatory risk."

To comply with regulations for new drug and biologic license applications, the FDA noted that sponsors must provide reports of certain pharmacogenomic investigations, though the extent and format of those reports will depend on the relevance and application of the information.

For updating new drug and biologic license applications, results of nonclinical or clinical pharmacogenomic investigations on known or probable valid biomarkers must be submitted in annual reports, while results of other types of biomarkers or pharmacoepidemiologic and observational studies can be voluntarily submitted.

"The FDA, in some areas, outlined what kind of validation and experimental design and information they need," said Dennis Gilbert, chief scientific officer at Applied Biosystems Inc., of Foster City, Calif. "I think it helps people focus in areas where their technology or applications of the technology need to be driven toward."

He also told BioWorld Today that the guidelines point the way "to a renewed partnership between industry tool providers and pharmaceutical companies," especially relative to genomics technologies such as sequencing, gene expression and genotyping.

The guidance attempts to distinguish between pharmacogenomic tests that could be considered probable or known valid biomarkers and possibly appropriate for regulatory decision making, and other less-developed tests that are either observational or exploratory biomarkers that are insufficient for regulatory decisions. Although most pharmacogenomic measurements are not yet considered valid biomarkers by the FDA, certain markers, such as those for drug metabolism, are well established "with clear clinical significance." The agency conceded that the distinction between tests appropriate for regulatory decisions and those that are not will change over time as the science evolves, and throughout the development of such tests, the agency said it would continue to seek public comment in evaluating biomarker validity.

"That's a difficult issue because of the complexity of the science," Webster said, "and also because we're moving into new territory here to a large extent."

Ed Abrahams, the executive director of the Personalized Medicine Coalition, a public policy advocacy organization in Washington, called the FDA guidance "a point on a continuum," adding that there will be a continued evolution in terms of the agency's "willingness and interest in using pharmacogenomics as a tool to expedite the development of drugs that treat disease."

He and Reed said further guidelines will focus on the hand-in-hand development of therapeutics and pharmacogenomic-based diagnostics, which Abrahams called the "basic underpinning of personalized medicine," as well as guidelines on the use of microarrays and DNA analysis. Reed also pointed to a need for a standard bioinformatics infrastructure to align the industry with the agency for data submissions.

Concurrent with releasing its final guidance, the FDA unveiled a pharmacogenomics website linked to the agency's homepage. Also, next month the agency is hosting a pharmacogenomics workshop in partnership with the Drug Information Association to focus on integrating pharmacogenomics in clinical trials for new drugs, biologics and associated devices. The agency also expects the gathering to address ways to translate pharmacogenomics into medical product development and clinical practice.

Most expect the new guidelines to foster further pharmacogenomic-based drug development. "I think you're going to see development gather momentum with this guidance," Abrahams said.