The US Food and Drug Administration may issue a series of white papers outlining the most up-to-date knowledge on drug/diagnostic codevelopment, a top agency official said at a recent conference.
Since the FDA issued its first white paper on Rx/Dx codevelopment in 2005, the agency's understanding of the issues involved has "significantly increased," said Janet Woodcock, director of FDA's Center for Drug Evaluation and Research, at the "FDA-Industry IVD/Companion Diagnostic Drug Roundtable" in Washington, DC, last month.
The meeting was sponsored by the agency and industry groups AdvaMed, the Pharmaceutical Research and Manufacturers of America, and the Biotechnology Industry Organization. Several drug sponsors and in vitro diagnostics manufacturers attended the meeting, at which participants discussed the difficulties of simultaneously developing a drug and companion diagnostic product.
According to Woodcock's presentation, which was posted on AdvaMed's website, while the agency and industry still face a "very complex set of problems" related to Rx/Dx codevelopment, the codevelopment strategy itself "is no longer theoretical.
"Various external groups [are] working on documents to contribute to the discussion," Woodcock said in her presentation, adding that the "agency may publish a number of white papers on these topics."
The FDA seems to favor Rx/Dx codevelopment as a pathway to getting pharmacogenetic information into drug's label. However, drug and diagnostic makers have pointed out several barriers to such a strategy, including divergent timelines, funding issues, identifying a demand, and added risk [see PGx Reporter 12-13-2006].
The FDA has long discussed its intent to issue guidelines on the regulatory framework for developing companion diagnostics, but has made little progress since issuing the initial white paper nearly four years ago. In an effort to make progress on the stalled guidance, the FDA last year formed a multi-center, multi-disciplinary working group to discuss potential guidance on Rx/Dx codevelopment [see PGx Reporter 11-19-2008].
[ pagebreak ]
At the March meeting, the agency did not provide a timeline for when it planned to issue a draft guidance on Rx/Dx codevelopment.
"While many challenges were brought to light at the meeting, it was unclear what FDA’s next steps will be" regarding Rx/Dx codevelopment, wrote Stephanie Devaney, genetic research analyst at Johns Hopkins University's Genetics and Public Policy Center, in the center's bi-monthly newsletter.
"Despite their disagreements over the contents of the [Rx/Dx codevelopment white paper], meeting participants seemed to agree that a guidance document is needed soon to help inform the process of co-developing drugs and diagnostics," Devaney wrote in the newsletter.
Drug Makers Speak Out
In an effort to illustrate the difficulties drug makers face in marketing a drug to a genetically targeted subpopulation, Bristol-Myers Squibb and Amgen discussed at the meeting their experiences garnering updates to the labels for two colorectal cancer drugs to include KRAS testing information. Both companies have submitted retrospective data — BMS for its drug Erbitux and Amgen for Vectibix — as part of their applications to the FDA to update their drugs' labels with gene-response data [see PGx Reporter 12-17-2008].
Although European regulators have narrowed the indication for Vectibix and Erbitux to include treating only those patients whose tumor carries the wild-type KRAS gene, and the National Comprehensive Cancer Network in the US has updated treatment guidelines recommending pre-treatment KRAS testing for CRC patients, the FDA has refused to make such a recommendation absent results from prospective clinical trials.
The FDA is specifically concerned about the validity of biomarker analyses derived from a trial that has not met its primary endpoint; whether the convenience sample is representative of the overall treatment population; the reliability of the marker; and the consistency of the treatment effect across sites.
[ pagebreak ]
At the meeting, Patricia Keegan, director of the Division of Biologic Oncology Products in FDA’s Center for Drug Evaluation and Research, noted that the "optimal approach" for the companies would be to "conduct an adequate and well-controlled trial, prospectively designed to assess efficacy in subgroups based on KRAS testing by validated assay."
The "pragmatic approach" according to Keegan, however, would be to conduct retrospective analysis in a large, randomized, well-controlled trial using an assay with "acceptable analytical performance" and a pre-specified analysis plan.
The FDA has said it is currently in the process of ironing out with Amgen and BMS retrospective data submission requirements.
At the meeting, Peter Collins, DxS senior vice president of business development, discussed the barriers to marketing its TheraScreen KRAS assay in the US as a companion diagnostic to Vectibix and Erbitux.
The FDA has previously indicated that it considers the risk profile of a companion diagnostic to be the same as that of the drug. As a result, DxS is in the process of submitting a modular pre-market approval application to the FDA for the TheraScreen KRAS assay [see PGx Reporter 03-04-2009].
DxS signed an agreement in December with Amgen to market its KRAS assay in the US as a companion diagnostic for Vectibix. Therefore, TheraScreen is considered a Class III high-risk device requiring PMA.
Yet, laboratory-developed KRAS tests are currently being marketed in the US by companies such as TrimGen and others. However, absent regulatory approval from the FDA or an update to drug labels, the DxS KRAS test is only available for sale in the US as a "research use only" product and DxS is barred for competing for market share in the KRAS testing space.
"There is definitely an uneven playing field here," Collins told Pharmacogenomics Reporter this week. "We have to go through an exhaustive, expensive process for regulatory approval with no guarantees, while some labs are launching homebrew tests and medical practice is moving on."