NEW YORK (GenomeWeb) – The US Food and Drug Administration today released a draft guidance on the principles of drug and diagnostic codevelopment as a practical framework for sponsors advancing a therapeutic alongside a companion test and for agency staff reviewing these products.
"Codevelopment of IVD companion diagnostics and therapeutic products is critical to the advancement of precision medicine," the agency stated in the draft guidance. "FDA seeks to facilitate innovations in precision medicine by providing sponsors with a set of principles that may be helpful for effective codevelopment and in fulfilling FDA's applicable regulatory requirements."
The ideal Rx/Dx codevelopment scenario for the FDA is when the need for a companion test is identified in the earliest phases of drug development and the two are advanced and launched on the market at the same time.
The FDA articulated this favored scenario when it finalized guidance on companion diagnostics two years ago. However, in the codevelopment and companion diagnostic guidances the FDA also recognized that contemporaneous evaluation of the drug and test may not always be possible, and allows for the test to come later when, for example, the drug is indicated for a life-threatening condition for which no other treatment exists.
The FDA typically considers companion diagnostics Class III, high-risk devices that require premarket approval. In the codevelopment guidance, however, the agency acknowledged there may be scenarios where a companion test is a Class II, moderate risk device that can come to market following 510(k) clearance or a de novo request.
The processes for advancing a drug and test are markedly different, but the guidance provides a chart for how sponsors can align these programs and at what points they should seek input from the agency. The agency recommended that sponsors have understanding of both therapeutic and diagnostic development and both parties should be present during meetings with FDA's drug and diagnostics divisions.
The FDA defines a companion diagnostic as a test that is necessary for the safe and effective use of a drug. As such, the agency recommended that the analytical performance of the CDx should be established before using it in clinical trials for the drug. When there is insufficient data on the risks that an investigational new drug poses to patients, the FDA may place the study on "clinical hold." The agency noted that uncertainty about the analytical validity of the companion diagnostic could trigger such a hold.
If the test being studied alongside a drug hasn't been previously approved by the agency for that specific intended use, the sponsors must apply for and receive an investigational device exemption. The FDA outlined in the draft guidance the types of information that should be included in an IDE application and the circumstances in which IVDs are exempt from this process. The document also discusses factors test makers should consider with regard to using training sample sets, the impact of test design changes, and IVD bridging studies.
The agency, for some time now, has been cautioning sponsors that physicians are increasingly pre-screening patients to determine their eligibility for biomarker-informed clinical trials, and that this practice may be problematic for Rx/Dx codevelopment. "Prescreening may result in a biased clinical trial population that does not represent the population that would be selected by the IVD companion diagnostic in real-world testing," the agency stated. "Thus, planning to enroll subjects into a trial based on confirmation of a local test is strongly discouraged."
The agency suggested sponsors ask clinical sites to submit samples from all potential candidates for testing instead of just those who have been pre-screened by a local test. This will enable evaluation of "the true performance of the IVD," FDA said, and ensure the intent to treat population isn't skewed.
The draft guidance also suggested that FDA will be flexible on a number of fronts, by allowing the use of clinical trial assays in early studies, the demonstration of analytical validity on a representative set of markers for NGS tests gauging many genes, and the utilization of contrived samples to study certain markers when actual patient specimens are unavailable.
However, the FDA noted in the guidance that sample acquisition is "critical to a successful codevelopment strategy," and encouraged sponsors to obtain specimens from all subjects tested for enrollment. This will ensure that sponsors can still validate a companion test for commercialization if changes are made to the IVD as studies are ongoing, or if the clinical trial assay from early studies doesn't end up being commercialized.
Drugmakers, meanwhile, could conduct different types of biomarker-guided trial designs, according to the guidance — one in which patients are randomized into two treatment arms based on positive and negative biomarker status, and another trial in which only biomarker-positive patients are randomized to treatment arms.
The first trial design is the most informative, according to the FDA, in terms of gauging the prognostic and predictive value of the biomarker. However, the agency has certainly granted accelerated approval to precision drugs based on studies enrolling only marker-positive patients when there is evidence to suggest that marker-negative patients will not respond to treatment. The agency also discussed the scenarios in which drugmakers could retrospectively evaluate patient response based on a biomarker after having completed a prospective study.
Based on how the test was used in a drug trial, it may receive a claim for being predictive of treatment benefit, for monitoring dosage or treatment discontinuation, or for selecting patients for the clinical trial. "For a selection claim, if the major efficacy trial demonstrates adequate safety and effectiveness of the therapeutic product within the population selected by the IVD, the IVD is considered to be 'clinically validated' in that it selected a population that benefits from the therapeutic product," the agency noted.
Although the breast cancer drug Herceptin (trastuzumab) was the first therapy the agency approved in 1998 with a companion diagnostic, HercepTest, in recent years, the list of Rx/Dx codeveloped products has grown rapidly, particularly in oncology. The recommendations in codevelopment the draft guidance are no doubt distilled from the agency's experiences approving these products.
The FDA's drug and diagnostics divisions "have worked hard to develop internal processes to get oncology drugs through FDA review," Amy Miller, executive VP of the Personalized Medicine Coalition, told GenomeWeb over e-mail. "We suggested that the FDA use that experience to draft a 'how-to' guide for sponsors and for FDA staff in other review areas, and they are responding to that request with this draft guidance."
The codevelopment draft guidance has been a long time coming, and over the years the agency has communicated the principles in this document to stakeholders at workshops and industry conferences. Miller added that the PMC will take a close look at the draft document and suggest improvements to it.
Members of the public have 90 days to comment on the draft guidance.