The US Food and Drug Administration (FDA) has released a new final guidance document on clinical pharmacogenomics that it says should help sponsors to better conduct premarket evaluations in early-stage clinical studies.
The guidance, Clinical Pharmacogenomics: Premarketing Evaluation in Early Phase Clinical Studies, was first released as a draft guidance in February 2011. Fairly extensive by guidance document standards at 20 pages in length, the draft guidance provided recommendations on "when genomic information should be considered to address questions arising during drug development, and in some cases, during regulatory review."
FDA added that pharmacogenomics approaches are part of an "evolutionary process" unique to each product, but that general principles of study design, data collection and data analysis were more broadly applicable.
Identifying genomic differences between patients has been of increasing interest to regulators and companies alike in recent years. Regulators, in particular, have been wary of approving products that seem to have a small contingent of so-called "super responders" who respond highly favorably to a product, while other treated patients might obtain little or no benefit after side effects are taken into account. Both industry and regulators have looked to genomic markers as one way to distinguish between patients who will respond well and those who will not. Some recent approvals, such as Herceptin (trastuzumab) for HER2-positive breast cancer and Kalydeco (ivacaftor) for cystic fibrosis patients with the G551D mutation, exhibit this approach.
It should be noted that not every product adopting this approach has been successful. Roche's Avastin (bevacizumab), a HER2-negative therapy for extending progression-free survival (PFS) in patients saw its indication for that revoked in late 2011 after FDA said new data showed its toxicities outweighed its benefit. The drug had been granted accelerated approval based on an interim analysis of surrogate survival endpoints in 2008.
But FDA's guidance makes clear that the approach is one that can substantially benefit nearly all areas of the regulatory approval process.
"Genetic differences between individuals can affect virtually all aspects of a disease and its treatment, such as the rate of disease occurrence, the risk of disease progression or recurrence, the drug or drug class most likely to result in benefit, the therapeutic dose, the nature and extent of beneficial responses to treatment, and the likelihood of drug toxicity," FDA explained.
Regulators identified that three categories of factors were likely to account for the majority of differences: genes affecting the drug's pharmacokinetics, genes affecting the drug's targets and efficacy, and genes predicting the occurrence of disease development, also known as prognostic markers.
But development of these targeted products requires companies to surmount numerous challenges, the greatest of which may simply be identifying the genomic targets likely to account for variations of efficacy, toxicity or progression.
"Ideally, consent for DNA collection should be obtained from all participants in clinical trials," FDA explained. "An effort should be made to collect genetic samples at enrollment and/or at baseline to avoid potential bias associated with delayed collection," including drop-outs, patients who are non-compliant with study protocols or those who do not complete the study.
Even once targets are identified, companies may still need to develop companion diagnostic tools-often regulated as medical devices-to test patients for the presence of those genes.
Regardless of the challenges, FDA makes clear that companies stand to benefit greatly if they focus on pharmacogenomics in "early stage"-phases I and II-clinical trials. The agency explained that early studies can identify populations who may require different dosing strategies, populations who might respond differently, the proposer dose range for particular biomarkers, and high-risk groups liable to suffer from serious adverse events.
If found to be clinically relevant, information about the relevance of genetic markers should be included in the product's labeling, FDA advised.
"A 'Pharmacogenomics' subsection should be created in the 'CLINICAL PHARMACOLOGY' section and should include details on the clinically relevant information on the effect of genetic variations affecting drug therapy," the guidance explains.
Changes to the Draft Guidance
The final guidance reportedly makes a number of "clarifying changes"-usually smaller language tweaks to clarify intent-to the draft guidance, and adds some entirely new content as well.
"FDA added content to describe when pharmacogenomics (PGx) studies are warranted, including circumstances when full sample ascertainment is expected to evaluate a specific hypothesis. In addition, a number of topics were further elaborated, including targeted sample collection, sample retention, genotyping approaches, pooled analyses, dedicated prospective PGx studies, genetic substudies, and safety pharmacogenomics," FDA wrote.