Skip to main content
Premium Trial:

Request an Annual Quote

FDA Highlights Use of Genomic Markers in Draft Guidance for Enrichment Strategies for Drug Trials


The US Food and Drug Administration issued a draft guidance in late December outlining various enrichment strategies, including those involving genomic and proteomic markers, that drug developers can use to gain insights into the safety and efficacy of drugs.

In the draft guidance, entitled "Enrichment Strategies for Clinical Trials to Support Approval of Human Drugs and Biological Products," the agency defines "enrichment" as the "prospective use of any patient characteristic to select a study population in which detection of a drug effect (if one is in fact present) is more likely than it would be in an unselected population."

The definition encompasses pharmacogenomically guided trials in which researchers investigate whether patients with specific genomic characteristics have a differential response to a treatment compared to those without such markers.

Among the enrichment strategies included in the guidance, the FDA includes sections discussing prognostic and predictive enrichment strategies using genomic and proteomic markers. As an example, the FDA cites Genentech's Herceptin (trastuzumab), indicated for the treatment of metastatic breast cancer in patients with tumors overexpressing the HER2 protein.

"In the treatment of metastatic disease, when added to either of two background regimens, trastuzumab increased survival by a mean of about five months, about three to four times the effect that would have been expected in an unselected population, assuming no response (which a modest amount of testing showed was the case) in the HER2-neu negative patients," the FDA wrote in the guidance.

"Enrichment thus allowed a modest-sized study to show a striking effect and directed treatment to the population that could benefit," the agency stated, noting that enriching adjuvant studies of Herceptin with HER2-overexpressing breast cancer patients became even more important in order to limit drug-related cardiotoxicity in patients unlikely to respond to treatment.

Other examples of enrichment strategies include a trial investigating Novartis' Gleevec (imatinib) in patients who had their gastrointestinal stromal tumors resected and had c-Kit expression; testing breast cancer patients for estrogen receptor status and progesterone receptor status before starting endocrine therapy; and testing melanoma patients for BRAF mutations to identify potential responders to Roche's Zelboraf (vemurafenib).

"When proteomic and genetic markers are used in an enrichment strategy, adequate characterization of the test for the marker is critical," the FDA notes in the draft document. "An inaccurate assay will undermine an enrichment effort if the study aims to demonstrate superiority or non-inferiority of the test treatment. It is also important to gain as much information as possible about the marker-response relationship (sensitivity and specificity)."

The agency acknowledges that although industry is increasingly employing genomic enrichment strategies to predict response to drugs, there are still many cases where the link between genetic patterns and patient outcomes is only made after the study is completed.

"Markers discovered this way will have credibility problems related to the post-facto nature of the finding and will almost always need confirmation in a prospectively planned enriched study," the agency said, noting that this type of prospective confirmation of the gene marker is difficult to do when the overall study fails. Such instances make it critical that drug developers routinely collect and preserve patient specimens for future genomic analysis, the agency advised.

Additionally, the agency suggests several strategies for conducting genomic subset analysis as part of an ongoing randomized trial in a genomically undifferentiated patient population. For example, the FDA suggests splitting the randomized trial into two halves in order to search for genomic markers linked to response in the overall study population in the first half of the trial in an unblinded fashion, and then enroll predicted non-responders and responders in the second half of the study according to the original enrollment criteria. At the end of the trial, researchers can gauge the efficacy of the drug in the overall population and in the genetically identified subset.

"The study shows evidence of effectiveness if either analysis is positive. When the responder population is a small fraction of the total population, but exhibits a large response, this design can improve the chance of detecting a treatment effect," the FDA said in the draft document. "It also retains good power for the overall study if the drug is more broadly effective."

The FDA urges industry players to understand the performance metrics of the screening measures gauging genomic and proteomic markers, noting they are not as well established as "phenotypically well-known prognostic factors such as blood pressure or cholesterol." For example, the agency advises industry players to establish the accuracy of the test and its sensitivity and specificity to the best extent possible. Additionally, the FDA tells companies to develop patient selection criteria that aren't overly inclusive, because the study will lack specificity to gauge the differential response between the enriched and non-enriched population; or overly exclusive, because the study will lack sensitivity to identify some patients who might benefit from treatment.

Furthermore, for predictive enrichment trials, the FDA explains the need to collect some data on marker-negative populations. "Experience suggests that the selected enrichment factors often do not prove to precisely dichotomize patients into subpopulations that will and will not respond, so that it is usually desirable to obtain some information on the marker-negative population to assess performance of the factor," the agency states in the document, adding that when there are no alternative treatments, investigating patients unlikely to respond to treatment "raises ethical issues."

For example, in post-marketing requirements, the FDA has asked Pfizer to collect data on how patients without ALK rearrangements respond to the non-small cell lung cancer drug Xalkori, a drug currently approved only for patients with ALK-positive tumors. In another case, however, due to the established knowledge that most patients with certain KRAS mutations don't respond to EGFR-inhibiting monoclonal antibodies, Bristol-Myers Squibb and Eli Lilly could not conduct a prospective analysis in KRAS-mutated and wild-type patients. Instead, the sponsors conducted retrospective analysis of two previously conducted studies to garner a new indication for Erbitux as a first-line treatment for metastatic colorectal cancer in KRAS wild-type patients (PGx Reporter 7/11/2012).

FDA's Center for Drug Evaluation and Research, the Center for Biologics Evaluation and Research, and the Center for Devices and Radiological Health together developed the guidelines presented in the document. The public has two months to provide comments on the draft guidance.