Using Biomarkers to Guide Therapy, Coverage Decisions

In August 2008, as part of its Critical Path Initiative, the FDA released a table listing genomic biomarkers that have established roles in determining drug response.1 This initiative, started in 2004, is aimed at identifying “patients likely to benefit from a treatment and patients more likely to respond adversely to a product,” according to the FDA’s Office of Management, Budget Formulation and Presentation.

Currently, the FDA reports that approximately 10% of approved drugs include pharmacogenomic information on their labels, representing a significant increase in the inclusion of such label information over the past decade.

The concept behind identifying genomic biomarkers is that they can play an important role in differentiating between responders and nonresponders to therapy, thus avoiding toxicity and minimizing the need for trial-and-error drug dose adjustments to optimize a drug’s efficacy. Using genomic profiles, physicians and other health care providers can identify patients who will (or will not) respond to a given drug, who metabolize drugs faster or slower than average, and who are at high risk for drug toxicity.

The use of clinically tested biomarkers to guide drug therapy decisions may prove to be an important step on the road to personalized medicine. It is likely that the number of tests available to predict patients’ responses to specific pharmaceuticals will continue to grow. The table compiled by the FDA lists validated genomic biomarkers, along with links to pharmacogenetic information.2 For each biomarker and drug, the table indicates whether pharmacogenomic testing is considered by the FDA as “required” for optimal therapy, “recommended,” or for “informational purposes only.”

For example, the package insert (PI) for Herceptin (trastuzumab) states: “Detection of HER2 protein overexpression is necessary for selection of patients appropriate for Herceptin therapy.” In addition, the FDA lists the ERBB2 (formerly HER2) test as required in patients beginning Herceptin therapy. Yet 5 of the 21 health plans recently polled (Rxperts data on file, 2008) state that they do not require or even check that ERBB2 levels have been measured before allowing coverage of Herceptin.

In another example, the PI for warfarin reads: “Certain single nucleotide polymorphisms in the vitamin K epoxide reductase complex 1 (VKORC1) gene (especially the c.-1639G>A) have been associated with lower dose requirements for warfarin.” The PI also states: “Analysis suggested an increased bleeding risk for patients carrying either the CYP2C9*2 or CYP2C9*3 alleles.” The FDA lists both the VKOR variant test and cytochrome P-450 2C9 isozyme (CYP2C9) variant genotype test as recommended before prescribing warfarin. However, not 1 of the 12 health plans surveyed (Rxperts data on file, 2008) requires or even checks to see whether either test has been performed before approval of coverage of warfarin. In fact, some payers, including Medicare, do not cover the cost of the CYP2C9 or VKOR variant test when ordered.

Managed care pharmacy and medical directors talk about the potential benefits of value-based formularies and the need to improve drug response rates and avoid adverse events and their related costs and are very willing to require prior authorization for products for reasons such as high cost or potential off-label use. Why are plans not equally willing to impose prior authorization on products for which testing is available to ascertain whether the patient will respond to the drug or to determine whether a lower or higher dose of the prod-uct might be needed to obtain an appropriate clinical response?

Over the next few years, it is expected that more pharmacogenetic tests will be available to assist in determining the appropriate dose and response rates for individual patients for a growing number of medications. With more tests available and rising drug costs, at some point it is possible that the value and even the optimal dose of each drug for an individual patient will need to be determined by a test. If plans are serious about ensuring that the right drug for the right person is prescribed and want to continue to put controls in place to manage drug appropriateness, they should be incorporating pharmacogenetic testing as part of their drug approval systems now. This will enable them to find ways to streamline the process while the list of required tests is still relatively small rather than wait until the number of available tests explodes, which is almost certainly on the horizon.

References:

References
1. Beals JK. FDA releases list of genomic biomarkers predictive of drug interactions. Medscape. August 1, 2008.
2. FDA Center for Drug Evaluation and Research. Table of valid genomic biomarkers in the context of approved drug labels. http://www.fda.gov/cder/ genomics/genomic_biomarkers_table.htm. Updated September 10, 2008. Accessed February 27, 2009.