Personalized Medicine Requires Regulatory Change

Bob Roehr

June 23, 2008

June 23, 2008 (Washington, DC) — The ongoing explosion of information generated by recent revolutions in the "-omics"— genomics, proteomics, metabolomics — requires changes to be made in the way that therapies are developed, approved, and paid for if the potential of personalized medicine is to be realized.

The American Association for the Advancement of Science and the Food and Drug Law Institute are sponsoring a series of meetings through 2009 to help facilitate these changes. The first of these meetings was held here on June 20.

The future of medicine is in identifying biomarkers that predict disease progression and response to therapy early in the course of disease, hopefully before the onset of symptoms, when perhaps the disease is more amenable to change, Gregory Downing, DO, PhD, director of the personalized healthcare program for the Department of Health and Human Services, said during the meeting.

"This represents a surrogate for a clinical endpoint," said Dr. Downing. One goal is to gather data and make informed decisions short of clinical trials, but that depends on validating those biomarkers.

Mark McClellan, MD, PhD, former head of the US Food and Drug Administration (FDA) and the Center for Medicare and Medicaid Services, said, "The goal is a more solid model of disease progression and a more solid understanding of when treatments are having an impact on particular patients."

The best example of this growing understanding is HIV, with the validation of CD4 cell count and viral load as biomarkers. Dr. McClellan said that the scientific foundation of understanding of the disease is so strong that it has allowed drugs to be developed and approved in record time, and none of these have been withdrawn from the market.

Analysis of the virus for resistance to classes of therapies has become standard of care, and the entry inhibitor maraviroc even carries a label indication for use of a tropism assay to ascertain whether the virus uses the CCR-5 cell receptor and therefore is susceptible to the drug.

In another example, a potentially fatal hypersensitivity to the drug abacavir has been linked to the HLA B*5701 genetic allele. A diagnostic test that identifies the allele has been cleared by the FDA, and the abacavir label has been modified to include use of this tool as a screen for safety. Its use has been integrated into clinical practice.

Shiew-Mei Huang, PhD, from the FDA Center for Drug Evaluation and Research, said that those are 2 of the 11 examples in which the FDA has begun to incorporate pharmacogenetic information in drug labeling and relabeling.

Dr. Huang spoke at length about the blood thinner warfarin, which has a narrow therapeutic window. Two gene variants have been identified as making patients particularly sensitive to bleeding when they are receiving warfarin.

The FDA cleared a diagnostic test for these genes last September, but adoption of the test into clinical practice has been slow. One participant thought that the delay is because treatment guidelines have not yet been revised to incorporate the test, which affects reimbursement.

Dr. McClellan said therapies that are targeted for a genetically specific patient population are a double-edged sword. On the one hand, such trials are likely to be smaller, more rapid, and less ambiguous; on the other, there are likely to be increased, unidentified risks associated with off-label use in populations other than the study sample. This increases the need for postmarketing surveillance.

New Sources of Information

"The whole postmarketing environment is about to change significantly," Dr. McClellan told Medscape Pathology & Lab Medicine. Under legislation passed in 2007, "the FDA now has more explicit authority around compelling phase 4 trials, along with the ability to impose civil monetary penalties when they don't get compliance," he added.

"The bigger change is that we are moving from an era when drug companies and product developers controlled the postmarket information and evidence, to an era when that will come from other sources," such as active surveillance through electronic medical records (EMRs) and those who directly deliver care, Dr. McClellan said. Perhaps more important, it will expand the available data from primarily that of efficacy in a controlled situation to include efficiency when the intervention is used in a broader population.

Two major impediments to EMRs are the lack of standardization of those systems and slow implementation by smaller medical practices, said Kavita Patel, MD, a senior health advisor to Senator Ted Kennedy, chairman of the Health, Education, Labor, and Pensions Committee.

Dr. Patel said the federal government is trying to create incentives to resolve these issues. The role of Congress "is not to legislate or overdefine [clinical practice] but to do what we can at the policy level to provide infrastructure to deliver the science."


Legacy coding and payment systems for reimbursement, created long before the appearance of genetic tools, can be a significant barrier to the adoption of new diagnostic tests into clinical practice, said Bruce Quinn, MD, PhD, a health policy expert and former administrator of the California Medicare program. "They work great for things that don't change, like a glucose test or a thyroid hormone test...but the more innovative the product, the more walls it runs into," he said.

"If the reimbursement isn't there, you won't get the products to the market in the first place, because [industry] works backwards and the investment becomes very fragile." He added that "1980 conventions are colliding with 2020 medicine. I think around 2010 something is going to snap."

Dr. McClellan said reimbursement reform must focus on the value of a therapy in a particular patient. He pointed to the example of bortezomib (Velcade) used to treat multiple myeloma: It can be very effective, but only in a subset of patients.

In the United Kingdom, the National Institute for Health and Clinical Excellence decided that the therapy was not cost-effective on a population basis and declined to approve it for inclusion in the formulary used by the National Health Service. That led to the sponsor negotiating an unusual "refund" scheme in the United Kingdom, in which the National Health Service is reimbursed for the cost of the drug in those patients who do not respond to it.

Steve McPhail, president of the genetics diagnostics company Expression Analysis, Inc, sees intellectual property concerns as a potential roadblock to future development of diagnostics. "There are over 40,000 patents on specific genomic sequences of pathogen- or cancer-related mutations," he said. "I think we need to change patent law and put this all out in the public marketplace."

However, Allen D. Roses, MD, director of the Deane Drug Discovery Institute at Duke University, Durham, North Carolina, thinks the opposite is true for drugs. Extending the life of patents on the chemical entity, he said, "is the only way you are going to get prevention trials done."


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.