Where's the infrastructure for precision medicine?

The remarkable advances over the past 15 years that began with the first draft sequence of the human genome can’t be brought to full fruition without an infrastructure to support biomarker tests for molecularly targeted therapies, according to a new report from the National Academy of Medicine (NAM).

See also: Pharmacogenomics: Right med, right dose, every time

To establish that infrastructure, major changes are needed in policy and practice, said the expert panel that wrote the report, titled “Biomarker tests for molecularly targeted therapies: Key to unlocking precision medicine.”

Those changes range from better consensus as to what constitutes evidence of a test’s clinical utility to electronic health records that can reflect those tests and allow for “rapid learning.”

Common standards

The panel called on the Department of Health and Human Services to facilitate the development of common evidentiary standards for tests, perhaps through the convening of multi-stakeholder bodies. The current lack of such standards is a significant drawback to getting tests into clinical practice. At the same time, biomarker tests with limited evidence could lead to patient harm, said the report.

Evidentiary standards and study approaches should be developed to simultaneously accommodate various types of decisions, including clinical, regulatory, coverage, and reimbursement decisions, as well as guideline recommendations, the panel said.

Partly because of the variety of considerations, an extensive range of stakeholders should be involved in utility studies, the report suggested, with participants to include patients, public and private payers, federal agencies, guideline developers, test developers, pharmaceutical companies, molecular pathologists, clinical laboratory geneticists, and research funders.

See also: Pharmacogenomics could boost community pharmacy sales

Evidence over time

Critical to the process, the panel asserted, is the concept of the evolution over time of evidence for test utility, something that is not always considered.

The report envisions a rapid learning system that “would systematically collect and analyze data on biomarker tests, molecularly targeted therapies, patient management, and outcomes.”

Consistent process

David Veenstra, PharmD, PhD, professor in the Pharmaceutical Outcomes Research and Policy Program at the University of Washington, called the proliferation of these tests quite impressive. The great challenge, he noted, is that there is no way to generate gold-star evidence for everything, “so we need some kind of process by which we can make reasonably consistent decisions about when we can move forward with implementation vs. evidence generation.”

The fact that this must be an ongoing process is one of the most important considerations. What that will look like with so many stakeholders involved is the question many people are struggling to answer, he noted.

“I think that it is really about identifying some common processes and concepts, frankly, at a pretty high level,” said Veenstra.

Important to the process, he said, is the concept of “risk sharing,” meaning there is a decision to move forward with a technology but those involved generate evidence while using it.

Data and decisions

In that vein, the new report noted that complete results of complex biomarker tests are not usually available in electronic health records. And yet the ability to track patients’ long-term outcomes is key to understanding the risks and benefits of the tests and the therapies.

The report also called the development of one or more large, integrated databases on patient outcomes is critical to the progress of precision medicine.

The report’s authors called upon FDA and the Centers for Medicare and Medicaid Services to begin integrated, concurrent reviews of regulatory, coverage, and reimbursement decisions for such tests as in vitro diagnostics, laboratory-developed tests, multi-analyte tests, and any corresponding molecularly targeted therapies.

FDA should develop a standardized label format for biomarker tests, said the report, with FDA or lab-accrediting bodies approving the label for each biomarker test. Labels should have separate ranking systems, such as a four-star system, for evidence of clinical validity and clinical utility for each intended use, said the report. It also called for expedited revision of labels as more evidence develops.

The report is available free from the website iom.nationalacademies.org. NAM (previously the Institute of Medicine) is a private but highly influential body that frequently receives federal and other funds to examine issues for the federal government and other bodies.

Kathryn Foxhallis a healthcare journalist based in the Washington D.C. area.