Login to Your Account



Money, Science Issues Lead to Biomarker Bargain Basement

insight_cover_8-19-13.jpg

By Anette Breindl
Science Editor

Biomarkers would seem to have it all – they could save money for the health care system by saving patients ineffective care and its attendant toxicity risks. But considering how useful they are, there are a surprisingly low number of biomarkers in widespread clinical use. "The way it's been going for the last 30 years has resulted in very few validated biomarkers," Dan Hayes told BioWorld Insight.

Even biomarkers that have validity – that is, they reliably measure what they purport to measure, and can separate patient populations – do not necessarily have proven clinical utility in the sense of demonstrably improving treatment decisions and outcomes.

That, in turn, collectively leads to the worst of both worlds for the health care system, which spends money it doesn't have for things it doesn't need.

"There's no question we over treat people," Hayes said, partly because, in the absence of good biomarkers, patients want to be over treated in the hopes of being in the small minority that will benefit from a drug.

Hayes is an oncologist at the University of Michigan Comprehensive Cancer Center. In the July 31, 2013, issue of Science Translational Medicine, he and his colleagues suggest that the dearth of clinically useful biomarkers is no accident.

Lip service aside, "we don't value biomarkers the way we value therapies," Hayes said. This attitude creates a vicious cycle of low funding and low-quality science leading to poorly validated biomarkers that lack commercial potential. As a result developers cannot get reimbursement to recoup their costs, and this serves as a deterrent for future development.

Breaking the Cycle

Hayes said that the cycle can be broken, but only through addressing all aspects of the issue: "We can't fix one of them and not fix all of them," he said.

The fixes suggested by his team not only include higher scientific standards, but also more enforcement by the FDA, which currently lets CLIA-certified laboratories use tests that have not been through regulatory approval, as long as those tests are not sold to other institutions.

"There are a lot of lab-developed tests that have not been to the FDA," he said, which leads to two problems. Without central oversight, the quality of such local tests varies. At the same time, they are a deterrent for developing FDA-approved tests, since they preempt the market for such products.

Hayes is well aware that asking the FDA to be more involved will not please everyone. "We are not asking for regulation for regulation's sake," he stressed.

In his opinion, such regulation will be a net positive for commercial development. And in his experience, he said, investors too "are not asking for less regulation. They are asking for consistent regulation that defines what the playing field is."

Such regulation, in turn, would lead to products with higher financial value, which in turn would spur new investment – an alternative "virtuous cycle."

Colin Williams, director of product strategy for life sciences at Thomson Reuters, agreed with the notion of a current vicious cycle in biomarker development "to a degree," he told BioWorld Insight. "If you can't get reimbursement," for biomarkers used in the clinic, "then people won't invest. And if people don't invest, then the science doesn't keep up."

But not all biomarkers, he added, are ever headed toward a career as an FDA-approved, revenue-generating test in the first place. "There are value drivers for biomarkers not just in the clinical setting."

Research Tools

A significant number of biomarkers "are being developed as research tools, and for the value they can bring in internal decision making . . . sometimes they are never intended to be clinically approved biomarkers," Williams said.

Thomson Reuters (owner of BioWorld) recently published a report, Developments in Clinical Trials from Cortellis Clinical Trials Intelligence, that included an update on biomarkers in development. That report looked at many different sources, including the scientific literature, patent databases, and clinical trial registries, to identify potential biomarkers, and Williams said that "we've seen an exponential growth in the publication of entities that could be biomarkers."

Many of those biomarkers come from patent databases, suggesting that at least someone sees potential commercial opportunities for them. Intellectual property issues, as always in biotech, loom large for the space. Myriad Genetics Inc. is the poster child for biomarker patent issues. Its BRCA tests were partly invalidated by the Supreme Court earlier this year. (See BioWorld Today, June 14, 2013.)

All business challenges aside, Williams pointed out that part of the reason biomarkers are difficult to develop is the same reason drugs are difficult to develop: biology is complicated.

The presence, absence, increase or decrease of any one gene mutation or protein – even in a BRCA gene – does not provide "a simple, binary, yes-or-no-answer" as to whether a patient will develop a disease or respond to a drug, he said. "That's what makes it challenging."

He predicts that just like therapeutics development is heading toward combination therapies, biomarkers, too, will need to look at a combination of factors to make useful predictions. The first steps along this path, which include looking at gene panels rather than just one gene, are already being taken by companies such as Genomic Health Inc., as well as Myriad.

Ultimately, Williams said, "the value is going to be in multiple types of biomarkers working in combination. For example imaging, genomic variation and proteomics, to give a more 'personalized' diagnosis and view on treatment options."

Editor's note: For a copy of the report from Cortellis Clinical Trials Intelligence, click on the link: Developments in Clinical Trials.