American Association for Clinical Chemistry
Better health through laboratory medicine
April 26, 2007


Improving the Use of Cancer Biomarkers
By Deborah Levenson


Researchers have long sought new ways to diagnose cancers at early, curable stages and select the best therapies for individual patients. While biomarkers present an opportunity to enhance detection and treatment of cancer, unclear regulatory guidance and a spotty patchwork of standards for their use has led  to some compromised patient care, according to a new report from the Institute of Medicine (IOM). The authors call for both a molecular diagnostics specialty under CLIA and federal reforms aimed at better coordinated regulation of existing oncology biomarkers and research on new assays, plus improved reimbursement. This issue of Strategies examines IOM’s report, and what  laboratory directors should know about its recommendations.

In recent years, knowledge about the biochemical pathways underlying several cancers has expanded exponentially, but translation of that knowledge into more effective patient care and better outcomes remains a challenge. While it’s widely believed that biomarkers can improve detection, drug treatment, and the safety and efficacy of care by giving physicians tools to tailor treatment to individual patients, relatively few oncology biomarkers have moved from the discovery phase to routine clinical use, according to a recent report from the IOM. One major reason, the report emphasizes, is that research activities have been "piecemeal and unorganized." The potential of some of the few clinically available cancer biomarker tests has been marred by quality problems, the report adds.

"The minimum generic standards set by CMS under CLIA are not adequately tailored to the complexities of molecular diagnostics, and private-sector accreditation is voluntary and limited in scope," the report points out, adding that CMS has dropped plans to create a specialty for these high-complexity tests. Widespread concern about the validity and inaccuracy of estrogen receptor tests based on immunohistochemistry underscores the need for better regulation. So do similar concerns about the Hercep Test (Dako Corporation, Carpinteria, Calif.), a commercial immunohistochemistry kit that identifies which breast cancer patients can benefit from the drug Herceptin, and fluorescent in situ hybridization to measure HER2 gene amplification, the report notes.

“The (IOM) group had real concern that the best tests are not being applied uniformly so there’s inappropriate treatment and withholding of treatment that would be helpful,” explains Harold L. Moses, MD, Chair of the IOM committee that issued the report and Professor of Cancer Biology, Medicine and Pathology, and Director Emeritus, Ingram Comprehensive Center, Vanderbilt University Medical Center, Nashville, Tenn. "Lack of regulation leads to uneven quality of tests.”

Spotty Oversight
The report criticizes the lack of uniform, consensus standards for demonstrating the validity of biomarkers. “The FDA and CMS have some authority over diagnostic tests, but oversight has been variable and unpredictable and, in many cases, inadequate to ensure the safety, effectiveness, and value of test on the market,” the report says. While some federal agencies, especially the Food and Drug Administration (FDA), have taken some initial steps toward this goal, IOM calls for more action:

  • The National Institutes of Health (NIH), FDA, the Center for Medicare and Medicaid Services (CMS), and the National Institute of Standards and Technology (NIST), representatives from academia, the pharmaceutical and diagnostics industries, and payors, led by NIST or another agency, should create standards for biomarker development, validation, qualification and use.
  • FDA should clarify its authority over biomarker tests linked to clinical decision making and consistently apply clear guidelines for oversight, and
  • FDA or the Federal Trade Commission should monitor and enforce molecular diagnostics’ marketing claims.

Moses recommended that the laboratory community take ownership of oversight via collaboration among the more than one dozen organizations that have existing standards and guidelines on biomarker development and use. “It might be best over the long term for a group – maybe the College of American Pathology – or a group of lab organizations to police their own industry, rather than let FDA and CMS do it for them,” he suggested.

A lack of uniform standards may become more problematic as biomarker tests “move toward looking at multiple markers at the same time,” as the breast cancer tests OncoType (Genomic Health, Redwood City, Calif.) and Mammaprint (Agendia, Amsterdam, The Netherlands) do, Moses predicted.

George Klee, MD, PhD, Professor of Laboratory Medicine and Chair of the Division of Experimental Pathology and Laboratory Medicine at Mayo Clinic in Rochester, Minn. praised IOM’s call for biomarker standards, but urged caution regarding its suggestion that FDA clarify its authority over biomarkers used in clinical decisions. “I’m concerned that clarification could be so prescriptive that FDA interferes with the best practice of medicine and leaves clinicians without the flexibility needed to serve individual patients,” he explained. Klee has research grants from Beckman Coulter and has licensed technology to the company, and has received grant funding from Invitrogen.

Coverage of Existing Biomarkers
Other recommendations focus on the methods and processes needed for evaluation and adoption of biomarker tests, with an eye toward better payment by Medicare and other insurers. While diagnostic tests historically have been adopted into clinical practice with relatively little assessment of their value to physicians and patients, payors are now demanding more evidence of effectiveness when making decisions about coverage. To facilitate collection of such data, faster access to the market, and appropriate pricing, IOM suggests that CMS:

  • Modernize its process for evaluating, coding, and pricing diagnostic tests, using their longitudinal data to asses value;
  • Oversee development of consensus guidance on coverage decisions;
  • Establish panels to determine appropriate coverage of new and homebrew tests;
  • Develop criteria for temporary, conditional coverage of new biomarker tests with other payors; and
  • Study biomarker tests given conditional coverage, although doing so may be difficult for private insurers that must administer benefits according to terms of benefit plans that often exclude experimental items.

These recommendations’ time has come, Klee agreed. “It’s hard to move a test into practice without some way to get reimbursed for its costs. We do need a rapid, facilitated mechanism to bring along coverage of these new biomarkers as they’re proven to be efficacious.” He suggested that CMS offer reimbursement in exchange for data from clinical trials conducted in clinics. “Controlled, grant-funded trials often yield results that differ from what you’d get in an actual practice setting,” he explained.

Federal and Industry Role in Research
Calling for a “more organized, comprehensive approach to discovering biomarkers and developing novel technologies” among federal agencies including NIH, NCI, FDA, and NIST, IOM suggests that NCI oversee discovery activities at all four agencies via a highly directed, contract-based program. It “would provide incentives for technology development and large-scale discovery that are currently absent from academia’s career structure and reward system,” the report explains. 

It also suggests that nongovernmental funders, including industry, help establish international public-private consortia to generate and share pre-competitive data and methods of validating and qualifying cancer biomarkers for specific uses. “The costs, uncertainties and risk of developing biomarkers have historically made this work unappealing to pharmaceutical companies and diagnostic companies alike,” the report notes. It points out that precedents for public-private consortia exist, most notably the Single Nucelotide Polymorphism (SNP) Consortium. Created by ten large pharmaceutical companies and the U.K. Wellcome Trust philanthropy, the consortium mapped 1.8 million SNPs and is now studying some of them in populations.

A recurrent theme in many of IOM’s recommendations is the need for more standardization and validation of research activities, said Klee.  Operating standards are generally well defined and standardized in clinical settings, but are less so in the research arena, he observed. “We need more standardization, validation, and oversight of research to ensure that data are robust enough to match the standards everyone likes to see on the clinical side,” he added.

Please help us improve Strategies by rating how useful this article is to your job.