January 15, 2012 (Vol. 32, No. 2)

Necessary Actions Include an Evidence Base as well as Physician, Patient, and Payer Education

More than a decade after the successful mapping of the human genome, clinical genomics is starting to permeate important parts of patient care and ripple throughout the entire U.S. healthcare system.

This increased adoption is partly due to growing acceptance of useful interventions, but it is also due to loose regulatory oversight that allows marketing of unproven tests or tests of no value that, one expert has warned, “threaten to overwhelm the health system.”

Actual usage and cost are impossible to ascertain because individual tests are not billed separately with Current Procedural Terminology (CPT) codes; the most recent data comes from 1996 from a mailed survey to labs. Nonetheless, the evidence strongly suggests that while use is growing rapidly, it too often contributes to wasteful spending and even patient harm through problems such as inaccurate cancer diagnoses.

As controversies over BRCA testing have shown, so-called personalized, predictive, preventive, and precision medicine can also be political, profit-, and plaintiff-driven and perplexing to patients and professionals alike. It is time for the policy community, long fluent in the argot of DRGs and billing codes, to acquire similar proficiency in the language of DNA and genetic codes. That process should start by examining what is hype, what is hopeful, and what is actually starting to make a genuine difference in patient care.

Although the field has fallen far short of the transformational therapeutic impact once widely predicted, genetic testing routinely guides therapy in several cancers and in HIV disease, and use of costly, targeted anticancer drugs is rising sharply. Six in ten primary-care physicians have ordered a genetic test, primarily for diagnostic purposes.

Pharmacogenomics (PGx) has become an established part of drug discovery. The Food and Drug Administration has approved biomarker information related to clinical usage for 76 unique drugs affecting medications in 18 therapeutic areas, including heart disease, depression, and pain, with a preponderance of oncology drugs.

However, government oversight does not appear to be keeping pace. An advisory committee to the Secretary of Health and Human Services recommended better regulation of genetic tests in 2008, but that committee was disbanded in 2010. Also in 2010, the FDA cited “public health concerns” about genetic tests and signaled its intent to regulate, but it has not done so with the exception of the small market for direct-to-consumer (DTC) tests.

A voluntary Genetic Testing Registry is set to be launched soon by the National Institutes of Health; the advisory committee had recommended a mandatory one. The Centers for Disease Control and Prevention’s Office of Public Health Genomics, which leads assessment of the evidence base, had its 2012 funding request slashed to under $1 million from $12 million. Just 1.8% of grants in cancer genetics for 2007 by the National Cancer Institute went to translational research, and a scant 0.6% of publications dealt with translational issues.

A sweeping update in CPT billing codes in 2012 is expected to include more than 90 gene-specific and genomic procedures in the category of commonly performed tests, making it much easier to order and be reimbursed for them. Yet primary-care physicians are confused: while 98% agree genetic testing is useful, just 10% feel adequately informed about using it.

Although professional groups are working to improve guideline development and dissemination, the for-profit sector, led by large pharmacy benefits managers, is much more rapidly building an implementation infrastructure. As with diagnostics, clinical genomic therapeutics also has a DTC component that skirts traditional gatekeepers.

Michael L. Millenson

Health Information Technology

In addition to these challenges, issues in integrating clinical genomics into routine care include health information technology (HIT) systems unable to handle genomic information in a structured way and the necessity for comparative effectiveness research that examines identifiable sub-populations. This field also brings awkward questions about genetic definitions of race and religion into the exam room and can add to concerns about underserved populations.

The buzz surrounding clinical genomics is similar in some ways to the early enthusiasm for HIT and should sound a warning about expectations. As clinical genomics grows in importance, it should be held to the same standards as other interventions. Particularly because of its complexity, one overarching imperative is for improved transparency of information to all stakeholders. This includes:

  1. Data on evidence. A recent JAMA commentary noted “the thin line between hope and hype in biomarker research.” There are large gaps in knowledge, either because studies have not been done or not been made public by those doing them.
  2. Data on implementation. NIH could demonstrate that its forthcoming registry serves constituencies beyond clinical laboratories by making available comprehensive information to be used by researchers and clinicians. The government could also fund more reliable research on benefits and harms, both clinical and economic, of genomics in actual practice. At present, the literature is largely filled with a sparse supply of oft-repeated anecdotes that provide little useful guidance.
  3. Data on regulatory actions. The FDA could regulate laboratory-developed tests (LDTs), report on what LDT tests have been approved, and initiate post-marketing surveillance on safety. FDA data on PGx and drug labels could also be examined.
  4. Data for payment. Payers could be much clearer about what evidence they require. The Medicare Payment Advisory Commission could counterbalance advocacy groups by giving Congress objective advice about appropriate implementation and reimbursement. Evidence-based criteria will become critical as expensive medications targeting small subgroups of patients, particularly those with cancer and other serious diseases, become more common.
  5. Data for consumers. Congress and federal agencies could give the public information and tools to deal with a technology that often raises uncomfortable ethical, clinical, and economic questions. For example, prenatal genetic testing is becoming easier and less expensive, as is testing newborns for predisposition to a long list of diseases that may or may not be clinically relevant.

Finally, as the tools of DNA discovery enter mainstream medicine, they need a different descriptor. People are more than the sum of their genes, terms such as “genetically personalized medicine” or “personalized genetic medicine” represent an important acknowledgement that genetic knowledge is just one component of the individualization of care.

Developing the evidence base and educating doctors, patients, and payers on how to use that evidence will take time, patience and money. As clinical genomics moves from bench to bedside and becomes a significant component of the U.S. healthcare system, it is an investment well worth making.

Michael L. Millenson ([email protected]), president of Health Quality Advisors, is a senior policy consultant to the Urban Institute.

Previous articleScientists Identify New Target for Apicomplexan Pathogens
Next articleMedicago Taps Cellectis to Boost Therapeutic Protein Production in Tobacco Plants