Will focusing on comparative effectiveness leave personalized medicine research in the dust?

That’s what Francis Collins, director of the National Institutes of Health Director, cautioned might happen while presenting at a colloquium on personalized medicine hosted by the American Association for the Advancement of Science. "We need to be mindful of the goal of comparative effectiveness research and not lose all that we have gained in understanding how individuals differ and how that could be factored into better diagnostics and preventive strategies," Collins said at the meeting.

Comparative effectiveness means comparing the efficacy of at least two treatments -- a drug versus surgery or several competing drugs -- with respect to a specific condition. The goal of comparing different approaches is to evaluate the benefits and risks of each, and to determine if a particular treatment is worthwhile and for what types of patients.

An article in Genomeweb breaks down Congress’s investment of $1.1 billion in comparative effectiveness research: “$400 million to the NIH, $300 million to Agency for Healthcare Research & Quality, and $400 million to the Office of the HHS Secretary to create the Federal Coordinating Council for Comparative Effectiveness Research.”

However, Collins’s comments raises some concerns expressed by personalized medicine advocates that prioritizing comparative effectiveness is not such a good idea, and could represent a step back in personalized medicine research.

Is this focus on comparative effectiveness research a good idea? And will it slow or halt adequate personalized medicine research?

Victoria Stern, Editorial Intern, The Scientist

As usual, Collins is correct. Comparative effectiveness research is promoted by politicians as a way to cut health care costs. Once a particular treatment is judged "most effective" in some way which is likely to involve its cost, everyone will be expected to be treated this way. This may be a reasonable approach in some cases but even now there are multiple treating options for many disorders. No one treatment is found to be the best and the choice depends upon the patient, the physician, the subtype of the disorder, the clinical setting and numerous other factors. Would we want to treat all cases of breast cancer the same way? We are finding that this is a very heterogeneous disorder with new treatments being developed faster than comparative effectiveness research can deal with them. This is likely to be the typical situation for other disorders as we learn more about them.

I see both of these as complementary approaches. CE research will identify populations for which treatments are effective or not. Then PM can be used to study why the therapy did not work in a subset that was identified in the CE trial.

A sophisticated computational algorithm, applied to a large set of gene markers, has achieved greater accuracy than conventional methods in assessing individual risk for type 1 diabetes. (show full discovery on the labslink.com, Oct 10, 2009) A research team led by Hakon Hakonarson, M.D., Ph.D., director of the Center for Applied Genomics at The Children's Hospital of Philadelphia, suggests that their technique, applied to appropriate complex multigenic diseases, improves the prospects for personalizing medicine.

We need a bit more discussion of the comparative effectiveness of the methods to implement research findings or results in practice so that one maintains enough flexibility for individual variation and innovation. Perhaps one way is to require an extra "activation energy" for deviation from the norm (e.g., a written and filed justification, or a decreased reimbursement). Both personalized medicine and genomics have the potential to greatly reduce costs or to stifle treatment depending on how they are used. We might need more study of optimizing regulations so that they don't get out of hand, but preserve the best intent. Too many seem written in stone, and then over-implemented.

cdb/UF