Is comparative effectiveness antithetical to personalized medicine? A report posted on Reuters suggests that Francis Collins, the new head of the National Institutes of Health and champion of the Human Genome Project, thinks so. Reuters quoted Collins predicting “a potential collision [between personalized medicine and comparative effectiveness],” at a forum sponsored by the American Association for the Advancement of Science.Collins worried out loud that comparative effectiveness studies will lump all people together when considering whether a novel drug works better than drugs already on the market. Collins’s concern is that comparative effectivness studies may overlook the fact that the novel drug may work better for smaller subpopulations of people than an available treatment, even if it’s not more effective in a disease population overall. Such distinctions may “get lost in the wash by considering everybody equivalent, which we know they are not,” Collins was quoted as saying. As an example of the power of genetic tests in pinpointing such subpopulations, Collins cited a predictive test produced by Genomic Health that determines the likelihood of recurrence in women that have had breast cancer, based on their genetic profiles.

Dividing patients into subpopulations is hardly a novel concept. It’s been a long time since anyone lumped together all cancer patients, or diabetes patients, or patients with cardiovascular disease. I’d even say that some sort of personalized medicine is built into many drug discovery programs these days, as drug companies have come to grips with the notion that the days of the one-pill-fits-all blockbusters are mostly behind them. At the same time I am not prepared to believe that Dr. Collins is conflating the notion of personalized medicine (also called “targeted therapeutics,” with treatment targeted to specific subpopulations of patients) with “individualized” medicine, where each patient receives medication tailored for their unique genome and none other (with the possible exception of autologous cell donation). So again, I must wonder whether Dr. Collins isn’t setting up a false dichotomy. I see no reason for comparative effectiveness to be pitted against personalized medicine. Rather, I can foresee a comparative effectiveness approach that amends the Utilitarian notion of “the greatest good for the greatest number,” to the “greatest good for the greatest number for whom it will do the greatest good.”