This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American
WASHINGTON—Many doctors and medical researchers applauded a new federal focus on comparative effectiveness research that was boosted through the 2009 stimulus package and codified with the signing of the health care reform bill in March. This shift will support—and in many cases require—work to establish which treatments work best.
The changing policies, however, present "a conundrum" for cancer work, Amy Abernethy, an oncologist at Duke University Medical Center in Durham, N.C., said here Tuesday at the annual meeting of the American Association for Cancer Research. Actually, they present at least a couple.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
The first challenge is that comparative effectiveness research, by necessity, focuses on broad populations of patients. But as Abernethy pointed out, "in oncology, we take care of individuals." Newly prioritized research might turn out new results about treatments that are best for most breast cancer patients with a defective HER2 protein, for example, but many other individuals with different genetic factors might still need trial-and-error treatment until research finds what works best for them.
Another big quandary is that though zeroing in on the most medically effective and cost-effective treatment for individual patients is a respectable goal, the current literature on treatments leaves many researchers and clinicians wanting. Few studies (less than a third) published recently in major medical journals have added to the overall comparative effectiveness literature, according to a March 2010 paper in The Journal of the American Medical Association.
And the problem is even worse for assessing off-label uses of cancer drugs (uses for approved drugs that have not been specifically okayed by the U.S. Food and Drug Administration (FDA)), Abernethy pointed out. Some 50 to 70 percent of all cancer drug prescriptions are for off-label uses, she noted, and although the FDA allows such uses in the oncology world—and even permits reimbursement—it requires backup from the literature to justify these decisions. However, off-label studies are often poorly conducted and frequently completed without a control population, Abernethy said, leaving doctors, patients, insurers and the government in the dark about the actual efficacy of a lot of very frequently prescribed meds.
"How can we, as a society, be confident we are paying for effective treatment?" Abernethy asked. And as more people with preexisting conditions, including cancer, will be eligible for coverage under health care reform, the overall treatment bills could escalate without firm evidence that the expensive drugs are really making them better.
With finite resources for expensive, well-run clinical trials, many researchers and clinical oncologists are hoping that better data gathering and sharing will soon fill in some of the blanks. The federal government's support for improved health care information technology (through health care reform and stimulus funds) should help providers and patients alike quickly access the latest data about particular treatments and patient subtypes. And to truly boost knowledge in the field, the data networks should allow doctors and patients to contribute their own findings and experiences, Kenneth Buetow, of the National Cancer Institute's Center for Bioinformatics in Rockville, Md., said at the meeting. These resources would help "share information in a partnership with our patients," Buetow said.
The growing web of data available through linked research centers and through patient electronic medical records will also help power better clinical trials when they are feasible. To eventually find the best treatments for particular patients, testing specific patient populations—down to genetic variants—is considered a must. And electronic medical records will help researchers find patients who might be best suited to specific trails. "We need to find the right patient with the right target for what we're testing," John Niederhuber, director of the National Cancer Institute, said in an address here on Monday. And because much of the data can be gathered and input remotely, "the trial will go to the patient," allowing patients to be enrolled as soon as they are diagnosed and increasing the likelihood they will stick with a trial for the long haul. With real-time subject monitoring, or close to it, Niederhuber noted that the current practices of infrequent checkups and screenings will seem clumsy and crude. And as the information gap between treatment and research continues to shrink, keeping data lodged in hard-copy records or images is "just not going to get us anywhere," he said.
This sort of connected cancer research and treatment world would resemble what Buetow described as the "e-ecosystem" being created through caBIG (the cancer Biomedical Informatics Grid—an information network), through which scientists, practitioners and patients will be able to access and share data. For researchers and even clinicians, "the days of holding all of your information and insight close to your vest are numbered," Buetow said.
With the field of cancer research poised for what Francis Collins, director of the National Institutes of Health, described here Tuesday as "a true revolution," the pressure is on to have the information infrastructure to support it. Buetow was cautiously optimistic about the government's ability to boost research. "We never seem to have enough time and money to do it right, but we always seem to have time and money to do it over," he said of other federally funded projects. "This time, I hope we do it right."
Image courtesy of Wikimedia Commons, National Naval Medical Center/Seth Rossman
