|
This story is from the category Health
Date posted: 14/07/2010 As a genre, personalized medicine has yet to deliver many individualized treatments. But progress has been more tangible on the diagnostics side. Some of that progress has come from Tethys Bioscience, a startup in Emeryville, CA, that is working on tests for some of the country's most prevalent diseases. Their first diagnostic, a blood test called PreDx, determines a person's risk of developing diabetes over a five-year period. The test is already changing the way some physicians diagnose and treat their patients. People with prediabetes don't regulate blood sugar the way they should; of the 57 million or so people in the U.S. with this syndrome, only about 10 to 15 percent develop the full-blown disease. With proper diet and exercise, the trajectory toward diabetes can be slowed or even reversed. But the current standard for identifying diabetes risk, the fasting blood-glucose test, can't determine which 10 to 15 percent are most likely to get the disease. As a result, doctors can't easily figure out which patients to concentrate on. There are a few tests to determine which patients are high-risk, but these tend to be expensive, resource-intensive, time-consuming, or all three. Tethys's simple $250 blood test can be almost as accurate. "The PreDx risk score allows a physician to determine which subjects to focus their interventions on," says Tethys president R. Michael Richey. Suddenly, a patient who might have once been told that his lab results indicate he's on the verge of diabetes could be told the likelihood of disease progression as a percent. Someone else with similar fasting blood-glucose results might be told that if she doesn't lose weight and start exercising, her risk of developing diabetes is quite high. Richey and his colleagues at Tethys developed this "risk stratification" approach by hunting down blood-sample banks from long-term studies of large groups of patients with known health outcomes. The researchers tested these blood samples for as many potential diabetes markers as they could. Then they looked for protein combinations that differentiated those who developed full-blown diabetes from those who didn't. "We ran them through complex algorithms and found the smallest number of markers that, when taken together, would accurately predict incidence of diabetes," Richey says. See the full Story via external site: www.technologyreview.com Most recent stories in this category (Health): 03/03/2017: Interactive health apps may inspire healthy behaviors, but watch the tone |
|
|||||||||||||||||||||||||||||||||||||||||||||||||
