For the first time, a blood test appears to predict with an impressive level of certainty whether patients will develop Alzheimer's disease within three years. Researchers with the Georgetown University Medical Center, along with colleagues at the University of Rochester School of Medicine and elsewhere, came up with the provocative new diagnostic after a trial involving 525 initially healthy elderly people.
The journal Nature Medicine details their discovery and work. Media outlets including MedPage Today, CNN and the Daily Mail in the U.K. summarize the accomplishment.
Much more work is needed here, but the early results point to a test that could someday serve as a preclinical diagnostic for Alzheimer's that's relatively easy to use. With that in hand, clinicians could, in theory, start treating patients to combat Alzheimer's before it creates too much neurological damage. There's a long way to go before we reach that point, however. The study sample did generate a 90% specificity and sensitivity rate as far as determining which patients in a particular subgroup would develop Alzheimer's within three years and which ones would remain normal. But as MedPage Today noted, the study produced a 5% rate of conversion from normal cognition to mild neurodegenerative impairment or Alzheimer's. As the story noted, if that reflects a real-world population, then the test shows a 35% positive predictive value, versus a 90% minimum screening test result that clinicians typically want for these kinds of tests.
If further testing can boost the positive predictive value and continue to show high levels of specificity and sensitivity, then the test could be a major advance for Alzheimer's care. The quest for a viable Alzheimer's drug has been frustrating and unsuccessful, and so drug developers are now pursuing treatments with an early diagnosis in mind. The research team's test could enable and even accelerate work in this area, once it generates more data from a wider patient pool.
That next test is already in the cards, Howard Federoff, a professor of neurology at Georgetown University Medical Center, said in a statement.
"We consider our results a major step toward the commercialization of a preclinical disease biomarker test that could be useful for large-scale screening to identify at-risk individuals," Federoff said. "We're designing a clinical trial where we'll use this panel to identify people at high risk for Alzheimer's to test a therapeutic agent that might delay or prevent the emergence of the disease."
Basically, the team came up with a blood test of 10 lipids as an outcome of their evaluation of patients over the course of their 5-year study, a subgroup of which was found to have low levels of these lipids who later developed cognitive impairment and Alzheimer's. MedPage Today noted that the model worked better in predicting which patients would convert from normal neurocognitive function to impaired levels, versus being able to simply distinguish between healthy and impaired patients.
The researchers said their method of screening for lipids may eventually work better than existing biomarker tests that screen for more accepted tags such as tau or amyloid beta proteins, because they rely on more expensive, invasive and time-consuming diagnostic methods such as cerebrospinal fluid tapping or MRI/brain amyloid imaging.
- read the release
- here's the journal abstract
- check out MedPage Today's story
- read CNN for more
- here's the Daily Mail's take