quinta-feira, 6 de janeiro de 2011

Alzheimer's blood test 'most accurate' so far

The blood of patients with the brain disease contains antibodies not found in healthy people.

Ewen Callaway
A new screen could be the most accurate test yet for Altzheimer's disease.PASIEKA / SCIENCE PHOTO LIBRARYA new blood test diagnoses Alzheimer's disease by sensing molecules produced by the immune systems of people with the neurodegenerative condition.
So far, the test has been applied to just a small number of blood samples, but if proven on a larger scale, the assay could help diagnose Alzheimer's disease in combination with other tests, says Thomas Kodadek, a professor of chemistry at the Scripps Research Institute in Jupiter, Florida. It could also be used to identify patients for trials of experimental Alzheimer's drugs, he adds. His team published its results online today in Cell1.
Currently, the only way to conclusively distinguish Alzheimer's from other dementias by examining the gnarled plaques and tangles of protein found in the brains of those with the condition. This can only be done after death. A furious search is under way for earlier, less-invasive tests using brain scans, blood draws and spinal taps, for instance.
Globally, over 35 million people suffer from Alzheimer's. There are no effective treatments for the disease or proven means of preventing it.



Most trawls for blood biomarkers typically whittle down a list of potential molecules to a few that differ between people with a condition and healthy people. For instance, Tony Wyss-Coray's team at Stanford University in California screened 120 proteins involved in cell communication and found 18 of the proteins present at higher levels in the blood of people with Alzheimer's disease than others2.



However, such 'candidate' screens rely on assumptions about which molecules are likely to be affected by a disease. They also rely on sensing vanishingly small levels of proteins and other molecules.



Slide screen

Antibodies — immune molecules in blood that stick to specific foreign and disease proteins — offer a solution, says Kodadek. If specific to a disease, they are likely to be found in high levels in those with the condition, yet be absent in healthy people.



Typically, a scientist would fish out such antibodies using the disease proteins they attack. However, Kodadek's team reasoned that antibodies should also recognize other kinds of molecules. They created glass slides coated with thousands of differently shaped peptoid molecules — chemical cousins of peptides that are likely to be recognized by antibodies.



“One could imagine our test being used a cheap front line screen.”

Thomas Kodadek

Scripps Research Institute

As a proof of principle, Kodadek's team searched for antibodies produced by mice that have a multiple sclerosis-like condition caused by their antibodies attacking the fatty sheaths surrounding nerves. Among the 4,800 peptoids on the slide, several latched onto antibodies in the blood of mice with the disease, but this did not happen for controls. "It was a night-and-day difference, it wasn't a subtle thing," he says.



Next, the researchers tested the peptoid slides on blood from six likely Alzheimer's patients, six similarly aged healthy people, and six patients with a different neurodegenerative condition, Parkinson's disease.



They identified three peptoids that recognized antibodies from people suspected to have Alzheimer's. Tested against 16 different people with the condition, each peptoid proved more than 93% accurate at diagnosing Alzheimer's, meaning they missed only one Alzheimer's case out of 16.



Antibody issue

In unpublished work, Kodadek says his team has applied the test to about 300 people, correctly diagnosing 98% of the people thought to have Alzheimer's. Meanwhile, the test proved 95% accurate at distinguishing healthy people from those suspected to have Alzheimer's, yet Kodadek doubts all 5% are true false positives. "I'm really pretty sure what we're seeing is pre-symptomatic Alzheimer's," he says.



Current tests for Alzheimer's relying on brain imaging and spinal tap accurately diagnose about 90% of patients, says Kodadek. "One could imagine our test being used a cheap front line screen, and at least if you test positive there, you could go for these more expensive things that couldn't be used as a general screen."



ADVERTISEMENT





Kaj Blennow, an expert on Alzheimer's biomarkers at Sahlgrenska University Hospital in Mölndal, Sweden, calls the findings "very promising" but says they need to be replicated in a larger group of patients, including those at early stages of Alzheimer's and with other forms of dementia.



Wyss-Coray says it is still an open question as to whether or not people with Alzheimer's produce antibodies that are specific to their condition. However, the small number of peptoids unique to those with the disease could point to a very specific immune response against an unknown disease molecule. "If true, that would most certainly change the current view of this disease."



Kodadek's team is hoping to find the molecule or molecules recognized by the Alzheimer's antibodies. They also plan to apply the peptoid screen to discovering antibodies against various cancers.



A company he is involved with, Opko Health based in Miami, Florida, plans to work with a large pharmaceutical firm to identify people suitable for its clinical trials. "If we're ever going to get an Alzheimer's drug, we need a much better early diagnostic. Otherwise these trials are doomed to failure," Kodadek adds.



References

Reddy, M. M. et al. Cell 144, 132-142 (2011).

Ray, S. et al. Nature Medicine 13, 1359-1362 (2007).

Comments

If you find something abusive or inappropriate or which does not otherwise comply with our Terms or Community Guidelines, please select the relevant 'Report this comment' link.



Comments on this thread are vetted after posting.



There are currently no comments.



Add your own comment

This is a public forum. Please keep to our Community Guidelines. You can be controversial, but please don't get personal or offensive and do keep it brief. Remember our threads are for feedback and discussion - not for publishing papers, press releases or advertisements.



You need to be registered with Nature to leave a comment. Please log in or register as a new user. You will be re-directed back to this page.



Log in / register