麻豆传媒

A blood test for Alzheimer鈥檚 might be just two years away.

at King鈥檚 College London and his colleagues have identified 10 proteins in blood that can predict who will develop Alzheimer鈥檚 disease a year after having mild memory problems. Its accuracy is almost 90 per cent. That could prove a huge boost for researchers seeking treatments.

So far, trials of Alzheimer鈥檚 drugs are thought to have failed because they have been given too late in the course of the disease to halt progression. The new blood test will initially be used to identify those people with mild cognitive impairment who are likely to get Alzheimer鈥檚 disease and so might be good candidates for clinical trials to find drugs that halt disease progression.

鈥淗aving a blood test is a really big step forward,鈥� says team member Ian Pike of Proteome Sciences in Cobham, UK. 鈥淭he most important thing we can do is get the correct patients into clinical trials so we can tell, for example, whether it is a drug that is slowing the progression of the disease or the fact that we just happen to have a group of patients who have a slow progressing form of the disease.鈥�

Feared diagnosis

鈥淭his [blood test] is a technical tour de force,鈥� says , director of research at the Alzheimer鈥檚 Research UK charity. However, he remains cautious about its use beyond clinical research. For every 10 people who take the test, one will get an incorrect result. 鈥淎lzheimer鈥檚 is the most feared diagnosis, so we have to be careful, particularly in the absence of any treatment,鈥� he says.

The researchers are also calling for caution until larger trials have taken place. 鈥淟ots of blood tests said to be the next big thing haven鈥檛 come to anything 鈥� we have to replicate these results with larger numbers,鈥� says team member now at the University of Oxford.

Hye and his colleagues analysed 26 proteins in blood from 1,148 people, including 476 people with Alzheimer鈥檚, 220 with mild cognitive impairment and 452 elderly healthy controls. Almost half of the participants in each group also had an MRI brain scan.

Further clues

The team identified 16 proteins that were strongly associated with brain shrinkage in people with mild cognitive impairment or Alzheimer鈥檚. A further analysis showed that 10 of these proteins could predict whether individuals with mild cognitive impairment would develop Alzheimer鈥檚 within a year. Some of these 10 proteins were associated with tau and amyloid proteins 鈥� both found in damaged brain tissue in Alzheimer鈥檚. This may yield further clues as to the cause of the disease.

Other blood tests have claimed to be able to predict Alzheimer鈥檚. For example, at Georgetown University in Washington DC, published a paper in March, describing a test that uses the level of 10 lipids in the blood to identify people who will go on to get Alzheimer鈥檚 two or three years later with an accuracy of 96 per cent.

However, the UK team says it is confident that the newer test will be easier to translate into clinical practice because it uses simple technology to measure proteins in the blood, whereas lipid measurement is more challenging.

As with any test for Alzheimer鈥檚 鈥� a disease for which there is no cure 鈥� there is always the question of whether people would want to know the result. Lovestone thinks some will. 鈥淚 have people with memory problems come to me and want to know what鈥檚 happened to them. I have to say 鈥榮ee me in a year and I鈥檒l tell you whether you鈥檝e got dementia or not鈥� 鈥� and that鈥檚 grim. So there are people who want to know.鈥�

鈥淚 think it will give patients and family members a degree of clarity of what the future may hold,鈥� Pike says, 鈥渨hich allows you to plan for the inevitable.鈥�

Journal reference: Alzheimer鈥檚 & Dementia, DOI: