Scientists are a step closer to developing a blood test for Alzheimer’s disease following the publication online this month in Neurology of a new study that found four biomarkers showed consistent results across three independent groups of patients.

Current methods for diagnosing Alzheimer’s are based mainly on clinical symptoms that often have to be confirmed with expensive PET scans, or by testing for beta-amyloid protein in samples of cerebrospinal fluid with a procedure that can be painful and distressing.

Over the last few years, studies have come out suggesting that a blood test for Alzheimer’s is around the corner, but the problem is the results don’t hold up as well when researchers try to repeat them in other groups of patients.

Now in this latest study, lead author William Hu, assistant professor of neurology at Emory University School of Medicine in the US, and colleagues, found four of the 190 markers they tested for were consistently associated with the diagnosis of very mild dementia, mild cognitive impairment and Alzheimer’s disease in three independent clinical cohorts.

The researchers worked with 600 participants in two cohorts at the University of Pennsylvania and Washington University, St. Louis, where some of Hu’s collaborators are based. They then validated the results in the multicenter Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort.

Hu told the press:

“Reliability and failure to replicate initial results have been the biggest challenge in this field.”

But now, he added:

“We demonstrate here that it is possible to show consistent findings.”

Hu and his collaborators started by looking at 190 proteins and peptides in blood samples from the 600 participants at the two institutions. The participants were healthy volunteers and patients diagnosed with Alzheimer’s disease or mild cognitive impairment (MCI).

MCI is a cluster of slight but measurable lessening of reasoning and memory skills that precedes the onset of Alzheimer’s, and is often considered a potential early sign of the disease. However, not everyone who has MCI will develop Alzheimer’s.

The researchers found 17 of the 190 compounds they tested for had significantly different properties in people with MCI or Alzheimer’s.

So they checked those 17 potential markers against data from 566 participants in the ADNI cohort, and found four of them showed consistent properties across all three cohorts. These compounds were: apolipoprotein E, B-type natriuretic peptide, C-reactive protein and pancreatic polypeptide.

The researchers also found changes in levels of the four markers were consistent with measurements from the same patients of beta amyloid protein from cerebrospinal fluid samples that had previously been linked to Alzheimer’s.

The results grouped together people with MCI who were at high risk for Alzheimer’s, and people with full-blown Alzheimer’s.

This was a key finding in the study, because one of the problems researchers are coming across in their search for a reliable blood test for Alzheimer’s is finding markers that are sensitive enough to capture the physiological changes that are common to both MCI and AD.

This study has only taken one step closer though, because there still remains the considerable challenge of showing these markers are unique to Alzheimer’s (the problem of “specificity”). The cohorts only contained small numbers of patients with non-Alzheimer’s dementia.

Another problem, said Hu, is that the three cohorts had different proportions of people with MCI and Alzheimer’s, which makes it difficult to pinpoint the changes that relate to one as opposed to the other.

The study has, however, made another significant contribution in that the researchers have identified ways in which to make sure a test is reliable.

The results now need to be repeated with larger groups, including patients will different forms of dementia.

“In the meantime, the combination of a clinical exam and cerebrospinal fluid analysis remains the best tool for diagnosis in someone with mild memory or cognitive troubles,” said Hu.

Written by Catharine Paddock PhD