Alzheimer's disease is diagnosed clinically by using tests that measure memory, language skills and other brain functions and by excluding other causes of dementia. A definitive diagnosis can only be made at post-mortem. At present it is impossible to predict which patients who have mild memory loss which does not significantly affect their everyday activity (mild cognitive impairment; MCI) will slowly progress to Alzheimer’s disease.

The brain controls many body functions by releasing proteins, and immune and inflammatory mechanisms which alter plasma protein concentrations have been implicated in Alzheimer’s disease. The Stanford-led research workers first examined the concentration of 120 of these proteins in samples from 43 Alzheimer’s patients and 40 controls without dementia using an ELISA (enzyme-linked immunoadsorbent assay). Computer analysis found that 18 proteins showed differences in concentration that ‘clustered’ in the two groups. An equation was derived that correctly classified 95% of Alzheimer’s patients and 83% of controls.

The equation was then applied to a second set of samples from 42 Alzheimer’s patients, 11 patients with other dementias and 39 non-demented controls. Ninety percent of the Alzheimer’s results were positive and 88% of the results from the other two groups were negative.

The workers also examined specimens from some other patients: 21 with other neurological diseases and 16 with rheumatoid arthritis. They reported a good separation from all Alzheimer’s samples. (However, examination of Supplementary Figure S2b published online shows that 5 of the ‘other neurological diseases’ samples fell within the Alzheimer’s cluster and were well separated from the rest. Their diagnoses were not specified.)

The researchers finally examined 47 specimens that had been taken and stored from patients who were diagnosed with MCI (mild cognitive impairment) from 2 to 6 years before a final diagnosis was made for this study. Of the 20 with a final diagnosis of Alzheimer’s, 91% were positive, and all 8 of those who developed other dementias were negative. Seventeen patients still had a diagnosis of MCI, 7 positive and 10 negative, at the time of the study.

Rebecca Wood of the UK Alzheimer’s Research Trust, said “This is exciting research and the results and accuracy levels are very promising. However, further research is needed with a larger sample group to confirm these findings and determine if a simple, accurate blood test for Alzheimer’s can become a reality.”