Recent reports of breakthrough research on a potential blood test to predict the onset of Alzheimer disease before symptoms lead to a clinical diagnosis may actually be less promising than they appear on the surface, according to the scientific director of the Alzheimer Society of Canada.

"This just didn't switch me on. There is some caution attached," says Dr. Jack Diamond.

The research, conducted at Stanford University and published in the fall in the British journal Nature Medicine, looked at proteins in blood plasma that are used by cells to communicate with each other. Eighteen of the 120 proteins screened showed a consistent abnormality in people with Alzheimer's and the researchers hoped this chemical signature could be a marker for early diagnosis of the disease.

They tested the theory in a group of people with Mild Cognitive Impairment (MCI), which is known to be a risk factor in the development of Alzheimer's, following them for up to five years and comparing blood tests with samples taken several years earlier. The results flagged the patients who developed Alzheimer's with about 90-per-cent accuracy.

Diamond says his reservations have to do with the numbers. It was a small sampling of 47 patients and the results were seen in about half who developed Alzheimer's, leaving the other half as a control group who did not develop the disease.

"We already know that more than 50 per cent of people with MCI will go on to develop Alzheimer's - in fact, the percentage is actually between 70 and 80 per cent - but it can take up to 10 years. So if the group had been followed longer than the five years, chances are others in that group would have developed the disease," Diamond says. "So I am leery about the control group."

The Stanford researchers allow that their studies are small but maintain that the concept is promising and they recommend further, larger-scale trails. And Diamond says that "any biochemical changes in tissues outside the brain in Alzheimer patients are worth investigating.'

He adds that other promising research is continuing, including work by B.C. neuroscientist Dr. Neil Cashman on blood tests for "misfolded proteins," fragments of which accumulate in the brain as the plaques that are seen in Alzheimer patients. There are also now more sophisticated imaging techniques that can see these plaques on brain scans, leading to more definitive diagnosis than was available in the past.