Testing for Alzheimer's disease

At present there are no readily available diagnostic tests for Alzheimer’s disease or tests to reliably predict high risk of developing the disease in the future. However many research groups are working in this area and some progress is being made. Currently the only way to diagnose Alzheimer’s disease with certainty is retrospectively by examining the brain under the microscope after death. A diagnosis of dementia is made by assessing cognitive function using a variety of simple mental status tests or more complex neuropsychological tests. Together with the results of a physical examination, some laboratory tests to rule out things like kidney or liver disease, vitamin deficiency or thyroid disease and often a brain scan to rule out strokes or other physical brain disorders, a diagnosis of Alzheimer’s disease as the cause for the dementia can be made with nearly 90% certainty.

Many researchers are looking for laboratory tests that can identify markers specific to Alzheimer’s disease. There are now some tests that show a person has an increased risk for Alzheimer’s disease or that they probably have Alzheimer’s disease. Some of the approaches are described below.

Genetic testing; the Apo E genes have been known to be associated with risk of Alzheimer’s disease for many years. People with one or especially two copies of the apo E4 allele have a higher risk of developing the late-onset Alzheimer’s disease (the common form of the disease). However not all people with Alzheimer’s disease have any apo E4 genes and not all people with apo E4 alleles get Alzheimer’s disease so the test is of very limited use. There are also other genetic markers for specific types of early-onset Alzheimer’s disease but this is much less common than the late-onset type.

CSF markers; there are a number of substances that can be measured in cerebrospinal fluid (CSF) that are linked to Alzheimer’s disease. However since collection of a CSF sample requires a lumbar puncture (spinal tap) this is not a simple way to develop a diagnostic test.

The blood biomarker panel approach. A number of these types of studies have been published in recent years. In this approach a large number of substances are measured in blood and then statistical analysis is used to see which ones may be risk markers or diagnostic markers. A study published in 2007 caused some initial excitement when it was announced that measurement of a panel of 18 cell communication proteins in blood could diagnose Alzheimer’s disease with 90% accuracy and predict the disease 2 to 6 years prior to clinical onset. As yet however this approach has not led to any clinically useful test.

Very recently more studies of this type have been published and two of these are from Australia. One Australian group measured 174 different substances and found that 18 were most useful. Measuring these and adding in the patient’s age and body mass index gave an accuracy of about 85% in the diagnosis of Alzheimer’s disease. A US group publishing at the same time as the Australian study also found useful biomarker profiles but with little overlap in the markers that were found to be useful. A second Australian group analysed data from a large US study and found a group of 11 markers with a similar diagnostic accuracy after starting with 190 blood proteins. A group in the US took a similar approach and from the 190 different proteins narrowed their list down to a panel of four proteins that could be used as a diagnostic test. Dr William Hu from the group acknowledges that although the results are significant, "A blood test to identify underlying Alzheimer's disease is not quite ready for prime time given today's technology.” The fourth recent study also from the US, took a different approach and measured special types of fats called ceramides in the blood of their subjects. This study followed 99 women for 9 years and found that high levels of ceramide in the blood increased the risk of developing Alzheimer’s disease by almost 10 times.

What can we make of the results so far? A useful blood test for Alzheimer’s disease should be both highly accurate and cheap. None of the tests described above meet either criterion. An ideal test would be one that measures one substance that is unique to Alzheimer’s disease and is easy and cheap to perform. Tests like this are actually quite uncommon in laboratory medicine, most tests have sensitivity and specificity both less than 100% and thus results need to be interpreted along with other relevant clinical information.

Watch this space. The test we are waiting for may be just around the corner.

« Back to News
Testing for Alzheimer's diseaseLatest Testing for Alzheimer's disease RSS

Was this page helpful?