A blood test for Alzheimer’s?

What increases our likelihood of suffering from a certain disease or condition? That’s obviously a key question for a lot of health-related research. On the other side of some diagnosis, attention is focussed on what might cure or reduce the symptoms of a disease or condition. We see a lot of stories in the media on these two points: things that will give us some disease or other, and things that might cure us of some disease or other. But there’s another equally interesting area: how might we diagnose people earlier, before they develop the symptoms of some disease or condition? There are many conditions where we want to not only find better treatments, but also develop news ways to diagnose people earlier to get them started on the treatments we currently have. In that regard, a recent Daily Mail headline ran “Blood test that can predict Alzheimer’s”, following with “Elderly could be given early warning”. This was similarly reported by the BBC under the banner “Blood test can predict Alzheimer’s, say researchers”. Before considering what that might mean for a disease like Alzheimer’s where there is currently no cure, let’s explore the story.

What did the media say?

A number of media outlets reported a recent study that had developed a blood-based test for Alzheimer’s. Before going much further, Alzheimer’s disease “is the most common cause of dementia… The term ‘dementia’ describes a set of symptoms which can include loss of memory, mood changes, and problems with communication and reasoning” (the Alzheimer’s Society have some very good information if you want to learn more about Alzheimer’s and other forms of dementia). The Daily Mail article described the study as showing that a “simple blood test” predicted if a person might develop Alzheimer’s over the next 3 years, with 90% accuracy. The piece clearly refers to the study team, and where the work was conducted and published (more on the actual study below). One of the key things the article does, however, is flag up the ethical angle with this development. Without a cure, what is the benefit of knowing you might develop a disease? The Daily Mail piece touches on some of these concerns by reporting that a general opinion is that diagnosis may be too late for our current treatments to have their full effectiveness, and that if earlier diagnosis were possible, a small delay in the onset of symptoms due to earlier treatment might be very welcome. The BBC also reported a feeling that many drugs are not as useful as they could be because of when they are often prescribed—too late in the process of deterioration to have an impact. The BBC also highlighted how larger, longer-term work is need on the new blood test to see if the current finding can be replicated and, indeed, how many years before Alzheimer’s is diagnosed that these blood markers can be identified.

What about the actual research?

The study was led by Dr Howard Federoff at Georgetown University, and published in Nature Medicine. The researchers note that current diagnostic tools for Alzheimer’s are invasive (such as drawing spinal fluid) or expensive (including different types of brain imaging), and so a blood-based test would be a valuable development. The study recruited 525 individuals who were generally healthy at baseline, and aged 70 or older. This included 74 individuals with Alzheimer’s or a mild cognitive impairment (a change in cognitive function defined primarily by a memory impairment), with 28 of these having received a diagnosis over the course of the study. To develop the blood test, however, 53 of these individuals were looked at initially, matched with 53 individuals who didn’t develop Alzheimer’s or memory impairment. The researchers then examined blood samples from these individuals to look for differences in biomarker profiles (differences in the levels of various lipids and metabolites). This suggested that 10 markers in the blood were present in different amounts in the Alzheimer’s/memory impairment group versus the control group. The researchers then took the blood samples of 40 other individuals from the original sample (but not those looked at in the “discovery phase”). Using their 10 markers, they then tried to classify these 40 individuals as either Alzheimer’s/memory impaired or healthy, but without looking at their other clinical data. This is referred to in the literature as a “blinded” procedure. Their analysis suggested that using the blood markers, they were 90% accurate in classifying individuals. Neither of the media reports were able to go into too much detail on the full methods of the study, but the BBC report did highlight the fact that the testing was done on much smaller subsamples from the over 500 people recruited. And what about the 10 blood markers? The researchers report that some of them are fatty compounds that have important roles in the structure and function of cell membranes, for example. The accumulation of these markers in the blood of those who developed Alzheimer’s, they suggest, might be driven by the degradation of cells in the brain.

The bottom line.

The study is obviously of interest to other researchers, clinicians and the public generally, as it offers the possibility of starting treatments earlier that might add an extra few months (or years) before the onset of the symptoms of Alzheimer’s (however, the ethical questions raised go beyond Alzheimer’s and other dementias, being relevant to any progressive or degenerative condition where the diagnostic tools might outstrip our treatment options). The researchers were also careful to highlight the need for more work to further develop the test to ensure it is of value in the population more generally. But, without more than a delay in symptoms, the big question remains: would you take the test?

Mapstone, M., Cheema, A. K., Fiandaca, M. S., Zhong, X., Mhyre, T. R., MacArthur, L. H., Hall, W. J., Fisher, S. G., Peterson, D. R., Haley, J. M., Nazar, M. D., Rich, S. A., Berlau, D. J., Peltz, C. B., Tan, M. T., Kawas, C. H., & Federoff, H. J. (2014). Plasma phospholipids identify antecedent memory impairment in older adults. Nature Medicine. doi: 10.1038/nm.3466