A Neurologist’s Deep Dive Into Blood Biomarker Tests for Alzheimer’s

Here’s the funny thing about Alzheimer’s biomarker blood tests: They can’t actually determine whether or not a person has Alzheimer’s. So what are they good for? They can signal the presence of biomarkers, which indicate Alzheimer’s risk. Better understanding your Alzheimer’s risk could inspire science-backed lifestyle changes to help prevent Alzheimer’s.

In the meantime, these tests’ results can be confusing to interpret — and people still have questions about how the landscape will evolve in the coming years. 

Federal University of Rio Grande do Sul neurology researcher Wagner S. Brum has been studying how these biomarker blood tests work, and how they could change the process of Alzheimer’s diagnosis in the near future. As an MD-PhD candidate, his research focuses on advancing the clinical interpretation of Alzheimer’s disease fluid biomarkers used in these blood tests. In particular, he has developed a two-step workflow to implement blood biomarkers in clinical practice and has been working on the clinical validation of plasma biomarkers. 

Brum joined Being Patient Founder Deborah Kan in a conversation about what we know about these blood tests and, ultimately, how they could aid diagnosis in the future. Read or watch the full conversation below.

Being Patient: What do blood tests for Alzheimer’s biomarkers tell us?

Wagner S. Brum: When we have a clinical analyte that we measure in blood in any routine area of medicine, we can always go all the way and try to think about the underpinning mechanisms. [For example], we’re going to take the two main classes of biomarkers and fluids traditionally used for Alzheimer’s, which are biomarkers measuring amyloid and tau in the fluids. These are also the two proteins that are canonically defining the disease. 

Dementia Biomarkers 101: What Does Beta-Amyloid Do In the Brain?

Alzheimer’s is understood as a disease [where] we have amyloid and tau accumulation in the brain. It makes sense to measure these proteins in the fluids. From a clinical standpoint, what these tests are good [at] telling us is whether or not an individual has an elevated level of amyloid in the brain. The blood biomarkers do more than a blood test measuring tau tell us about amyloid. There are more nuances to it. 

“From a clinical standpoint, what these tests are
good [at] telling us is whether or not an individual
has an elevated level of amyloid in the brain.”

Individuals that have more tau in the brain will have higher levels of tau in the blood. The most immediate clinical need when evaluating patients, given the current understanding of [the] disease, is to know whether a patient that presents to you with cognitive symptoms is within the Alzheimer’s disease continuum, which is defined by the presence of amyloid. We know that the blood tests are elevated in people with amyloid in the brains. 

Being Patient: In past interviews with neurologists, they’ve explained to me that people have amyloid in their blood, and it will cross over to the brain. Is there good amyloid and bad amyloid? What exactly are the blood tests actually picking up in terms of amyloid?

Brum: We take these amyloid plaques that we have in the brain as the center of focus in amyloid metabolism so that we can understand the biomarkers. There is one amyloid peptide that is 42 amino acids long. That is the one that is the most interesting and more clinically relevant within the Alzheimer’s field. 

What happens is that this peptide crosses the blood-brain barrier and goes into the blood. Since amyloid was defined as the disease-defining feature of Alzheimer’s back in the 80s, measuring amyloid in the blood has always been a big goal in the field. There have been lots of advances in that area, but what have we found right now?

[For example,] if you take a spinal fluid test, you have your brain, and you have the vessels that feed your brain. Let’s say that the spinal fluid is almost like the saliva around your brain. The things that happen there will go there, and they will get washed out into the blood. It’s a liquid that bathes your brain. If you measure this amyloid protein in the spinal fluid in a patient with Alzheimer’s that has plaques and compare it to a healthy patient, you will see that it is reduced. 

It’s a bit counterintuitive in the beginning. We know that having high blood sugar is bad, and we know that having high cholesterol is bad, but with amyloid in the fluids, it’s the opposite. Having low [amounts] is bad because it means that it is accumulating into blocks in the brain. 

“We know that having high blood sugar is bad, and
we know that having high cholesterol is bad, but
with amyloid in the fluids, it’s the opposite. Having low
[amounts] is bad because it means that it is
accumulating into blocks in the brain.”

Spinal fluid [is] a very clean fluid, and it’s very valuable for us [who] work with neurochemistry to measure processes there. You will see that Alzheimer’s patients have a very clear reduction in amyloid in the spinal fluid; let’s say it’s something that goes around 80 percent with not much overlap. In the blood, the problem is that [like] most brain proteins, they are present in the blood in a much lower concentration than they are in the spinal fluid. 

We need good technology to measure that, and right now, we have a wide array of different analytical platforms to measure amyloid in the blood. They all tell us around the same thing: amyloid in the blood is only reduced by 10 percent in Alzheimer’s patients. It’s a much tighter range of reduction than what we see in this spinal fluid. 

On top of that, there is a complicating factor that is more than just thinking of the central nervous system; it’s important to consider that amyloid is also produced outside of the brain in the periphery. We know that hepatocytes, which are cells composing our liver, and we know that platelets, which are part of our blood tissue that help us to coagulate, we know that … [amyloid is] to some extent produced in the periphery too.

“It’s important to consider that amyloid is
also produced outside of the brain in the periphery.”

What we’ve been discussing now is that amyloid in the blood is a test that has a robustness issue because the range is too low. There will always be analytical [errors] when you measure something. If you change maybe a little bit of your reagents, and you have certain variations, it’s very easy to get to 10 percent. It’s not necessarily good [or] bad. 

Being Patient: Are blood tests diagnostic tools, or are they moreso indicators that more follow-up is needed? Or can we actually diagnose Alzheimer’s disease today with a blood test?

Brum: To answer that question, we have to go to the controversial topic of defining what Alzheimer’s disease is. There’s a lot of debate. There are two main clinical guidelines. One of them points to a more biomarker-centered way of defining Alzheimer’s. 

[In] a more biomarker-defined perspective of the disease you do not take fully into account the clinical picture to define it. If you get a blood biomarker result that is indicative of an abnormality, then from that point of view, you will be considering that the person is within the Alzheimer’s continuum. 

In the other line of thinking, Alzheimer’s [is] a little bit against this biomarker-focused way of seeing things. In that case, you don’t consider having amyloid necessarily a problem. You understand that it puts the person at a higher risk, but you will only diagnose a patient with Alzheimer’s disease when they present symptoms characteristic of the disease and have biomarker evidence. 

To answer your question a bit more to the point, I think we have enough evidence as it is to say that the blood tests —of course, they are not widely available yet — but one day they will become important tools in the diagnostic workup of Alzheimer’s disease patients.

Being Patient: Will we get to a point where we can actually take a blood test that will tell you what type of dementia you have? As the technology improves, could we discover new biomarkers in the future?

Brum:  That is the dream of every neurologist. Right now, we have reliable and good blood tests for Alzheimer’s. We still don’t have a good blood test for frontotemporal dementia, Lewy body dementia, [and] vascular dementia. What we think of, with the blood biomarkers now, is that an abnormal result indicates that you have Alzheimer’s pathology in your brain. 

Based on that information, the physician evaluating the patient will be able to say with a good level of certainty that the person has Alzheimer’s pathology. It’s up to a more complex diagnostic process to determine if that is what is really contributing to that patient’s syndrome. 

Dementia is a clinical syndrome, right? Dementia is when an individual has substantial cognitive impairment to the extent that his independence in daily life activities is harmed, and the person has less independence. Alzheimer’s is one of the causes of dementia. If we take fever as an example, we know that fever can be caused by bacteria, fungi, [and] metabolic problems by viruses. In some cases, these causes of fever can show up together. With Alzheimer’s and blood tests, now, when a clinician is evaluating a patient, he needs to do a thorough examination of the patient’s clinical picture and understand the information that the blood tests provide. 

“With Alzheimer’s and blood tests, now, when a
clinician is evaluating a patient, he needs to do a
thorough examination of the patient’s clinical picture and
understand the information that the blood tests provide.”

Blood tests shouldn’t be used to say, “Oh, you have Alzheimer’s.” The blood test will give you an indicative of pathology, and then the neurologists doing the diagnostic workup of the patient need to understand, “Okay, based on the patient’s demographic profile, the history, the symptoms, do I think that this is a case where it’s a typical Alzheimer presentation, and that the pathology may be driving the symptoms? Or is it a case that maybe it’s just a patient that has a lot of cardiovascular problems?” You can see that it has a lot of [atherosclerosis], so maybe that patient’s amyloid Alzheimer’s pathology is just a coagulant. 

Being Patient: We have a question for our audience. Does evidence exist yet on how many years ahead of showing symptoms, blood tests could show a positive result? Would we be able to know ten years before any symptoms if we are, in fact, on the road to Alzheimer’s?

Brum: I think that this is a very important thing to discuss. Often, when some research study points to something along those lines, we see very strong statements in the media and in the news about it. For instance, blood tests may detect Alzheimer’s 20 years before symptoms and all that. But in real life, this is a little bit more complicated, because what do we see in the studies? 

The evidence does suggest that these biomarkers start to increase for a number of years before symptoms onset. The thing is that the studies work on a group level. We see that there is a group of patients that are increasing their biomarker levels on average. These average increases are happening many years before the symptoms, but these are on average. There will be people there who will [have] lower biomarker values; there will be people who will be in a gray zone; there will be more people getting abnormal, so [this drives] the average up. 

When the blood tests reach the clinic, I think they will definitely [be used] alongside other important risk factors, such as family history, the age of the patient, the patient’s cardiovascular health, and genetic information as well. The blood test will be another important element that will inform us of future risks. 

This is an important disclaimer: I still do not hold my medical degree. I dedicated my life to studying Alzheimer’s blood tests, but I’m still not a registered doctor. I’ll be soon. If I were in the future, evaluating a patient, and the patient would have a series of risk factors and a positive blood test, I would understand that this is an indicator of greater risk, but it’s not determinant. This is the tricky part of a disease like Alzheimer’s.

Being Patient: Prior to blood tests, the two ways to definitively diagnose Alzheimer’s were through a PET scan and a spinal tap. Are the blood tests as accurate as cerebral spinal fluid in determining biomarkers?

Brum: This is actually one of the areas in which I conducted what I consider to be, for the moment, my main contribution to the discussion with blood tests. I think that we have enough evidence that points [out] that there are better-performing and worst-performing biomarkers. If you take the best-performing ones, you will see that their performance is similar to cerebral spinal fluid and to positron emission tomography in detecting pathological changes. 

We’ve been working with spinal fluid tests since the mid-90s and since 2004 with amyloid PET. We have a long history with these biomarkers. We know how to interpret them better. We know what may be affecting results. Blood is a little bit newer. Also, the blood is a more massive fluid because you have a lot of things going on related to all the different organ systems in your blood. 

“I think that we still need to know more
about blood biomarkers there in terms of
their longer-term operating characteristics.”

For instance, the blood, you cannot see through it; it’s red and you have to do a lot of centrifuging to separate different fractions of the blood. Whereas the spinal fluid— it is as clear as water. I think that we still need to know more about blood biomarkers there in terms of their longer-term operating characteristics. It’s still something a little bit new. 

Being Patient: We need more data to understand, right?

Brum: They do seem to be as accurate or very close to accuracy. I think it’s a good approach to interpreting biomarkers to determine if a person has Alzheimer’s changes. Because with the gold standard test, the imaging tests and the spinal fluid, we usually tend to classify people into positive and negative.

We can do that with blood tests, too, but with blood tests, we do have more overlap between the groups. That’s not just for amyloid, that we mentioned before. Better biomarkers than amyloid in the blood, such as phospho-tau 217 also present a little bit higher overlap between groups in the blood than in the CSF. 

We know that the biomarker in the blood is reliable enough. We’ve seen a lot of data, but [we can] take a little bit more cautious and humble approach towards the technology that we don’t have decades of experience with. We could have patients that are clearly abnormal, patients that we know are very likely to have Alzheimer’s pathology in the brain, you can have those who are clearly negative, and then you have those in the middle where you have a little bit more overlap. In these individuals, you could forward them to do a spinal fluid test, or a PET.

Being Patient: Is plaque the only biomarker target currently being used with blood tests? I know tau is being developed, but it isn’t used in clinics, as far as I know.

Brum: As I mentioned before, when I was talking about the challenges of measuring amyloid in the blood, this is one of the reasons that the field, at some point, started to invest a little more in measuring tau in the blood. Coming back to the disease definition, we know that having amyloid plaques is fundamental for you to be considered within the Alzheimer’s continuum. But, the amount of plaques you have in your brain is not so important to determine your cognitive symptoms as the amount of tau tangles that you have in your brain.

Dementia Biomarkers 101: What is Tau?

Being Patient: Is that because tangles mean that you’re further into the disease?

Brum: Yes, exactly. Then, we have these two main targets: amyloid and tau. We saw that measuring amyloid in the fluids was a little bit difficult, even the best performing [tests] have important limitations. The field has invested a little bit more in trying to detect tau in the blood, and it was a fantastic surprise. 

Measuring tau in the blood, specifically phosphorylated tau, is the way that we see tau in Alzheimer’s. Let’s think of our neurons as [having] a command center. Then, they have a big tail in which they will carry information to otter neurons. Tau is the protein that acts as the stabilizer of these rails, let’s say. 

These rails, and in Alzheimer’s, these proteins get hyperphosphorylated a lot, and a lot of phosphates are added to this. This causes instability. You have neuronal dysfunction, but these phosphorylated tau species also leak into the fluids.

In the end, we had a lot of success in the field with measuring phosphorylated tau in the blood. Then, what this was related to, but tau in the blood tells us. Phosphorylated tau 217 is the most promising biomarker for clinical implementation. 

“Phosphorylated tau 217 is the
most promising biomarker for
clinical implementation.”

People with elevated p tau 217 are likely to have amyloid in the brain. Not only that, a person who is amyloid positive but has more tau in their brain will also have higher levels. It will increase further their phospho-tau levels. 

Being Patient: Is p tau 217 a more accurate biomarker? Is that measuring both amyloid and tau?

Brum: Yes. This is also a matter of a lot of discussion in the field. There are people who believe that tau increases in response to amyloid, and there are people who just think there are a [few] different processes, and tau is still tau. From a clinical standpoint, I guess the best way to synthesize what we have now of evidence is that both amyloid and tau in the brain contribute to higher tau levels in the blood.

Being Patient: As these tests become more readily available and people choose to use them more, we’ll have a lot more data to understand exactly what they can tell us. It’s been used in trials, and we know there are different degrees of accuracy. It’s not a diagnostic tool on its own, at least right now, but it does help us create a better picture. It’s just one more thing to check off to get a more accurate diagnosis.

Brum: Exactly. My grandma has Alzheimer’s, and as [of] right now, without more technology, the most you get is a brain scan. You think: I have a test for diabetes, I have a test for metabolic syndrome for problems with lipids, and I don’t have anything to check how the brain is working. It’s reassuring for both clinicians and families of affected patients to also have something that gives you some direct information [about] what’s going on in the brain.

Katy Koop is a writer and theater artist in Raleigh, NC.