Podcast – Blood-based Biomarkers for Dementias

Voice Over:

Brought to you by dementiaresearcher.nihr.ac.uk in association with Alzheimer’s Research UK, Alzheimer’s Society, Race Against Dementia, and the Alzheimer’s Association, bringing you research, news, career tips, and support.

Dr Nicholas Ashton:

Welcome to the Dementia podcast, bringing together early career researchers and leaders from the field to discuss their research, our hot topics, and share tips about careers and career progression. My name is Nicholas Ashton. I’m from the University of Gothenburg and it’s my pleasure to be guest hosting this exciting podcast about biomarkers and we’ll mainly be discussing blood biomarkers today and their incredible advancement in the last few years. So, let’s meet our guests. So, we have two, I would say, of the experts in the world of biomarkers. Not just fluid biomarkers, but also imaging, as well. So, we have Professor Oskar Hansson, a professor of neurology at London University and the PI of the BioFINDER study.

And we have Professor Henrik Zetterberg, who’s a professor of neurochemistry in Gothenburg and several places around the world, spanning many continents. So, it’s a pleasure to have you both here. Thanks for accepting the task of putting up with me as a host. This is the first time I’ve ever hosted anything in my life, so we’ll see how it goes. So, for many people that, or many people do know you guys, but for those that maybe don’t, Oskar, do you just want to explain in a nutshell your work and your research?

Professor Oskar Hansson:

Yes, sure, sure. So, as I said, I’m a professor of neurology, always been very interested in neurogenerative disorders. Started out with Huntington and Parkinson’s diseases, but then got more and more interested in Alzheimer’s disease. Just as I said, I’ve been working quite a lot with fluid biomarkers together with Henrik and Kai and so on, but also MRI and PET imaging, especially in the last 10 years. So yeah, I guess that, in short, is me.

Dr Nicholas Ashton:

And Henrik?

Professor Henrik Zetterberg:

Yes, I started in this research world 20 years ago now, and my clinical training is as a clinical chemist. So, I am a laboratory specialist. That’s my medical specialty. When I did my residency in clinical chemistry, I got to meet Kai Kanov, who came, he was also working as a clinical chemist here in Gothenburg and he came to our research group to learn how to clone genes. I did studies in molecular biology; more specifically how Epstein-Barr virus can transform lymphocytes into cancer cells. Then Kai came to our research group to learn how to clone genes and we started to discuss clinical chemistry. The fascinating thing for me when I spoke with Kai was that in clinical chemistry, as a clinical specialty, I mean we have liver tests and cardiac markers and kidney markers, but then we didn’t have very good markers for brain diseases in bio fluids.

Of course, imaging was big already then, but developing vastly. So, I was thinking, we started to discuss some collaborative projects and then I was thinking that perhaps this could be my future field. Since then, I’ve been working with clinical neurochemistry. So that has been my subspecialty as a clinical chemist and also done academic research here in Gothenburg and other places in the world like London and where I also run the UK Dementia Research Institute fluid biomarker lab.

Dr Nicholas Ashton:

Very nice. So, a lot of wealth of experience here. So as full disclosure, I work for Henrik, and I work a lot with Oskar as well. So, if this was a political debate, it would be banned because it’s going to be the most biased blood biomarker supporting podcast you’ll hear, I think. But I think it’s my role to play devil’s advocate today. So, I’m going to pick holes in my own work and that won’t be too hard, I’m sure. But I want to ask you a, to be honest here, this is an honesty question to start with. So let’s go back 10 to 15 years ago and you are in conferences and me and other [inaudible 00:04:22] PhD students are showing this explorative blood biomarker work using mass spectrometry and different things, and we are coming up with these models where they could say [inaudible 00:04:32] work two individuals that have mainly worked in CSF and PET imaging, let’s say 10 years ago, did you really think that there would be a blood test for Alzheimer’s disease to the level that we have currently showing in our research today?

Professor Oskar Hansson:

I can simply say, I can just say no. To be honest, we had done some work together with Kai and Henrik on plus maybe [inaudible 00:04:55], it didn’t look too exciting even though there was maybe some signal there. Of course, things could improve, but my gut feeling was very wrong. I thought that we would never get any blood tests that could work at the same accuracy as CSF biomarkers. So that was what I was thinking 10 years ago.

Dr Nicholas Ashton:

Do you agree, Henrik?

Professor Henrik Zetterberg:

I could add that I had such a big respect for the blood-brain barrier and my view on the blood-brain barrier was that it was super tight, and I was thinking that proteins will not… There are too many steps between the brain tissue and the blood for this to be successful. I must say, I thought some of… I mean, you presented some work, Nick, on complement proteins in blood and there were also genetic associations, for example, with complement factor age where there also seemed to be a protein concentration difference in blood and those things, I was thinking that perhaps this could be meaningful in some way, but I thought that relationship with brain tissue changes would be much looser than we now see with these novel tests that you, Nick and Oskar and other people, have developed. I’m positively surprised that it works so well.

I also actually think it’s quite fascinating that the small amounts of proteins released from neurons in the brain eventually can be measured in the relatively large blood volume together with a lot of other proteins that are in much, much higher abundance, thinking about the plasma proteins. There were also some reproducible blood biomarker signals early on. For example, transthyretin came up over and over as slightly altered in Alzheimer’s disease. It is an interesting protein from an amyloidogenic perspective, but the differences were always a little bit small. For me, it was easier to think that this is probably something systematic, some systematic difference between people with dementia or a cognitive disorder perhaps in a pre-dementia state and a biomarker change rather than a direct effect of the pathological processes in the brain parenchyma and the blood protein. But again, that was wrong, just like Oskar said, it turned out to be wrong.

The AB story was interesting because AB was possible to measure with several immunoassays, but when we tried to validate those assays in spike recovery experiments, [inaudible 00:07:40] dilution linearity, it was quite clear that the assays didn’t work that well. But they were the best we had. The spike recovery was lousy, below 50% for some of the assays. If you diluted the sample, the signal got higher and higher, which didn’t make sense either, up to a certain amount of dilution. So, there were many warning signs that when we tried to measure tau and AB and Neurofilament light with existing ELISAs and got some results, they were not very reliable until we got better assays that were more sensitive and could measure these proteins without a lot of matrix effects from the blood [inaudible 00:08:24].

Dr Nicholas Ashton:

Did you have maybe this realization moment, this where were you moment when you realized, “Oh, this is going to work, actually. I believe this now.” Was it with Neurofilament light or was it the [inaudible 00:08:35] data or when did you…

Professor Henrik Zetterberg:

It was the Neurofilament light data. The was such a fun project. We had gotten in contact with Quanterix, actually Quanterix, a guy called Dave Wilson. He emailed me and Kai with, it was almost like a commercial email from Quanterix, a small startup company, and he said, “We have a new technology, which allows for ultrasensitive biomarker concentration measurements, and we noticed that you work on brain diseases, you might be interested.” Then he explained the Simoa technology very briefly as a compartmentalized detection reaction based on the sandwich ELISA principle. I was reading it and I was thinking that “Ah, I see how this could work,” and then we said that “Let’s try to do some pilot experiments.” We started out with tau, total tau, and then Quanterix rapidly made a total tau assay that worked in acute brain diseases.

We started out with the most severe brain disease we could think about and that was hypoxic injury following out-of-hospital cardiac arrest where we thought that this would be a massive release and it was in poor outcome patients, whereas good outcome patients released much less. So, then it felt like this was working, but of course, it was a bit extreme. But then we took NFL antibodies together with human diagnostics and put them onto Samoa and it worked. That project worked in-house here with the home brew kit. It took four weeks until we knew that this assay was working and such a fast… I mean, those weeks were amazing in the lab when a guy called Victor Liman here did the experiments and could see that transferring the human diagnostics ELISA onto the single molecular array platform achieving ultra-sensitivity would give a blood test with correlation with CSF for Neurofilament light. But then I guess you must have had both of you, Nick, and Oskar, must have had quite cool experiences with the phospho-tau project.

Dr Nicholas Ashton:

Oskar, do you remember when you first saw the phospho-tau results in BioFINDER for the first time, for instance?

Professor Oskar Hansson:

Yeah, I actually do, because that was a very specific moment and it was actually a revelation, to be honest. So I was actually in Stockholm to give a lecture and you [inaudible 00:10:59] sitting next to me and I was just going to go up and give a talk and then [inaudible 00:11:03] here in my group had sent [inaudible 00:11:05] with phospho-tau two and seven in BioFINDER too, and I just opened the file and looked at it and then I was looked at it and then I looked at it, “This is so great.” You see it’s well above 0.9, we have a blood marker for detecting Alzheimer’s disease.” I actually remember I showed it to you, Henrik, and I said, “Look at this, look at this,” and I just almost forgot I was going to give a talk five minutes later. That was actually one of those moments in my life that I even remember everything, what was happening around me. So yeah, it was very special.

Dr Nicholas Ashton:

We had something similar here in the lab. So, Thomas Karikari developed the 181 assays here in our lab and we measured the triad cohort for our first paper. We were completely blinded. So, when we set the data away and then we had this phone call like two hours later from Pedro Rosanetto in Canada, who is the lead of the cohort, saying, “I’m out of a job,” because he’s a tau imaging guy. He said, “This looks fantastic, I’m out of a job. This looks amazing.” So that was [inaudible 00:12:10]. Oh okay. So, we didn’t actually see any [inaudible 00:12:12]. We just had this phone call from a panicked professor over in Canada saying that he’d lost his job because there was such a good blood biomarker. So, I mean, these are the early studies. So, what we’re talking about here is these 2020 papers and there were a few earlier as well which gave a hint that this was going to work. But there were papers from the BioFINDER study here in Gothenburg and also in the US showing that this phospho-tau marker was really specific if you had Alzheimer’s disease pathology.

I guess we’ve now moved on and there are many studies that are replicating these findings and now, we are really talking about the details about these assays and there’s some debate about which assay we should use, and which context should we use them. But there are different contexts. So, Oskar, as someone that sees patients on a more regular basis, how does now a blood test or a test for phospho-tau, for instance, impact a clinical setting? Like a patient that comes to see you now, how do you see this happening in two to three years’ time? Or are you doing it now, for instance?

Professor Oskar Hansson:

Yeah, so we’re not yet doing it, but actually we will start in two months doing it in a prospective manner. What we first will do is in our memory clinic, it’s a secondary memory clinic, so still quite heterogenous population. So, the idea is to start doing this in all patients that come in, new patients from a primary care. Initially, we will actually continue doing cerebrospinal fluid measurements on everyone as well. That’s just a little bit to get a feeling for how well it works. But I hope that we quite quickly, at least within one year, can stop doing CSF in the huge majority of individuals. So, what I’m hoping for, depending on how accurate the test will show out to be, this will be phospho-tau 217 is that maybe those with very low levels of a phospho-tau in blood, they will be negative in the [inaudible 00:14:13] majority on CSF. So, we don’t need to do any CSF. Then we have those with very high levels and hopefully, they will be to the large degree also positive on CSF above maybe 90%.

But then we’ll have people that are very close to the cutoff, maybe 10, 20, maybe 30% of the patients where I still think that we will probably do cerebrospinal fluid measurements. This will still be, I guess, dependent on if everybody sees modified therapies around because then, of course, it’s even more important to prove that the patient had amyloid pathology in the brain. So, then we might be even more cautious and maybe do CSF still in everyone before we start such a treatment. But I think we need to see. Some of the new [inaudible 00:14:56] that have been developed for hospital show such high accuracy that maybe in the future, we will not need to do that before starting a [inaudible 00:15:04] therapy. I think that would be the dream and also will make it much easier for clinics to do not do lumbar puncture as our clinic do. So that is my goal, my dream. But also of course, use in primary care. But that will take a little bit more time, I think. We need to evaluate a little bit more closely. I don’t know what you think, Henrik.

Professor Henrik Zetterberg:

No, I agree with all what you said here, Oskar. So, I work in a clinical lab, but in Sweden, it’s so fortunate because we have the possibility of taking research grade assays and validate them for clinical use. We can do that as clinical chemists and then we do it according to standard operating procedures. Then we are, of course, also examined by accreditation board in Sweden, one of those bodies are called, it is called Severik. Then we put our license to practice on the table. That’s how it works. But we have not yet, we are still making up our minds exactly on what phospho-tau form to go for in clinical practice. In clinical practice, it’s a little bit more complicated than in research studies because in clinical practice, you have to be able to run the assay week by week and maintain stability in your measurements in relation to the studies where you establish the reference limits and cut points and you monitor this with internal control samples and then you can see if you start to drift a little bit.

But if the drift needed for giving a systematic effect on the diagnostic accuracy, if that drift is small, if the difference between cases and controls, for example, is relatively small, then it can be very tricky to maintain the assay in a good enough shape for clinical practice. So, we have to look at the full change and we have to look at analytical variation and we also have to look at biological factors and the variability of the biomarker. If I would start to discuss these biomarkers from these aspects, biological variability is relatively low for phospho-tau, super low for the AB42/40 ratio. It’s amazingly low for that ratio, and confounding factors seem to be a little bit lower, the influence of such factors on AB42/40 than on phospho-tau. But the difference between cases and controls, [inaudible 00:17:32] positive individuals, negative individuals for AB42/40 is really small compared with phospho-tau. So, there are pros and cons with these markers and most of the pros lean towards the phospho-tau tests over AB42/40, but exactly which phospho-tau form one should peak is still a little bit open.

Although phospho-tau 217 often looks a little bit better than the other ones, but it’s a little bit harder to measure. So that is the [inaudible 00:18:02] chemist’s viewpoint on this issue. But we hope to be able to be up and running in clinical practice in clinical laboratory practice during the spring in parallel with the study Oskar management also to prepare in this case Sweden and neighboring countries. So, we can also potentially help Northern Europe if needed for the advent of disease modifying treatments. There are many other groups that also work on this. So, I think we will see a quite rapid development of these blood tests into clinical laboratory practice. They are actually already up and running in some of the US’S labs and Neurofilament light we have done in clinical chemistry within our lab and in several European labs for more than two years now.

So, it’s going fast but, of course, one needs to think about what these biomarker results mean and keep the mindset that the biochemical test in itself will never be diagnostic. It has to be interpreted with a clinical picture with other examinations and it’ll be exciting to be a memory clinic doctor in the future I’m sure because there will be even more interesting tests one can do on the patients, including cognitive tests and imaging and blood tests when needed [inaudible 00:19:18].

Dr Nicholas Ashton:

I think that the aspect about clinical routine and therapeutic trials and I think it’s very translatable, but I also think for researchers that are listing here that may be not in biomarker work, they can use these tests in their own research. This is a cheap way to say that maybe it’s an inflammation drug that you’re studying or a completely different profile that you can use this blood tested dichotomized patients in a cheap way that when you didn’t have CSF and imaging biomarkers. So, I think this is useful for not only in clinics and therapeutic trials, but us as researchers to use these tests in our day-to-day research pattern. I just want to go on onto a point here about you, Oskar, Henrik, that you said about clinical symptoms. Now, I often try to communicate these in my talks that phosphorylated tau does not mean Alzheimer’s disease dementia, it means Alzheimer’s disease pathology.

So, I don’t know, Oskar, if you want to explain what you think phosphorylated tau is reflecting on the biological level. Is it amyloid? Is it tau? I mean, I guess that you’ve done a lot of research to say that the correlation, but with symptoms, is quite weak, I would say. It’s not perfect. It reflects more of the biology that’s going in the brain rather than the symptoms that someone exhibits, for instance.

Professor Oskar Hansson:

Yeah so, to me I guess phospho-tau is reflecting amyloid-induced hyper-phosphorylation and secretion of tangle tau. So, it’s something very connected to that, the amyloid pathology in the brain. But still, it’s not just reflecting the number of plaques because we still see not a perfect correlation either between the number of plaques and the phospho-tau levels and it’s slightly different between the different phospho-tau species that we have looked into. Yeah, no. So, some are definitely more associated with tau tangle load than others are. I think we need to be a bit careful there, not using just the word phospho-tau, but actually for example 217 seems to be more related to tangle pathology maybe than [inaudible 00:21:25], and maybe the others like 205 and so on that are even more related to tangle pathology. So, I think in the future, we will know much more of that when it comes to that. But I also want really to reinforce what Henrik said because, as a clinician, we’ve been working with CSF in our clinic for maybe 20 years and we always look at the results in the context of a symptomatology of each patient.

So, it’s not unrare, no, or it’s not rare, I mean, that the patient with clear FDD, for example, also have some amyloid pathology in the brain. When we have done a tau PET scan, for example, that is usually negative and we know that this individual have probably that [inaudible 00:22:06] tau or TDP in the brain and also some amyloid, but it’s not amyloid causing the symptoms that I really think that the clinicians still need to think about this and not just look at the test, just as Henrik said, because that will be important. Because removing, I guess, amyloid in the brain of a person with actually FDP pathology driving symptomatology will not be [inaudible 00:22:27] so helpful.

So there, actually, there is maybe some role for still for tau PET imaging but is much more associated with the symptomatology of a patient, not only below, as you know, also the region where tau is, [inaudible 00:22:43] that fits very well with the symptomatology, you know it’s Alzheimer’s disease. But I still hope that in the future, we cancel that with more tau tangle-related markers so we can look at the level of these markers and say that this individual has a lot of taus in the brain. It must be related to that patient’s symptomatology.

Dr Nicholas Ashton:

I think it’s really very clear to, like we’ve not invented something new here. We should treat blood tests as we would do with CSF tests. It just means that it’s easier to get tests to more people essentially is what we’re trying to communicate here, I think.

Professor Oskar Hansson:

Yes, but also still optimistic with all these new tau isoforms that you and others are now developing assays for that might be more related to the tau tangle portfolio. But when we normally [inaudible 00:23:36] CSF, I think that could be helpful here, actually. So, we do not need to do tau PET imaging in these cases where we feel a bit more uncertain if AD pathology is actually driving the pathology or the [inaudible 00:23:49] pathology.

Professor Henrik Zetterberg:

I could also add that when we talk about the different phospho-tau forms, again thinking about the clinical laboratory practice situation, it’s really nice when you are a clinical chemist that look at data and the things do not really fit. It’s good that there are other assays around because then I can call a colleague in another lab and ask them, “Could you help us doing a measurement with your assay on this particular sample where we are uncertain about,” perhaps it’s an unusual high level that pops up in an individual where it’s unexpected somehow. So therefore, again, very much for this view that we work in this field together, developing different assays that might give slightly different associations with brain pathology in different disease stages and they could complement each other. In CSF, we have had this fortune in the Alzheimer diagnostics field to look at total tau, phospho-tau, AB42 and AB40 and particularly looking at AB42/40 ratio, and then this becomes a pattern and if one biomarker is a little bit off, we would never draw any strong conclusions on that.

If all three are positive, it really indicates that something is going on, although one can never exclude that this is preclinical Alzheimer’s disease and that there is something else that is causing the patient’s symptoms. So yeah, it reiterates what you just said here now, that the biomarkers have to be interpreted humbly and as markers of brain changes, but not as markers that explain the whole situation for a patient. I’m also thinking about the different diagnostic modalities we have at hand with CSF and advanced imaging also. Then I remind myself about, again, the hepatologist’s view on liver disease. So perhaps the primary care physician notice says liver and enzymes that are a little bit up, they discuss alcohol, they discuss some other things, deliver the enzymes are still abnormal. Then eventually, the patient is referred for evaluation at the specialty clinic and then the hepatologist has a host of different things he or she can do and that spans from ultrasonography and some different examinations of the biliary ducts and then eventually he or she might even do a liver biopsy and look in the tissue.

So, I really think we should have that mindset also for these new biomarkers and look at them as [inaudible 00:26:27] chemistry test for brain diseases with all the pros and cons or strengths and limitations that exist for these biomarkers. When one compares the diagnostic performance of these biomarkers that we are discussing now with other clinical chemistry tests, they are top tests, they are really, really good. The diagnostic performance of some of the kidney biomarkers we use a lot or the, again, liver disease tests and also some of the cardiac biomarkers, they have similar AOCs as the ones we have now, have at hand. But one needs to always keep limitations and so in mind. Thyroid hormones are the same also. We have really seen some analytical problems throughout years, and they have been solved more and more, but are also effects of general somatic disease on thyroid hormone levels and this could be regarded as confounders, although it also could have to do with how the thyroid is regulating the metabolism in the [inaudible 00:27:36] in general.

So, there are lots of things to consider and discover when we now go for more broad scale use of these biomarkers in context where we perhaps wouldn’t have used them before also because they will be more easily accessible. This is where the primary care study will be super interesting, Oskar, when perhaps some individuals with relatively mild symptoms and a lot of other problems will be tested then as will, of course, challenge the biomarkers the most.

Dr Nicholas Ashton:

So let me ask this question. So, this fits quite nicely into this. We had a lot of questions, but this one stood out. This is from Eddie Rukati, and he is doing research in Tasmania by looks like it. He wanted to ask a question about clinical cutoffs for blood biomarkers. So, this is his question, will there be a one-size-fits-all cutoff for these tests? There’ll be some maybe things about round robins and certified reference materials Henrik wants to maybe comment on about that. Will age, sex, and genetic APOE tests have an influence on these results? Will they factor in at all? That was the second part. The third part, would it be more useful to look at someone’s intra-individual change as a marker rather than their cutoff for the whole population? So, if someone increases by 20, 30% over a year, would that be more useful or more interpretable than, so what is someone’s baseline level phospho-tau is different to someone else’s, would the increase be more [inaudible 00:29:05]? So, it’s three, I thought, really nice questions. I don’t know who wants to start with those.

Professor Oskar Hansson:

I can maybe just start quite briefly and, Henrik, you can fill in. As a physician, now so I’m not a chemist, so as a physician I don’t have a problem that there are different cutoffs for different tests, to be honest, because there’s so many different tests and as a primary care physician, you will probably order a phospho-tau test maybe 10 times a year or something and you will anyway not remember the cutoff. So as long as the laboratory provides the cutoff to you, that would be fine. Henrik can explain more if it’s important from a chemist’s perspective to have the same catalyst. But from a physician, I don’t really see that that is needed. When it comes to adjustment for sex and age and APOE, when it comes to phospho-tau and AB42/40, I don’t see that either because they’re so tightly associated with the AD pathology in the brain and if you have a high performing test, I don’t see why the cutoff should vary with these demographic variables. So, I think it would be the same cutoff independent of these.

Of course, it is more common with AD pathology the older you are, and more common if you have an APOE4, if you’re an APOE4 carrier. But the cutoff should be the same to me, so it shouldn’t make a difference. The last we’ve changed over time, I guess it’s only interesting in those that have values very close to the cutoff or I guess just below the cutoff. Still, we know that these tests AB42/40, phospho-tau, at least with phospho-tau we’ve been talking about today change maybe 10 years before clinical onset. So, if a patient comes in and have symptoms, it should already be abnormal.

But still, if you have a test result close to the cutoff, I, as a physician, would probably ask a patient to come in three months later and I would take a new test and I would see, “Okay, is it now clearly elevated or not?” But that may be more test, retest variability more than change over time, I guess. It might be different I guess if we have a marker that is more tangled related that we assume is changing more during the symptomatic phase of the disease. It might also be different if we are thinking about pre-symptomatic Alzheimer’s disease, which is maybe not what we are thinking about use of now in clinical practice.

Professor Henrik Zetterberg:

No, I agree completely with everything you said here, Oskar. There is actually one example of a test where it really helped to normalize for inter-individual differences in biomarker release, and that was [inaudible 00:31:39]. So once this was something that [inaudible 00:31:43] talked about a lot already in 2006, I think. We were so tired of him talking about this ratio because he really advocated for the use of AB42 over 40 compared with AB42 alone. But then I felt it was a bit confusing because AB40 was not systematically changed in Alzheimer’s disease, but everything was explained when you put AB42 concentrations on the Y axis and AB40 on the X axis because then you see that people with amyloid pathology, they have a different correlation between AB42 and 40 compared with people without. That really made the separation between the groups much better. So, when we started to use AB42 over 40, we couldn’t normalize for this inter-individual difference.

We still do not understand that inter-individual difference why some are low producers, and some are high producers and those who are low producers, but amyloid negative, they are closer to the cut point for amyloid positivity. So, they will more rapidly go below that cut point and become biomarker positive, CSF biomarker positive for amyloid. But then again, the ratio corrects for this. For phospho-tau, I also think that the disease association, at least in symptomatic disease, is so high that we won’t need to do these things. There are some groups that speak for doing a type of ratio which, for example, Randa Bateman’s team in St. Louis with the occupancy ratio, when his team compares phospho-related and non-phospho-related fragments of tau. It’s really interesting and it might be good, but in blood, there is always a risk that one starts to measure peripheral big tau also with that approach. But I think we should keep an open mind and see what eventually turns out to be the best.

Change over time, I like it a lot from this perspective if I know that I have Alzheimer’s disease in my family and would go and check my biomarker levels every second year at my GP’s office, of course it would be quite cool if we could define clinical relevant delta change where one could consider start of treatment and that delta change in itself might not have to be related to a cutoff for diagnostic positivity. But I think the studies that are planned now, both in your place, Oskar, and in some US, centers will reveal if this will be possible or not or viable, but a clinically meaningful way forward also. It would put less pressure on us establishing good reference limits and cut points if this delta change could turn out to be of value. Then if I should now, I move towards back to the first question if I should just address this. I think with universal cutoffs, they are impossible without a really good reference standard. This is really, again, core standardization in clinical chemistry.

So then one would the meter in Paris, one would have a defined set of samples stored at very nice conditions in multiple aliquots that would last over 10 years for kit vendors to be able to use them to calibrate their calibrators. For example, a phospho-tau 181 reference material. It could be plasma with spiked in CSF for plasma with spiked in peptides that cover the 181 phospho-related site and some other parts of tau so that the assay will work. They can be created in different ways and it’s quite a lot of job or work behind these reference materials. You have to have a nice, a really good reference method to value assigns the reference materials. Most likely the different phosphoforms will have to have phosposite-specific reference materials. So, one could ask, “Is it worth the effort?” Perhaps at least for now, it would be fine with laboratory-specific cut points. But then the laboratories, they have to establish their cut points in a good clinical material and then maintain their measurement stability over time in relation to those experiments. That’s really important.

If the laboratory as such can’t do that, then the laboratory has to approach a companion laboratory that has such data and/or collaborate, of course, and get access to samples to which they can validate the cut points. But somehow each laboratory then has to validate their level of measurement to be able to import reference limits and standards. That can be hard work instead of having the possibility of relying on the clinical chemistry companies and reference materials and [inaudible 00:37:02] standardization. The advantage with [inaudible 00:37:04] standardization, of course, is that one could compare results on a global scale, and one could even do epidemiological studies around the globe and look at prevalence of phospho-tau positivity [inaudible 00:37:19], for example, relate to genotype information and exposure data and a lot of interesting things.

It’s also easier to interpret clinical trials if there had been fully standardized tests, then one could directly compare the percent reduction of a certain anti-amyloid antibody with another anti-amyloid antibody. So, there are advantages with full standardization, that’s for sure. But I still wonder if it would be worth the effort. Now, what we need to do is to firstly finalize the standardization projects for total tau and phospho-tau 181 for the CSF tests, the most available tests. But then I think the International Federation of Clinical Chemistry and Laboratory Medicine might take on one or two phospho-tau markers, especially if the drugs become available [inaudible 00:38:14].

Dr Nicholas Ashton:

I think that we have very good assays in CSF, and we always use different cutoffs, we always use different assays, and I think the advantage, of course, of CSF is they all work great. One of the maybe disadvantages of plasma is that we have quite a range of performances of phospho-tau assays in blood. We’ve done some studies in Gothenburg, [inaudible 00:38:32]. You’ve also done similar things showing that the choice of assay that you use or the choice of biomarker you choose is very, very important here. Our first-generation assays were excellent in showing that this was proof of principle, this is going to work, but our newer assays are the ones that are more likely to be in clinic, I think.

This is where I think that maybe not standardization, but at least the consensus in the biomarker field would be more useful to say, “Okay, well these are the top assays that we recommend you use in the lab because they give us the best diagnostic performance, I think would be most useful.” We are running out of time, but I want to ask you these quick-fire questions because I think some are interesting and I want to know some of the answers to these as well. So, I’ve scanned them. So, some of them are about science and some of them are not. So just short answers for you both. Keep an open mind. But the first one is an interesting one. So, what was your first paper that you published as a first author, and what was it about?

Professor Henrik Zetterberg:

Okay, I can start. 1998, it was about [inaudible 00:39:39] regulation of [inaudible 00:39:41] virus, nuclear antigen1 gene in lymphoma.

Dr Nicholas Ashton:

And which journal was it in, Henrik?

Professor Henrik Zetterberg:

Journal of General Virology.

Dr Nicholas Ashton:

Oskar?

Professor Oskar Hansson:

Yes. I think it was about exitotoxicity in a mouse model of Huntington’s disease. I think it was published in ’99 in PNAS, if I’m not mistaken.

Dr Nicholas Ashton:

This is also a fun one, I think, well you’re both music fans so I think you’ll both enjoy this. If you could join and play with a band in a song, live on stage, just once, which band would you be, and which song would you pick? So, you’re allowed to join any band in history for one song.

Professor Henrik Zetterberg:

I know. Should I go?

Professor Oskar Hansson:

Yes, yes.

Professor Henrik Zetterberg:

Motorhead, Ace of Spades.

Dr Nicholas Ashton:

Motorhead, Ace of Spades.

Professor Henrik Zetterberg:

Motorhead, Ace of Spades.

Professor Oskar Hansson:

Okay. For me, it’ll probably be Primal Scream maybe Higher than the Sun.

Dr Nicholas Ashton:

Very nice, very good. Okay, and then the next one, so if you weren’t in science or medicine, where do you think you would’ve ended up? Not where you wanted to be, where do you think you would’ve ended up?

Professor Oskar Hansson:

Aye, aye, aye. I know I applied to physics at the same time as med school, and I actually wanted to work with theoretical physics more than med school. But since I was allowed into med school, I didn’t want to lose opportunity. So, I guess physics somehow.

Dr Nicholas Ashton:

Physics.

Professor Henrik Zetterberg:

When I was a child, I wanted to be a garbage truck driver.

Dr Nicholas Ashton:

There’s still hope, Henrik. You can still do it.

Professor Henrik Zetterberg:

But I’m not sure, actually. I’m red and green color-blind. I don’t think I would qualify.

Dr Nicholas Ashton:

This is a nice one. So, who is your scientific or academic idol, I guess? Do you have one?

Professor Henrik Zetterberg:

Yeah, I have one, E.R. Klein, George Klein, E.R. Klein was called in Swedish, professor of tumor biologist at Karolinska who always wrote Popular Science essays in parallel with his work as a tumor biologist and immunologist and having a broad… He had such a broad and interesting view on science and society and his hardcore science, the molecular biology was doing. Actually, it plays a role in my choice of career because when I was 17, I got one of his books from my father. He had heard him on the radio, and it was a book gift from my father to son, I didn’t read it at first, but eventually, I read it and it’s called the Atheist and the Holy City. It’s a wonderful, wonderful book and it is translated to English also, so it’s possible to find it if anyone is interested.

Professor Oskar Hansson:

I’m not sure if I have one research idol like that. But yeah, I’m very impressed by Colleen Masters, for example, that have done so beautiful work and still is fantastically sharp and gives very good comments. So yeah.

Dr Nicholas Ashton:

Very good. So, this is a very simple one. Do you have a favorite movie?

Professor Henrik Zetterberg:

Let’s see here. What the English name could be? I really like a Finnish director. His name is, they are actually brothers, [inaudible 00:43:00], the life of the Bohemians. The real name of the movie is in French, but [inaudible 00:43:13] Bohemian or something like that. The movie is taking place to a large extent in Paris, but they speak a lot of Finnish in it and broken French with Finnish. I think one can find it. It’s one of the most beautiful movies I’ve seen in the whole of my life.

Professor Oskar Hansson:

I’m not sure. To be honest, I have a very short memory, so I like a lot of movies, but I don’t remember the titles. So yeah.

Dr Nicholas Ashton:

Here’s a very strategic question. Someone was very smart when they wrote this one. So, they said, when you’re looking to get a new PhD student or a postdoc, what’s the three characteristics which impress you the most? Someone’s looking for a job here. So, any characteristics of a perspective employee that you look for in their CV or something like that?

Professor Oskar Hansson:

I can say one thing and then you can say, Henrik. So, to me, the absolute most important is intrinsic motivation so that I get a feeling that a person really loves this and really want to do it, not because of a career but because of curiosity and so on. But that, to me, is the number one most important thing.

Professor Henrik Zetterberg:

I couldn’t agree more. I was just about to say engagement, but that’s the same thing you are thinking of, Oskar, I think. So formal education, it doesn’t really matter. I have to say that, of course, it’s good if one knows something and comes with some knowledge to the team that we might need. But if somebody’s engaged and interest in gaining knowledge, then that is the most, most important thing one could think of and has this type of driving force towards that. It really can be… I think that’s the only real trait that I feel I’m looking for all the time.

But then of course, it’s important that also the person who comes likes the place they’re coming to, so that one gets this type of synergies, and the responsibility is on both sides somehow. So, I think if people are looking for job, it’s also a good idea to come and test to work in a place before making long-term commitments and so on. There are plenty of opportunities to do that also by smaller grounds. So, I think it’s a relatively good time to look for jobs now in the Alzheimer research field, actually.

Dr Nicholas Ashton:

Just a few more. So, someone’s really interested to know how many emails you both get a day. Are we in the hundreds or are we in the thousands? Just give us a ballpark.

Professor Henrik Zetterberg:

We are approaching the thousands now. It’s really… I am about 500 emails per day and the spam filters work nowadays. I mean, it’s emails that are, of course, they’re not always something I need to act on, but there are many emails, but it has actually grown a lot the last two years for me. I don’t know what it’s like for you, Oskar. So, if I don’t reply, send an email again?

Professor Oskar Hansson:

Yes.

Professor Henrik Zetterberg:

Or find my phone number.

Professor Oskar Hansson:

Exactly. Yeah, yeah, yeah.

Dr Nicholas Ashton:

Then the final one, I think this is… So, this person says that as an inspiring principal investigator, what’s the biggest challenge of being the senior scientist in your lab? What’s the biggest challenge that you face as being a principal investigator?

Professor Oskar Hansson:

I guess to me it’s because I really want to work with science. I love supervision, discussing results with a PhD student, postdocs, and others in the lab. So that’s what I really want to spend my time on. But so, my biggest challenge is to find enough time for that and keep that and not get too much dragged into administration or a lot of other things that is not directly related to research. Because that is what I really love, and I want to continue doing.

Professor Henrik Zetterberg:

My biggest struggle this autumn has been that I’ve been absent too much and traveling too much. I have to cut down on that, I think. But this has been really fun also. But of course, this is not fair to the people you work with closely. So, I feel a bit bad about that this autumn, actually. But it’ll get better, I’m sure.

Dr Nicholas Ashton:

Is this a public apology to me, Henrik?

Professor Henrik Zetterberg:

It’s a big public apology to anyone I’ve been working with.

Dr Nicholas Ashton:

So, I think we are running out of time, but there’s two questions here that we’re going to ask, and I want to ask them to you. So, let’s try and keep them brief. But the first one is about other neurogenerative disorders. So, we haven’t discussed it, but [inaudible 00:48:02] blood biomarkers that don’t really work for other tauopathies or other neurogenerative disorders. They are specific for Alzheimer’s pathologies. But have you seen any data or promising data published of any biomarker, whether it’s imaging or CSF or skin, saliva that may work for other tauopathies or other types of neurogenerative disorders? Is there anything promising out there that you’ve seen in conferences or data that you’ve seen?

Professor Oskar Hansson:

To me, I guess [inaudible 00:48:30], maybe not yet blood, I want to see that replicated a little bit more, but skin has been shown several times now, which I think is very encouraging. But also, the blood if that can be replicated, would be fantastic. I don’t know, there have been some preliminary also with skin and immunochemistry and so on I think for other diseases. So that could be maybe something, yeah, I don’t know what you think, Henrik?

Professor Henrik Zetterberg:

[inaudible 00:48:57] looks really promising. It’s a difficult test, of course, because when you try to do it, you spike in recombinant protein that is prone to aggregate in itself. So, one could get false positives, but it’s starting to look promising. TDP43 has one paper where this basic technology also has been applied [inaudible 00:49:17] 2020 in brain communications. I think there is hope for that, and some other research groups are trying to do this for different tauopathies, and TM106B, this novel protein of [inaudible 00:49:35] that we know very little about but seems very interesting. There might also be potential for [inaudible 00:49:41] amplification of aggregates. Then we perhaps could figure out what is the clinical meaningfulness of those pathologies.

Then I’ve been really impressed by [inaudible 00:49:54] in blood, which [inaudible 00:49:55] and his team have developed. I don’t know if that will be in, by any means, disease-specific or so it’s way too early to say that, but I really think one important topic for research for the future is to try to address the non-Alzheimer’s neurogenerative dementias more from fluid and imaging biomarker standpoint. Because now as you mentioned, Oskar, if one has a very low phospho-tau concentration, but still have cognitive symptoms, then one would need to think about these other neurodegenerative diseases and then, of course, it would be helpful if there were some positive tests for them as well. Of course, Neurofilament light could pinpoint a neurodegenerative aspect, but not in a disease-specific [inaudible 00:50:41].

Dr Nicholas Ashton:

So, I think that people should not be discouraged if they’re young and getting into the biomarker if they’re saying, “Oh, have I missed the biomarker boat?” I mean, this is such a small aspect that we’ve been working on. Alzheimer’s disease can encompass so many different types of, I mean we’re just tackling amyloid and tau at the moment. All the other types of proteinopathies and vascular dementia and all this, there is a lot of space to work on in terms of fluid and imaging and biomarkers.

Professor Henrik Zetterberg:

Imagine all the non-neurodegenerative [inaudible 00:51:15] diseases, the psychiatric diseases. Of course, they will be very hard from a fluid biomarker standpoint since there is no, the pathologists are not overt. But there must be a lot of interesting things in neuropsychology and the imaging and perhaps also including biomarkers for those. So yeah, I would definitely be super keen to get into this field also as of now.

Dr Nicholas Ashton:

Okay. My final question, and it’s a long question. It definitely comes from a blood biomarker skeptic. I won’t list… Basically, the question is, what is the biggest problem that we face getting a blood biomarker from where it is now to a routine test in the clinic? He or she makes a lot of suggestions, but I won’t mention them, but how do you see it as people that work with it every day? What is the number one challenge outside of it proving itself as a diagnostic? Let’s say it works well. What’s the biggest thing? Is it getting physicians to trust it and what it understands? Is it actually the logistics around getting the blood test into the clinic or something else?

Professor Oskar Hansson:

Yeah, so I guess briefly, I think in memory clinics that are used to using [inaudible 00:52:24] or CSF, they already know how to interpret these types of results. But then I don’t think it’s education, I think it’s more like Henrik said, standardization of the tests that actually work on a daily basis over time and that the cutoffs are stable. But we have seen that some assays actually fulfill that already now. So, I’m, to be honest, very optimistic that phospho-taus, certain phospho-taus can be used in memory clinics very soon in a reliable way. If you move out into the primary care, I think besides how samples are shipped and so on, that can definitely be sold. It’s probably education.

So, this what we have, all of us know now that phospho-taus are changing five, 10 years before symptom onset. It’s very important for them to know so they don’t diagnose people just based on the phospho-tau result, even though maybe the patient actually has depression now and might have symptomatic [inaudible 00:53:15] 10 years from now. So, I need to know about those things and how to use a test maybe mainly to refer people to another instance like a memory clinic for further evaluation. So, it’s different challenges, yes, depending on where it will be used in the healthcare system. It might also be very different to different countries, of course.

Professor Henrik Zetterberg:

I agree with all you said, Oskar, and I was actually thinking about the last point you mentioned now, because for biotech companies and clinical chemistry diagnostic companies, they need, of course, to get some revenue and then to be able to do that in the US, for example, they have a lot of regulatory hurdles to cross. In Sweden, again, it’s easier to set up also research use only test in clinical practice. But then we do it not as, the companies can’t do it, but the clinical chemists can. In some other parts of Europe, it is like this as well. But some parts of Europe are much more US-like. Then we have, of course, all the other, if you look into Asia and South America, there are also region-specific regulatory hurdles to consider.

I believe that could be a little bit problematic. But with the advent of treatments, I think there will be forces that are strong enough to push all these forward so that… Because I’m convinced the tests and the biomarkers work, and that’s such a good starting point. So, I am actually also very positive. If I was very skeptical 10, 15 years ago to blood tests, I’m now much more positive. I think that these developments will go rapidly, especially if we see additional approvals of disease-modifying drugs.

Dr Nicholas Ashton:

Great. So, I’m afraid that’s all we have time for today. If you want to know more about this topic, you can visit the Dementia Researcher website. You’ll find a full transcript of the blog today and the biographies on Henrik and Oskar, and much more on the topic generally from Dementia Researcher. So, I hope everyone has found this informative. I have, of course. I really want to thank Henrik and Oskar for giving their time for us and explaining some complicated details in simpler forms. Also, we now know a little bit more that Henrik is wanting to be a garbage truck driver. Oskar wants to star as a part of Primal Scream, so we can mention that next time in our conferences when we meet together. So, thanks for your time, Henrik and Oskar, and everyone for listening. Thank you.

Professor Henrik Zetterberg:

Thank you, Nick.

Professor Oskar Hansson:

Thank you. Thank you.

Voice Over:

Brought to you by dementiaresearcher.nihr.ac.uk in association with Alzheimer’s Research UK, Alzheimer’s Society, Race Against Dementia, and the Alzheimer’s Association, bringing you research, news, career tips, and support.

END