How blood-based and plasma biomarkers are transforming clinical neurology

My name is Henrik Zetterberg. I'm a professor of neurochemistry, and I work at the University of Gothenburg and the University of Wisconsin–Madison. My background is actually molecular biology. I was very interested in that as a teenager, but then I started to study medicine. Medicine, clinical medicine, and laboratory medicine became, to me, the perfect marriage, actually. So that's when I started in laboratory medicine. At the time, there was very little for brain diseases, so I entered the field in the early 2000s and worked with cerebrospinal fluid biomarkers, expanded the breadth of those markers, and then transferred many of them into blood tests.

In regards to blood-based biomarkers, for a while I actually thought it wouldn't work because our assays were not sensitive enough. We could measure key molecules for neurodegenerative brain diseases in cerebrospinal fluid, which is much closer to the brain tissue. But thanks to methodological advances and improved antibodies, we could start to build research-grade assays that eventually turned out to be sensitive enough for blood analysis. And that was the breakthrough.

The development of clinically useful biomarkers is a relatively long journey. It starts with a very exciting discovery phase, where you develop new assays that might work. Occasionally, you see something, and then you try to figure out what it means. Then you have to make the assays that you use in your research laboratory reproducible, and you have to replicate your findings in collaboration with clinical colleagues who have samples.

Once you have a well-validated and replicable assay that works in a research setting, interactions of course start at conferences like this, with industry. Moving a research-grade assay into a clinical-grade assay requires a lot of work, especially longitudinal stability and high precision of the assay. Those two parameters are very important, as is the important work of trying to find out how robustly the biomarker relates to the pathophysiological process you're interested in, in real-world settings with patients who come to the clinics.

The complementary role of these biomarkers, thinking about the more easily accessible lab biomarkers and more advanced biomarkers through CSF analysis and PET measurements of brain pathology, is that they are complementary in nature. We could think about it very much like we think about other clinical chemistry tests. A hepatologist who is interested in liver diseases might start by looking at the patient clinically, doing some simple blood tests, and figuring out if there seems to be something going on with the liver.

Then there is a stepwise diagnostic procedure where a diagnosis could be made with relatively simple tools in clear-cut cases. But eventually, in difficult cases, the hepatologist might need to do a liver biopsy. So I look at it like this: you have the blood biomarkers that can be used potentially even in primary care, when a patient complains about memory. If there is a positive signal, the primary care physician will know that there is something going on here: “Let's talk to my specialist colleague.” Then the specialist colleague can evaluate the patient further.

If there remains uncertainty, the specialist can decide to do a CSF analysis, confirm the analysis, or even do, for example in Alzheimer's, an amyloid PET scan. Eventually, we will reach a diagnostic accuracy that might be—or actually seems to be—very similar to what you get from a neuropathological examination, which was what we had to rely on 20–30 years ago.

Neurodegenerative diseases are actually super complex. We know a lot about Alzheimer's disease and the core pathophysiological features of the disease, and we can measure those with blood biomarkers, including tau and neurodegeneration. But there are often, especially in a slightly older person, additional processes that are ongoing at the same time, and for those we still do not have validated biomarkers.

So I think this will be a very exciting field. There will be continuous development of additional biomarkers, and my vision with all this is that we will have a panel of, say, five, 10, and perhaps 15 biomarkers, hopefully accessible in blood, that can help us diagnose the underlying molecular pathophysiological processes that contribute to the patient's symptoms. It will be an exciting field to follow.

Moving research-grade assays for these blood biomarkers that we now use in clinical practice into clinical tests requires a lot of standardization work. There are also currently several assays to measure the same analyte, and they are not calibrated to each other. They sort of measure the analyte of interest in a correlative manner, but at different scales. Bringing these measurements together through external calibration from certified reference materials that have been value-assigned by certified reference methods is a big thing, and a core element of clinical chemistry practice.

That work is ongoing. It's not completed, and it doesn't prohibit the use of the biomarkers now, but it will be easier in the future when those projects have been completed, because then patients can go to different clinics and get the same result back. If you are a lab leader running a clinical lab, then from my point of view it is very important to jump on this and implement the new biomarkers that have been quite well validated, the biomarkers for Alzheimer's disease pathophysiological processes. Then I also think one should prepare for the advent of additional biomarkers, because a lot is happening in this field right now.