Biomarkers for neurological diseases used to be few and far between, but better technology is making it easier for researchers to track brain health by measuring molecules. This means less invasive testing, earlier diagnosis, faster drug development, and—hopefully—more effective treatments.
Biomarkers—molecules that indicate the presence of a disease or dysfunction—are becoming increasingly instrumental for confirming diagnoses, choosing the best treatments, and monitoring disease progression. One exception is biomarkers for neurological conditions. Neurological biomarkers are present in cerebral spinal fluid (CSF), but rarely—or at undetectable levels—in blood. The brain is closely protected by its own private “security guard,” the blood–brain barrier, which shields it from harmful substances circulating in the bloodstream. Unfortunately for diagnostic purposes, this barrier has also made the brain’s chemistry inaccessible to a convenient blood test. Neurological biomarkers can be studied using CSF, but this requires an invasive and painful lumbar puncture procedure.
However, these elusive neurological biomarkers are now coming into view. Recent advances in detection, as well as the comprehensive power of clusters of biomarkers, or biomarker signatures, are making the brain more accessible and neurological diseases more treatable. Diagnosing and treating neurological disorders—such as chronic traumatic encephalopathy (CTE), Alzheimer’s disease, Parkinson’s disease, autism, and major depressive disorder—is likely to become easier with the recent advent of neurological biomarkers detected in blood.
Redefining baseline with greater sensitivity
The recent ability to detect neurological biomarkers in the blood is due in part to technological advances in detection. For example, Quanterix’s Simoa (single-molecule array) technology bumps up sensitivity by digitizing an ELISA (enzyme-linked immunosorbent assay), a highly effective method of determining binding between two molecules. Quanterix screens patient samples with 720,000 microscopic beads coated with capture antibody (for the biomarker of interest), incubates the beads with a capture antibody and mixes them with a fluorescent marker, then spreads the beads into 216,000 isolated microchambers. The resulting high signal-to-noise ratio makes for very responsive detection.
About half of Quanterix’s current applications are linked to neurology, and many researchers using their platform measure biomarker concentrations in both CSF and blood. “Typically, we see a concentration difference of about 1:100 or 1:500, so we think it’s a breakthrough to now be able to look at brain health noninvasively,” says Kevin Hrusovsky, CEO of Quanterix in Lexington, Massachusetts. “We can see a single femtogram per mL of a biomarker, which is sensitive enough to detect biomarkers in saliva and breath condensate as well.”
A benefit of greater sensitivity means earlier detection of biomarkers for disease, but it also brings up the issue of how to define baseline biomarker presence. If a biomarker is detected at all, does that mean disease is present or will develop—or is biomarker presence below a certain level considered “normal” for baseline? Hrusovsky believes that biomarkers for disease might be present at low levels earlier than we currently understand. “We hope this will enable scientists to detect disease long before symptoms appear, early enough that it’s still very treatable,” he says.
One example is their work on biomarkers for concussion and CTE. People with mild concussions can display no symptoms despite damage to the brain, and development of CTE is the cumulation of many subconcussive impacts to the brain. Quanterix can measure biomarkers in blood, such as neurofilament light (NFL), a biomarker for neuronal damage that indicates head trauma. Using Quanterix’s platform, an international team headed by Kaj Blennow and Henrik Zetterberg at the University of Gothenburg in Mölndal, Sweden, recently found that the biomarker NFL increased in college football players—but not contact-free control athletes—during a football season (1). One day, Hrusovsky hopes that parents and coaches can use handheld testing devices on the sidelines at soccer or football games to assess the impact of players’ head injuries noninvasively. “Hopefully such early detection and monitoring can prevent progression to more complex conditions like CTE,” he says.
Biomarker signatures for understanding nervous system dysfunctions
Biomarkers signatures, detected with panels of high-quality antibodies, are another powerful tool for studying neurological diseases and conditions. Kristian Doyle, assistant professor of immunobiology and neurology at the University of Arizona College of Medicine in Tucson, uses biomarkers to study how the immune system deals with dead brain tissue after a stroke. The immune system removes dead brain tissue after a stroke by a process known as liquefactive necrosis, but the pathophysiology of this process is largely unknown.
Yet this information is important because liquefactive necrosis may be neurotoxic. “We use multiplex immunoassays to characterize the inflammatory milieu within chronic stroke infarcts at the stage of liquefactive necrosis, and also to characterize how this milieu is altered by common stroke comorbidities,” says Doyle. Because over 10 million people survive a stroke each year, Doyle hopes that biomarkers will help them track the progression of liquefactive necrosis, “and begin to tailor treatments that mitigate the secondary damage caused by this process,” he says.
Another link between inflammation and neurotoxicity is studied by Alysson Muotri, professor of cellular and molecular medicine and director of the Stem Cell Program at the University of California, San Diego School of Medicine. The Muotri lab uses induced pluripotent stem cells (iPSCs) from patients with autism and schizophrenia to look for biomarkers of these conditions.
The Muotri lab uses induced pluripotent stem cells from patients with autism and schizophrenia to look for biomarkers of these conditions.
Recently Muotri’s lab began studying the cytokine interleukin-6 (IL-6) as a biomarker, because evidence suggests that chronic exposure to elevated cytokines may be neurotoxic, with high levels linked to depression, autism, and schizophrenia. “The difference [among brain disorders] might be that cytokines act on specific types or subtypes of neurons, or in a specific brain region,” says Muotri. His lab differentiates iPSCs from autistic patients into neurons and glial cells, which they suspect may be releasing cytokines at levels higher than normal in autistic patients.
As IL-6 is also involved in immune inflammatory pathways, Muotri suspects a link between autism and in utero exposure to—perhaps not even infection with—Zika virus. “Our prediction is that the inflammation resulting from Zika exposure is enough to create a neurotoxic environment that could rewire the way the brain is formed,” he says. “We see that in mice, so we think that some Zika-exposed kids will develop autism or have intellectual disabilities.”
Larger biomarker signatures can be detected with technology from CDI Laboratories, which offers microarrays of functional human proteins (over 20,000 on a single array) to test the antibodies present in human liquid biopsy samples, such as blood, serum, plasma, CSF, or tissue lysates. The resulting “autoantibody profile” is a useful tool both for research and for informing diagnoses or prognoses of patients. “We’ve worked in the field of biomarker discovery for various neurodegenerative diseases such as multiple sclerosis, neuropsychiatric lupus, Alzheimer’s, and Parkinson’s,” says George Dorfman, director of business development at CDI Laboratories, a spin-off company located in Baltimore, Maryland, and Mayaguez, Puerto Rico that was developed from research at the High Throughput Biology Center at Johns Hopkins University, also in Baltimore.
CDI’s platform is especially helpful in constructing panels for biomarker discovery, because researchers can begin by using patient samples or banked samples to compare immune profiles of cohorts that show particular symptoms (or no symptoms in the case of control samples). “This gives us an underlying candidate biomarker panel that provides information on the subsequent clinical outcome or therapeutic efficacy, which can be validated to yield a final panel, then translated into say, an ELISA-based kit or some other immunodiagnostic format in the clinical setting,” says Dorfman. “In the case of multiple sclerosis, as a patient progresses through steps of the disease, their body generates novel antibodies or higher existing antibody titers against certain proteins, such as myelination proteins. Our panels can detect these, to give an idea of what patients’ disease progression might look like, and provide a signature that can be translated into a subsequent test or even an FDA-approved diagnostic.”
In a recent publication, CDI’s technology was also used to develop an autoantibody profile for neuropsychiatric lupus (2), a valuable diagnostic tool for a disease that typically lacks clear clinical signs.
Nucleic acid tools in detection
Although multiplexed immunoassays are frequently used to detect biomarkers, they can be complicated by the possibility of antibody cross-reactivity, which happens when an antibody binds to similar or multiple antigens. Using nucleic acids can mitigate or eliminate this problem, especially with the high degrees of multiplexing seen in the platform of Olink Proteomics, headquartered in Uppsala, Sweden.
Olink’s Proximity Extension Assay (PEA) detection technology is based on antibody probes linked to DNA tags. Upon binding to target proteins, the probes are close enough that their DNA tags form a new reporter sequence for the target. The sequence is read by quantitative PCR, so that the signal is proportional to the protein concentration in the sample. Each DNA reporter sequence acts as a unique barcode for a biomarker. Their technology can multiplex up to 1,000 biomarkers in less than one drop of blood, according to president and CEO Jon Heimer. “The minimum sample volume required is only 1 µL for analysis of 92 proteins, which has been proven to be a great advantage for applications with limited sample volumes, such as pediatric applications, fine-needle biopsies, and tumor microbiopsies,” he says.
The large microarrays are also well suited for finding new biomarkers. Olink recently collaborated with Douglas Galasko, professor of neuroscience at the University of California, San Diego, where they assayed biomarkers from patients with Alzheimer’s disease and healthy controls. “We identified over 100 biomarkers that differed significantly between the healthy people and the Alzheimer’s patients,” Heimer says. Such clear differences in biomarker signatures are an important step in identifying new disease markers, and perhaps therapeutic targets.
Nucleic acids are also helping scientists to detect biomarkers, such as Abcam’s Fireplex (formerly Firefly) particle technology, which profiles biomarker microRNAs (miRNAs) directly from biofluid samples, without RNA purification. Hydrogel particles barcoded with unique fluorescent signatures enable detection of multiple biomarkers simultaneously. A research group headed by Gustavo Turecki, professor of psychiatry at McGill University in Montreal, Canada, recently used the platform to show that miRNAs can be used as prognostic markers for therapeutic responses to an antidepressant medication in patients with major depressive disorder (3).
Many neurological researchers still want to use CSF samples, says Dan Pregibon, general manager of platform innovation at Abcam, which is headquartered in Cambridge, United Kingdom. Abcam is working with medical experts on developing specific guidelines to evaluate methods for CSF sample collection and processing prior to miRNA analysis, with the aim of maximizing detection. “The method supports detecting microRNAs directly from very small volumes of CSF to potentially get an indication of different types of diseases, including amyotrophic lateral sclerosis (ALS), Alzheimers, Huntingdon’s, and traumatic brain injuries,” says Pregibon.
Putting it all together
As the number of biomarkers increases and the types of biomarkers expand, the amount of information that researchers must organize becomes overwhelming. “It should be no surprise that data scientists spend up to 80% of their time managing—not evaluating—data,” says Scott Marshall, managing director of translational informatics and diagnostic sciences at Precision for Medicine in Frederick, Maryland. The company’s biomarker data management platform, PATH, is designed to integrate any type of biomarker data with clinical annotations, which is key for making associations across assays. “The real power of biomarkers comes when you link this information to clinical data,” says Marshall.
Their biomarker data management platform supports translational research and biomarker-guided drug development, and places no limit on the number of biomarkers that can be tracked. “It can handle multiple biomarker technologies simultaneously, such as complex flow cytometry, next-generation sequencing, immunosequencing, epigenetic profiling, and other types of asssys measuring biological variation as well,” says Marshall. Their type of “translational informatics” tool is more efficient than producing reams of data “without a strategy to get actionable insights from them,” he says, “and it costs only a fraction of the assays themselves.”
Research groups use Precision for Medicine’s platform for neuro-related applications that range from disease pathogenesis to developing complex signatures that are predictive of treatment response. For example, the platform was recently used in a study of therapeutic treatment of major depressive disorder, involving analysis of genomic and transcriptomic data. The result is a genomically defined subset of patients with a clear likelihood of clinical improvement. “Such a signature can now be assessed by an assay, which can then be developed into a companion diagnostic or complimentary diagnostic to more successfully target the proper patient group,” says Marshall.
Biomarker data management platforms become even more essential as different types of biomarkers are analyzed together, such as proteins and miRNAs. And combining types of biomarkers is likely to increase their clinical utility. “Diagnostics will become more and more robust as we understand the interplay between microRNAs, proteins, DNA, and messenger RNA,” says Pregibon. Clinical decision-making will also benefit from an increase in the number of well-characterized biomarkers—especially for neurological diseases, where the blood–brain barrier has mostly blocked the brain from the reach of assays until recently. “The opportunity to leverage biomarker-driven targeted therapies means that the patients who are more likely to respond to therapies are getting them faster,” says Marshall. “To me, that is the power of biomarkers.”
- J. M. Oliver et al., J. Neurotrauma 33, 1784–1789 (2016).
- C. Hu et al., PLOS One 10, e0126643 (2015).
- J. P. Lopez et al., Nat. Commun. 8, 15497 (2017).
Newly offered instrumentation, apparatus, and laboratory materials of interest to researchers in all disciplines in academic, industrial, and governmental organizations are featured in this space. Emphasis is given to purpose, chief characteristics, and availability of products and materials. Endorsement by Science or AAAS of any products or materials mentioned is not implied. Additional information may be obtained from the manufacturer or supplier.