August 8, 2023 | A point-of-care (POC) diagnostic test for long COVID is under development that would be a godsend to the millions of Americans still suffering the health consequences of SARS-CoV-2 months after being infected. In the absence of any sort of diagnostic test or evidence-based therapy, they are understandably frightened and demoralized by their condition, according to Douglas Fraser, M.D., Ph.D., a scientist at Lawson Health Research Institute and a critical care physician at London Health Sciences Centre (LHSC) in London, Ontario.
Startlingly, long COVID is thought to strike up to 20% of acutely ill COVID patients with a constellation of new or lingering symptoms—including fatigue, brain fog, difficulty breathing, and cardiovascular symptoms ranging from chest pain and arrhythmias to sudden cardiac death—but it remains a diagnosis of exclusion with an unknown biological basis, he says. Several published papers have even suggested it might be a psychosomatic ailment.
Fraser and his colleagues have been putting that hypothesis to bed with a string of studies, the most recent of which used targeted proteomics and bioinformatics to understand the cells and signaling pathways involved in the manifestation of long COVID (Journal of Translational Medicine, DOI: 10.1186/s12967-023-04149-9). Those data were used to search drug databases for therapies known to be safe for human use and target pathways that are hyperactive, for potential repurposing in future clinical trials.
Researchers measured over 3,000 proteins in blood plasma samples collected from long COVID outpatients at St. Joseph’s Health Care London and acutely ill COVID-19 inpatients at LHSC, as well as a group of healthy volunteers. As in the earlier studies, vascular transformation biomarkers were found to be significantly elevated in those with long COVID, says Fraser. Patterns also emerged suggesting why those blood vessels are changing.
The expression of the protein biomarkers was determined with proximity extension assays, which Fraser describes as a technology combining proteomics and genomics that has been well exploited by Olink, which curated markers for the study. The method involves tagging an antibody with a nucleotide sequence before binding to a protein, enabling the signal to be amplified in a linear fashion (using next-generation sequencing or polymerase chain reaction) to a measurable level, he explains.
Multiple bioinformatics tools powered by artificial intelligence were then used to break that complex system into its component cell types and signaling mechanisms, as well as by specific organ. The analysis found that people with long COVID have prolonged inflammation associated with changes in their immune cells and blood vessels that could lead to problems in the heart and brain.
Collectively, study results point to vasculo-proliferative process in long COVID that is probably initiated either by prior hypoxia (low levels of oxygen in body tissues either locally or systemically) and/or stimulatory factors such as cytokines, chemokines, and growth factors. A vascular proliferative state associated with the hypoxia-inducible factor 1 (HIF1) pathway could mark the progression of patients from acute COVID-19 to long COVID—and contribute to changes in the organ-specific proteome.
Implicated biomarkers suggest patients have a lot of inflammation in their heart, driven by extracellular remodeling of cardiovascular blood vessels and the glue that holds cells together to help the organ function properly, says Fraser. There is also evidence of brain dysfunction, as suggested by markers of how cells talk to each other at the level of the synapse, as well as proteins implicated in language impairment and alterations in various ion channels.
An association was also found between long COVID and elevation of an amyloid precursor protein, which is thought to be an early marker of Alzheimer’s disease and neurocognitive decline, he adds. This could be the biological basis for patients experiencing brain fog, headache, and ongoing neurological symptoms.
Among the bioinformatics pipelines used in the study was CYBERSORT, an analytical tool out of Stanford University that allowed the researchers to understand the relative proportion of various cell types and how they’re changing based on the circulating proteins found, says Fraser. In this way, they learned that two specific cell types are changing with long COVID—natural killer cells, which go from an active to a resting state, and neutrophils, which get activated to form extracellular traps.
These cellular changes suggest a chronic form of inflammation is at work and blood vessels that are still inflamed are trapping neutrophils that then release enzymes and DNA. Extracellular traps are valuable when the body wants to snare an infectious agent and release substances to kill the bacteria, he explains. But with long COVID, the process is more likely to result in microclotting in the blood vessels.
“My overall impression is that it’s a system that has been highly dysregulated and injured and is trying to come back in a healing way but is having some ups and downs as it is working through that process,” Fraser says. Since blood vessels serve to connect all the body’s systems, inflammatory proteins and signals in one area will be carried to other areas. “Depending on how the virus affected you initially, you are going to have susceptibility [to injury] in a variety of different tissues.”
“Right now, the diagnosis of long COVID is rather subjective,” says Fraser, based on an initial SARS-CoV-2 infection and having persistent or new symptoms related to it three months later. “Typically, if you see a physician, you will also have lots of tests done that would be negative.”
The potential disease markers being analyzed here are proteins released by cells that play a role in the body’s immune response to viruses, he says. As proven out in an earlier study, damaged and inflamed blood vessel cells that take up the SARS-CoV-2 virus go looking for healthy tissue so they can optimize the job they need to do (e.g., gas exchange in the lungs) once individuals come out of the acute COVID-19 phase.
“It is pretty clear that long COVID can create a lot of different symptoms in a lot of different organ systems, and that has created a lot of confusion,” notes Fraser. “But the one system in your body that connects all your organs and tissues are your blood vessels... and they are not static. They are always growing and looking for new and healthy tissue, and they are also retracting from rotten and dead tissue, so it’s a very dynamic system.”
Day-to-day vasculature functioning involves multiple other systems in the body that routinely release proteins because of their interaction, Fraser says. When people have an acute infection with SARS-CoV-2, one of the affected regions is the lungs, where alveoli and the small blood vessels surrounding them are exchanging oxygen and carbon dioxide. “If they are not well matched, that process doesn’t work well.”
Of the 16 vascular transformation biomarkers measured, 14 of them (e.g., ANGPT1, MMP1, VEGF-A) were grossly elevated, says Fraser, published last year in Molecular Medicine (DOI: 10.1186/s10020-022-00548-8). This suggests that angiogenesis (the development of new blood vessels) may be common to patients with prolonged and diffuse symptoms and modulators of that process may have therapeutic efficacy.
A panel combining the top three vascular transformation markers showed strong disease specificity, with virtually no overlap with any other investigated condition, he adds. It is now being explored as a panel as a long COVID diagnostic test.
In another study that published earlier this year in Molecular Medicine (DOI: 10.1186/s10020-023-00610-z), Fraser and his team used a variety of artificial intelligence techniques to help narrow down the number of biomarkers seen in long COVID patients to the most important proteins, the tissues where those were expressed, and what the various combinations revealed. They identified 119 highly relevant proteins and two optimal models with nine and five proteins, respectively.
Drugs associated with the HIF1 pathway with therapeutic potential against long COVID, all approved for human use and some still under clinical development, are listed in the paper. These include anti-anemia and anti-ischemia drugs targeting PHD; an anti-inflammatory drug targeting NOS2; cancer drugs targeting VEGFR1, VEGFR2, TOP1, ANGPT2 and ANGPT1/2R; and a drug for a group of metabolic disorders targeting TFRC.
The research group has already patented hundreds of proteins that seem rather specific for long COVID, reports Fraser. A licensing agreement has been signed with a diagnostics company to create a laboratory-based test as well as a POC diagnostic.
“We are hoping to have the point-of-care diagnostic prototype completed and ready for clinical trials in the upcoming months,” he says. Once approved, the POC test could be available direct-to-consumer as well as in the offices of family physicians and at urgent care clinics, hospitals, and emergency rooms.
Fraser’s message to individuals struggling with long COVID is to know that good scientists worldwide are doing their best to work on the problem using state-of-the-art technologies and are quickly making discoveries that could bring an end to the suffering soon. At the very least, he says, the wait won’t be long for a definitive diagnostic test for long COVID.
This should offer emotional reassurance to patients, since it is likely symptoms of long COVID will clear up with time if not because of a specific intervention, says Fraser. A diagnostic test will also help ensure they get the medical support they need from an insurance standpoint. Moreover, providers could start worrying less about diagnostic lapses that can lead to a liability claim.
Once long COVID is a diagnosable condition, physicians will also no longer feel compelled to order so many tests to rule out other possibilities, Fraser points out. “Our estimate suggests that a diagnostic test for long COVID could save up to or more than $385,000 per patient in unnecessary [tests] and lost income.”