NEW YORK – As the COVID-19 pandemic continues with no sign of letting up, researchers across the world are looking into ways to help physicians treat and manage patients with the virus more effectively.
A new study led by researchers at the Royal College of Surgeons in Ireland's University of Medicine and Health Sciences in Dublin makes the case that measuring and comparing certain interleukin biomarkers could provide information about how severe a patient's response to SARS-CoV-2 will be in advance.
Researchers developed a measurement, the Dublin-Boston score, to predict how severe the infection will be on day seven after measuring the patient's blood for four days and monitoring IL-6 and IL-10 levels. IL-6 is a mostly pro-inflammatory biomarker, while IL-10 is mostly an anti-inflammatory marker, so senior author Gerry McElvaney, a professor of medicine at RCSI, noted that making it a ratio instead of just measuring one marker or the other allowed for a more complete picture of the patient's inflammatory response. "Rather than concentrating purely on the high pro-inflammatory levels, we thought we'd look at the inability to mount an anti-inflammatory response," he said.
Patrick Barth, a spokesperson for diagnostics and pharma giant Roche, which has US Food and Drug Administration Emergency Use Authorization for a test for IL-6, said that because IL-6 is released particularly early during a severe infection, it can help physicians determine which patients are severely ill.
The RCIS study, published in The Lancet's translational research journal EBioMedicine this month, was performed on relatively fresh samples from patients in the hospital, which McElvaney said allowed for more accurate measurements. Previous measurements of IL levels relied on samples that had been stored for months to years and had been frozen and unfrozen, which could have affected the levels, he added.
To measure the levels, researchers used an enzyme-linked immunosorbent assay. The levels were compared on the first day a patient arrived at the hospital to day two, and on day two to day four. McElvaney said the better result was over a longer time period – by day four researchers had "extra insight into the overall situation," by getting a view of how a patient progresses over time.
"We'll be able to say with a fair amount of accuracy what's going to happen by day seven," McElvaney said. Knowing what will happen in advance can help hospital staff appropriately escalate and de-escalate care, especially when it comes to allocations of beds in emergency rooms and intensive care units.
To get the score, researchers measured IL-6 and IL-10 ratios the first day patients were in the hospital and on the fourth day they were in the hospital. They then subtracted the first ratio from the second, multiplied that by two, and rounded it off to the nearest whole number, McElvaney said. There's a five-point scale, ranging from -2 to 2, with -2 signifying normal biomarker levels and 2 being worst. A one-point upward change on the scale leads to a 5.6 times chance of a worse prognosis, making it more likely a patient will be on a ventilator, McElvaney said.
IL-6 by itself has some useful predictive value, which has led companies like Roche and Beckman Coulter to develop immunoassays to measure the marker. Shamiram Feinglass, Beckman Coulter's CMO, noted that using IL-6 could help risk stratify patients, and knowing their IL-6 levels could allow clinicians to try to mitigate risk and could act as "an early warning system."
More hospitals and national healthcare agencies are including IL-6 in their guidelines to identify patients with severe complications early, Roche's Barth said. IL-6 is part of the Chinese Clinical Guidance for COVID-19 Pneumonia Diagnosis and Treatment published by the China National Health Commission, for example, and it is included in the Italian Society of Infectious and Tropical Diseases' handbook for caring for patients with COVID-19.
Feinglass said Beckman's test can help with admitting triage, and he noted that some clinicians are using it on the frontlines as early indicators. Beckman's immunoassay runs on its Access instruments and takes about 20 minutes to return results, she said.
"A lot of labs measure interleukin-6, not that many measure interleukin-10," McElvaney said. IL-6 is barely detectable in a healthy person, and the lower an IL-6 level, the more difficult it is to detect change, he said. The body's response to inflammation is partly determined by high levels of IL-6, while IL-10 is an attempt to mitigate inflammation in the body, leading really sick people to have a blunted anti-inflammatory response and less IL-10 than they should, he said. That is the reason the ratio increases as a patient gets sicker.
COVID-19 sets off a "chain of inflammation" in the body, resulting in an increase in cytokines or a "cytokine storm," although McElvaney emphasized the importance of measuring the balance of cytokines. The increase in cytokines isn't only a sign of inflammation, but also propagates inflammation, co-author and Gerry's son Oliver McElvaney said. When the balance of cytokines is off, patients experience fever and increases of C-reactive protein, and it can alter patient's blood cell counts, he added.
The ratio is "a picture in time," and shouldn't be used just to treat the rise in interleukin levels, Gerry McElvaney continued. "The answer is not to necessarily decrease IL-6, because doing so may have unintended consequences," he said. "It's more about the underlying cause and what's driving it, Oliver McElvaney added. The ratio can give an idea of the timing of an inflammatory response and how quickly severity of the illness will increase. The levels are impacted by treatments as well, such as steroids, so measuring the interleukin levels can help clinicians understand how well a given treatment is working for a certain patient.
Clinicians can also "get a feel for the patient that is stable right now but that might deteriorate rapidly," Oliver McElvaney added.
Although this specific study was only performed on patients sick enough to come into the hospital, Gerry McElvaney said measuring the interleukin biomarkers could be done on asymptomatic patients. Oliver McElvaney added that people with -2 scores could probably stay in their communities without spreading disease but cautioned that that was "purely speculative."
Gerry McElvaney noted that the researchers are preparing to test the score on asymptomatic patients and said he hopes other researchers test the score to get more data. There is a need for standardized, multicenter trials to ensure the score is accurate and to refine it further, he said.
"We think this is a very useful tool, certainly in hospitalized patients, and it'd be even more interesting if it was useful in people who weren't hospitalized," Gerry McElvaney said. It can also be used for other viral infections, since inflammatory responses are common in infected patients, although the biomarkers might not be IL-6 or IL-10. "Cytokines don't exist in isolation … they're always interacting with other cytokines," he said.
Feinglass added that knowing a patient's IL-6 levels can help physicians determine the logistics of caring for a patient and whether a crash intubation will be necessary – an emergency procedure that can add stress on the patient and provides less time to plan for their care. Knowing IL-6 measurements can also potentially stave off ventilation and help clinicians figure out if other interventions would work for a patient, such as steroids.
Not a silver bullet
There were some limitations to the study, Gerry McElvaney said, including the fact that it was only performed at a single site on a small number of patients. However, a benefit of testing at a single site meant the patients could be regularly measured and researchers knew all patients had been treated the same way. In addition, the laboratory is based inside the hospital, so there was little potential for mishaps when transporting samples.
Joshua Hayden, the chief of chemistry at Norton Healthcare in Louisville, Kentucky, who has studied IL-6 and IL-10, said the study was interesting, but that in order to be used in the real world, there would need to be high-quality and widely available IL-6 and IL-10 tests, which isn't a reality in the US, especially for IL-10. He added, "There are still a very small number of labs offering this testing in the US."
In addition, he said there are questions about the performance of these tests, specifically regarding inter-assay variability. "Interleukin tests are well known to show very poor agreement," he said. "Whether the authors' findings with this one particular IL6/10 test is applicable to other IL6/10 tests is unclear."
Hayden also said there have been a number of observational studies that have shown elevated IL-6 and IL-10 levels are "a poor prognostic factor in COVID-19 patients." He did note, however, that there are reliable IL-6 tests that have provided "measurements we can trust and can now put to use," for COVID-19.
The main benefit of the score is its simplicity, Oliver McElvaney said. "We didn't aim to make the score super complicated. We wanted it to be as simple and as clear-cut as possible."
The point of the test is to allow physicians to estimate severity at the patient's bedside and "actually affect patient care," Gerry McElvaney added. He said he expects to see the score utilized worldwide eventually, and that he was particularly interested in changes in the score based on treatment. "This will be a good way to figuring out if [patients] are trending in the right direction," Oliver McElvaney said.
Although there is still significant work to be done to determine the full scope of clinical utility for the Dublin-Boston score and interleukin measurements, Hayden said the question of predicting severe COVID-19 infections is essential.
"We have, understandably, been laser focused on developing COVID-19 diagnostic testing," he said. "Unfortunately, we have not paid enough attention to developing COVID-19 prognostic testing. … It is tremendously important we develop these prognostic tests and give our providers more tools for predicting which patients will experience a severe course."