For nearly two years, public health systems have been playing a game of Russian roulette when it comes to severe COVID-19. Some people get extremely sick or die. Others escape with few or no short-term symptoms.
We do have some ways of predicting who might get very sick. As the Centers for Disease Control and Prevention reports, people at high risk are usually above the age of 65 or have underlying medical conditions or are people of color who are not well served by the U.S. healthcare system.
But scientists are identifying other markers of COVID-19 risk that could predict hospitalization better than surface-level characteristics. A new study in EMBO Molecular Medicine suggests that there are “fitness” proteins on the surface of certain cells that can predict who might progress to a severe case of COVID-19 should they be infected.
They’re called Flower-lose proteins (or hFwe-Lose, scientifically) and they’re usually expressed in cells that are nearing the end of their lifetimes. In this case, the scientists were able to find it through a simple nasal swab—theoretically opening the door for people to be tested for it at the same time they are tested for COVID infection itself.
In a study involving nasal swab samples from 283 people, the authors of the new paper were able to predict who would end up hospitalized with about 84 to 88 percent accuracy.
“It’s like a pre-existing condition,” lead study author Rajan Gogna, a cancer biologist and assistant professor at the University of Copenhagen told The Daily Beast. “In our finding, your chances of telling a person to go home when they should be in the hospital are virtually nil.”
Biomarkers that can predict the course of disease are helpful when patients turn up in hospitals, said Faraaz Shah, an intensivist at UPMC and assistant professor of pulmonary, allergy, and critical care medicine at the University of Pittsburgh who was not involved with the study. He has treated COVID-19 patients in the ICU. “It's encouraging to have results like this, recognizing that there still is a lot of work to go before we can take them to something that helps at the bedside,” he told The Daily Beast.
Flower-lose proteins are like a “fitness fingerprint” for a cell, explained Gogna. Cells that express significant amounts of them are essentially signaling to nearby cells that they’re struggling.
Gogna’s theory is that these struggling cells are easier for the COVID-19 virus to infect. And once the infection takes hold, they’re more likely to die as the virus sets off a cascade of changes inside the body. These clumps of dead cells would then lead to the respiratory distress seen in severe COVID patients.
Gogna and his team tested his hypothesis by first looking at lung tissue samples from 11 people who had already died from COVID. They found that there were clusters of dead cells present in those tissues, suggesting that cells had died in packs, rather than at random.
When the team analyzed those cells, they found that they expressed higher levels of Flower-lose proteins, suggesting they were “unfit.”
This finding alone wasn’t especially useful—taking lung tissue samples from people is an extremely invasive procedure, for the living and dead alike. In a follow-up study, the authors found that Flower-lose proteins were detectable via nasal swabs (the kind that reach back into the uppermost part of your nose and throat).
So the next step was to figure out whether the presence of the protein actually could predict how well people fared against COVID-19. The team analyzed nasal swabs taken from 200 people who had recently tested positive for COVID-19 and reached out to healthcare providers. Based on their early nasal swab samples, the team accurately predicted that 72 people would end up in the hospital after COVID infection (86 people actually were hospitalized). In short, the Flower-lose method was about 84 percent accurate in predicting hospitalization.
The study also followed 80 patients who had just tested positive for COVID. In this case, the Flower-lose method was about 88 percent accurate in predicting who ended up in the hospital.
In theory, you could get a second swab done when you go for a COVID-19 test, and test for this biomarker to determine your risk level for severe infection. In practice, however, it would more likely be used by doctors to inform treatment once you show up at a hospital with symptoms.
“I don't really see this as being something that would be in an over-the-counter kit,” Jaquelin Dudley, a professor of oncology and molecular biosciences at the University of Texas at Austin who was not involved in the study, told The Daily Beast.“I think this would really be a hospital-based test.”
One of the biggest questions is whether Flower-lose proteins are really a better predictor of severe infection than other well-known risk factors, like age or underlying conditions.
High levels of Flower-lose proteins actually correlate with some known risk factors. “If you are old, obese, or have bad health, the cells in your body tend to increase the levels of Flower-lose, which is a signal of low quality of cells,” said Eduardo Moreno, another study author and director of the cell fitness lab at the Champalimaud Center for the Unknown in Portugal.
But the new findings suggest that Flower-lose proteins are a slightly better metric than other factors (like age) for predicting severe COVID. Gogna argues this analysis could be used to make specific predictions about an individual’s prognosis.
Biomarkers like Flower-lose proteins that predict disease outcomes with more specificity are called prognostic biomarkers. Shah would like to see more biomarkers that can inform how patients will react to certain treatments—predictive biomarkers—but he acknowledged prognostic biomarkers certainly have a role to play.
“It's surprising to us. We will have a patient who's older with multiple health problems, who's gotten COVID and has done well. Then we'll have a patient who is relatively younger, healthier, and they don't make it out of the ICU,” he said.
“[Prognostic biomarkers], I think, are important because it would help us guide our care.”
Ultimately people probably won’t be lining up for nasal swab tests to determine their COVID-19 risk. But for those who do end up getting sick, another biomarker could help chart the course of their treatment, and perhaps be the difference between life and death.
Editor's note: This article originally misspelled Jaquelin Dudley's first name. We regret the error.