NEW YORK – Less than a month after receiving 510(k) clearance from the US Food and Drug Administration for its MeMed BV test and its MeMed Key platform, Israeli diagnostic firm MeMed is working to commercialize a new test to predict whether patients will have severe responses to COVID-19 infection.
The MeMed COVID-19 Severity test uses the same core technology as the BV test, which distinguishes between bacterial and viral infections, cofounder and CEO Eran Eden said. While the MeMed Key platform was originally developed for the BV test, Eden said the firm switched its focus to COVID-19 after the BV test was finished.
The test "tries to address the clinical problem" of whether a patient will develop a severe outcome from the virus, helping guide clinicians about sending a patient home or admitting them to the hospital, Eden said.
There's "a lot of uncertainty" surrounding the issue, since physicians don't want to admit patients to a hospital if they don't need to be admitted, but they also don't want to risk a patient deteriorating at home, Eden added. Twenty percent of patients are admitted to the hospital unnecessarily, while 8 percent aren't admitted when they need to be, he said.
The test looks at three soluble proteins — TNF-related apoptosis-induced ligand (TRAIL), interferon gamma-induced protein-10 (IP-10), and C-reactive protein (CRP) — to determine a patient's immune response to the virus. That response can predict outcomes up to 14 days in advance, with a severe outcome meaning a patient is likely to end up in the intensive care unit or die, Eden said.
The proteins measured in the severity test are the same as those measured in the BV test, just used in a different way, Eden said. The TRAIL protein, which Eden called the "pillar protein in the signature," goes up in a normal viral infection and is "extremely low" in severe patient outcomes. There is a "big differential" between patients with a competent viral immune response and patients that will experience a severe outcome, he said.
IP-10 is used in infectious diseases and increases when there's a lung injury. CRP, meantime, is a marker that indicates hyper-inflammation. The combination of the three proteins allows the test to "get something more extensive" than it would using just one or two of the proteins, Eden said.
Niv Mastboim, director of product innovation at MeMed, noted that other markers the team looked at weren't as helpful in determining severity. By developing an "all new signature" that draws proteins from many different biological pathways, MeMed's test is a "more comprehensive tool," he added.
Around 15 percent of people with SARS-CoV-2 have a dysregulated immune response, Mastboim said, and it's "tricky to separate who will have a severe outcome." But, he also noted that the team wanted to predict the severity of an infection rather than just confirming that a patient was experiencing a severe outcome.
The point-of-care platform can return a result from a serum sample in approximately 15 minutes, inputting the measurements of the proteins into an algorithm that produces a COVID-19 severity score.
The score ranges from 0 to 100, with 0 to 20 indicating a very low likelihood for severe outcome; 20 to 40 indicating a low likelihood of a severe outcome; 40 to 80 indicating a moderate likelihood; and 80 to 100 indicating a high likelihood.
A key goal for the test was making it "intuitive" and easy to use, which is why the company decided to go with a consolidated algorithm that scored patients from 0 to 100, Mastboim said.
Making the platform and test intuitive allows physicians to incorporate the tool into the "very hectic [emergency department] workflow," he added.
In an abstract presented at the American Association for Clinical Chemistry's Annual Scientific Meeting last month, the company explained how it developed the algorithm. To create the model, the firm analyzed 518 COVID-19 patients from the US, Germany, and Israel, 113 of whom had a severe outcome, with a machine-learning algorithm it developed.
The company used regularized logistic regression to integrate the three proteins into a single score.
The researchers found the COVID-19 severity score has an area under the curve, or AUC, of .86, meaning it outperforms other COVID-19 severity stratification tests in terms of accuracy. Many other severity stratification tests rely on interleukin-6, which has an AUC of .77, according to the abstract.
MeMed is currently working on multinational, multicenter studies to further validate the test, which are either completed or close to completion, Eden said. It is also working with the FDA to determine the regulatory pathway for the test. Although the MeMed Key platform is already cleared, the new cartridge and algorithm will need to receive regulatory approval.
"The burden of proof is on us," Eden said. He added that the firm can't put a concrete timeline on when it will receive clearance for the test from the FDA, although the test received CE marking in July.
Sergey Motov, an emergency medicine physician at Maimonides Medical Center in New York who participated in one of the studies, said the test's "unique combo of three proteins" makes it more accurate and noted that the test also helps clinicians determine which specific therapy would be best for a patient, depending on their likely outcome.
"I believe the test is invaluable to acute care providers," Motov said, adding that it would be helpful for providers outside of the emergency department throughout the hospital as well, including providers in the ICU determining treatment for COVID-19 patients.
"It allows us to provide superb, tailored, individualized care," he said.
Currently, Eden said the plan is to roll out the test in hospital emergency rooms, although there could be other use cases for the test in the future.
The severity test is "part of a bigger picture," he said, as the company has a universal severity test for all infections currently in development. That test would provide insight outside the context of COVID-19 and could help with future pandemics, he added. It's "a solution to the scenario of a fever," Eden said.
MeMed plans to use its commercialization process of the BV test to roll out the severity test as well, focusing primarily on the EU and US. Right now, the company is looking at providing the test directly to private hospitals, although Eden said there has been interest from the public sector, including public health hospitals.
The firm is also looking at launching the test at urgent care centers, Eden added.
Eden said the company did question the value of developing the test if the pandemic ends, but said the Delta variant has shown that new strains of COVID-19 are "probably here to stay" and the virus likely won't go away.
"It's a long-term game," he said. "We need to learn to live side by side [with COVID-19]."