A sweat sensor developed at the University of Texas can predict whether a person will develop an excessive immune reaction known as a cytokine storm that characterizes the worst COVID-19 cases.
The researchers, who reported their results this week at the American Chemical Society spring meeting, think the sensor also has potential to help patients with other conditions or infections such as flu, which can also result in a cytokine storm.
At the moment the sensor is worn as part of a wrist band, a bit like a smart watch, but the team thinks it could be developed further and made into a plaster like patch to give more flexibility.
With something like a cytokine storm, it can happen fairly quickly and its important to treat a patient’s symptoms as soon as possible to minimize damage to the body. Anything that can speed up this process is beneficial.
“Especially now in the context of COVID-19, if you could monitor pro-inflammatory cytokines and see them trending upwards, you could treat patients early, even before they develop symptoms,” says Shalini Prasad, Ph.D., a researcher at the University of Texas in Dallas, who led development of the sensor and is presenting the work at the meeting.
Many of the biomarkers found in blood are also found in sweat, including inflammatory proteins such as cytokines, but levels are generally lower. This means that an efficient sensor needs to be able to pick up very small amounts of the markers it is looking for.
This can be a challenge, but Prasad and team have already gained experience in this area by developing a similar sensor that detects two proteins found in sweat during inflammatory bowel disease flareups. This work is currently being commercialized by a startup Prasad founded called EnLiSense, which is also working on the cytokine storm sensor.
The new sensor includes electrodes coated with antibodies that can bind to seven inflammatory proteins that are seen at high levels in patients experiencing a cytokine storm. They include: interleukin-6 (IL-6), IL-8, tumor necrosis factor-α (TNF-α), TNF-related apoptosis-inducing ligand, IL-10, interferon-γ-induced protein-10 and C-reactive protein.
When proteins in the sweat bind to the sensors the electrical current changes and gives off different signals to a reader. This then connects with a smartphone app that estimates levels of the different proteins in the sample. It’s a continuous process as sweat gradually works its way through the sensor and makes way for new sweat to take its place meaning that measurements can be taken over time.
This new sensor is still at the prototype stage, but in a test on six healthy people and six with influenza the levels of cytokines measured by the sensor were comparable to those measured in blood samples taken for comparison. Prasad thinks this sensor has real potential to see which COVID-19 patients are at risk of developing an excessive immune reaction.
“Access to COVID-19 patients has been a challenge because healthcare workers are overwhelmed and don’t have time to test investigational devices,” Prasad says. “But we’re going to continue to test it for all respiratory infections because the disease trigger itself doesn’t matter — it’s what’s happening with the cytokines that we’re interested in monitoring.”
Notably, one thing that came out of the testing is that it is important to test ‘passive’ rather than ‘active’ sweat as levels of inflammatory proteins such as cytokines can change while we exercise and distort any measurements designed to diagnose illness.
Prasad and her team found their sensor is very sensitive and could even detect cytokines present in the low-picogram-per-ml range in people taking anti-inflammatory medication. It was also able to be used continuously for up to 168 hours (around a week) before the sensor strip needed to be replaced.
The next step for the researchers is to test the sensor in more people and validate their initial findings in patients with a range of different respiratory infections.