Researchers at the University of Michigan are developing a highly multiplexed digital immunoassay meant to serve as a rapid bedside diagnostic.
The technique, called pre-equilibrium digital ELISA, or PEdELISA, enables more continuous monitoring of treatment response, disease progression, and related processes by being able to quickly and affordably measure multiple biomarkers, such as cytokines, at numerous time points.
"The platform is particularly useful for timely monitoring and intervention of the diseased conditions of critically ill patients," said Katsuo Kurabayashi, a professor of mechanical engineering at the University of Michigan and in whose lab PEdELISA is being developed.
Digital ELISAs, such as Simoa from Quanterix, generally achieve significantly higher sensitivities — up to approximately a thousandfold — than the traditional single-plex ELISA but incur high instrumentation costs and take up more physical space than is often practical for bedside diagnostic use.
PEdELISA, in contrast, is more miniaturized and, in several recent publications, achieved a sensitivity of between 10 fM and 1 nM for up to 12 multiplexed analytes, with sample-to-answer times ranging from 30 to 40 minutes.
Yujing Song, PEdELISA's lead developer and a Ph.D. candidate in the Kurabayashi lab, said that he and his colleagues aim to strike a balance between high accuracy, high throughput, fast turnaround, and low cost.
"We believe 'sensitivity' is not everything in the existing clinical diagnostic market," Song said in an email.
"In fact," he continued, "it may even be the least important once the platform has gone below a certain threshold."
Alejandro Chavez, a pathologist and researcher at Columbia University who is not involved with the method's development, agreed with that sentiment, mentioning that assays that push the current limits of sensitivity require studies to prove the clinical utility of such ultralow analyte measurements.
"Right now," he commented, "one of the issues we face is that most biomarkers cannot be measured at super low concentrations. Because of this, it isn't clear to the field the contexts in which they will be most useful."
PEdELISA uses a polydimethylsiloxane (PDMS)-based microfluidic chip containing 10 to 16 parallel sample detection channels on a glass substrate, arranged in an array of hexagonal biosensing patterns. This conformation enables each pattern to contain 43,561 femtoliter-sized microwells, which fit into the field of view of a full-frame CMOS sensor through a 10x objective lens.
Magnetic beads conjugated with capture antibodies occupy specific locations on the chip. These bind to the target protein, after which a second detection antibody binds to a different location on the target protein. This complex is later labeled with an enzyme that triggers a fluorescent reaction, by which the analyte binding interaction is measured.
A convolutional neural network algorithm assists in accurately identifying the millions of potential true interactions from background noise.
Uniquely, the algorithm cuts down analysis time while maintaining precision by running two neural networks in parallel, simultaneously assaying for both targets and defects.
Sample incubation periods ranging from 15 seconds to 9 minutes — optimized according to a given analyte's clinically useful dynamic range — also contribute to PEdELISA's rapid turnaround time.
Although the assay itself provides results within 30 to 40 minutes, other factors may affect the total time from sample acquisition to answers in the clinic.
"In the practical operation of our test starting from a patient blood draw and ending with data delivery to physicians, a larger amount of time was spent on sample processing, transport, and team coordination, as well as biosafety and disinfection protocol observation in handling COVID-19 samples," the researchers wrote in a study, in which they used PEdELISA to monitor COVID-19 patients for changes in cytokine levels during a cytokine storm.
Song, Kurabayashi, and their colleagues first tested PEdELISA by using it to monitor changing levels of cytokine biomarkers in response to CAR-T cell cancer therapy across different time points.
"In addition to CAR-T patients," Kurabayashi said, "we have applied PEdELISA [to] monitoring the cytokine storm of severely ill COVID-19 patients, blood testing-based food allergy screening, and functional phenotyping of brain cells coupled with on-chip image cytometry. I believe PEdELISA is highly poised for facilitating time-sensitive therapeutic strategies to treat highly acute life-threatening illnesses and diseases by permitting "on-demand" biomarker testing."
The group filed an international patent application in November and is currently working to fully automate the PEdELISA process and refine the chip design for mass production. They are currently looking for both future customers of the technology, and for potential collaborators in the in vitro diagnostic field, who can help commercialize PEdELISA.
"We will finalize our commercial partner quite soon," Kurabayashi stated.
In the meantime, the team is also exploring the use of PEdELISA in contexts beyond blood samples.
"We are actively exploring measuring biomarkers in saliva, bronchoalveolar lavage fluid, cerebrospinal fluid, and other body fluids with our clinical collaborators," Kurabayashi said.