Researchers from the Georgia Institute of Technology have come up with a novel, non-invasive way to monitor anti-tumor responses to immune checkpoint inhibitors using biosensors that are released into urine.
The biosensors track the activity of protease enzymes, which play a key role in T cell cytotoxicity and tumor biology, and are produced in response to treatment with anti-programmed cell death protein 1 (αPD1) antibodies.
Although immune checkpoint blockade with treatments such as αPD1 antibodies has transformed cancer therapy in recent years, response rates may be below 25% for many tumor types. On top of this, it can be difficult to assess therapy response in a timely manner—radiographic response assessment is not carried out until after the first 8- to 12-week course of therapy and tumor biomarkers can be difficult to detect as they are diluted in serum.
Gabriel Kwong, associate professor in the Wallace H. Coulter Department of Biomedical Engineering at Georgia Tech and Emory University, and colleagues therefore set out to develop a non-invasive way of measuring response and resistance to immune checkpoint blockade in the early stages of treatment.
They used their expertise in biosensors to create a fluorescently labelled peptide substrate that is selective for granzyme B—a serine protease released by cytotoxic T cells. They then coupled this sensor to αPD1, which carries it to the tumor environment.
To understand how the biosensors work, “we first have to understand what happens with treatment,” Kwong told Inside Precision Medicine. “Immune checkpoint therapy works by reinvigorating T cells that, through the release of potent enzymes called proteases, kill cancer cells. Our biosensors are activated by these proteases to release a signal that is then concentrated into urine for detection.”
The sensors are detected in urine by mass spectrometry and when the researchers tested them in a mouse model of colorectal cancer they found that sensor levels began to increase after the second treatment, given on day 10, and were inversely associated with tumor volume with a high degree of sensitive and specificity.
The researchers also “showed that before immune checkpoint therapy began to reduce tumor burden in mice, we were already able to detect differences by our urine test,” Kwong remarked.
However, the team points out in their Nature Biomedical Engineering paper that granzyme B expression on its own is not a specific biomarker of immune checkpoint blockade response but instead a general biomarker of T cell cytotoxicity that could also be elevated as a result of immune-related adverse events or reactivation of latent viral infections. Moreover, the granzyme B sensor was unable to differentiate between resistance mechanisms that cause loss of T cell cytotoxicity.
To address this, Kwong and co-workers designed a panel of 14 probes to improve the ability to assess response and resistance to immune checkpoint blockade using signature analysis.
“We showed that with artificial intelligence, we can train computers to pick up patterns in these signatures to differentiate not just when the therapy is not working, but also to identify the pathway that prevents the therapy from working,” Kwong remarked.
In spite of the positive findings, he acknowledged that it will be several years before the biosensors undergo clinical testing.
“We will first have to demonstrate safety in patients and show that repeat dosing is safe,” Kwong said. “Afterward, then we can design a clinical study to demonstrate whether our biosensors can report on patient responses.”
