NEW YORK (360Dx) – Researchers at Georgia Institute of Technology and Emory University are developing nanosensors that can be injected intravenously to produce fluorescence in urine that can be measured outside the body to guide immunotherapy decisions.
The nanosensors consist of nanoparticles conjugated with a peptide substrate specific for protease granzymes. In preclinical studies, the approach is providing diagnostic signals in urine far stronger and more abundant than can be achieved by blood tests for similar indications, said Gabriel Kwong, an assistant professor of biomedical engineering at Georgia Tech and Emory University, who is leading the platform's development.
Kwong said in an interview that the researchers anticipate commercializing a platform that clinicians could potentially use to make better and more effective decisions about immunotherapy treatments.
While immunotherapies have opened up new potential pathways to treat cancers, some cancers have shown the ability to develop resistance to such treatments. Such non-responders need to be detected quickly, and Kwong and his colleagues are looking to commercialize a platform that can better classify different forms of resistance, so that clinicians can do a better job of selecting treatments that are most effective.
The National Institutes of Health has awarded $1.8 million to the team to support their work to develop the platform for this application.
The system uses nanoscale structures with bristles made of short amino acid strands that have fluorescent reporter molecules. Made from biocompatible materials, the nanosensors accumulate in compromised tissue present in cancer tumors.
Importantly, the sensors have short amino acid strands that mimic cancer cell strands that play an important role in enabling cancer cell death by immunotherapy, Kwong said.
When patients receive immunotherapies, T cells — either engineered or already within the body — secrete granzymes that cut target amino acid strands and kill the cancer cells. While cutting the target amino acid strands that trigger cancer cell death, the granzymes also cut the nanosensor strands. That process releases the reporter molecules that travel through the kidney’s filtration system and enter the urine, producing fluorescence that can be analyzed to determine the intensity of the immunotherapy’s attack on cancer.
Kwong said that he anticipates commercializing the nanosensor platform within "three to five years" through Glympse Bio, a company he cofounded in 2015 with Sangeeta Bhatia, who heads the Cancer Nanomedicine Center at the Massachusetts Institute of Technology.
The firm currently is validating the platform in preclinical studies using mouse models, and expects to advance to human clinical trials. Afterward, Glympse Bio plans to apply with the US Food and Drug Administration for clearance of the system as a drug-diagnostic device.
Glympse Bio is also developing the protease nanosensor platform to detect cancers and infectious diseases, among other indications and applications. The firm is conducting its first clinical trial to validate the platform in patients with nonalcoholic steatohepatitis (NASH). And, in a recent study published in Nature Biomedical Engineering, Kwong and his colleagues described the potential for using the platform in early detection of transplant rejection.
In preclinical studies, the platform's signals are hundreds of times stronger than is being produced by diagnostic approaches that measure tumor fragments, tumor DNA, or circulating cells in blood, Colin Buss, a graduate student and researcher in Bhatia's MIT lab, said in an interview.
Buss noted that he is familiar with the work of Kwong and his colleagues, but he is not involved in the development of the platform or employed by Glympse Bio.
Stronger sensor signals may enable early detection in a number of applications and guide immunotherapy decisions, Buss said.
"We know that immunotherapies work for subsets of patients, and a lot of research is being done into why that might be and which patients are likely to respond to specific treatments," he said. "By being able to quickly monitor a patient's response to an immunotherapy treatment, the nanosensor platform could allow patients to get the best treatment possible in as short a timeframe as possible."
Because clinicians need more effective tools than are available through diagnostic imaging, researchers are developing in vitro diagnostic platforms, including those that use liquid biopsies, sequencing, and other diagnostic methods to improve immunotherapy selection.
Imaging-based methods, including CT, PET, and MRI, which is currently used most frequently, is not ideal, Buss said, in part because its sensitivity is lower than is desired by clinicians.
Tumor shrinkage is currently an important measure of treatment success, but when a tumor is infiltrated by T cells, it can swell, providing a false impression that cancer is growing.
Clinicians can encounter "counterintuitive responses whereby a tumor that is responding well to an immunotherapy might show up brighter or slightly larger on imaging because of an immune reaction against the tumor," Buss said. Then, the clinician might mistakenly take the patient off an immunotherapy that's actually working well for them.
With colleagues, Kwong anticipates publishing a paper soon that describes the first use of the platform to guide immunotherapy. His group is also investigating the potential to use the platform to develop bespoke diagnostics — assays that are customizable for specific disease indications and applications but use the same technological approach.
The human genome produces 550 proteases, the enzyme relevant to detecting and combating disease, Kwong said. His team is developing a nanosensor panel that differentiates T-cell activity in tumors over their activity in fighting colds or other indications, Kwong said.
The group has experimented with multiplex panels that included up to 100 different proteases, so getting to 550 is not as daunting as it sounds, he said.
The probes could be combined into a cocktail, he said, to detect various indications, and use machine learning to analyze disease fingerprints in the urine signals.