Cancer Instagram

A snapshot of immune system proteins predicts patients’ response to immunotherapy.

Checkpoint therapy blocking PD-1, a cellular surface marker that dampens the immune response, has revolutionized cancer care, enhancing overall survival in 33 to 40 percent of melanoma patients. However, it does not work in most patients and can have adverse effects and considerable costs to the health care system. Thus, it is crucial to identify patients likely to benefit from blocking PD-1 before administering therapy. MUSC Hollings Cancer Center researcher Carsten Krieg, Ph.D., an assistant professor in the MUSC Department of Microbiology and Immunology and the Department of Dermatology, and colleagues are using a novel technology to address that issue.

The goal was to see if a blood test for biomarkers could identify those who are likelier to benefit, while allowing “non-responders” to begin other treatments without losing time, says Krieg. “It’s a decision instrument for physicians and for the health care system.”

The technology, a single-cell mass cytometry – aptly named Helios because the machine can heat plasma up to 6,000 degrees Celsius, levels comparable to temperatures found on the sun – allows Krieg to generate an ‘Instagram’ of a patient’s immune system to identify the molecular details of the immune cells. “This is possible because there’s artificial intelligence, machine learning combined with algorithms that can make a very complex system easy to visualize,” he says.

Image provided by Dr. Carsten Krieg

“MUSC has already invested in this technology by obtaining a second-generation instrument,” notes Zihai Li, M.D., Ph.D., chair of the Department of Microbiology and Immunology and co-leader of the Cancer Immunology Program at Hollings. “Dr. Krieg’s arrival will catalyze the rapid development and application of this technology to benefit the patients in the state of South Carolina.”

To date, researchers have tried to use patient-derived killer T-cells to identify biomarkers; however, this approach is limited by the small sample sizes that can be attained. Krieg and others in the field are starting to widen their approach by collecting peripheral blood mononuclear cells from a
simple blood draw, so-called liquid biopsies.

The cells are stained with metal-conjugated antibodies that target surface and intracellular proteins. This allows for sensitive detection of more than 30 proteins on a single cell over millions of blood cells.

The stained cells are placed in Helios and are ionized. Because the metals placed on each antibody weigh differently, the resulting ions can be separated into different pools. Using artificial intelligence-guided bioinformatics and expert-guided analysis, researchers can create “immune instagrams,” or simplified visualizations of the blood immune response in a tumor patient.

Using this technology, Krieg and his colleagues confirmed that, as intended, T-cells respond to anti-PD-1 immunotherapy. They showed that the frequency of classical monocytes, a type of immune cell, in the peripheral blood correlated with a patient’s response rate: the more monocytes in the blood, the better the response to anti-PD-1 therapy.

Generating immune profiles enables clinicians to sort patients towards immuno-therapeutic regimens to which they are most likely to respond, thus supporting precision medicine.

Here at MUSC, we hope to use the technology to help clinicians monitor success in their clinical trials or tell them to switch the immunotherapy,” says Krieg. He wants to continue his work in melanoma as well as look at using this technique for other cancers, including head and neck, gastrointestinal and lung cancer.

Not only will this technology allow clinicians to identify the subset of advanced melanoma patients who will respond positively to anti-PD-1 therapy, it may also identify alternative targets for patients who would not respond. Additionally, since this approach uses a simple blood test, samples could be taken throughout the treatment regimen. This would allow clinicians to monitor any changes to patients’ profiles that might require changes to their treatment.

Krieg sees the technology is ushering in a new era in cancer research. Looking back only a couple of decades, chemotherapy was the main course of treatment, he said. Then science advanced to sequence genomes and find mutations in the DNA for which specific inhibitors were developed.

“After DNA and the transcriptome, the exciting thing about this technology is the next step in evolution: we’re now looking at the proteome.”