NEW YORK – A team led by researchers from Cedars-Sinai Medical Center has developed a test to determine a patient's prostate cancer status using extracellular vesicles (EVs).
The assay relies on a combination of a microfluidic device, gene expression, and digital PCR to isolate and purify the EVs and detect gene signatures that indicate whether prostate cancer is localized or has metastasized to other regions of the body. Its development and application to 40 patient samples was explained in a paper published last month in Nanotoday.
The assay, dubbed the EV Digital Scoring Assay, builds on previous work done by members of the team using aspects of the technology to detect circulating tumor cells, but Edwin Posadas, medical director of Cedars-Sinai's urologic oncology program and an author of the paper, said that difficulties purifying and preserving the circulating tumor cells led the researchers to focus their efforts on extracellular vesicles. EVs "contain tons of information," such as DNA, miRNA, and other proteins, that can be detected and used for diagnosis, he said, and are also found in higher volumes in blood than circulating tumor cells.
The team, which also includes researchers from the University of California, Los Angeles and the Chinese Academy of Sciences, modified the original platform they had been using to detect circulating tumor cells to work with EVs, providing a purer signal and cutting out some of the extra noise that can be found in plasma, Posadas said.
Us of the EV Digital Scoring Assay involves several steps, he noted. Starting by incubating the plasma with capture agent antibodies, the sample is then run through a microfluidic chip, called the EV Click Chip, with a nano-engineered surface that contains multiple wires with a second capture substance to separate out the EVs associated with prostate cancer. The EVs in the plasma then bond with the second substance and are stuck to the surface of a chip, after which a detergent is applied that releases the vesicles into a droplet digital PCR system. An 11-gene panel developed by the researchers then quantifies the genes of interest and provides a numerical score that indicates whether the cancer is localized or metastatic. The test takes about four hours to run.
The panel was developed by looking at previous research from other groups, such as the Human Protein Atlas, showing which genes are upregulated in prostate cancer. Of the 120 genes originally selected, a set of eight were chosen that have a high expression in prostate compared to other tissues and an additional six genes from a separate gene set associated with metastatic prostate cancer were chosen. Three of those 14 genes from the two sets overlapped, resulting in the 11-gene panel.
The researchers noted in the paper that the majority of EVs circulating in the blood of cancer patients originate from immune cells rather than prostate cancer cells, so they chose genes that are expressed at low levels in immune cells to minimize background noise.
One of the benefits of the test is the ease of using blood and plasma samples compared to tissue biopsies — a main benefit of liquid biopsy tests in general, according to Posadas. Blood samples can also be collected noninvasively multiple times to measure therapy progression, while multiple tissue biopsies are burdensome to patients, he said.
In the Nanotoday paper, the team applied its test to 40 patients that had either localized or metastatic prostate cancer and found that the numerical scores calculated by the test were higher in metastatic samples, with the score able to distinguish metastatic cancer from local cancer with an area under the receiver operating curve of 0.88 at the optimal cutoff.
A use case of the test that Posadas cited is for patients that have undergone surgery but still have elevated levels of prostate-specific antigen in their blood, meaning there is still some cancer remaining. The researchers can "detect the footprint of cancer," even when there is no prostate gland left, and get a sense of whether the cancer has strayed from the prostate bed and spread to the rest of the body, Posadas said. If the cancer has spread, that affects the type of treatment a clinician could use — instead of harsh radiation therapy focused on the prostate area that risks harming the bladder or rectum and won't eliminate the cancer outside of that area, a patient with metastatic cancer may undergo a different type of treatment focused on the rest of the body, he said.
The researchers were also able to show in the Nanotoday paper that the test detected metastasis that was missed by traditional imaging in three patients.
Determining metastasis isn't the only application for the technology, Posadas noted. As the prostate cancer field moves more toward biomarker-specific therapies, the gene panel could be adapted to look for resistance genes to predict a patient's response to therapy. The gene panel is flexible and could be expanded with more genes or swapped out for different genes, depending on the clinical question being asked, Posadas added. Right now, the team is looking at gene signatures for androgen sensitivity that could be used as predictors of therapy response.
Previous work done by the team and published in a 2020 Nature Communications paper applied the test method to hepatocellular carcinoma, and Posadas said that the test could be further applied to other types of cancer — specifically breast, lung, and pancreatic cancer — as future options. It will take refinement to adapt the gene panel to other cancers, since the genes currently being used are disease- and situation-specific, but the same informatics pipeline can be used to develop the new panels, he said.
Hon Leong, a translational urology researcher at Sunnybrook Hospital in Toronto who has worked with EVs in prostate cancer, said via email that distinguishing localized and metastasized prostate cancer is challenging when using cell-free DNA "because most tumors in the localized setting won't be expected to release significant amounts of cfDNA into the bloodstream." In contrast, the EVs Posada's team is detecting are truly from the prostate or prostate cancer tumor, he said.
Although it's too early to say what the clinical utility of the test would be, "the method and technology used to enumerate prostate cancer EVs is very promising," Leong noted, adding the method is particularly unique because some of the protein biomarkers the researchers are looking at have never been shown to be present in EVs before. "Their ability to detect these proteins is impressive, especially when considering how small EVs are," he said.
Leong and his team are also working on similar tests looking at different proteins, such as STEAP1 and polysialic acid, that are found on prostate cancer EVs, he noted.
With commercialization of the EV Digital Scoring Assay still a step or two away, the team's focus right now is on working with academic partners and further amassing data on the assay. Posadas said that the test is being used by the NRG Oncology cooperative group and being incorporated in large clinical trials, which will demonstrate its value and build confidence in its efficacy.
"The technology is great, but until we run patient samples that would show it where the rubber hits the road, what we have is a lot of potential," he said.
The researchers particularly want to work with banked plasma samples, although there are some issues when using banked or archival samples compared to fresh ones, he noted. Namely, differences in plasma preparation, such as the speed at which the samples were centrifuged or how they were stored, can affect the signal-to-noise ratio and the test's readout. That's not a problem in prospective samples, since the researchers can uniformly prepare the plasma to eliminate variability. Unlike archival samples, however, certain clinical endpoints — such as mortality risk — are missing from prospective samples.
By using plasma samples, the test is also able to be more broadly used with patients in rural settings or other geographic regions where testing a tissue sample may be more difficult. EVs are also able to be transferred to large medical centers like Cedars-Sinai for testing more easily than circulating tumor cells or other sample types because they stay more intact and are able to be preserved, Posadas said.
As a result, the team sees this test as a potential "bridge to disseminating genomics … into places it hasn't been able to go," he said.