Exosome Profiling Pinpoints Cancer Type

A recent study indicates that analyzing the protein contents of exosomes and related extracellular vesicles can distinguish tumors from noncancerous tissue—and that profiling vesicle proteins in plasma may reveal cancer type (Cell 2020;182:1044–61.E18). The findings suggest that such vesicles could provide the basis for liquid biopsies to detect cancer and monitor treatment response.

Normal cells and cancer cells shed extracellular vesicles packed with proteins, nucleic acids, and other molecules. Researchers have been investigating whether proteins in tumor-associated vesicles could serve as biomarkers for diagnosing cancer. However, whether these vesicles reflect the protein content of tumors and thus can be used in a diagnostic context has remained unclear.

In the study, David Lyden, MD, PhD, of Weill Cornell Medicine in New York, NY, and his team focused on extracellular vesicles and particles (EVP), a category that includes exosomes and similar structures. They performed a proteomic analysis on human EVPs isolated from 426 normal and cancerous samples, including cell lines, tissues, and body fluids such as lymph and plasma. They found that the EVPs contained an average of more than 800 unique proteins.

The researchers then assessed whether EVP proteins are distinctive in cancer by comparing EVPs derived from pancreatic or lung tumors with those from adjacent tissue. In both types of cancer, certain proteins were much more abundant in EVPs than in adjacent tissue. Additionally, in both malignancies, EVPs from plasma carried unique combinations of proteins.

To investigate whether EVP proteins are useful for diagnosis, the scientists analyzed samples from 18 cancer types. They found that a panel of 16 EVP proteins distinguished tumor tissue from nontumor tissue with 90% sensitivity and 94% specificity. Moreover, by profiling EVPs from tumor samples, they could distinguish pancreatic, colorectal, and lung cancers, as well as melanoma. The team also analyzed plasma-derived, cancer-associated EVPs, which they discovered came from tumors, the organs where the tumors originated, distant organs, and dysregulated immune cells. Using plasma EVPs, cancer could be detected with 95% sensitivity and 90% specificity. In addition, by analyzing 30 proteins, the researchers could discriminate among five types of tumors, including mesothelioma and breast cancer.

“We' ve established an exosome protein profile that allows us to diagnose cancers at an early stage,” says Lyden. EVP proteins have advantages over other markers used in liquid biopsies, he adds. For example, circulating tumor DNA is often swamped by nontumor DNA, whereas EVPs from plasma can be enriched and easily isolated from other blood components to reveal potential biomarkers. EVP profiling could also enable doctors to track the progress of therapy and pinpoint the tissue where a cancer originated, Lyden says.

“This paper is a major breakthrough,” says Karoliina Stefanius, PhD, of The University of Texas Southwestern Medical Center in Dallas, who wasn' t connected to the study. “They found bona fide tumor-specific markers” for the vesicles, and thus the study provides an important reference for the field, she says.

Before the study, “nobody had shown that the protein complement of vesicles derived from the plasma was the same as the tumor of vesicle origin,” adds Theresa Whiteside, PhD, of the University of Pittsburgh in Pennsylvania, who also was not involved in the study. Because the researchers could analyze tumor, nontumor, and plasma samples from the same patients, “this paper finally provides validation of this concept,” she says. “This is on the road to making the vesicles a clinically useful liquid biopsy.” –Mitch Leslie

For more news on cancer research, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/CDNews.