Scientists at the University of Wisconsin–Madison have identified blood-based fingerprints – human protein markers – associated with the pre-cancerous forms of colon cancer that are most likely to develop into disease. They say their findings are a promising start to what could ultimately lead to a new blood test for the cancer.
The lab of biochemistry professor Michael R. Sussman, in collaboration with researchers at the McArdle Laboratory for Cancer Research and UW–Madison School of Medicine and Public Health, used a technique called mass spectroscopy to isolate biomarkers in mouse and rat models of the disease and then test patient blood for the same markers. The researchers presented their newest advancements in the spring of 2019 in a study published in the Proceedings of the National Academy of Sciences.
“The most commonly used technologies such as colonoscopy are highly invasive or utilize stool samples for testing, which may not be appealing to patients,” Sussman says. “Because colon cancer is highly curable if detected early enough, setting up tests for the earliest signs of colon cancer to provide as many early diagnostic options as possible is critical. That is the logic behind what we were trying to do.”
Since nearly all cases of colon cancer are curable if caught early enough, this should make screening tests straightforward. However, this is not the case because colon cancer screening suffers from a paradoxical combination of low compliance rates and what is called ‘over-screening’ with colonoscopies.
While the gold standard for colon cancer screening is a colonoscopy, patients must complete a day-long prep to empty their bowels before undergoing an invasive procedure – factors that contribute to low screening compliance. On the other hand, research from UW–Madison shows many of the polyps found and biopsied during a colonoscopy are regressing or static and not at risk of being harmful. A blood test that can detect if a polyp is growing or cancerous would give a better indication that a colonoscopy is needed.
“It would not be meant to replace a colonoscopy in any way,” says Melanie Ivancic, the study’s lead author who started in the Sussman Lab as a graduate student and was then a postdoc. “But the blood test could serve as a pre-screen to detect polyps that have the greatest propensity to turn into cancer.”
This work in the Sussman Lab was started more than a decade ago by Ed Huttlin while he was a biochemistry graduate student developing quantitative mass spectrometry methods to identify differentially expressed tissue proteins in animal models of colon cancer. After Huttlin graduated, he went on to continue his research using mass spectrometry for understanding basic mammalian cell biology at Harvard University.
Ivancic continued the research in blood serum as part of the Integrated Program in Biochemistry (IPiB) in Sussman’s lab. IPiB is the joint graduate program of the Department of Biochemistry and Department of Biomolecular Chemistry. Over the years, their work found blood protein biomarkers in mice and rats and wanted to look for them in humans.
Collaborations with the School of Medicine and Public Health gave them access to patient blood that was taken from patients with known growing polyps as determined by CT colonography. This procedure, also known as a virtual colonoscopy, requires a similar bowel prep but provides a non-invasive image of the colon with needed follow-up for removal if it reveals an area of concern. Blood was taken both before colonoscopy where the polyp was removed and three weeks after polyp removal to compare pre- and post-polyp biomarker levels.
Left to right: Jennifer Pleiman, Melanie Ivancic, and Michael Sussman.
The analysis of the biomarkers in the Sussman Lab utilized a technique called mass spectrometry with methods previously developed in the lab. Mass spectrometry is a method of finding molecular identities of proteins based on molecular weights. Typically, they analyze molecules or sets of molecules they extract from biological fluids or tissues. In this case it was blood proteins.
The instrument used is known as a triple quadrupole mass spectrometer and performs three steps. It first targets a mass of interest and filters out the ones they are not interested in measuring. The second quadrupole fragments them into even smaller pieces, while the final quadrupole is another filter that isolates exactly what they are interested in measuring and determines its amino acid sequence and thus, molecular identity. These steps allow their method to have both the proper sensitivity and selectivity to get an accurate result. Similar methods are used already in the clinical world for drug analysis, but mass spectrometric analysis of blood proteins in the health industry is still an emerging and new area.
Their results allowed them to quantify 19 protein biomarkers across a cohort of patient samples. Four proteins were able to predict known growing polyps in pre-cancerous patients. Interestingly, the size of the polyp, which was determined by CT colonography, did not influence expression.
“It’s the biochemistry going on in the body causing the polyps to grow that we are detecting in these biomarkers,” she explains. “Very broadly, the biomarkers tended to be circulating blood proteins and a dominant place they are made is the liver, although there are exceptions. We are hypothesizing that there is a whole body response to the presence of the polyp that we are detecting.”
Ivancic adds that in blood after the polyps were removed the levels of several of the biomarkers were returning to normal. This shows the biomarkers were dependent on the presence of the polyp and means they could potentially be used for monitoring recurrence or treatments, but that is still a far-off application, she says.
“This study is the first peek at the possibility that there will be blood markers for a minimally-invasive procedure,” says Bill Dove, professor of oncology and genetics with the McArdle Laboratory for Research and Carbone Cancer Center who was also part of this study. “They do exist.”
Sussman adds that their next steps in the project have yielded promising data from an analysis of 150 patients that shows how the levels of these biomarkers go up as the cancer progresses. A second manuscript describing this work is undergoing peer review.
Although far from commercialization, their findings provide a proof of concept for a future test that could use these biomarkers to determine patients’ need for a colonoscopy. Their ultimate goal is to provide another option to screen for colon cancer and help diagnose the disease when it is most treatable.
“We believe that using our test in combination with other marketed tests, such as the stool DNA tests developed by Exact Sciences in Madison, WI, will provide an orthogonality in which the tests complement each other,” Sussman says. “The reasons you might get a false positive from one kind of test are different from the false positive you might get from a blood test. Being able to have multiple options can be very helpful and increase opportunities for catching colon cancer early.”
The project was supported by grants from the National Institutes of Health, the National Cancer Institute and the Wisconsin Alumni Research Foundation. Sarah Perdue contributed reporting in a previous story on this work.
Author Perry Pickhardt is co-founder of VirtuoCTC, consultant for Bracco and Check-Cap, and shareholder of SHINE, Elucent, and Cellectar. Authors Melanie Ivancic, Perry Pickhardt, Mark Reichelderfer, Michael R. Sussman and Bill Dove are inventors on patent application PCTUS2015065049 submitted by The Wisconsin Alumni Research Foundation that covers the quantitative proteomic analysis, animal model, and patient resource design described in this report.
Read more on Sussman and his work in the UW–Madison Department of Biochemistry: