NEW YORK (GenomeWeb) – New biomarkers discovered in a proteomics study of children with inflammatory bowel disease can distinguish between two forms of the disease requiring different treatments, according to a new study.
Using panels of proteins identified with mass spectrometry, scientists from the University of Ottawa were able to distinguish between Crohn's disease and ulcerative colitis in a cohort of 99 children who had not yet been treated for any form of IBD. One panel — which included epidermal fatty acid-binding protein, visfatin, UDP-glucose 6-dehydrogenase, mitochondrial leucine-rich PPR motif-containing protein, and inorganic pyrophosphatase — successfully distinguished CD from UC with an accuracy of 80 percent. A second panel, using proteins found in 95 percent of the patients, was able to distinguish IBD patients from healthy controls.
The researchers successfully validated their results with enzyme-linked immunosorbent assays (ELISA) using two proteins each from the IBD-control and UC-CD panels in a subset of nine patients.
Led by first author Amanda Starr, a research associate at the University of Ottawa, and senior author Daniel Figeys, the scientists published their results last month in Gut.
IBD is a condition on the rise all around the world, especially in children and in more economically advanced countries, Figeys told GenomeWeb. While its incidence is approximately one in 100,000 kids, he said that Canada has seen a 10 percent increase in new cases each year.
UC and CD are the major disease subtypes and manifest differently in adults. UC starts at the rectum and advances proximally, whereas CD can occur anywhere, and in patches throughout the intestinal tract.
But UC and CD are both much more aggressive in children than in adults making them appear visually similar and rendering endoscopy, the primary diagnostic method, less informative. Because the treatments for UC and CD are different, it's important to have markers that can differentiate cases, Figeys said. "If you're not sure, you can go a long time with a specific treatment that's not working."
If the right treatment isn't applied, the disease can progress in a devastating manner, delaying development into the teenage years.
While there exist molecular markers, such as C-reactive protein and hemoglobin, "some of the markers currently used in the clinic are not identified in the majority of patient populations," Starr said. Existing biomarkers are often simply linked to inflammation, which cannot be used to tell UC from CD, or even to tell IBD from other inflammatory conditions.
The new study suggests that many of the biomarkers in the UC-CD panel are involved in metabolism, especially of fatty acids. While it's not an entirely new finding to link metabolism to IBD and it's unclear if the biomarkers represent a biological mechanism contributing to the disease or are simply a side effect of it, they seem to work.
To that end, Figeys is pursuing intellectual property based on the biomarkers. He's applied for a provisional patent and filed for a patent application. In addition, Biotagenics, a firm he co-founded, is in the process of licensing the biomarkers from the University of Ottawa.
"The goal is to make these [panels] available as tests," he said, even though there is more validation that needs to be done.
While the study was performed using mass spectrometry, the idea is to first move forward with an ELISA-based test, Starr said. However, mass spec-based testing could also be developed.
In the study, the researchers used a Thermo Fisher Scientific Orbitrap Elite mass spectrometer. "The higher resolution it provides is very important to be able to identify and quantify the changes in the proteins that are present," Figeys said. "We would not do that type of experiment on a lower-resolution mass spectrometer. That would just be a waste of sample." Each tissue sample is approximately the size of the ball on a ballpoint pen.
"Those are very complex samples so you want to have a higher resolution and a system that can do high-throughput analysis as well," he said.
Even more important than high resolution is the robustness of the mass spec system, Figeys said. "You cannot put those samples onto a system that would have a success rate of 70 percent," he said. "You need a 99 percent success rate because the cost of failure is too high."
Starr noted that the study collected 122 biopsies from 100 patients. One patient was excluded from the study due to poor data quality. Some of the CD patient biopsies were non-inflamed.
In addition to diagnosis, the biomarkers showed promise in monitoring disease progression. "There's some very solid correlation with the stage of the disease," Figeys said of some of the biomarkers. He's planning to move forward to study that in a different set of patients.
"If you have good markers, you could use them to see if the treatment is changing aspects of the disease," he said.