UMN NEWS
Interdisciplinary Team Works to Improve Prediction Techniques
Oral cancer is the sixth most common cancer worldwide, and one of the more deadly ones. Its long-term survival rate — 50 percent — hasn’t improved in the past 30 years. A University of Minnesota research team is working on an interdisciplinary project to improve these statistics, using a mix of medical research and computer science.
The research group analyzes saliva and uses it to help identify proteins, or biomarkers, for the early detection of oral cancer. The goal is to identify the proteins that lead to oral cancer and create a method to diagnose the disease in its earliest stages.
The team’s work on a new three-step process for improving this technique was published in the journal Molecular & Cellular Proteomics. This new method can detect three to four times more proteins than other methods, and they are working to improve this even more.
“Survival depends on early diagnoses,” says Wu. “Our study takes a novel approach to improving survival rates that haven’t changed for 30 years.”
The research group is led by Timothy Griffin, an assistant professor in the University’s Department of Biochemistry, Molecular Biology and Biophysics, and includes researchers from medicine, dentistry, computer science and engineering, and the School of Public Health. The National Institutes of Health is funding the four-year study.
As part of the project, John Carlis, a professor in the University’s Department of Computer Science and Engineering, and doctoral student Getiria Onsongo use databases and data modeling techniques to analyze the medical data. They work to find effective ways to pull out and visualize desired information from a seemingly insurmountable volume of information.
Baolin Wu, an assistant professor in the U’s School of Public Health, is responsible for study design. Using complex statistical models, Wu will determine how many proteins — of the thousands found in human saliva — to pinpoint for study. He also analyzes data on how the proteins interact with each other to potentially lead to cancerous cells.
While it’s uncertain how many proteins might prove to be culprits, it is certain that prevention is the key to reducing rates of oral cancer. “It’s actually twice as common and three times as deadly as cervical cancer, ovarian cancer, and melanoma,” says Nelson Rhodus, a University professor working on the project who is also director of the Division of Oral Medicine in the School of Dentistry.
However, studies have shown that when oral cancer is diagnosed in early stages, the survival rate jumps to 80 percent. “Survival depends on early diagnoses,” says Wu. “Our study takes a novel approach to improving survival rates that haven’t changed for 30 years.”
Adapted from an article in Advances, fall 2007, a publication from the School of Public Health.
