That his work as a student in the Master of Science in Biomedical Informatics (MScBMI) program would lead to changes in the practice of gastroenterology was not anything Dr. Thomas Lu could have imagined before he started.
For his capstone project, Thomas Lu, MD, a gastroenterologist at Indiana University Health Southern Indiana Physicians, evaluated whether proton pump inhibitors (PPIs) affect the response to infliximab therapy in patients with inflammatory bowel disease (IBD). The two medications have been prescribed together for years.
“Infliximab is very widely used, and we have lots of data available for it, since it was first approved in 1998,” Dr. Lu says, noting it was the first biologic medication ever approved for IBD treatment. “The second medication was the PPI, which we pretty much use like water now. There were a total of 50 million PPI prescriptions filled last year in the United States.”
The conclusion he arrived at was both surprising and significant: The drug combination decreases remission rate, indicating that using the PPI in combination with infliximab is a risk factor.
“Finding that patients with IBD taking PPI were less likely to achieve remission with infliximab therapy is an important result,” Dr. Lu says. “We prescribe this commonly and believe it is safe. Now that we know PPI is a risk factor, we will be very careful about prescribing it, and we will go back and see if there is a clear indication for those who are being prescribed it now.”
Those two classes allowed me to meet my goals,” Dr. Lu says. “I wanted to learn how to analyze big data myself, and now I can do that. It also means I can be the bridge between clinicians and bioinformaticians, which is also very important since most clinicians do not have a bioinformatics background.
Thomas Lu, MD, MScBMI Student
Clinical Skills Meet Python
Dr. Lu first learned about the MScBMI program while he was a gastroenterology fellow at the University of Chicago. After working in a predominantly clinical setting for the first year of his fellowship, his daily activities began to depend more and more on the analysis of data.
“With all the new clinical trial data coming out, a significant part becomes managing and analyzing data,” he says. “I never had a chance to acquire programming skills, so I had to send out that work for bioinformaticians to do. But when I learned about UChicago’s Biomedical Informatics program, I realized I could learn the skills to do the bioinformatics work myself.”
Dr. Lu found the coursework to be a perfect complement to his clinical biology background. Since he was focused in particular on gaining programming skills, he was especially pleased with the courses that taught him the basics of Python and R programming.
“Those two classes allowed me to meet my goals,” Dr. Lu says. “I wanted to learn how to analyze big data myself, and now I can do that. It also means I can be the bridge between clinicians and bioinformaticians, which is also very important since most clinicians do not have a bioinformatics background.”
Historical Data, New Breakthroughs
For his capstone project, Dr. Lu contacted Dr. Atsushi Sakuraba, an assistant professor of medicine in gastroenterology at UChicago, who had sponsored capstone projects for the MScBMI program in the past. In fact, Dr. Lu was able to build on work completed by Erika Lin and Trevor Peters that emerged from their capstone work with Dr. Sakuraba. MScBMI instructor Matthew Dapas also made key contributions.
“They did a lot of back-work, like getting the data set approved and the protocol and dataset loaded,” Dr. Lu says. “My project used different data, but thanks to their work the data set was already there, and I was able to reuse some of their codes too.”
Dr. Lu’s project analyzed patient-level data from five randomized controlled studies. All the patients had moderate-to-severe IBD and were treated with infliximab, while 147 patients were also taking a PPI and 889 were not. While it was known that the use of PPIs early in life leads to an increased risk of developing IBD later on, Dr. Lu’s analysis, which led to the discovery that PPIs affect the outcome of IBD therapy, was not known and therefore a significant contribution.
Dr. Lu further emphasizes that, due to an absence of funding, working with historical data is the only way a study like this can be conducted. “No pharmaceutical company has an incentive to fund a study that looks at a commonly prescribed medication used together with infliximab to see whether it’s harmful or not, since we’re not making new clinical indications. That’s why you can only do this type of study using existing data.”
After writing his capstone paper, Dr. Lu submitted his project to Gut, a premier journal in the field of gastroenterology with a high level of impact on the field. Not only was it accepted, but it went on to become the subject of several news pieces in publications for medical professionals. For Dr. Lu, that put the finishing touches on an MScBMI experience that not only met but exceeded his expectations.
“First, it was the program and the significance of the computer programming exposure it gave me,” he says. “And then it was the assistance we get with the capstone experience. It is just incredible that that led to a high-impact factor publication with a potentially practice-changing effect. I never would have done this without the program.”
The Graham School will not be admitting new students to the Master of Science in Biomedical Informatics (MScBMI) in Autumn 2024. The University will take this opportunity to consider future programming in the Biological Sciences Division (BSD). Please see the BSD website for more information about their offerings.
