By Candace Gwaltney
Oct. 9.2025
Dr. Bian talks about GLP-1 and cancer risk studies, plus using big data in research
Using 10 years of electronic health record data, researchers at the IU Simon Comprehensive Cancer Center found that patients taking GLP-1s (Glucagon-like peptide-1 receptor agonists) had a reduced risk for some cancers. Led by Jiang Bian, PhD, the retrospective study found taking GLP-1s was associated with lowered risks for endometrial, ovarian and meningioma cancers. Bian and colleagues published these findings in the journal JAMA Oncology this summer.
Bian is the inaugural Walther and Regenstrief Endowed Chair in Cancer Informatics at the Regenstrief Institute and chief research information officer at the IU Simon Comprehensive Cancer Center. He’s also a member of the cancer center’s Cancer Prevention and Control research program.
Dr. Bian answers questions and his research, role with the cancer center, and the future of big data and AI in cancer research.
Q: Your recent study has gained a lot of attention. In your own words, what is the significance of your GLP-1 research?
A: We have had quite a few of publications in the last two years on GLP-1s, with the latest three focusing on GLP-1s and cancer risks. The other two looked at cancer risks in the diabetes population using Medicare data and survivorship.
GLP-1 is having a ‘metformin moment,’ another drug that was originally approved for one thing but now showing effectiveness across many diseases. The first approval for GLP-1 was in 2005, and the drug has gone through several generations. It helps with obesity, type 2 diabetes, cardiovascular and renal functioning benefits. There has been different reporting on concerns with mental health, so those aspects are being studied. We also published in JAMA Neurology on GLP-1’s protective effects for Alzheimer’s diseases and in the journal Movement Disorders for Parkinson’s disease.
Personally, I’ve been on GLP-1, and I’ve lost about 35 pounds in the last year and half. As a patient, I share the same concerns about possible long-term risks. What are the neurodegenerative disease risks? What are the cancer risks? These questions led us to the retrospective study on cancer risks, using OneFlorida+ electronic health records for more than 80,000 patients.
We have been looking at obesity-related cancer for a while. Ten to 15 years ago, obesity wasn’t even considered a cancer risk. Now, in the last few years, researchers are broadly pointing to obesity as a huge risk factor and the underlying mechanisms are clearer, related to inflammation and cellular information.
That’s why I think it’s important to explore this real-world data to understand cancer risks.
Q: What should the public know about this study and your other work?
A: As a patient, I’d say: be cautious but not pessimistic. Most GLP-1 use is for approved indications—diabetes, obesity, and now obstructive sleep apnea. These go through clinical trials, and the clinical evidence is clear that the benefits outweigh risks for approved uses.
Still, we lack long-term data related to cancer risks and reduction. Cancer and neurodegenerative diseases are aging-related, so we need more research to understand long-term impacts.
Q: What’s next for your cancer research?
A: Our study found protective effects for three cancers and a slight increase in kidney cancer risk. These findings can help guide future randomized controlled trials.
We used a large electronic health records (EHR) dataset—observational, not randomized—for this study. We tried to reduce bias, but it’s not conclusive. Randomized controlled trials (RCT) are ideal but expensive and hard to conduct for long-term cancer outcomes.
Combining real-world data with RCTs is key. Real-world data is generalizable; RCTs are controlled. Both have strengths and weaknesses.
Q: You’re speaking at the cancer center’s Metabolic Research Symposium on Nov. 25, organized by the Cancer, Obesity, Metabolic-Musculoskeletal Pathways: Advancing Scientific Solutions (COMPASS) Working Group. Can you preview your talk?
A: I’ll present all three of our GLP-1 papers. I’ll discuss data sources—structured data like diagnoses and labs, and unstructured data like radiology and pathology reports.
I’ll also talk about using AI to identify subgroups that benefit most and explore mechanisms—whether cancer protection is direct or mediated by improved diabetes or cardiovascular health.
Q: You are the chief research information officer for the IU Simon Comprehensive Cancer Center. Can you tell us more about your role and its efforts?
A: My background is in data infrastructure—making data connected, standardized, and harmonized so that it’s useful for research and clinical operations.
As chief research information officer at the cancer center, I’m working to build a cancer data science program and a cancer clinical informatics shared resource to support investigators. For example, we are looking at how to screen patients for cancer clinical trials, leveraging large language model AI and EHR models. This could allow us to identify and pre-screen patients for trials and then do targeted recruitment.
We aim to avoid redundant or duplicated data efforts across IU Health, IU School of Medicine, Regenstrief, and CTSI by creating shared infrastructure.
Q: How is big data and AI changing cancer research?
A: I’m cautiously optimistic. There’s hype and skepticism. I think AI is going to play a huge role in cancer—not just in research but also in cancer care.
We’ve already made progress—trial matching is a good example. I’m excited looking at the next five to 10 years, looking at how AI is going to evolve. We are narrowing the gap between the hype around AI what concretely AI data science solutions can solve. That’s the exciting part—solving real problems.
Q: Anything else you’d like to add?
A: All this research and these publications are a team effort. I truly believe a scientific problem needs to be sorted out by a multidisciplinary team with equal say in the research process. It’s team science.
