How AI May Help Address Overtreatment in Gastrointestinal Cancer Care

Dr. William A. Hall sat down with CURE for a live interview at the 2026 ASCO Gastrointestinal Cancers Symposium to discuss the challenge of overtreatment in oncology and the growing role of artificial intelligence (AI)–driven biomarkers in refining response assessment and personalizing cancer care.

In the interview, Hall, who is a professor and chair of Radiation Oncology and the medical director of Froedtert Radiation Oncology, explained how AI may help some individuals safely avoid unnecessary treatment. He also holds the Bob Uecker Endowed Chair in the Department of Surgery within the School of Graduate Studies at the Medical College of Wisconsin in Milwaukee.

Transcript

Many patients worry about overtreatment. How might AI-driven response assessment and emerging biomarkers help some individuals safely avoid additional chemotherapy or surgery, without compromising the chance for a durable cure?

Overtreatment is something that we really focus on a lot in the cancer space. We do not want patients to be overtreated. It is a huge priority, and it is a really delicate balance that we walk as oncologists: we really want to cure them, and achieving a long-standing, durable cure is a massive priority, but we want to do that with just enough treatment to achieve that cure so that they are not receiving one single bit of extra treatment.

In order to do that optimally, we are really going to need to rely on very sophisticated biomarkers going forward into the future. We have been talking about this for many, many years as oncologists; the concept that precision biomarkers must guide our therapies going forward is not new. However, what I think is very new and very exciting is the ability to analyze, collect, and deploy biomarkers at a faster rate than we have ever seen before, and much of that is enabled by some of the sophisticated computational technologies that artificial intelligence has brought forward.

Many patients are probably familiar with common artificial intelligence tools that they may even use in their daily lives, such as Siri on the iPhone, ChatGPT, Gemini, or other common AI tools. You can see in your own experience that some of these tools can summarize data, provide information, and make typical tasks that would take a long time much faster. To a certain extent, it is going to be similar for AI-based oncologic biomarkers.

Now, we have a much higher standard to set and a much higher bar that we need to meet as oncologists to make sure that these tools are reliable, safe, and, most importantly, are directing us in the right direction for patient care. Broadly speaking, they are going to enable us to further refine the treatment that patients are receiving to make sure that we do not overtreat and that we do not recommend more treatment than is necessary, while simultaneously achieving that cure that we all really want for the patient.

Transcript has been edited for clarity and conciseness.

For more news on cancer updates, research and education, don’t forget to subscribe to CURE®’s newsletters here.