Cancer-related studies were among the nearly 6,000 abstracts presented at Digestive Disease Week® (DDW) 2025, including research on artificial intelligence (AI) in patient communication, polyp detection, and colonoscopy prep.
Use of AI in Patient Communication
AI outperformed physicians in answering gastrointestinal cancer questions, with oncologists preferring ChatGPT’s responses nearly 80% of the time, according to a new study (Abstract 149).
Researchers analyzed 15 real patient questions posted to Reddit’s AskDocs forum and had 12 board-certified oncologists evaluate the responses from both ChatGPT and verified physicians for quality, empathy, and readability.
The AI-generated answers were rated significantly higher in quality (80% vs 35%) and empathy (82% vs 18%), though they were judged more difficult to read based on the Flesch Reading Ease Score. The findings highlight the potential for large-language models to support patient communication.
Endoscopic Resection of Noncancerous Colorectal Polyps
More patients with noncancerous colorectal polyps are avoiding surgery thanks to the growing use of endoscopic removal techniques, according to a large U.S. study (Abstract 375).
Researchers analyzed data from more than 1 million patients who had polyps detected during colonoscopy between 2015 and 2023, comparing rates of endoscopic vs surgical removal and associated outcomes. They found that the rate of endoscopic resection nearly tripled—rising from 0.26% to 0.67%—while surgical removal declined from 0.45% to 0.35%.
The shift not only reduced the need for invasive surgery but also led to fewer serious complications and a significantly lower 30-day mortality rate.
Role of Fasting in Colonoscopy Prep
A low-fiber meal just 2 hours before starting colonoscopy prep may be safe after all, according to a new randomized controlled trial (Abstract Mo1192). Researchers found that allowing patients to eat before bowel cleansing did not reduce the effectiveness of preparation or negatively impact their experience.
In the study of 525 patients using a split-dose polyethylene glycol regimen, 81.4% of those who ate achieved optimal bowel cleansing—nearly identical to the 83.6% in the fasting control group. Patient comfort and tolerance were also comparable, suggesting greater flexibility in precolonoscopy dietary guidelines could be on the horizon.
