GI Side Effects of Immune Checkpoint Inhibitors Linked to Colon Adenoma Risk
FROM ACG 2025
PHOENIX — — potentially compounding their cancer burden to include a risk for colon cancer, new research showed.
“The cancer population is already at a higher baseline adenoma risk, and our findings show that ICI-mediated diarrhea and colitis compounds this,” said first author Tanvi Gupta, MD, Department of Internal Medicine, The University of Texas Health Science Center, Houston, in presenting the findings at American College of Gastroenterology (ACG) 2025 Annual Scientific Meeting.
“The study supports a risk-stratified approach to surveillance colonoscopy perhaps within 1 year of ICI-mediated diarrhea and colitis, or earlier if risk factors present,” she said.
ICIs, such as PD-1 and PD-L1 inhibitors, have emerged as highly effective front-line treatments for various cancers, however, diarrhea and colitis are among their most notorious off-target adverse effects, believed to be caused by excessive inflammatory activity from the therapy.
Interestingly, prolonged courses of colitis of more than 3 months have in fact been associated with improved long-term outcomes and survival related to the cancer being treated.
However, prolonged inflammation of the intestine is not without key risks, including compromised colonic mucosa, and such complications are particularly concerning considering indications that the colitis can persist for years even after ICIs are discontinued, the authors noted.
With a previous small study from the researchers at MD Anderson suggesting a higher risk for adenomas associated with ICI-mediated diarrhea and colitis within 1 year of diagnosis, Gupta and colleagues investigated the effects in a larger retrospective study, enrolling 248 patients treated for cancer at MD Anderson from 2010 to 2025.
All patients had developed ICI-mediated diarrhea and/or colitis, confirmed by colonoscopy within 90 days of onset, and all underwent a subsequent colonoscopy at any time point.
The patients, who had a median age of about 63.1 years, were 57.9% men and 90.3% White individuals. Of the patients, 43.7% had been treated with either PD-1 or PD-L1 inhibitors, whereas 42.5% received combination therapy including CTLA-4 inhibitors.
Patients’ predominant cancer types were melanoma and genitourinary malignancies. About 65% had cancer stage IV.
They were compared with a group of historical control individuals who had been treated with ICIs but did not develop diarrhea/colitis, and who had normal baseline and follow-up colonoscopies.
Overall, 71, or nearly 30% of the patients, developed adenomas on follow-up colonoscopies at least 6 months later, with more than 50% of the adenomas developing rapidly, within 7.5 months of ICI-mediated diarrhea and colitis onset. Rates subsequently declined over the following 6 years.
Of 210 who developed ICI-mediated diarrhea and colitis and had baseline and 1-year colonoscopies, those with baseline polyps had an increased risk for detection of adenomas on follow-up endoscopy compared with those with no baseline polyps (33.9% vs 15.9%; P = .004).
However, compared with the patients who were treated with ICIs but did not develop diarrhea and colitis (n = 31), those who did develop the side effects had a higher risk of developing adenomas, even if they had no prior history of polyps (n = 139; P = .002).
Likewise, among those with active histological inflammation at baseline, the risk for adenoma development was higher among those with baseline adenomas vs no adenomas (32.6% vs 15.8%; P = .014), but the risk was higher even among those with no baseline polyps but who did have active inflammation compared with those with ICI treatment but none of the gastrointestinal (GI) side effects (P = .003).
Of those who did develop adenomas, tubular adenomas were the most common types of adenomas, increasing from a median of 46.4% of polyps per patient at baseline to 87.5% at a follow-up endoscopy.
And the size of adenomas increased, with the rate of those under 5 mm at baseline of 100% dropping to 83.3% at the follow-up colonoscopy.
Overall, the findings underscore the role of ICI-mediated diarrhea and colitis in adenoma development.
“We know the cancer population is at a higher baseline risk of adenomas, and ICI-mediated diarrhea and colitis compounds this,” Gupta said.
Importantly, the findings indicate that “histological inflammation drives mucosal injury and adenoma development, regardless of baseline polyps.”
Findings Raise Questions of Surveillance
Danny Issa, MD, an interventional endoscopist and assistant professor of medicine of David Geffen School of Medicine at UCLA, who co-moderated the study, noted that, while longer-term studies are needed, “it’s an interesting study and eye-opening for the many patients on these medications now — but we do need more data.”
“The key question raised is are these patients at a higher risk for colon cancer? It’s possible, and it’s important for clinicians to keep that in mind,” he told GI & Hepatology News.
Further commenting, session co-moderator Sita S. Chokhavatia, MD, AGAF, a gastroenterologist with Valley Medical Group in Paramus, New Jersey, noted the alarming context of ICIs already being used for an existing cancer in the first place.
“So they have one type of cancer and now these patients may be at a higher risk for getting adenomas linked to colon cancer, and so an important question is how often to follow-up on these patients,” she said.
The study’s senior author, Yinghong Wang, MD, PhD, a professor and director of the Oncology-GI Toxicity program at MD Anderson, noted that at MD Anderson, “we routinely offer the first surveillance colonoscopy 1 year after the index ICI-mediated diarrhea and colitis.”
“The following surveillance interval will be based on the finding of this first surveillance colonoscopy,” she told GI & Hepatology News.
Wang recommended that others treating ICI-mediated diarrhea and colitis should follow suit, “given the higher incidence and rapid development of colonic adenomas during this time frame.”
Gupta, Wang, and Chokhavatia had no disclosures to report. Issa reported relationships with Boston Scientific and Eli Lilly.
A version of this article appeared on Medscape.com.