S3128 Upper Gastrointestinal Immune-Mediated Toxicity: ICI Management Strategy and Factors Associated With Unfavorable Outcomes

Introduction: The widespread use of immune checkpoint inhibitors (ICIs) as cancer therapy has led to an increase in upper gastrointestinal (GI) adverse events, notably gastritis, enteritis, esophagitis, and overlapping presentations. Balancing the efficacy of the ICI and the management of its toxicities is a major challenge for health care providers, who must determine whether to continue ICIs, adjust dosage, or discontinue the treatment. Furthermore, the identification of early predictors of unfavorable outcomes, notably recurrence or persistence of disease, is critical to guide risk stratification, personalize follow-up, and management. Methods: We retrospectively analyzed 529 patients with upper GI immune-related adverse events (irAEs). Patients were grouped based on ICI status after the index toxicity event: ICI continued or resumed after the episode and ICI discontinued. Clinical outcomes included remission and unfavorable outcomes (recurrence or persistence of symptoms). Groups were compared using Chi-square tests and logistic regression. Results: Of 529 patients, 216 (40.8%) continued or resumed ICI therapy and 313 (59.2%) had ICI discontinued after the index event. Recurrence occurred more frequently in the continued group (11.1% vs 5.4%; P = 0.041), while persistence of toxicity findings until last follow-up was comparable (4.6% vs 3.5%). There was a trend of ICI discontinuation among patients in an inpatient setting, suggesting that continuation or resumption of ICI may have been selected for outpatient cases. Factors associated with increased risk of unfavorable outcomes included overlapping symptom presentation (odds ratio [OR] 3.1; 95% confidence interval [CI] 1.63-5.83), acute inflammation on histology (OR 1.82; 95% CI 1.03-3.21), steroid use (OR 2.24; 95% CI 1.31-3.82), selective immunotherapy (SIT) use (OR 1.98; 95% CI 1.07-3.64), and ICI being continued after the index event (OR 1.90; 95% CI 1.12-3.24). These factors remained significant in univariate analysis and can inform future risk stratification models. Conclusion: Post-toxicity ICI continuation after upper GI irAEs was associated with a higher rate of recurrence compared to ICI being discontinued. Patients with overlapping upper GI phenotypes, acute inflammation on histology, or those who required steroids or SIT in the first event are at higher risk for recurrence or symptom persistence. In addition to early identification of high-risk patients, clinical decision-making should balance oncologic benefits with toxicity severity to individualize care.