Document resource
Introduction Immune checkpoint inhibitors (ICIs) have transformed cancer outcomes but frequently cause ICI-induced colitis (ICI-c), resembling IBD. ICI-c incurs significant morbidity and premature discontinuation of ICI therapy. The rapid expansion of ICI therapy has made ICI-colitis a significant clinical challenge. Treatment often requires prolonged systemic immunosuppression yet data on infection risk and clinical impact remain limited. Emerging evidence suggests a role for topical corticosteroids in milder disease.Methods We retrospectively analysed consecutive patients with endoscopically confirmed ICI-c at 2 UK centres. Clinical data were extracted from electronic records. The primary outcome was infection occurring after initiation of immunosuppressive therapy and within three months of treatment completion. Associations with treatment exposures, clinical outcomes and cancer outcomes were analysed using chi-squared testing and multivariable logistic regression.Results Among 259 patients (53% male, median age 63), cancers included melanoma (43%), renal (19%), lung (16%) and others. ICI regimens included anti-PD-1/L1 monotherapy (38%), dual anti-CTLA-4/PD-1 (44%) and others. Endoscopy showed a Mayo Endoscopic Score of 0 in 33% (n=86), 1 in 47.5% (n=123), 2 in 12% (n=31), and 3 in 7.3% (n=19). Clinical remission rates were 45.5% (n=108) of those receiving systemic corticosteroids (cs), 57.5% (n=61) with infliximab (IFX) and 83% (n=45) receiving topical steroids.No infections occurred in the 9 patients who did not receive systemic immunosuppression. Among the remainder 250, 63 (25.2%) developed at least one infection requiring ≥24 hours of antibiotics (78 infections total). Infections included respiratory (34.6%, n=27), genitourinary (25.6%, n=20), unknown source (14.1%, n=11), skin/soft tissue/MSK (14.1%, n=11), gastrointestinal (6.4%, n=5), and others.Infection was not associated with age, ICI regimen, or endoscopic severity. Infection occurred in 15.4% of all patients treated with cs alone, after a median of 44 days, and in 17.5% of all patients treated with cs and IFX after a median of 33 days. Exposure to at least one dose of intravenous methylprednisolone nearly doubled infection risk (OR 1.9, 95% CI 1.03–3.32, p=0.04).Infection was associated with hospital admission for ICI-c (OR 4.25, 95% CI 2.10–8.14, p<0.0001) with the majority of admissions (57%) preceding infection. Two infection-related deaths occurred. Infection was not independently associated with overall or progression-free survival.Conclusions Infections occur in one quarter of ICI-c patients receiving systemic immunosuppression, frequently requiring antibiotic therapy, and are strongly associated with hospitalisation. Judicious use of systemic immunosuppression is warranted, with greater consideration of topical steroids in selected patients.