Mucosal inflammation predicts response to systemic steroids in immune checkpoint inhibitor colitis

BACKGROUND: Immune-related colitis is a common, often serious complication of immune checkpoint inhibition (ICI). Although endoscopy is not strictly recommended for any grade of diarrhea/colitis, emerging evidence suggests that endoscopic evaluation may have important therapeutic implications. In this retrospective study, we sought to comprehensively characterize the clinical and histologic features of ICI-induced colitis with a specific focus on evaluating the prognostic role of endoscopy. METHODS: Data were collected from the medical records of 130 patients with confirmed ICI-induced colitis. In a subset of patients (n=44) with endoscopic and pathologic data, endoscopic data were scored using the Mayo Endoscopic Score (MES) with scores ranging from 0 (no inflammation) to 3 (colonic ulceration). The impact of infliximab on antitumor outcomes was evaluated using progression-free survival (PFS) and overall survival (OS). RESULTS: We identified 130 patients with ICI-induced colitis across two institutions. All patients were treated with corticosteroids. Additional and/or alternative immunosuppression was employed in 59 cases, with 52 patients (42%) requiring at least one infusion of infliximab 5 mg/kg. Endoscopic assessment with biopsy was performed in 123 cases of suspected colitis (95%), with 44 cases available for MES tabulation. Presence of ulceration (MES 3) was associated with use of infliximab (p=0.008) and MES was significantly higher in patients who received infliximab compared with those who did not (p=0.003) with a median score of 2.5; conversely, those with an MES of zero rarely required secondary immunosuppression. Notably, symptoms of colitis based on Common Terminology Criteria for Adverse Events grade had no association with endoscopic findings based on MES classification. After adjustment for baseline patient and disease characteristics, there was no significant difference in steroid duration or cancer-related outcomes in patients treated with infliximab. CONCLUSIONS: In our study, we demonstrate the association of endoscopic features, specifically the MES, with immunosuppressive needs. Importantly, we also show that MES was not related to severity of patient symptoms. The data suggest that endoscopic features can guide clinical decision-making better than patient symptoms, both identifying high-risk patients who will require infliximab and those who are likely to respond to initial corticosteroids.