P067 Immune Checkpoint Inhibitor Colitis in a Community-Based Hospital System

Immune checkpoint inhibitors (ICIs) have significantly changed the oncologic treatment landscape for a wide array of cancers over the past few decades. While these medications can produce a durable remission for patients they also come with a risk of inflammatory toxicities. With the widespread use of immune checkpoint inhibitors there have been increased reports of immune-related adverse events (irAEs) including severe gastrointestinal toxicities such as colitis. While multiple studies have assessed ICI colitis in tertiary care settings little is reported on the characteristics and treatment outcomes of ICI colitis in a community-based population. Additionally, the utility of gastroenterologists in the diagnosis and management of ICI colitis is poorly defined. The aim of this study is to report the characteristics and treatment outcomes of ICI colitis in a community-based hospital system.

This is a single-center retrospective case-series of patients with ICI colitis. Charts were reviewed from patients in the Prisma Health Upstate Network from 2/1/2016 to 12/31/2020. Potential cases of ICI colitis were identified using a diagnosis code of "toxic colitis", "indeterminate colitis", "unspecified colitis", and "noninfective gastroenteritis and colitis" (ICD-10 K52.1, K 52.3, K 52.89 and K52.9) as well as a prior infusion (within 12 months) of pembrolizumab, nivolumab, cemiplimab, atezolizumab, avelumab, durvalumab, or ipilimumab. Charts were abstracted for demographics, tumor type, checkpoint inhibitor used, mode of colitis diagnosis, endoscopic severity of inflammation, and treatments used. Study was approved by the Institutional Review Board at Prisma Health Upstate.

The mean (±SD) age for the 37 patients was 64.1 ± 13.2 years. The patient population was predominantly male (62%) and Caucasian (100%). The cancer type in the patient population were melanoma (43.2%), non-small cell lung cancer (18.9%), renal cell carcinoma (10.8%), and other (27%). Patients with ICI colitis were most-commonly treated with ipilimumab (35.1%), nivolumab (35%), pembrolizumab (10.8%), and other (13.5%). The mean time to colitis onset was 25.7 ± 29.7 weeks. Over 56% of the patient population reported abdominal pain and 45.9% were hospitalized. Only 64.8% were evaluated by a gastroenterologist. Cross sectional abdominal imaging by computed tomography (CT) was obtained in 20 (54.0%) and showed colitis in 13. 54% obtained a colonoscopy with inflammation severity stratified into normal (25.0%), mild (40.0%), moderate (20.0%) and severe (15.0%). Endoscopic extent showed proctitis (0%), left-sided disease (5.0%), right-sided disease (5.0%), pancolitis (55.0%), and isolated ileal disease (25%). Histology showed 90.0% with active inflammation and 10.0% with microscopic colitis. Outcomes were resolution with steroids (64.8%), antibiotics (5.4%), biologic (10.8%), surgery (5.4%), hospice (8.1%). Of those that received biologic therapy, 6 received infliximab and 1 received vedolizumab.

As immune checkpoint inhibitors gain approval in more cancer treatment ICI-colitis is being seen in a community-based setting. The most common treatment were steroids and often required a prolonged course. Biologic use in our series (18.9%) was much lower than prior reports from academic tertiary referral centers (>50%). Only 64.8% of patients saw a gastroenterologist for ICI-colitis symptoms. Early gastroenterology referral in the community setting could help to identify more severe disease and patients necessitating escalation to biologic therapy.