Real-world insights into immune-related adverse events (irAE) in gastrointestinal (GI) malignancies.
Abstract
545 Background: This retrospective review aimed to examine the incidence and baseline characteristics (BLC) that irAE risk in patients with gastrointestinal (GI) cancers undergoing immune checkpoint inhibitor (ICI) therapy at our institution. Methods: GI cancer patients who received at least one dose ICI-only between 1/1/2017 and 7/31/2021 at the Ohio State University were included in the study. The BLC including hematological and chemistry labs were extracted with immune-related adverse events (irAE) details. Statistical analyses included descriptive statistics, chi-square test, logistic regression, and matched pairs t-tests were performed. Results: Our analysis included 198 patients (14 anal (Al), 65 hepatobiliary (HPB), 46 lower (LGI) and 74 upper (UGI)) with irAEs reported in 37 patients (19%). Median age of irAE-group was 67 years (range: 26 to 88) with 54% males and 86% Caucasians. Most of them (n=36) in had single agent ICI. Pneumonitis was most frequently (n=7) noted followed by dermatitis (n=), colitis (n=5), hypophysitis (n=5), and hepatitis (n=4). Relatively rare ones were hypothyroidism (n=3), nephritis (n=2), encephalitis (n=2), diabetes mellitus (n=1), neutropenia (n=1), and myasthenia gravis (n=1). In irAE-group, 18 (49%) required hospitalization (≥ grade 3) and 5 (14%) had irAE-related deaths. Complete recovery from irAEs was reported in 20 patients (54%), with 13 patients (35%) restarting ICI therapy. The irAE incidence in HPB group was significantly higher than other GI malignancies (HPB vs. LGI vs. Al vs. UGI = 46% vs. 24% vs. 16% vs. 14%, p=0.004). History of asthma (41% vs. 16% (p=0.004) and female gender (26% vs. 15%, p=0.04) increased irAE-risk; presence of liver (9 % 22%, p=0.04) and > 3 lung (8 vs. 21%, p=0.04) metastasis reduced the risk. Compared to non-irAE group, the irAE-group had significantly lower baseline (mean) whole blood cell count (WBC: 6.3 vs. 7.6 K/uL, p=0.02), neutrophil count (NC: 4.2 vs. 6.8 K/uL, p=0.04), and neutrophil-lymphocyte-count (NLR: 5.1 vs. 9.3, p=0.03), and lower albumin (3.7 vs. 3.5 mg/dL, p=0.02). A significant rise in NC (2.1 K/uL, p=0.0008), NLR (by 2.5, p=0.004), WBC (by 2.1 K/uL, p=0.002) from baseline was noted at irAE-incidence. Heterogeneity in the primary tumors and treatments received is a notable limitation of this analysis. Conclusions: Overall, the risk of irAEs in patients with GI cancers is relatively low, and certain BLC can predict their incidence.
Article Details
Journal Info
Journal of Clinical Oncology
Lippincott Williams & Wilkins
Authors (4)
Ashish Manne
The Ohio State University Comprehensive Cancer Center, Columbus, OH
Samuel Paul
The Ohio State University Wexner Medical Center, Columbus, OH
Eric Min
Pannaga Malalur
The Ohio State University, Wexner Medical Center, Columbus, OH