E0382 - Characteristics, Treatment and Outcome of Patients With Bowel Perforation After Immune Checkpoint Inhibitor Exposure

Tuesday, October 25, 2022

3:00 PM – 5:00 PM ET

Location: Crown Ballroom

Has Audio

Presenting Author(s)

Antonio Pizuorno Machado, MD

University of Texas Health Science Center
Houston, TX

Antonio Pizuorno Machado, MD¹, Malek Shatila, MD², Hao Chi Zhang, MD³, Anusha S. Thomas, MD², Yinghong Wang, MD, PhD²
¹University of Texas Health Science Center, Houston, TX; ²MD Anderson Cancer Center, Houston, TX; ³UT MD Anderson Cancer Center, Houston, TX

Introduction: Exposure to immune checkpoint inhibitors (ICI) can predispose to immune-related adverse events (irAE) involving the gastrointestinal tract. The association between ICI and bowel perforation has yet to be elucidated. We aim to describe the clinical course, complications, treatment and outcomes of patients with bowel perforation on ICI therapy.

Methods: This is a retrospective study of adult cancer patients with bowel perforation that occurred between the first dose of ICI treatment and up to 2 years thereafter, at MD Anderson Cancer Center between 1/1/2010 and 4/30/2021. Patients’ clinical courses, imaging, treatment and outcomes related to bowel perforation were collected and analyzed.

Results: Of the 13,991 patients that received ICIs in the study period, 90 (0.6%) inclusion criteria . A majority of our sample were male (54.4%) with melanoma cancer type (23.3%), and PD-1/L1 inhibitor treatment exposure (58.8%). Onset of perforation usually occurred after a median of 4 ICI treatment cycles. The most common symptom was abdominal pain (95.5%) followed by abdominal guarding (20%) and fever (16.6%). The colon was the most common location for the perforation (37.7%) followed by small bowel (30%), and rarely the stomach. Twenty-eight patients (31%) had evidence of of enterocolitis, diverticulitis, appendicitis and 6 (6.6%) patients had luminal cancer involvement at the time of perforation. Overall hospitalization rate related to perforation was 95.5% with mortality of 15.5% on the same admission. 95% of our sample received antibiotics with 38% of patients requiring surgery or a procedure by interventional radiology. Forty-six patients (51.1%) had perforation related complications (such as sepsis, fistula, abscess), which was associated with higher mortality rate of 30%.

Discussion: Bowel perforation after ICI treatment has not been well-studied. Our findings suggest the low incidence of 0.6%, with 40% patients having co-existing bowel abnormalities as the potential contributing factor. Patients with bowel perforation had aggressive disease course, high rate of hospitalization, complications, and mortality. Early recognition and prompt intervention is critical to improve the patients’ outcome. Future studies are still warranted to further investigate the cause, predictive markers and optimal treatment for this patient population.

Figure: Figure 1.
1A, 1B, 1C; CT images with stomach perforation after initiation of ICI
1C, 1D, 1E; CT Images with colonic contained microperforation

Table 1. Perforation-related characteristics of patients treated with ICIs (n=90).
Characteristic	No. of patients (%)
Median overall duration of ICI treatment before perforation, cycles (IQR), n=90	4 (2-7)
History of perforation before ICI	0 (0)
Previous history of bowel disorders	0 (0)
Clinical symptoms
Abdominal pain	86 (95.5)
Guarding	18 (20)
Fever	15 (16.6)
Perforation distribution
Colon	34 (37.7)
Small bowel	27 (30)
Stomach	2 (2.2)
Esophageal	0 (0)
Appendix	11 (12.2)
Potential reasons for perforation
Active inflammation of the gut	28 (31)
Infection of the gut	0 (0)
Cancer involvement	6 (6.6)
Iatrogenic (surgery, or IR or scope)*	2 (2.2)
Co-existing bowel condition at the time of perforation
Diverticulitis	21 (23.3)
Appendicitis	4 (4.3)
Enterocolitis	7 (7.7)
Micro-perforation or contained perforation on imaging	4 (4.3)
Complication of perforation, no. (%)
Fistula	3 (3.3)
Abscess	10 (11.1)
Free air	15 (16.6)
Sepsis	35 (38.8)
Treatment, no. (%)
No treatment	0 (0)
Antibiotics	86 (95.5)
Antibiotics plus surgery	20 (22.2)
Antibiotics plus IR procedure	14 (15.5)
Cancer treatment after ICI termination, no. (%)	19 (21.1)
No further cancer treatment	71 (78.8)
Continued with ICI treatment	11 (57.8)
Switched to other non-ICI treatment	8 (42.1)
Outcomes
Hospitalization, no. (%)	86 (95.5)
Days of hospitalization, median (IQR), n=86	7 (5-15)
Death on the same admission	14 (15.5)
All-cause mortality	71 (78.8)
ICI: immune checkpoint inhibitor, IQR: interquartile range, IR: interventional radiology, mo.: months, SD: standard deviation. Footnote: Treatment option: 4 patients (4.3%) underwent both IR procedure (drainage) and Surgery at a later date.

Table: Table 1: Perforation related characteristics of patients treated with ICIs

Disclosures:

Antonio Pizuorno Machado indicated no relevant financial relationships.

Malek Shatila indicated no relevant financial relationships.

Hao Chi Zhang indicated no relevant financial relationships.

Anusha Thomas indicated no relevant financial relationships.

Yinghong Wang: AzurRx – Consultant. MabQuest – Advisory Committee/Board Member. Sorriso – Consultant. Tillotts – Consultant.

Antonio Pizuorno Machado, MD¹, Malek Shatila, MD², Hao Chi Zhang, MD³, Anusha S. Thomas, MD², Yinghong Wang, MD, PhD². E0382 - Characteristics, Treatment and Outcome of Patients With Bowel Perforation After Immune Checkpoint Inhibitor Exposure, ACG 2022 Annual Scientific Meeting Abstracts. Charlotte, NC: American College of Gastroenterology.