Immune related adverse events: Timing, identification patterns, and impact on survival.

Abstract

e14705 Background: Increased use of immune checkpoint inhibitors (ICI) has resulted in a rise in immune related adverse events (irAEs), which many non-oncology clinicians may underrecognize. Controversy exists surrounding whether irAE presence has an impact on survival. Our primary goal was to evaluate the timing between detection to diagnosis of irAE, and determine the department responsible for making the irAE diagnosis. We also assess the incidence of single vs multisystem irAEs, impact of irAE on survival, and assess whether standard irAEs (s-irAEs) those diagnosed within 90 days of initial ICI administration vs delayed irAE (d-irAEs) those diagnosed >90 days impacts survival. Methods: We performed an IRB approved retrospective study of patients 18 years and older who received ICIs for a malignancy at Gundersen Health System between January 1, 2011 and December 31, 2021. Statistical analysis included Chi-square, Fisher’s Exact, and Wilcoxon rank sum tests and Kaplan-Meier survival analysis. All analysis was completed in R version 4.2.3 with a p-value <0.05 considered significant. Results: Our study included 664 patients who were 99.6% white, 60% male, and an average age of 67 years. Median Charlson Comorbidity Index was 10. There were 245 (36.8%) patients with at least 1 irAE, and 62 (9.3%) with a multisystem irAE. While 28 (11%) irAEs were initially detected by non-oncology providers, only 10 (4.1%) irAEs were diagnosed by a non-oncology department. Median time (days) from symptoms to diagnosis was 3 for those diagnosed in non-oncology departments and 0 for those diagnosed in oncology (p<0.01). A total of 142 (58.0%) s-irAEs were detected with grades 1-2-3-4-5 accounting for 21.1%-54.9%-19.7%-1.4%-2.8%. There were 103 (42.0%) d-irAEs with grades 1-2-3-4-5 accounting for 23.3%-53.4%-17.4%-5.8%-0%. Of the d-irAEs, 17 (16.5%) occurred >365 days from initiation of ICI. The 3 most common irAE in this patient population included thyroiditis (34%), dermatitis (21%), and pneumonitis (12%). Our data shows a significant survival difference in patients with an irAE compared to those without (p<.01). No significant survival difference was noted between patients who had s-irAE vs d-irAE (p=.069). Conclusions: Our results show that irAEs may be underrecognized and often not formally diagnosed by non-oncology providers. However, median time to diagnosis is short within this context. This may be due to non-oncology providers feeling uncomfortable making the final decision, which emphasizes the importance for oncologists to finalize irAE diagnoses. We did see a positive impact on survival with irAE presence. While oncologists often clinically expect irAEs to surface within the first 6-10 weeks of treatment, our study demonstrates that 42% of irAEs occur after 12 weeks from ICI initiation. There is no difference in survival between s-irAEs vs d-irAEs, and overall grade distribution between these two groups appears similar.