Characteristics of end-of-life (EOL) chemotherapy (CTx) received by patients with advanced cancers and association with EOL emergency department (ED) and intensive care unit (ICU) care.

e19222 Background: Use of EOL CTx is an established quality metric in patients with advanced malignancies but less is known about which types of CTx are most commonly used and association with ED and ICU utilization. We sought to describe the different types of EOL CTx and to quantify the frequency of EOL ED and ICU care associated with them. Methods: Patients in the cancer registry of an urban cancer center who died between January 1, 2018 and October 10, 2019, and ever received CTx were included. EOL CTx was defined as any CTx given within 30 days of death, while any ED visits or ICU admissions in the last 30 days of life were defined as EOL ED and ICU care, respectively. CTx was categorized by administration route (intravenous (IV), oral (PO), other), and by type (immunotherapy (IMT), non-immunotherapy biologics (NIB), other). We used Pearson’s chi-squared to measure associations between EOL CTx and EOL ED and ICU care, logistic regression to assess how CTx type modulates those associations, and Mood’s median test to compare median IMT doses between groups. Results: Among 390 eligible patients, 32% received EOL CTx, 30% EOL ED care, and 11% EOL ICU care. Most received IV CTx (78%), and 10% received IMT. Median age at diagnosis was 69 years (interquartile range (IQR) 62 - 77), and median days from diagnosis to death was 390 (IQR 185 - 665). Most common malignancies were pancreatobiliary (40%), other gastrointestinal (15%), lung (13%) and hematologic (6%). Patients treated with EOL CTx were significantly more likely to receive IMT (p = 0.03). Receipt of any EOL CTx was significantly associated with EOL ED care (p < 0.0001) and EOL ICU care (p < 0.0001). Subgroup analyses showed significant modulatory effect of IMT on association of EOL CTx with EOL ED care (b = -0.89, p = 0.046), but was not significant for ICU care (b = -0.67, p = 0.29). Median doses of IMT was 2.5 (IQR 2 - 3.8) among patients who were given EOL IMT and 4 doses (IQR 2 - 6) in those that discontinued IMT before EOL (p = 0.06). Conclusions: EOL CTx is associated with significantly increased rates of EOL ED and ICU care, which may indicate poorer quality of life. While rates of use of other CTx modalities did not significantly differ at EOL, patients were more likely to receive IMT within 30 days of death, which could be due to the belief that IMT is more tolerable or more effective than other CTx modalities at EOL. IMT at EOL is associated with a reduced risk of EOL ED care, but not ICU care. Further research on strategies to reduce EOL CTx and appropriateness of IMT at EOL is warranted.