272. HOSPITAL VARIATION IN PALLIATIVE TREATMENT FOR ESOPHAGEAL AND GASTROESOPHAGEAL JUNCTION CANCER: A NATIONWIDE POPULATION-BASED STUDY IN THE NETHERLANDS

Abstract Background Palliative care for patients with esophageal (EC) includes a large variety of treatments and data on variation of palliative care across centers are unknown. This study aims to evaluate treatment variation at the hospital level for EC and GC patients in the Netherlands. Methods A nationwide retrospective cohort study was conducted using population-based data from the Netherlands Cancer Registry. Patients diagnosed with inoperable advanced (cT4B) or synchronous metastatic (cM1) EC between 2015-2022 were included. Probability of receiving potentially life-prolonging treatment (palliative resection, chemoradiotherapy or systemic treatment) according to hospital of diagnosis was analyzed using multilevel logistic regression analysis. Hospitals were categorized in tertiles according to this probability (high, medium, low) and overall survival (OS) was assessed using cox regression analysis. Results 8,580 EC patients were included. Potentially life-prolonging treatment ranged from 23.5%-74.0% across hospitals of diagnosis (p<0.0001). After adjustment for patient and tumor characteristics, the probability of receiving these treatments was associated with hospital of diagnosis (p<0.0001). 13 of 79 hospitals demonstrated a significantly lower probability of giving potentially life-prolonging treatment. Patients diagnosed in hospitals with low probability of potentially life-prolonging treatment with had significantly worse OS than medium- (HR 0.85 (0.80-0.90) p<0.0001) or high-probability hospitals (HR 0.77 (0.73-0.82) p<0.0001). Conclusion This study shows substantial variation in palliative treatment for EC and GC care across Dutch hospitals, emphasizing the challenge of defining optimal treatment plans. Hospital of diagnosis was significantly associated with the probability of receiving potentially life-prolonging treatment and also with overall survival. These results suggest that appropriate selection of patients for treatment with potentially life-prolonging treatment leads to improved OS.