Real-world treatment patterns and outcomes for patients with neoadjuvant treatment between 2019-2022 for gastric or gastroesophageal junction cancer (GC/GEJC) in the US community setting.

345 Background: Treatment options for patients with early-stage GC/GEJC are still limited and the prognosis remains poor, despite recent therapy approvals. With the latest adoption of FLOT use in the US, it is important to understand patients’ characteristics, treatment patterns and outcomes in the US community oncology setting. Methods: This retrospective observational cohort study used electronic health records data from 278 adult patients with stage II-IVA GC/GEJC adenocarcinoma initiating neoadjuvant treatment (index) within The US Oncology Network between 1/1/2019 and 12/31/2022, based on Food and Drug Administration approval of FLOT in 2019. Patients were followed through 9/30/2023, last patient record or death, whichever occurred first. Descriptive analyses were conducted to evaluate patient and treatment characteristics. Kaplan-Meier was used to assess overall survival (OS) and event-free survival (EFS) from index. Adjusted Cox proportional hazard models were used to analyze the association of surgery as a time-dependent variable with OS and EFS. Results: Median (interquartile range [IQR]) age at index was 67.5 (58.2, 74.2) years; most patients were male (68.0%), and half were Caucasian (51.1%). Stage III was the most common initial stage (50.0%), followed by IIB (29.1%), and II (not otherwise specified)/IIA (19.8%). Most patients had an ECOG (0-1) 65.5%. Median (IQR) follow-up from index was 15.6 (8.6, 27.8) months. The most common neoadjuvant regimens were FLOT (51.1%), doublet (28.4%), and CROSS (18.0%). Median OS (months) (95% confidence interval [CI]) for patients with surgery was not reached (NR) (47.7, NR) and those without was 16.1 (8.6, 35.1). Patients with surgery had a longer median EFS (95% CI) (months) than those without 38.3 (25.0, NR) vs 5.6 (4.5, 6.5) (log rank p-value <0.0001). Adjusted hazard ratios (95% CI) for OS and EFS indicated that patients who received surgery were less likely to experience an event than those without, 0.252 (0.139, 0.456) and 0.268 (0.169, 0.426), respectively (p<0.0001). Conclusions: Study findings showed a high proportion of patients receiving FLOT since its approval in US, with a directional OS benefit similar to the FLOT4 trial. Also, receipt of surgery was a strong predictor for OS and EFS. While the results are promising, most patients who did not receive surgery progressed within 6 months and died a little over a year from index. However, given the small portion of patients who did not receive surgery, further research is needed to understand if there is an unmet need for this group. Variable Doublet CROSS FLOT N 79 50 142 Surgery Rate (%) 60.8% 62.0% 71.8% GC Rate (%) 79.8% <20% 85.9% OS, median (95% CI), months 26.0 (20.9, NR) 37.7 (27.8, NR) 52.1 (47.7, NR) EFS, median (95% CI), months 10.8 (8.4, 23.2) 20.9 (11.3, 26.3) 37.7 (18.0, NR)