Impact of immunochemotherapy administration sequence on overall survival in advanced esophageal and gastric cancers: a propensity score-matched multicenter analysis

BACKGROUND: The optimal sequence of administering immune checkpoint inhibitors (ICIs) and chemotherapy in advanced esophageal cancer (EC) and gastric cancer (GC) remains to be established. Understanding how this sequencing affects overall survival (OS) is crucial for optimizing treatment strategies. METHODS: This multicenter study enrolled 334 patients diagnosed with advanced EC or GC between January 2018 and October 2024. To reduce confounding biases, propensity score matching (PSM) was applied, resulting in 292 matched patients categorized into two groups, including those receiving ICIs first (IF) and those receiving chemotherapy first (CF). RESULTS: Following PSM, Kaplan-Meier analysis showed that the CF group had a significantly prolonged median OS compared to the IF group (16.80 vs. 14.60 months; p = 0.014). These findings were further corroborated in the unmatched cohort (16.60 vs. 14.60 months; p = 0.012). Multivariable analysis identified ECOG performance status, PD-L1 expression, antibiotic use, and line of ICI treatment as independent prognostic factors, whereas the immunochemotherapy administration sequence did not retain independent prognostic significance. Subgroup analysis revealed that among patients receiving first-line immunochemotherapy, those in the CF group had significantly longer OS (16.50 vs. 12.30 months; p = 0.004), with no significant difference observed for those beyond first-line treatment (p = 0.862). The objective response rate and disease control rate were comparable between the two groups. CONCLUSIONS: This study suggests that administering chemotherapy prior to immunotherapy appears to improve OS in patients with advanced EC and GC, particularly when initiated as first-line treatment. These findings highlight the need for further research into biomarkers and personalized treatment strategies to enhance patient outcomes.