Comparative outcomes of continuous infusion versus intravenous push followed by infusion 5-fluorouracil in patients with pancreatic and colon cancer: A propensity-matched analysis.

Abstract

e15735 Background: 5-Fluorouracil (5-FU) is a cornerstone of therapy for gastrointestinal malignancies and is commonly administered as continuous infusion or as intravenous (IV) push followed by infusion. Despite widespread use of bolus-containing regimens, real-world comparative data evaluating clinical outcomes between these administration strategies in pancreatic and colon cancer remain limited. Methods: We conducted a retrospective cohort study using the TriNetX US Collaborative Network, including adults (≥18 years) with pancreatic or colon cancer who received injectable 5-FU between January 2000 and January 2026. Patients were categorized into continuous infusion 5-FU versus IV push followed by infusion 5-FU. Capecitabine exposure was excluded. Outcomes assessed from 1 to 365 days after the index event included all-cause mortality, hospitalization, granulocyte colony-stimulating factor (G-CSF) use, blood transfusion, platelet count < 50×10³/µL, and platelet count < 10×10³/µL. Propensity score matching (1:1) was performed to balance demographics, comorbidities, procedures, concomitant therapies, and genomic variables. Time-to-event outcomes were analyzed using Kaplan–Meier methods with log-rank testing, with Cox proportional hazards models used for effect estimation. Results: After propensity score matching, 8,256 patients were included (4,138 per cohort) with well-balanced baseline characteristics. At 1 year, mortality occurred more frequently in the continuous infusion group compared with the IV push followed by infusion group (8.6% vs 7.5%; risk difference 1.1%, p = 0.068). Kaplan–Meier analysis demonstrated a trend toward improved survival with IV push followed by infusion (log-rank p = 0.053). The hazard ratio did not reach statistical significance (HR 1.16, 95% CI 1.00–1.35, p = 0.074), suggesting a modest and potentially time-dependent difference in survival. No statistically significant differences were observed between groups for hospitalization (log-rank p = 0.349), G-CSF use (log-rank p = 0.303), blood transfusion (log-rank p = 0.527), platelet count < 50×10³/µL (log-rank p = 0.286), or platelet count < 10×10³/µL (log-rank p = 0.525). Conclusions: In this large real-world analysis of patients with pancreatic and colon cancer, IV push followed by infusion 5-FU demonstrated comparable safety outcomes and a trend toward improved survival compared with continuous infusion alone. While differences did not reach statistical significance, Kaplan–Meier analyses suggest potential time-dependent survival differences that warrant further investigation. Prospective studies are needed to clarify the optimal 5-FU administration strategy.