This is a pooled analysis of three clinical trial datasets (NCT00384176; NCT00272051; NCT00305188) that were accessed from the Project Data Sphere platform. Kaplan-Meier survival estimates were used to evaluate the impact of antibiotic use on overall and progression-free survival and multivariable Cox regression models were employed to further assess this impact.
A total of 1446 patients were included in the current analysis. These include 108 patients who received antibiotics before the start of chemotherapy, 499 patients who received antibiotics after the start of chemotherapy, and 839 patients who did not receive antibiotics. Using Kaplan-Meier survival estimates, the use of antibiotics prior to the start of chemotherapy was associated with worse progression-free (P = 0.001) and overall survival (P < 0.001). Likewise, when multivariable Cox regression analyses were conducted, prior antibiotic use is associated with worse progression-free (HR for antibiotic use during chemotherapy versus antibiotic use prior to chemotherapy = 0.764; 95% CI 0.604-0.966; P = 0.024) and overall survival (HR for antibiotic use during chemotherapy versus antibiotic use prior to chemotherapy = 0.710; 95% CI 0.537-0.940; P = 0.017).
Antibiotic use before (but not following) the start of 5FU-based chemotherapy is associated with worse progression-free and overall survival among patients with metastatic colorectal cancer.