Impact of Thiazides and Fluoropyrimidines Interaction on Myelotoxicity and Other Adverse Events in Real-World Practice: A Retrospective Cohort Study.

[INTRODUCTION] The European Summary of Product Characteristics (SmPC) for 5-FU warns of significant granulocyte decline when combined with thiazides, cyclophosphamide, or methotrexate, based on a 1981 cohort of 14 patients. Despite limited evidence, drug-interaction checkers still flag this risk. We conducted a retrospective cohort study to compare myelotoxicity in patients receiving fluoropyrimidines with or without thiazides. Secondary objectives included assessing clinical relevance and overall safety.

[METHODS] A retrospective, single-center observational study was performed using electronic health records. The cohort included colorectal cancer patients treated with capecitabine or capecitabine-oxaliplatin, with a maximum of one prior chemotherapy line. Myelotoxicity was defined using local laboratory thresholds. Clinical relevance was assessed through predefined adverse events (AEs). All AEs and lab data were collected up to 1 month after chemotherapy completion.

[RESULTS] We included 192 patients (mean age 68.6 ± 13 years; 61.5% male); 37 (19.3%) were on thiazides at treatment start. Baseline characteristics were largely comparable between groups, although thiazide users were older. Median follow-up was 126 days (IQR: 84-158). Hemoglobin declined significantly in thiazide users at 1-3 months (-0.3 vs. +0.11 g/dL; p = 0.006) and > 6 months (-2.63 vs. -0.75 g/dL; p = 0.002). Other hematologic parameters showed no significant differences. Myelotoxicity occurred in 83.8% of thiazide users vs. 75.5% in controls (RR = 1.11; 95% CI: 0.94-1.31; p = 0.280). No differences were seen in febrile neutropenia, mucositis, infections, or other AEs. Logistic regression showed no association between thiazide use and myelotoxicity risk.

[CONCLUSIONS] Concomitant use of fluoropyrimidines and thiazides was not associated with increased myelotoxicity or clinically relevant adverse events. These findings suggest that the clinical impact of this interaction may be lower than currently assumed and should be confirmed in larger multicenter studies.

[TRAIL REGISTRATION] The protocol was submitted and approved at the Spanish Clinical Studies Registry (REEC ID: 0050-2024-OBS) and the EMA-RWD Catalogs (ID: 1000000181).