Tolerability of fluoropyrimidines appears to differ by region

Abstract

3514 Background: Limited information is available regarding regional differences in tolerability for fluoropyrimidines. This is an initial assessment of the Roche phase III databases on patients treated for metastatic colorectal cancer (MCRC) and adjuvant colon cancer. Methods: Retrospective multivariate analyses (logistic regression) were performed using pooled data from two identical studies (n=1189) in MCRC comparing capecitabine (X) monotherapy with i.v. 5-FU/LV (Mayo regimen) [Hoff et al. and Van Cutsem et al. JCO 2001], and from an adjuvant study (n=1861) comparing XELOX (X+oxaliplatin) with either the Mayo or Roswell Park 5-FU/LV regimens [Schmoll et al. ASCO 2005]. Treatment-related safety parameters evaluated were: all grade 3/4 AEs, grade 3/4 GI events (diarrhea, nausea, vomiting, stomatitis), and grade 3/4 neutropenia events (febrile neutropenia, neutropenia reported as AE or lab value). Dependent variables for 1st-line MCRC data were: US vs. non-US; and for adjuvant colon cancer: US vs. non-US, US vs. Asia, and RoW (rest of the world) vs. Asia. Factors in the adjusted relative risk model were: age, gender, PS, BSA, BMI, baseline creatinine clearance and treatment (X vs. 5-FU/LV; XELOX vs. Mayo vs. RP). Results: Region was significantly associated with grade 3/4 AEs and grade 3/4 GI events in both settings ( table ). There were no relevant interactions between region and type of fluoropyrimidine treatment. Reporting of grade 3/4 neurosensory toxicity was similar in the US and outside US. Conclusions: Differences in the rates of grade 3/4 toxicity were significantly associated with the region of the world in which the patients were treated. Patients treated in the US experienced the highest rates of grade 3/4 toxicity (see table ). This difference was observed for both X and for 5-FU. The reasons for these differences remain to be elucidated. Acknowledgements: C. Allegra, J. Bertino, J-Y Douillard, B. Gustavsson, G. Milano, M. O’Connell, Y. Rustum, J. Tabernero, J. Fagerberg, F. Gilberg, and F. Sirzen. [Table: see text] [Table: see text]