User-friendly therapeutic drug monitoring (TDM) software tools are a requirement for a wide use of dose individualization strategies in clinical practice. In order to obtain accurate and precise estimates of individual pharmacokinetic parameters, the software should offer the possibility of implementing population pharmacokinetic models.
The purpose of this study was to integrate a validated population pharmacokinetic model of fluorouracil (5-FU) into the TDM software tool MwPharm++ and to evaluate its performance in comparison with the reference software NONMEM for 5-FU dose adjustment in three different infusional treatment regimens. 
The integration of the validated model into MwPharm++ was carried out using the EDSIM++ platform incorporated in the program. Afterwards, AUC estimations obtained with MwPharm++ were compared with those obtained with NONMEM in a previous study using a dataset of 62 patients who received 5-FU-based chemotherapy. The clinical implications of the agreement between both programs were evaluated based on the total exposure of 5-FU as target criterion for pharmacokinetically-guided dose adjustment. 
The model parameter values of volume of distribution (V), clearance (CL), the residual error values and the influence of body surface area (BSA) on CL were successfully introduced. A good agreement between AUC estimates using both software tools was obtained based on median AUC, 5% and 95% percentiles and the Bland-Altman analysis showing a small bias of 0.3 mg•h/L. Differences in AUC values were related to the analytical imprecision of each program and the different algorithms used for the Bayesian estimation of individual parameters. In 81 % of the cases, MwPharm++ and NONMEM resulted in the same dose recommendation. 
In conclusion, the implemented 5-FU population PK model in MwPharm++ represents a valuable tool to guide 5-FU dose adaptation in clinical practice which is much easier to use than NONMEM. Its application in the oncology setting should be considered in future studies.