This study was initiated to assess the quantitative impact of patient anthropometrics and dihydropyrimidine dehydrogenase (DPYD) mutations on the pharmacokinetics (PK) of 5-fluorouracil (5FU) and to explore limited sampling strategies of 5FU.; We included 32 patients with gastrointestinal malignancies, receiving 46-h continuous-infusional 5FU and performed PK-sampling at baseline, 15, 30, 45 min, 1 and 2 h after the start of infusion and at the end of infusion, for 2 subsequent cycles. Plasma concentrations of 5FU, 5-fluorodihydrouracil (5FUH2), uracil (U) and 5,6-dihydrouracil (UH2) were determined using LC-MS/MS and submitted to population PK analysis using nonlinear mixed-effects modeling. Broad genotyping of DPYD was performed, and the potential impact of the DPYD genotype on the elimination of 5FU was assessed. Limited sampling strategies were evaluated for their accuracy to predict steady-state concentrations of 5FU (CSS(5FU)), using data simulations based on the final PK-model.; The area-under-the concentration-time curve of 5FU (AUC(5FU)) was found to be <20 mg h/L in 33 occasions (58 %), between 20 and 30 mg h/L in 17 occasions (30 %) and >30 mg h/L in 7 occasions (12 %). Men had a 26 % higher elimination of 5FU and a 18 % higher apparent elimination of 5FUH2. Accordingly, women had a higher AUC(5FU) compared to men (22 vs. 18 mg h/L, p = 0.04). No DPYD risk variants were found, and the DPYD variants detected (c.496A>G, c.1601G>A, c.1627A>G) were not significantly associated with the elimination of 5FU. Individual baseline UH(2)/U ratio was significantly associated with AUC(5FU) (R = -0.49, p < 0.001). Limited sampling strategies with time-points <3 h after the start of infusion were not adequate to predict CSS(5FU). Female gender was the only predictor of nausea/emesis in the multivariate model.; Gender-specific elimination of 5FU is supported by the present data and may partly explain the gender-specific association between DPYD risk variants and 5FU-specific toxicity.