Purpose: P-glycoprotein (P-gp; ABCB1) efficiently transports lipophilic amphipathic drugs, including the widely used anticancer drug paclitaxel (Taxol). We found previously that human multidrug resistance protein 2 (MRP2; ABCC2) also transports paclitaxel in vitro, and although we expected that paclitaxel pharmacokinetics would be dominated by P-gp, the effect of Mrp2 was tested in vivo. Experimental Design: We generated and characterized Mdr1a/1b/Mrp2 À/À mice, allowing assessment of the distinct roles of Mrp2 and Mdr1a/1b P-gp in paclitaxel pharmacokinetics. Results: Surprisingly, the effect of Mrp2 on i.v. administration of paclitaxel was as great as that of P-gp. The area under plasma concentration-time curve (AUC) i.v. in both Mrp2À/À and Mdr1a/ 1b À/À mice was 1.3-fold higher than in wild-type mice, and in Mdr1a/1b/Mrp2 À/À mice, a 1.7-fold increase was found. In spite of this similar effect, Mrp2 and P-gp had mostly complementary functions in paclitaxel elimination. Mrp2 dominated the hepatobiliary excretion, which was reduced by 80% in Mrp2 À/À mice. In contrast, P-gp dominated the direct intestinal excretion, with a minor role for Mrp2. The AUC oral of paclitaxel was 8.5-fold increased by Mdr1a/1b deficiency but not affected by Mrp2 deficiency. However, in the absence of Mdr1a/1b P-gp, additional Mrp2 deficiency increased the AUC oral another 1.7-fold. Conclusions:Thus far, Mrp2 was thought to mainly affect organic anionic drugs in vivo. Our data show that Mrp2 can also be a major determinant of the pharmacokinetic behavior of highly lipophilic anticancer drugs, even in the presence of other efficient transporters.Variation in MRP2 activity might thus directly affect the effective exposure to paclitaxel, on i.v. administration, but also on oral administration, especially when P-gp activity is inhibited.