Search

Purpose Erlotinib (Tarceva®, OSI-774) is a small molecule inhibitor of the epidermal growth factor receptor (EGFR) tyrosine kinase. As high-grade gliomas frequently show amplification, overexpression and/or mutation of EGFR, this drug has been tested in several clinical trials with glioblastoma patients, but unfortunately, with little success. As erlotinib is a known substrate of P-glycoprotein (P-gp) and Breast Cancer Resistance Protein (BCRP) we have investigated the effect of these ABC-transporters on the brain penetration of erlotinib. Study design Erlotinib (50 mg/kg) was given by i.p. administration to wild-type (WT), Mdr1ab-/- (single P-gp knockout), Bcrp1-/- (single Bcrp1 knockout) and Mdr1ab-/-Bcrp1-/- (compound P-gp and Bcrp1 knockout) mice. Drug levels in plasma and tissues were determined by reversed-phase high-performance liquid chromatography. Results Relative to Mdr1ab-/-Bcrp1-/- mice that are deficient for both drug transporters, the area under the concentration time curve in brain tissue (AUC)brain of erlotinib decreased significantly by 1.6-fold in Mdr1ab-/- mice where Bcrp1 is present (49.6 ± 3.95 versus 31.1 ± 1.7, μg/g*h; P < 0.01). In Bcrp1-/- mice, were P-gp is present, a more pronounced 3.8-fold decrease to 13.0 ± 0.70, μg/g*h (P < 0.01) was observed, which is close to the 4.5-fold decrease in the AUCbrain of erlotinib found in WT mice where both drug transporters are present (11.0 ± 1.35, P < 0.01). The plasma clearance of erlotinib was similar in mice deficient for P-gp and/or Bcrp1 compared with wild-type mice. In all other tissues the differences between the genotypes were negligible. Conclusions Both P-gp and Bcrp1 reduce the brain penetration of erlotinib. Although P-gp appears to be the most effective factor limiting the brain penetration of erlotinib, the highest brain accumulation was observed when Bcrp1 was also absent. Strategies to inhibit P-gp/BCRP in patients to improve delivery of (novel molecular-targeted) substrate agents, such as erlotinib, to the brain may be required for treatment of intracranial malignancies.

Summary The anticancer drug paclitaxel is formulated for i.v. administration in a mixture of Cremophor EL and ethanol. Its oral bioavailability is very low due to the action of P-glycoprotein in the gut wall and CYP450 in gut wall and liver. However, proof-of-concept studies using the i.v. formulation diluted in drinking water have demonstrated the feasibility of the oral route as an alternative when given in combination with inhibitors of P-glycoprotein and CYP450. Because of the unacceptable pharmaceutical properties of the drinking solution, a better formulation for oral application is needed. We have evaluated the suitability of various self-micro emulsifying oily formulations (SMEOF’s) of paclitaxel for oral application using wild-type and P-glycoprotein knockout mice and cyclosporin A (CsA) as P-glycoprotein and CYP450 inhibitor. The oral bioavailability of paclitaxel in all SMEOF’s without concomitant CsA was low in wild-type mice, showing that this vehicle does not enhance intestinal uptake by itself. Paclitaxel (10 mg/kg) in SMEOF#3 given with CsA resulted in plasma levels that were comparable to the Cremophor EL-ethanol containing drinking solution plus CsA. Whereas the AUC increased linearly with the oral paclitaxel dose in P-glycoprotein knockout mice, it increased less than proportional in wild-type mice given with CsA. In both strains more unchanged paclitaxel was recovered in the feces at higher doses. This observation most likely reflects more profound precipitation of paclitaxel within the gastro-intestinal tract at higher doses. The resulting absolute reduction in absorption of paclitaxel from the gut was possibly concealed by partial saturation of first-pass metabolism when P-glycoprotein was absent. In conclusion, SMEOF’s maybe a useful vehicle for oral delivery of paclitaxel in combination with CsA, although the physical stability within the gastro-intestinal tract remains a critical issue, especially when applied at higher dose levels.

Imatinib is transported by P-glycoprotein (P-gp) and Breast Cancer Resistance Protein (BCRP), however, the exact impact of these transporters on absorption, distribution, metabolism and excretion (ADME) of imatinib is not fully understood due to incomplete data. We have performed a comprehensive ADME study of imatinib given as single agent or in combination with the well known BCRP/P-gp inhibitors, elacridar and pantoprazole, in wild-type and P-gp and/or BCRP knockout mice. The absence of P-gp and BCRP together resulted in a significantly higher area under the plasma concentration-time curve (AUC) after i.v. administration, whereas the AUC after oral dosing was unaltered. Both elacridar and pantoprazole significantly increased the AUC of orally administered imatinib in wild-type but also in P-gp/BCRP knockout mice. This lower clearance was not due to a (further) reduction in biliary excretion. Fecal excretion was significantly reduced in P-gp and P-gp/BCRP knockout but not in BCRP knockout mice, whereas the brain penetration was significantly higher in P-gp/BCRP knockout mice compared to single P-gp or BCRP knockout or wild-type mice. In conclusion, P-gp and BCRP have only a modest effect on the ADME of imatinib in comparison to metabolic elimination. P-gp is the most prevalent factor for systemic clearance and limiting the brain penetration. The considerable drug-drug interaction observed with elacridar or pantoprazole is only partly mediated by inhibition of P-gp and BCRP and far more by the inhibition of other elimination pathways.

Trabectedin (ET-743, Yondelis™) is a novel anticancer drug currently undergoing phase II and III investigations. There are various and conflicting reports whether trabectedin is a substrate for P-glycoprotein (P-gp), an important factor in drug disposition and multi-drug resistance (MDR). We have now unambiguously shown that trabectedin is a P-gp substrate by investigating vectorial transport over monolayers of LLC-PK1 pig kidney and Madine-Darby Canine kidney (MDCK) cells and the mdr1a and/or MDR1 transfected subclones. We further characterized the cytotoxic effects and cellular accumulation of trabectedin in these cell lines as well as in a panel of other cell lines with high or moderate expression levels of P-gp. Trabectedin displayed the typical MDR phenotype only in highly P-gp expressing cell lines, but not in cell lines with expression levels more closely conforming to clinical samples, suggesting that P-gp will not confer resistance to trabectedin in cancer patients.