Heleen A. Bardelmeijer

Modulation of oral bioavailability of anticancer drugs: from mouse to man

Oral bioavailability of many anticancer drugs is poor and highly variable. This is a major impediment to the development of new generation drugs in oncology, particularly those requiring a chronic treatment schedule, a.o. the farnesyltransferase inhibitors. Limited bioavailability is mainly due to: (1) cytochrome P450 (CYP) activity in gut wall and liver, and (2) drug transporters, such as P-gp in gut wall and liver. Shared substrate drugs are affected by the combined activity of these systems. Available preclinical in vitro and in vivo models are in many cases only poorly predictive for oral drug uptake in patients because of a.o. interspecies differences in CYP drug metabolism and intestinal drug-transporting systems. Clearly, novel systems that allow reliable translation of preclinical results to the clinic are strongly needed. Our previous work, also using P-gp knockout (KO) mice, already showed that P-gp has a major effect on the oral bioavailability of several drugs and that blockers of P-gp can drastically improve oral bioavailability of paclitaxel and other drugs in mice and humans (Schinkel et al., Cell 77 (1994) 491; Sparreboom et al., Proc. Natl. Acad, Sci. USA 94 (1997) 2031; Meerum Terwogt et al. Lancet 352 (1998) 285). This work revealed, however, that apart from P-gp other drug-transporting systems and CYP effects also determine overall oral drug uptake. The taxanes paclitaxel and docetaxel are considered excellent substrate drugs to test the concept that by inhibition of P-gp in the gut wall and CYP activity in gut wall and/or liver low oral bioavailability can be increased substantially. In current studies we focus on the development of chronic oral treatment schedules with these drugs and on other drug transport systems that may play a significant role in regulation of oral bioavailability of other classes of (anti-cancer) drugs. The current review paper describes the background and summarizes our recent results of modulation of oral bioavailability of poorly available drugs, focused on drug transport systems and CYP in gut wall and liver.

Efficacy of novel P-glycoprotein inhibitors to increase the oral uptake of paclitaxel in mice

P-glycoprotein inhibitors can increase the oral bioavailability of paclitaxel. We have now explored the mechanisms that determine the efficacy of several novel P-glycoprotein inhibitors to increase the absorption of paclitaxel from the gut lumen of mice in both in vivo and in vitro experiments. The inhibitors studied were cyclosporin A, PSC 833, GF120918, LY335979 and R101933. Mass balance studies showed that GF120918 was the most effective inhibitor, resulting in almost complete uptake of paclitaxel. PSC 833 was slightly less effective, whereas cyclosporin A and LY335979 were moderately effective. R101933 had only marginal effects. These findings were in line with in vitro transport experiments using LLC-mdr1a cells. By studying the intra-intestinal kinetics of the agents we found that cyclosporin A, PSC 833 and GF120918 rapidly passed the stomach and traveled concurrently with paclitaxel through the intestines, whereas LY335979 and R101933 delayed stomach emptying. Moreover, these latter compounds appear to be more readily absorbed when released into the intestines thus reducing local intestinal concentrations. Due to their combined effects on absorption and metabolic elimination of paclitaxel, cyclosporin A and PSC 833 resulted in the highest paclitaxel levels in plasma. In conclusion, our models provide insight into the factors that determine the suitability of P-glycoprotein inhibitors to enable oral paclitaxel therapy and will be useful in selecting candidate inhibitors for clinical testing.

The co-solvent Cremophor EL limits absorption of orally administered paclitaxel in cancer patients

The purpose of this study was to investigate the effect of the co-solvents Cremophor EL and polysorbate 80 on the absorption of orally administered paclitaxel. 6 patients received in a randomized setting, one week apart oral paclitaxel 60 mg m–2 dissolved in polysorbate 80 or Cremophor EL. For 3 patients the amount of Cremophor EL was 5 ml m–2, for the other three 15 ml m–2. Prior to paclitaxel administration patients received 15 mg kg–1 oral cyclosporin A to enhance the oral absorption of the drug. Paclitaxel formulated in polysorbate 80 resulted in a significant increase in the maximal concentration (Cmax) and area under the concentration–time curve (AUC) of paclitaxel in comparison with the Cremophor EL formulations (P = 0.046 for both parameters). When formulated in Cremophor EL 15 ml m–2, paclitaxel Cmax and AUC values were 0.10 ± 0.06 μM and 1.29 ± 0.99 μM h–1, respectively, whereas these values were 0.31 ± 0.06 μM and 2.61 ± 1.54 μM h–1, respectively, when formulated in polysorbate 80. Faecal data revealed a decrease in excretion of unchanged paclitaxel for the polysorbate 80 formulation compared to the Cremophor EL formulations. The amount of paclitaxel excreted in faeces was significantly correlated with the amount of Cremophor EL excreted in faeces (P = 0.019). When formulated in Cremophor EL 15 ml m–2, paclitaxel excretion in faeces was 38.8 ± 13.0% of the administered dose, whereas this value was 18.3 ±15.5% for the polysorbate 80 formulation. The results show that the co-solvent Cremophor EL is an important factor limiting the absorption of orally administered paclitaxel from the intestinal lumen. They highlight the need for designing a better drug formulation in order to increase the usefulness of the oral route of paclitaxel

The Oral Route for the Administration of Cytotoxic Drugs: Strategies to Increase the Efficiency and Consistency of Drug Delivery

There is an increasing interest to administer cytotoxic drugs topatients by the oral route. Quality of life issues, treatmentadvantages and pharmaco-economics are major arguments in favorof oral therapy. However, low or moderate bioavailability incombination with considerable interpatient variability arefrequently observed which may reduce the feasibility of the oralroute for this class of drugs with a generally narrow therapeuticwindow. Until recently, investigators focused on absorptionenhancers which slightly damage the intestinal surface such assalicylates, methylxantines and surfactants to improve the oralbioavailability of drugs. To date, a shift can be seen towardsmore subtle mechanisms to enhance the absorption. This reviewarticle focuses on two important mechanisms that determine theoral bioavailability of cytotoxic drugs. These include thepresence of drug transporters in the intestinal epitheliumpumping drugs into the intestinal lumen, such as MDR1 typeP-glycoproteins, and first-pass elimination by cytochrome P450isoenzymes (e.g. 3A4 and 3A5) or other enzymes in the intestinesand/or liver. Currently preclinical and clinical studies arebeing performed to explore the feasibility of blocking thesetransporters/enzymes in order to achieve higher and less variablesystemic drug levels after oral dosing. This review gives anupdate of the results of these studies. It is concluded however,that further research to unravel the processes involved in oraldrug uptake is warranted to make the oral route a more efficientand consistent way of drug administration.

Cannulation of the jugular vein in mice: a method for serial withdrawal of blood samples

We have developed and validated a method that allows serial drawing of blood samples in freely moving mice using a cannula that is inserted via the jugular vein into the right atrium of the heart. The cannula was tunnelled subcutaneously to the head of the animal and attached to the skin by sutures, together with a metal spring, which was covered with PVC tubing for protection of the outer part of the cannula. Samples of blood up to 250 µl could be taken at serial time points. The blood volume in the circulation was maintained by replacement with an equal volume of blood obtained from donor animals. The applicability of this method of blood sampling for pharmacokinetic purposes was validated by comparing plasma concentrations–time curves in six cannulated animals after receiving an intravenous bolus dose of 10 mg/kg of the anti-cancer agent docetaxel versus the results in plasma samples obtained by cardiac puncture of non-cannulated mice. The presented method may lead to improved pharmacokinetic data produced from a reduced number of mice.

Metabolism of docetaxel in mice

Previous studies have shown by quantification of the parent drug and the known metabolites M-1, M-2, M-3 and M-4 that the mass balance of docetaxel in mice and humans is not complete. We therefore used reversed-phase high-performance liquid chromatography (HPLC) with photodiode array (PDA) detection and tandem mass spectrometry to trace and identify putative metabolites in the feces and bile of mice injected intravenously with docetaxel. HPLC-PDA revealed two metabolic products in the feces and more than ten potential new metabolites in the bile. Mass spectrometry was performed on docetaxel reference compound, on the known metabolites M-1, M-2, M-3 and M-4, and on HPLC eluate fractions containing metabolic products, six fractions originating from the bile and two from the feces. The mass spectra of the most abundant unknown metabolite in the bile and the feces were identical, and indicated that this structure contained a carboxyl moiety at the tert-butyl group. Under the conditions of storage this product degraded to metabolite M-4. All other unknown metabolites found in the bile samples were oxidized products, with the oxidations in both the C-13 side chain and the baccatin structure, the latter being a new finding.

Determination of cyclosporin A in human and mouse plasma by reversed-phase high-performance liquid chromatography

An isocratic reversed-phase high-performance liquid chromatographic method with ultraviolet detection at 227 nm has been validated for the determination of cyclosporin A in human and mouse plasma. The cyclosporin D analog PSC 833 was used as internal standard. Plasma samples were pretreated by liquid-liquid extraction with diethyl ether. A good chromatographic separation between cyclosporin A, the internal standard and two potentially interfering endogenous peaks was achieved using a stainless steel column packed with 5 microm Nova-Pak phenyl material operated at 72 degrees C, and a mobile phase consisting of acetonitrile-methanol-water (20:52:28, v/v/v). The calibration curve for cyclosporin A in human plasma was linear over the tested concentration range of 0.11 to 5.34 microM. Murine plasma samples (200 microl) were diluted up to a total volume of 500 microl with blank human plasma and the concentrations were read from the calibration curve prepared in human plasma. The lower limit of quantitation was 0.11 microM using 500 microl of human plasma and 0.28 microM using 200 microl of mouse plasma. The validation data showed that the assay is sensitive, selective and reproducible for determination of cyclosporin A. The applicability was demonstrated in a pharmacokinetic experiment where mice received oral cyclosporin A.