Hepatic Biotransformation of Docetaxel (Taxotereâ®)In Vitro: Involvement of the CYP3A Subfamily in Humans1

Home , Procarbazine

[CANCER RESEARCH 56. 1296-13(12. March 15. 1996] Hepatic Biotransformation of Docetaxel (TaxotereÂ®)in Vitro: Involvement of the CYP3A Subfamily in Humans1

FranÃ§oiseMarre,2 Ger-Jan Sanderink,3 Georges de Sousa, Claude Gaillard, Michel Martinet, and Roger Rahmani3

InstituÃ National Je la Sanie et de la Recherche MÃ©dicale,4l Bd. du Cap. BP 2078. 06606 Antikes Cedex Â¡F.M.. G. d. S.. K. R. l, and RhÃ´ne-Poulenc Rarer SA Recherche- DÃ'vi'lupiienu'nÃ.Centre de Recherche de Viity-Alfonville, 3 Digue d'Alfortville. 94140 Alfortville Â¡G-J.S., C. G., M. M.]. France

ABSTRACT involved in the metabolism of drugs, the CYP4 superenzyme family was recognized early as the chief contributor. Currently, at least 15 Docetaxel metabolism mediated by cytochrome P450-dependent nio- human liver CYPs involved in drug metabolism have been character nooxygenases was evaluated in human liver microsomes and hepatocytes. ized in terms of molecular, spectral, enzymatic, and immunological In microsomes, the drug was converted into four major metabolites properties (9). This multigene superfamily of hemoproteins is in resulting from successive oxidations of the tert-butyl group on the syn volved in the deactivation and/or activation of numerous endogenous thetic side chain. Enzyme kinetics appeared to be biphasic with a Vmaxand apparent AMIfor the high-affinity site of 9.2 pmol/min/mg and 1.1 ;IM. substances (e.g., steroids and fatty acids) and xenobiotics (e.g., drugs, carcinogens). CYPs are involved directly in a number of drug-induced respectively. The intrinsic metabolic clearance in human liver microsomes (VmaJKâ€žâ€ž8.4ml/min/g protein) was comparable to that in rat and dog hepatotoxicities and drug interactions. Some of these adverse effects liver microsomes, but lower than in mouse liver microsomes. are caused by genetic polymorphism. Others result either from drug Although the metabolic profile was identical in all subjects, a large interactions occurring between several coadministered drugs that are quantitative variation in docetaxel biotransformation rates was found in a metabolized by the same CYP. or from the specific induction of the human liver microsome library, with a ratio of 8.9 in the highest:lowest CYP in question. Thus, understanding the role of the CYPs involved in a given drug's metabolism and their regulation will be useful in biotransformation rates. Docetaxel biotransformation was correlated sig nificantly (0.7698; P < 0.0001) with erythromycin A'-demethylase activity, determining the range of interindividual variability in drug response but not with aniline hydroxylase or debrisoquine 4-hydroxylase. It was and avoiding undesirable drug interactions. This is a matter of par inhibited, both in human hepatocytes and in liver microsomes, by typical ticular concern in cancer chemotherapy, in which many prescriptions CYP3A substrates and/or inhibitors such as erythromycin, ketoconazole, may be encountered simultaneously. Therefore, since most anticancer nifedipine, midazolam, and troleandomycin. Docetaxel metabolism was drugs produce a number of side effects, a better understanding of the induced in vitro in human hepatocytes by dexamethasone and rifampicin, enzyme systems that process these compounds would be of great use both classical CYP3A inducers. These data suggest a major role of liver in designing optimal therapy protocols. cytochrome P450 isoenzymes of the CYP3A subfamily in docetaxel bio To date, research on the identification of specific CYP isozymes in transformation in humans. Finally, some Vinca alkaloids and doxorubicin the metabolism of cancer chemotherapeutic agents has been limited to were shown to inhibit docetaxel metabolism in human hepatocytes and a few compounds such as CYCLO or procarbazine, where rat hepatic liver microsomes. These findings may have clinical implications and microsomes were used (10). Except for some very recent studies on should be taken into account in the design of combination cancer chem Vinca alkaloids and paclitaxel (11-13), little data on the human otherapy regimens. enzymes involved in the metabolism of cancer chemotherapeutic agents are available. In that context, we evaluated the biotransforma INTRODUCTION tion of docetaxel by using microsomes and/or hepatocytes from hu man livers. The aim of the present report was to identify hepatic CYP Docetaxel (TaxotereÂ®,RP 56976; Fig. 1), a semisynthetic taxoid, is isozymes involved in docetaxel biotransformation in humans and to an antimitotic agent belonging to the spindle poison family (1). The evaluate possible metabolic drug interactions between docetaxel and drug promotes the assembly of stable microtubules in vitro and is an other anticancer agents, which are potentially associated with this inhibitor of cell replication (2). The compound has demonstrated taxoid. excellent antitumor activity against various types of solid tumors in preclinical models (3, 4) and is currently undergoing Phase II and III MATERIALS AND METHODS clinical evaluation. The pharmacokinetics of docetaxel has been stud ied after i.v. infusion in mice, rat, dogs, and cancer patients and Chemicals. Docetaxel (TaxotereÂ®;RP56976. batch PH12512), DZP. IMI, showed multiphasic disposition profiles with rapid initial tissue up INTO (RP60475), PB, RANI, and DOXO were obtained from the Centre de Recherche of RhÃ´ne-Poulenc Rorer SA (Vitry-sur-Seine. France). take and large volumes of distribution. In these species, the drug is [14C]Docetaxel (specific activity, 50 mCi/mmol) was synthesized by the Cen eliminated almost exclusively by hepatic metabolism and biliary tre des Etudes NuclÃ©aires(Gif-sur-Yvette, France). a-NF, /3-NF, ANI, BP, excretion, with quite similar metabolic profiles. The parent drug is the CIME. DEX, ERY, KETO, NADPH, NIF, propranolol, QUI, SPA. TP, TB. main circulating compound, and metabolites of docetaxel described and TAO were purchased from Sigma (Saint Quentin Fallavier, France). DBQ. thus far are much less cytotoxic (4-8). MDZ, and SKF525A (proadifen) were a gracious gift from Hoffmann LaRoche Metabolism represents a major factor of variability in drug response (Basel, Switzerland). ACET and QUIS were obtained from E. Merck; CAP and toxicity particularly for anticancer compounds that exhibit a narrow therapeutic index. Among the various hepatic enzyme systems 4 The abbreviations used are: CYP, cytochrome P450; DZP. dia/.epam; IMI, imipramine; INTO, intoplicine: PB. phÃ©nobarbital; RANI, ranitidine; DOXO. doxorubicin; a-NF, a-naphtoflavone; ÃŸ-NF. ÃŸ-naphtoflavone: ANI. aniline; BP, ben/o[Â«]pyrene; Received 5/2/95; accepted 1/15/96. CIME, cimetidine: DEX, dexamethasone; ERY. erythromycin: KETO. keloconazole: NIF, The costs of publication of this article were defrayed in part by the payment of page nifedipine; QUI, quinidine: SPA. spancine; TP. theophylline: TB. tolbutamide: TAO. charges. This article must therefore he hereby marked advertisement in accordance with troleandomycin; DBQ, debrisoquine; 3MC, 3-methylcholanthrene: MDZ. mida/.olam; 18 U.S.C. Section 1734 solely to indicate this fact. ACET, acetanilide; VP16, etoposide; QUIS. quinine sulfate; CAF. caffeine; CISPT, 1This work was supported in part by a grant from Ligue Nationale Contre le Cancer. cisplatin; CYCLO, cyclophosphamide; 5-FU, 5-fluorouracil: HB. hexobarbilal: VRB, * Present address: Laboratoire de Biochimie, FacultÃ©de Pharmacie. 2 rue du Dr vinorelbine; OME, omeprazole: IC^,,, 50% inhibitory concentration; VCR. vincristine; Raymond Marcland, 87025 Limoges Cedex. France. VBL, vinblastine: HPLC, high-performance liquid chromatography: PARA, paracetamol; 3 To whom requests for reprints should be addressed. HBSS. Hank's balanced salt solution.

1296

1200 T studies were performed to determine apparent K, values; microsomal fractions were incubated for 30 min with 1. 2.5, and 5 /UMdocetaxel and 0.5, 1, 2, 5, and 10 UM KETO. 1000 â€¢¿ Human liver microsomal fractions were also characterized by using specific substrates of several CYP isoforms. ERY iV-demethylase activity was deter mined by colorimetrie estimation of the formaldehyde formed after a 20-min 800 â€¢¿ incubation period of microsomal fractions ( 1 mg microsomal protein/ml) with 2 mM ERY (17). ANI hydroxylase and DBQ 4-hydroxylase activities were determined as described previously (18. 19). CPM 600 â€¢¿ Hepatocyte Isolation and Use of Primary Culture. A discarded human hepatic tissue fragment from a patient with secondary hepatic tumor (IU 46) and normal hepatic fragments (HL 45. 50, and 51) were obtained from the 400 â€¢¿â€¢¿ Sainte Marguerite Hospital (Professor P. Fuentes and Dr. P. Thomas. Mar seille, France). After the tissue was washed with HBSS. hepatocytes were isolated by conventional collagenase perfusion as described previously (20). 200 -â€¢ Using such a method, viability of cells, determined by erythrosine B exclusion test, was at least 80%. After washing (three times) with L15 medium, cells were resuspended in William's E medium supplemented with PCS (10%), insulin (0.1 lU/ml). 20 40 penicillin (50 lU/ml). streptomycin (50 tig/ml), and netilmicine (50 tig/ml). Hepatocytes were seeded in 6-well (800,000 cells/well) and I2-well (300.000 Fig. I. HPLC metabolic- pattern of docetaxel in liver microsomes. Incubation time. 60 cells/well) plates precoated with collagen type I (200 and 100 tig/well, respec min: on-line radioactivity detection. VI. alcohol derivate; V and Vil. cyclized aldehyde tively). Four hours later, the medium was renewed. derivatives: IV. carboxylic acid derivate. To screen for potential inducers of docetaxel metabolism, the medium was removed 24 h after seeding and replaced by William's E medium containing insulin, penicillin, streptomycin, netilmicine. and 5% FCS. Hepatocytes were from Prolaho: CISPT from Roger Bellori: CYCLO from Sarget; 5-FU from maintained in primary culture for 72 h in the presence of each of the following inducers: PB (2 mM), 3MC (40 /AM),ÃŸ-NF(50 /LIM).DEX (50 /AM),and RIF (50 Roche; HB from Boehringer; VRB from Pierre Fahre. OME from Research Biochemicals, Inc.; VPI6 from Sandoz; and VCR. and VBL from Eli Lilly. AIM).Medium was then removed and replaced by HBSS without inducer and containing [l4C]docetaxel. After 24 h of exposure, the medium was sampled Other chemicals were purchased from commercial sources and were of ana lytical or HPLC grade. and cells were scraped with water/methanol/hydrochloric acid (v/v, 0.1 N). The Preparation and Use of Liver Microsomal Fractions. Liver microsomes samples were analyzed using HPLC without further processing. were prepared from pools of livers from 6-week-old male OFI mice (23-28 g) To evaluate the potential inhibitory effect of characteristic substrates, cells were preinduced with RIF as described above. After a 72-h incubation period, and OFA Sprague-Dawley rats (180-200 g; IFFA-CREDO. Saint Germain- sur-1'Arbresle. France) and a liver from a male beagle dog ( 10 kg; Shamrock the medium was replaced by HBSS containing 5 AIM[l4C]docetaxel. Cells were incubated in the absence or presence of 5 or 25 JAMof each of the Farms. United Kingdom). Animals were maintained under standardized con following compounds: BP. TP, HB. DBQ, SPA. ANI. ERY. NIF. KETO. ditions of light and temperature. They had free access to food and water. Food MDZ, QUI. CIME, and SKF525A. Cells were then incubated under the usual was withdrawn 16 h before sacrifice. Human liver portions were from organ conditions. transplant donors or normal fractions of operative specimens oblained from Metabolic Drug Interactions. A number of anticancer agents and other hepatic segmentectomy of liver tumors. Characteristics of donors were re drugs potentially associated with docetaxel in cancer chemotherapy were ported previously (12). Microsomal fractions were prepared as described assessed for their inhibitory effects on docetaxel biotransformation. These previously (14). Microsomal protein concentration was determined according drugs included ARAC, CISPT, CYCLO, 5-FU, DOXO, INTO, VP 16, VCR, to the method of Bradford (Bio-Rad Protein Assay kit; Ref. 15) using bovine VBL. and VRB as anticancer agents and CIME. DZP. IMI, OME. PARA, and serum albumin as standard. Concentrations of CYP and cytochrome b5 were RANI as other drugs. Assays were carried out using HL 50- and HL 51- measured according to Omura and Sato (16). The incubation mixtures con cultured hepatocytes and human microsomal fractions in the optimal condi tained microsomes (2 mg microsomal protein/mil and NADPH (1 mM) in a tions described above. Final docetaxel concentration in the incubation medium potassium phosphate buffer (100 mM. pH 7.4). Incubations were carried out at 37Â°C in a shaking water bath under atmospheric oxygen. Reactions were was 5 AIM;the drugs were used at different concentrations that ranged between initiated by the addition of pure or isotopie dilutions of [I4C]docetaxel (50 0.25 and 1000 AIM.All assays were done in triplicate. mCi/mmol) and stopped by 1 volume of methanol. The samples were centri- fuged ( 10.000 X g, 2 min), and supernatant fluids were analyzed using HPLC without further processing. Each experiment was performed in triplicate. Incubations for analysis of enzyme kinetics were carried out at docetaxel concentrations from 2 to 100 /MM,with sampling at several time points between 0.5 and 60 min. The kinetics of the biotransformation of the substrate was monitored. This was more appropriate than analysis of metabolites because only one initial oxidation reaction occurred on the parent drug, but the primary metabolite was further oxidized (see "Results"). For interindividual variability studies in human microsomes, metabolism of docetaxel (5 /J.M)was measured in the microsomal fractions (n = 29) incubated for 30 min. Incubations for inhibition studies with human liver microsomes were carried out in triplicate for 60 min at a target concentration of 5 iiM docetaxel : R = CH3 docetaxel. The CYP substrates or inhibitors used were ACET, a-NF, CAF, TB, metabolite VI : R = CH2OH QUI, QUIS, ANI. ERY. KETO, NIF, TAO, and SKF525A. Concentrations ranged from I to 1000 AIM(Table 2). In each experiment, control incubations were included containing the appropriate vehicle, and reference incubations Â¡n which the substrate was added after enzyme inactivation with methanol. In Fig. 2. Structure of docetaxel and its main metabolite. *. site of 14C radiolabel. Arrow, case of inhibition. 1C,,, values were determined. Moreover, for KETO. kinetic position of metabolic oxidation. 1297

Table I Enzyme kinetics constants of docetaxel metabolism in liver microsomes Docetaxel metabolic profiles were analyzed with a high-pressure gradient Microsomal fractions (1 mg microsomal protein/ml) were incubated at 37Â°Cwith system consisting of Merck L6200 and L6000 pumps, a Waters 717 UltraWisp docetaxel (range. 2-100 JIM) for 0.5-60 min. Reactions were initiated by the addition of injector, a Merck L4000 UV detector, and a Berthold LB 506C continuous NADPH (I mini and stopped by methanol/hydrochloric acid (v/v, 0.1 N) before HPLC analysis. flow scintillation detector. Docetaxel and its metabolites were separated on a Merck RP Select B column (5 /j.m, 4 X 250 mm) thermostated at 25Â°C.The mobile phase solutions consisted of watenacetic acid:acetonitrile in the pro SpeciesMouse protein)258 (ml/min/gprotein)25 portions 95:0.2:5 (solution A) or 5:0.2:95 (solution B). The gradient was 26% Rat*Dog B (0-60 min), 26-35% B (60-80 min), 35% B (80-105 min). 35-100% B 22.243.5 6.48.51.1Clm" 3.55.18.4 (105-115 min), and 100% B (115-125 min). Flow rate was I rnl/min. Human*VW(prnol/min/mg 9.2"m(/J.M)10.3 The detection limit of the methods was 150 dpm (three times background " Intrinsic metabolic clearance (Vm.AJKm). counts), which corresponded in routine incubations to <0.05 JUM.The repeat '' Parameters of the first term of a two-enzyme model. ability of triplicate incubation assays was about 5%. Data Analysis. Biotransformation rates and enzyme kinetic parameters were calculated using iterative nonlinear regression analysis with the program Enzfitter (Elsevier-BIOSOFT. Cambridge, United Kingdom). IC50 concentrations were calculated from linearized plots of rate ' against HPLC Analysis. HPLC analysis for kinetic studies was performed with a Waters model 510 equipped with an automatic Waters 715 UltraWisp injector. inhibitor concentration. Separation was performed on an Ultrasphere ODS Beckman column (5 /urn, Correlation between the enzyme activities and docetaxel metabolism rate 4.6 X 250 mm). Elution was carried out at 1 ml/min by using a water:methanol was determined using linear regression analysis on MicroStat PC software. (v/v, 35:65) mixture, delivered isocratically. Radiolabeled compounds were Variance analysis (F test) was performed to validate the linear model hypoth detected by continuous flow liquid scintillation (Flow-One; Packard). Under esis and correlation was considered significant when the null hypothesis the above conditions, the retention time for docetaxel was 22.5 min. (slope = constant) could be rejected at the 5% risk level.

MOUSE RAT

300

-t 40 60 80 100 120 20 40 60 80 120 [docetaxel] [docetaxel] (MM)

DOG HUMAN

20 40 60 80 20 40 60 80 100 [docetaxel] (pM) [docetaxel] (\iM) Fig. 3. Interspecies variability of docetaxel metabolism: determination of Michaelis-Menten constants. Concentration-rate data were fitted to a one-enzyme model for mouse and dog liver microsomes and to a two-enzyme model for rat and human liver microsomes. , curves for the high- and low-affinity sites of these latter species. [Left cun-es (rapidly increasing at the beginning and leveling off first), high-affinity sites; right curves, low-affinity sites.) 1298

substrate ANI had no effect. The nonspecific suicide substrate SKF525A reduced docetaxel biotransformation by 40%. â€¢¿Bs:C Human hepatocytes from HL 45 and HL 46 were incubated for 24 h in the presence of 5 /J.Mdocetaxel and 5 or 25 JU.Mofeach potential inhibitor. The results (Table 3) confirm that only specific substrates or inhibitors of the CYP3A subfamily (KETO, NIF, and MDZ) inhibit the biotransformation of docetaxel to a significant extent. Induction of Docetaxel Metabolism. Human hepatocytes from HL 45 and HL 46 were preincubated for 72 h with the classical inducers PB, 3MC, j3-NF, and RIF. Docetaxel (5 /Â¿M)metabolism was I 7 9 11 15 18 21 23 26 28 30 32 34 36 39 5 8 10 14 17 20 22 24 27 29 31 33 35 37 evaluated subsequently in these induced hepatocytes. Table 3 illustrates that the biotransformation of docetaxel was HL microsome increased by the pretreatment of hepatocytes with RIF (+44% com Fig. 4. Interindividual variability of docetaxel oxidation in human liver microsomes. pared to control) and DEX (+41%), whereas PB (+8%), ÃŸ-NF Human liver microsornal fractions ( I mg microsomal protein/ml) were incubated at 37Â°C (+20%), and 3MC (0%) had less or no effect. for 30 min with docetaxel (5 /IM). Correlations with Monooxygenase Activities. An approach that has been used frequently to associate activities to a specific enzyme in human liver is to correlate the activity of a selected substrate toward RESULTS that of known substrate in a variety of human liver samples. We used Metabolic Pattern of Docetaxel in Liver Subcellular Fractions. a bank of 25 human liver microsomal samples that had been pheno- The metabolic pattern of docetaxel obtained after a 60-min incubation typed previously with respect to the relative level of various CYP- with human liver microsomes in the presence of NADPH is illustrated in Fig. 1. Formation of four metabolites, each more polar than the parent drug, was observed. The metabolites were identified by com Table 2 Effects of CYP Hthstrati's/inhibiton on dotetaxel (5 H.M)metabolism hy human parison with metabolites isolated from human feces by Gaillard and liver micro-fomes Microsomal fractions (2 mg protein/ml) were preincubated for 10 min with various colleagues (6-8) and synthesized by CommerÃ§on et al. (21). All in potential inhibitors at 37Â°Cbefore incubation with docetaxel (5 /J.M)for 60 min. vitro metabolites resulted from successive oxidations of the Ãe/7-butyl CYP group on the synthetic side chain (Fig. 2). The major metabolite (VI) isoenzyme corresponded to the alcohol. Metabolites V and VII are two oxazoli- CompoundACETa-NFCAFTBQUIQUISANIERYKETONIFTAOSKF525AMainaffinityIA21A21A22C8/92D6(2D6 dine-type compounds, resulting from cyclization of an unstable inter mediate aldehyde. Metabolite IV corresponds to the carboxylic acid. Following cyclization this compound may yield an oxazolidinedione derivative, the major docetaxel metabolite in human feces (7). Interspecies and Interindividual Variability of Docetaxel Bio- control)2E1/23A3/43A3/43A3/43A3/4(nonspecific)Concentration(/Â¿M)1000140050555010001010010250Inhibition{*)13-123-7-7718539944S340IC50((J.M)aâ€”1 transformation Rate. Docetaxel was incubated at concentrations (0.1-10)*130(30-300)''10(1 from 2 to 100 /IM with mouse, rat, dog and human liver microsomes in the presence of NADPH. Enzyme kinetic parameters show that -300)*â€” biotransformation rates were similar in the rat. dog, and human liver microsomes, but considerably higher in mouse liver microsomes " â€”¿.notdetermined. (Table 1). These rates are also shown in Fig. 3, with the best-fit curves Concentration range of compounds used to determine IC<Â¡()values. calculated according to the Michaelis-Menten equation. For mouse and dog liver microsomes, the best fit was observed with the classical equation for one enzyme site. For rat and human liver microsomes, a two-enzyme model appeared to be more appropriate. The parameters Table 3 Effects of inhibitory and inducer compounds on docetcixel (5 Â¿I.M)metabolism by human hepaiocytex in primary culture for the low-affinity site in rat and human liver microsomes are not Cultured human hepatocytes were incubated either with enzyme inducers (PB, 3MC, included because saturation was insufficient in the experimental range /3-NF, DEX. and RIF) for 72 h and then with docetaxel (5 /IM) for 24 h; or with potential inhibitors (BP. TP. HB, DBQ. SPA. ANI. ERY. NF. KETO. MDZ, QUI. and SKF525A) of substrate concentrations. and docetaxel (5 /IM) simultaneously for 24 h. Docetaxel biotransformation rates were studied in 29 human liver microsomal samples. Results (Fig. 4) indicate a significant quantita Compound3MCBP(3-NFTPPBHBDBQSPAQUIANIRIFDEXERYKETONIFMDZSKF525AConcentrationfjttfl405505200055555505055555Inhibition<%)26â€”29â€”2424274424â€”â€”3895907744Induction(%)0a+20â€”+8â€”â€”â€”â€”â€”+44+41â€”â€”â€”â€”â€” tive variability among individuals since docetaxel (5 /JLM)biotransfor mation rates ranged between 18 and 160 nmol total metabolites/ min/mg microsomal protein, while the mean rate was 89.5 nmol total metabolites/min/mg microsomal protein (coefficient of variation, 44.7%). Inhibition of Docetaxel Metabolism. The effect of several CYP inhibitors on docetaxel biotranstbrmation in human liver microsomes is shown in Table 2. Docetaxel metabolism in human liver micro somes was decreased most effectively by the CYP3A substrates and/or inhibitors ERY, KETO, and NIF. Among these compounds. KETO exhibited an IC50 value of about 1 /J.M,TAO about 10 Â¿K.M.and NIF about 130 /AM.The apparent K, value for KETO was 1.35 ptM. The CYP1 A2-inhibitor a-NF and substrates CAF and ACET, the CYP 2C8/9 substrate TB, the CYP2D6 inhibitor QUI, and the CYP2E *â€”¿,notdetermined. 1299

200 CYPs in docetaxel metabolism subject to variable regulation and/or

C genetic polymorphism. This study shows that isoenzymes of the CYP Â°s: r = 0.7698 subfamily CYP3A appear to be the major forms responsible for 150 (p < 0.0001) S F'x-E docetaxel biotransformation in humans. Docetaxel metabolism by Ã¶>'Eo e human cultured hepatocytes was induced by RIF and DEX. The 100 CYP3A substrates and inhibitors TAO, ERY, and NIF (23, 24) re X i= CO?, Eo duced docetaxel biotransformation in human hepatocytes and liver y Q- 50 microsomes. KETO, which had the same effect, is also known to be a very potent inhibitor of CYP3A (25). Moreover, the docetaxel metabolism rate was correlated strongly with ERY /V-demethylase 0 0.5 1 1.5 2 2.5 activity, a marker of CYP3A3/4 activity, but not with ANI hydroxy Erythromycin N-demethylase (nmol/min/mg P) lase (CYP2E1/2) or DBQ hydroxylase (CYP2D6). This shows that the interindividual variability in docetaxel oxidation is related to varia Fig. 5. Relationship between docetaxel oxidation and ERY W-demelhylase activity. bility in expression of the CYP3A subfamily. Indeed, the expression Human liver microsomal fraetions ( I mg microsomal protein/ml) were incubated at 37Â°C of CYP3A4 is known to be highly variable in human liver and with dwetuxel (5 P.M, 30 min) or ERY (2 mM. 20 min), r was determined using linear regression analysis. intestine, and the CYP3A5 isoform is thought to be polymorphic (24, 26). It is interesting that the main metabolizing enzyme of docetaxel is different from that of paclitaxel, which is transformed mainly by specific monooxygenase activities: ERY yV-demethylation (CYP3A CYP2C8/9 in human liver microsomes (13). subfamily), ANI hydroxylase (CYP2E subfamily), and DBQ This study provides some contradictory indications concerning the 4-hydroxylase (CYP2D6). The only significant correlation obtained number of isoenzymes involved in docetaxel metabolism. On the one was the biotransformation of docetaxel with ERY /V-demethylase hand, the biotransformation kinetics of docetaxel in rat and human activity (r = 0.7698. P < 0.0001). indicating that variability in liver microsomes was most consistent with a two-enzyme model. The docetaxel metabolism is likely caused by that in CYP3A levels lack of a zero interception in the correlation with ERY /V-demethylase (Fig. 5). In contrast, there was no correlation with either DBQ 4-hy may indicate that other minor CYP forms are involved in the metab droxylase (r = -0.3373, not significant) or ANI (r = 0.3535. not olism of docetaxel. On the other hand, none of the inhibitors of other significant) activities measured in the same samples. CYPs had a significant effect on docetaxel biotransformation in Metabolic Drug Interactions. Results of inhibition studies with human liver microsomes. However, biphasic kinetics in human liver various drugs are given in Table 4. DOXO. VRB, and VBL were able microsomes has been observed for other CYP3A substrates, and in to inhibit docetaxel biotransformation in vitro with ICS() values of one case evidence for substrate activation at high concentrations was about 20, 60, and > 100 /Â¿M,respectively. The anticancer agents 5-FU, given (27, 28). On the other side, the inhibition experiments in this CYCLO, CISPT, ARAC, VCR, VP16, and INTO and the anti-H2 study were carried out at a docetaxel concentration (5 /U.M),at which drugs CIME, RANI, as well as DZP, IMI, and PARA had no effect, almost all activity is catalyzed by the high-affinity site (Fig. 3). This even at high concentrations. For OME, weak inhibition occurred at the may explain the absence of inhibition by compounds with affinity for highest tested concentration (100 JAM). other CYPs. Cultured human hepatocytes were incubated in the presence of Among the tested inhibitors, QUI is a special case. This compound docetaxel (5 Â¡Â¿M)andeach anticancer drug (0.25-5 /AM).Results (data is a very potent inhibitor of CYP2D6 (KÂ¡< 1 /MM),hut metabolized not shown) were similar to those obtained with microsomal fractions. itself mainly by CYP3A with an apparent Km of about 30 JU.M(29,30). In our experiments the QUI concentration was 5 JU.M;therefore, only DISCUSSION an effect on CYP2D6 would be observable. As this was not the case, and as no difference between QUI and quinine (which is a much less The biotransformation of docetaxel was investigated in human hepatic in vitro models. Enzyme kinetics was compared with mouse, rat, and dog liver microsomes. Because previous studies did not Table 4 Effect* of drugs on docetaxei (5 pMt inelaholisin h\ human liver tnicrosomes indicate any direct conjugation reactions of docetaxel by rat and Microsomal fractions (2 mg protein/ml) were preincubated l'or IO min with various human UDP-glucuronosyltransferases or glutathione S-transferases,5 drugs at 37UC before incubation with docetaxel (5 JIM) for 60 min. this study focused on docetaxel biotransformation by the CYP-dependent monooxygenase system. Docetaxel biotransformation rates DrugARAC5-FUCISPTCYCLODOXOINTOVBLVCRVPI6VRBCIMERANIOMEDZPIMIPARAInhibition"(%)-13-12-8-2865445-29638II329-26-33IC50"((Â¿M)'â€”â€”â€”20â€”>100â€”â€”60â€”â€”â€”â€”â€” were more rapid in mouse liver microsomes than in those from other species. Apparent Km values were in the low micromolar range for all species. This finding appears to be in agreement with that of hepatic elimination of docetaxel in the isolated perfused rat liver being con centration dependent in the 5-50 /MMconcentration range (22). The metabolism of docetaxel followed a one-model Michaelis-Menten kinetic pattern in mouse and dog liver microsomes and a two-model kinetic pattern in rat and human samples. The biphasic linear kinetics suggested a multienzyme binding site for docetaxel. Docetaxel biotransformation by human liver microsomes showed a large quantitative interindividual variation. A variability factor of 9 was found between the lowest and the highest metabolic rates. This variability could be explained by the involvement of one or more " Inhibitor concentration. 100 /IM. ''Concentration range. I-HXX) /IM. s Unpublished data. ' â€”¿,notdetermined.

I3(X)

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1996 American Association for Cancer Research. HEPATIC CYP3A-MEDIATED DOCETAXEL METABOLISM potent inhibitor of CYP2D6) was found, no influence of the DBQ vitro experiments will occur in vivo is likely to depend on the dosage phenotype on docetaxel metabolism is expected in humans. and pharmacokinetics of both docetaxel and the coadministered So, taken together, docetaxel appears to be metabolized mainly by drug(s). However, the concentrations in the in vitro models cannot be CYP3A isoenzyme(s), especially at therapeutic concentrations, which compared directly to in vivo plasma concentrations since factors such are 5 Â¡JMorlower (5). Although the involvement of other isoenzymes as hepatic uptake or protein binding will be of importance. Concern at high docetaxel concentrations is not to be excluded, this study gives ing docetaxel, its low biotransformation rate in vitro is consistent with no definitive proof for this. The CYP3A family of CYP is thought to a rather low hepatic extraction rate in vivo (38). Its elimination is thus be the predominant family in human drug metabolism. It contains four likely to be dependent on drug-metabolizing enzyme levels and could, very closely related proteins (CYP3A3, CYP3A4, CYP3A5, and in consequence, be affected by inhibitors and inducers. On the other CYP3A7) which account for up to 60% of the total CYP expressed in hand, the low Km(app)of 1.1 /AMindicates a high affinity, which makes human liver. CYP3A5 is expressed polymorphically, having been docetaxel metabolism less susceptible to competitive inhibition. detected in only about 10-20% of adult human livers analyzed, Our data demonstrate clearly the involvement of liver CYP3A whereas CYP3A7 is the major CYP expressed in the human fetal isozymes in the biotransformation of docetaxel and significant meta liver. The most important hepatic isoenzymes in adults are CYP3A3 bolic drug interactions between this drug and a number of anticancer and CYP3A4, having only 11 amino acid substitutions, and no dif agents which could potentially be associated with docetaxel. Because ferences in catalytic activities have been demonstrated; thus, they are inhibition of docetaxel metabolism may result in higher effective commonly referred to as CYP3A3/4. They represent key enzymes exposure of patients, and. regarding the low safety margin of this class responsible for the metabolism of many clinically important drug of drugs, these findings may have both fundamental and clinical families like corticosteroids, antifungals, antineoplastic agents, immu- implications and should be taken into account in the evaluation and nosuppressants, macrolide antibiotics, sex hormones, opioid analge design of combination cancer chemotherapy regimens. sics, antiarrhythmic agents, vasodilators, etc. ERY /V-demethylation, cyclosporin metabolism, NIF oxidation, ACKNOWLEDGMENTS MDZ hydroxylation, testosterone 6ÃŸ-hydroxylation, and Cortisol 6ÃŸ- hydroxylation are probes of CYP3A3/4 catalytic activity. The levels We are grateful to A. Touzet for technical assistance. Special thanks are due of CYP3A3/4 vary highly among liver microsomal samples. Several to the clinical staff of the organ transplant units (Marseille. France). drugs (cortisol, DEX, RIE, PB, and carbamazepine) are inducers of CYP3A, some of which are likely to be coadministered in patients. REFERENCES This regulation is not yet elucidated completely, but occurs probably 1. Gueritte-Voegelein. F.. Guenard. D.. Lavelle. F., Le Goff, M. T.. Mangatal, L., and at the level of transcription and does not involve a known steroid Potier. P. Relationships between the structure of Taxol analogues and their metabolic receptor. In addition, several potent inhibitors of 3A-mediated metab activity. J. Med. Chem., 34: 992-998. 1991. 2. Ringel. !.. and Horwitz. S. B. Studies with RP 56976 (Taxotere). a semisynthetic olism in vivo have been characterized (e.g., KETO, verapamil, ethinyl analogue of Taxol. J. Nati. Cancer Inst.. 83: 288-291. 1991. estradici, TAO, gestodene, and narengenin). 3. Bissery. M. C., Guenard. D.. Gueritte-Voegelein, F., and Lavelle. F. Experimental antitumor activity of Taxotere (RP 56976. NSC 628503), a Taxol analogue. Cancer Combination chemotherapy has become standard in the treatment Res., 51: 4845-4852, 1991. of most malignancies due to a number of theoretical advantages and 4. Bissery. M. C., Nohynek, G.. Sanderink. G. J.. and Lavelle, F. Docetaxel (Taxotere*): proven superior clinical efficacy compared to single-agent treatment. a review of preclinical and clinical experience. Part I. preclinical experience. Anti- cancer Drugs. 6: 339-368, 1995. Protocols with three, four, or more drugs have been used in a variety 5. Bruno. R.. and Sanderink, G. J. Pharmacokinetics and metabolism of Taxotereâ„¢ of combinations, administered concurrently or sequentially. Yet, (docetaxel). Cancer Surv. 17: 305-313, 1993. 6. Gaillard. C.. Monsarrat. B., Vuilhorgne. M., Royer. L, Monegier, B., SablÃ©,S., many unrecognized drug interactions undoubtedly occur. Therefore, GuÃ©nard,D., Gires, P., Archimbaud, Y., Wright. M., and Sanderink. G. J. Docetaxel evaluation of interactions between anticancer drugs are of consider (Taxotere*) metabolism in thÃ©ratin vivo and m vitro. Proc. Am. Assoc. Cancer Res.. able interest with regard to the optimization of cancer chemotherapy. 35: 428, 1994. 7. De Valeriola, D., Brassine, C., Gaillard. C., Kettler, J. P., Tomiak. E.. Van Vreckem, Some drugs which may potentially be coadministered with docetaxel A.. FrÃ¼hling.J.. Frydman. A., Kerger, J.. Piccar!. M.. Chapelle. P., and Blanc, C. were tested for their capacity to modify docetaxel metabolism in Study of excretion balance, metabolism and protein binding of '4C-radiolabelled human liver microsomes. For most of these drugs the interaction with Taxotere (TXT) (RP56976. NSC628503) in cancer patients. Proc. Am. Assoc. Cancer Res., 34: 373. 1993. specific human CYP isoenzymes has not been studied previously. The 8. Vuilhorgne. M.. Gaillard, C.. Sanderink, G. J.. Royer. I.. Monsarrat, B.. Dubois, and role of CYP3A in the metabolism of some Vinca alkaloids was Wright. M. Metabolism of taxoid drugs. ACS (Am. Chem. Soc.) Symp. Ser., 583: 98-110, 1995. demonstrated recently (11, 12). As for the Vinca alkaloids, an absence 9. Neben. D. W., Adesnik, M., Coon, M. J., Estabrook. R. W., Gonzalez, F. J., of inhibition of CYP3A-mediated metabolism by 5-FU, cytarabine, Guengerich, F. P., Gunsalus. I. C.. Johnson. E. F.. Kemper. B.. Levin. W., Phillips. and cisplatin was observed for docetaxel. Inversely, the metabolism of I. Râ€žSato. R., and Waterman. M. R. The P450 gene superfamily: recommended nomenclature. DNA, 6: 1-11, 1987. docetaxel in human liver microsomes can also be inhibited by DOXO. 10. Guengerich. F. P. Roles of cytochrome P450 enzymes in chemical carcinogenesis and VBL, and VRB. DOXO can be metabolized by NADPH-CYP reduc cancer chemotherapy. Cancer Res.. 48: 2946-2954, 1988. Ãase(31), which may thus constitute an alternative inhibition site, 11. Zhou, X. J., Zhou-Pan, X. R.. Gauthier, T., Placidi, M., Maurel, P.. and Rahmani, R. Human liver microsomal cytochrome P4503A Â¡sozymesmediated vindesine biotrans along with the CYP3A enzymes. Only the effects of VCR and VP16 formation. Biochem. Pharmacol.. 45: 853-861, 1993. diverged, both compounds being inhibitors of vindesine and VBL 12. Zhou-Pan. X. R.. Zhou-Pan X., Placidi. M., Maurel, P.. Barra, Y. Rahmani, R. Involvement of human liver cytochrome P450 3A in vinblastine metabolism: drug metabolism but not of docetaxel. interactions. Cancer Res.. 53: 5121-5126, 1993. Among the other tested drugs, none except the already mentioned 13. Cresteil, T.. Monsarrat, B.. Alvinerie. P.. TrÃ©luyer,J. M.. Vieira, I., and Wright. M. CYP3A substrates/inhibitors were found to inhibit docetaxel biotransfor Taxol metabolism by human liver microsomes: identification of cytochrome P450 isozymes involved in its biotransformation. Cancer Res., 54: 386-393. 1994. mation. These results are consistent with the fact that in humans diaze- 14. Van der Hoeven. T. A., and Coon, M. J. Preparation and properties of partially pam and OME are metabolized mainly by CYP2C19 (32, 33), imipra- purified cytochrome P-450 and reduced nicotinamide adenine dinucleotide phosphate-cytochrome P-450 reducÃasefrom rabbit liver microsomes. J. Biol. Chem., 249: mine by CYP2D6 and CYP1A2 (34, 35), and PARA by several enzymes, 6302-6310, 1974. among them mainly CYP1A2 and CYP2E1 (36). The lack of inhibition 15. Bradford. M. A rapid and sensitive method for the quantification of microgram by CIME is consistent with the absence of an effect on another human quantities of protein utilizing the principle of protein dye binding. Anal. Biochem.. CYP3A-mediated reaction, cyclosporin oxidation (37). 72: 248-254, 1976. 16. Omura. T.. and Sato. R. The carbon monoxyde-binding pigment of liver microsomes. Whether the metabolic drug-drug interactions observed in these in I Evidence of its hemoprotein nature. J. Biol. Chem.. 239: 2370-2378, 1964. 1301

17. Nash, T. The colorie estimation of formaldehyde by means of the Hantzsch reaetion. in human liver microsomes by cytochrome P450IIIA4. Clin. Pharmacol. Ther., 46: Biochem. J.. 55: 516-521. 1953. 521-527. 1989. 18. Mieyal. J. J.. and Blumer, J. L. Acceleration of the autooxidation of human oxyhe- 2Â°.Otton, S. V.. Inaba. T.. and Kalow. W. Competitive inhibition of spancine oxidation moglobin hy aniline and its relation to hemoglobin-catalyzed aniline hydroxylation. in human liver hy ÃŸ-adrenoreceptor antagonists and other cardiovascular drugs. Life J. Biol. Chem.. 251: 3442-3446. 1976. Sci.. 34: 73-80, 1984. 19. Marre. F.. Fahre. G.. Lacarelle. B.. Bourrie. M.. Catalin. J.. Berger. Y.. Rahmani. R.. 30. Guengerich. F. P.. MÃ¼ller-Enoch. D., and Blair. I. A. Oxidation of quinidine hy and Cano. J. P. Involvement of the cytochrome P450I1D subfamily in minaprine human liver cytochrome P-450. Mol. Pharmacol.. 30: 287-295. 1986. 4-hydroxylation hy human hepatic microsomes. Drug Metan. Dispos.. 20: 316-321. 31. Cummings. J.. Allan. L.. Willmoll. N.. Riley. R.. Workman. P.. and Smyth, J. F. The 1992. enzymology of doxoruhicin quinone reduction in tumour tissue. BuKheni. Pharma- 20. Don. M., de Sousa. G.. Lacarelle. B.. Placidi. M.. Lechenc de la Porte. P.. Domingo. col.. 44: 2175-2183, 1992. M.. Lafont. H., and Rahmani. R. Thawed human hepatocytes in primary culture. 32. Bertillson, L.. Henthorn, T. K.. Sanz. E., Tybring. G.. Siiwc. J.. and Villen. T. Cryobiology, 29: 454-469. 1992. Importance of genetic factors in the regulation of dia/epam metabolism: relationship 21. Commercon, A., Bour/.at, J. D.. BÃ©/ard.D.. and Vuilhorgne. M. Partial synthesis of to .S'-mephenytoin. hut not debrisoquin. hydroxylation type. Clin. Pharmacol. Ther.. major human metabolites of docelaxel. Tetrahedron. 50: 10289-10298. 1994. 45: 348-355. 1989. 22. Gires, P.. Gaillard. C.. Sanderink. G. J.. Martin. S.. Piguet. V.. Kettler. J. P.. Nichele. G.. Martinet. M.. Chapelle. P.. and Bost, P. E. [l4Cl-Docetaxel (Taxotere") disposi 33. Andersson, T., Regardh. C. G.. Dahl-Puuslinen. M.. and Bertillson, L. Slow ornepra- zole metaboliz.ers are also poor .V-mephenytoin hydroxylators. Ther. Drug Moni!.. /2: tion in the isolated perfused rat liver. Eur. J. Drug Metah. Pharmacokinet.. /ÃŒMSuppI. 2).- 29. 1994. 415-416. 1990. 34. Skjelbo. E.. and Brosen, K. Inhibitors of imipramine metabolism hy human liver 23. Bahany. G., Larrey. D.. and Pessayre, D. Macrolide antibiotics as indiicers and inhibitors of cytochrome P-450 in experimental animals and man. In: G. G. Gibson (ed.). Progress microsomes. Br. J. Clin. Pharmacol., 34: 256-261. 1992. in Drug Metabolism. Vol. 12. pp. 61-98. London: Taylor and Francis. 1988. 35. Lemoine, A., Gautier, J. C.. Azoulay. D.. Kiffel. L., Belloc, C.. Guengerich. F. P.. 24. Guengerich. F. P.. Martin. V. P., Beaune, P. H.. Kremers, P.. Wolff. T.. and Waxman. Maurel. P., Beaune. P., and Leroux. J. P. Major pathway of imipramine metabolism D. J. Characteri/ation of rat and human liver microsomal cytochrome P-450 forms is catalyzed by cytochromes P-450 IA2 and P-450 3A4 in human liver. Mol. Pharmacol., 43: 827-832. 1993. involved in nifedipine oxidation, a prototype for genetic polymorphism in oxidative drug metabolism. J. Biol. Chem.. 261: 5051-5060, 1986. 36. Raucy. J. L., Lasker, J. M.. Lieber. C. S., and Black. M. Acetaminophen activation hy 25. Back. D. J.. and Tjia. J. F. Comparative effects of the antimycotic drugs ketocona/ole. human liver microsomes P450IIEI and P450IA2. Arch. Biochem. Biophys., 277: lluconazole. Â¡tracomi/oleand terbinafine on the metabolism of cyclosporin by human 270-283. 1989. liver microsomes. Br. J. Clin. Pharmacol., 32: 624-626. 1991. 37. Pichard. L.. Fahre. I., Fahre. G.. Domergue. J.. Saint Aubert. J., Mourad. G., and 26. Watkins. P. B. Noninvasive tests of CYP3A enzymes. Pharmacogenetics. 4: 171-187. Maurel. P. Cyclosporin drug interactions: screening for inducers and inhibitors of 1994. cytochrornc P-450 (cyclosporin A oxidase) in primary cultures of human hepatocytes 27. Kerlan. V.. Dreano. Y., Bercovivi, J. P.. Beaune. P. H.. Floch. H. H.. and Berthou. F. and in liver microsomes. Drug Metab. Dispos.. 18: 595-606. 1990. Nature of cytochrome P450 involved in the 2-/4-hydroxylations of estradiol in human 38. Bruno. R.. Vergniol, J. C., Montay. G.. Le Bail. N.. Frydman. A.. Clavel, M., and liver microsomes. Biochem. Pharmacol., 44: 1745-1756, 1992. Marty. M. Clinical pharmacology of Taxotere (RP56976) given as 1-2 hour infusion 28. Bargel/i. M. J.. Aoyanta. T.. Gonzalez. F. J.. and Meyer. U. A. Lidocaine metabolism every 2-3 weeks. Proc. Am. Assoc. Cancer Res.. 33: 261. 1992.

1302

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1996 American Association for Cancer Research. Hepatic Biotransformation of Docetaxel (Taxotere®) in Vitro: Involvement of the CYP3A Subfamily in Humans

Françoise Marre, Ger-Jan Sanderink, Georges de Sousa, et al.

Cancer Res 1996;56:1296-1302.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/56/6/1296

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/56/6/1296. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.