Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Cyclophosphamide (CPA¹) is an alkylating agent prodrug widely used over a large dose range in the treatment of malignancies and autoimmune disorders. CPA is extensively metabolized by cytochrome P450 (P450) pathways via 4-hydroxylation and N-dechloroethylation, with only 10 to 30% of a dose excreted in the urine as unchanged parent compound (Sladek, 1988; Moore, 1991). The predominant pathway, 4-hydroxylation, yields the active alkylating moiety and an equimolar amount of the urotoxic byproduct acrolein. N-dechloroethylation generates dechloroethylCPA and the nephrotoxic and neurotoxic byproduct chloroacetaldehyde (Fleming, 1997). The major P450 enzymes responsible for 4-hydroxylation of CPA appear to be CYP2B6, CYP3A4, and CYP2C (Chang et al., 1993; Roy et al., 1999; Huang et al., 2000). Among these enzymes, CYP2B6 was identified as the P450 isozyme with the highest intrinsic clearance for CPA 4-hydroxylation (Roy et al., 1999). This is in contrast to the findings of another group of investigators who found no evidence for a quantitatively significant amount of formation of 4-hydroxyCPA from CPA by CYP2B6 in a panel of 5 human liver microsomes (Ren et al., 1997). CYP2B6 contribution to CPA activation has been shown to vary significantly among liver samples due to differences in the expression of CYP2B6 protein levels in individual livers (Roy et al., 1999; Huang et al., 2000). Huang et al. (2000) reported that CPA 4-hydroxylation ranged from 0.1 to 0.6 nmol/min/mg of protein in a panel of 17 human liver samples and that CYP2B6 contributed to an average of 45% of total CPA 4-hydroxylation (range 10-70% depending on the CYP2B6 content of the liver), compared with 25 and 12% for CYP3A4 and CYP2C9, respectively. CYP3A4 was shown recently to be the P450 enzyme exclusively involved in the N-dechloroethylation of CPA (Huang et al., 2000). Given the high capacity of CYP2B6 for CPA 4-hydroxylation and negligible formation of dechlorethylCPA, patients with relatively high CYP2B6 activity would be expected to experience the greatest therapeutic response with the least amount of nonmyelosuppressive toxicity from an administered CPA dose. In contrast, CYP3A4 produces relatively equivalent amounts of active and inactive CPA metabolites, and therefore patients with relatively high CYP3A4 activity would be expected to experience good therapeutic response at the expense of toxicity.

A high degree of interpatient variability has been reported in the pharmacokinetics and systemic exposure of CPA in children and adults with cancer (Moore et al., 1988; Yule et al., 1996; Ren et al., 1998; Yule et al., 1999a). This variability could partially reflect differences in the expression of individual P450 isozymes. The CYP3A subfamily is the most abundantly expressed in human liver and small intestine, and interindividual differences in CYP3A expression exceed 30-fold in some populations (Watkins, 1995). A recent report characterizing the genetic basis of polymorphic CYP3A5 expression found that CYP3A5 was expressed in 60% of livers of African Americans compared with 33% for Caucasians, and that, when present, represented at least 50% of total hepatic CYP3A content (Kuehl et al., 2001). Although previous work focused on the relative importance of CYP3A4 to the CYP3A subfamily, CYP3A5 may be the more important contributor to interindividual and ethnic differences in metabolic clearance of a large number of CYP3A substrates. In addition, recent studies using improved immunoquantitative techniques have reported a greater frequency of detection and a higher contribution of CYP2B6 to total P450 content than previously estimated. In addition, 20- to 250-fold variability (Code et al., 1997; Stresser and Kupfer, 1999) and ethnic differences (Shimada et al., 1994; Kim et al., 1997) in CYP2B6 expression have been recognized.

Other sources of variation include concomitant treatment with P450 inhibitors (Yule et al., 1999b) or prior treatment with P450 inducers including dexamethasone (Yule et al., 1996) and anticonvulsants (Slattery et al., 1996). CPA itself has been shown to induce its own metabolism manifested by an increase in clearance following repeated administration of high doses at 24-h intervals (Chen et al., 1995; Slattery et al., 1996; Busse et al., 1997). CPA autoinduction appears to be through induction of P450 isozymes. Exposure of cultured human hepatocytes to 250 µM CPA resulted in increased 6-hydroxylation of testosterone and CYP3A4 protein (Chang et al., 1997). However, no increase in CYP2B6 protein was detected with this same treatment. In contrast, Gervot et al. (1999) reported induction of CYP2B6 protein and mRNA in hepatocytes treated with 1 mM CPA. To date, no studies have reported whether CPA-induced increases in P450 activity occur through activation of the pregnane X receptor (PXR), which has been identified as the primary mediator of drug-induced increases in CYP3A4 expression (Lehmann et al., 1998; Goodwin et al., 1999).

Although dexamethasone (DEX) is frequently coadministered with CPA as an antiemetic agent, information regarding the effect of this CYP3A4 inducer on CPA autoinduction is not available. Our laboratory has previously demonstrated significant increases in CYP3A4 activity with DEX administration in healthy volunteers and in microsomes isolated from human hepatocytes (McCune et al., 2000). CYP3A4 induction was highly variable, and the induction ratio (-fold increase in testosterone 6-hydroxylase activity compared with control) was inversely related to baseline activity, with the greatest increase in activity observed in healthy volunteers and human hepatocytes with the lowest baseline CYP3A4 activity (McCune et al., 2000). In hepatocytes, 2 to 250 µM DEX resulted in an average 1.7- to 6.9-fold increase in CYP3A4 activity, respectively. In contrast, a similar concentration range of DEX resulted in less than 2-fold induction of CYP2B6 activity in human hepatocytes (Faucette et al., 2001).

The purpose of this study was to characterize the concentration-response effects of CPA on the expression of CYP2B6 and CYP3A4, the major CPA-metabolic activator and deactivator, respectively, in primary cultures of human hepatocytes. In addition, the influence of factors such as coadministration of DEX and baseline CYP2B6 and CYP3A4 activities on the extent of the autoinduction of CPA was evaluated. Finally, the capacity of CPA and DEX, alone and in combination, to activate PXR was explored using transient transfection assays.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. NADP⁺, potassium phosphate, glucose 6-phosphate, glucose-6-phosphate dehydrogenase, magnesium chloride, EDTA, triethylamine, cortisol, bupropion, and cyclophosphamide were purchased from Sigma-Aldrich (St. Louis, MO). Hydroxybupropion (HBUP) was graciously supplied by Glaxo Wellcome Inc. (Research Triangle Park, NC). Testosterone and 6-hydroxytestosterone were purchased from Steraloids Inc. (Newport, RI). Formic acid was obtained from Fisher Scientific (Fair Lawn, NJ), and HPLC-grade acetonitrile and methanol were purchased from J. T. Baker, Mallinckrodt Baker Inc., Div (Phillipsburg, NJ). All other chemicals were of the highest grade commercially available.

Hepatocyte Isolation and Cell Culture. All hepatic tissues were obtained through qualified medical staff, with donor consent and with the approval of the University of North Carolina Hospitals Institutional Review Committee. Hepatocytes were isolated from human liver tissue procured through the Department of Surgery, University of North Carolina at Chapel Hill School of Medicine or nontransplantable donor livers using a modified two-step collagenase digestion method. Prior to perfusion of the tissue, the outer capsule on the cut surface was reconstructed using medical grade adhesive. This technique permitted the use of lower flow rates and enhanced both blood clearance and digestion of the liver. Encapsulated liver tissue (15-100g) was perfused with calcium-free buffer containing 5.5 µM glucose, 0.5 mM EGTA, 0.05% ascorbic acid, and 0.5% bovine serum albumin for 10 to 15 min at a flow rate of 14 to18 ml/min, followed by Dulbecco's modified Eagle's medium (DMEM) containing 0.5% bovine serum albumin, ascorbic acid (0.05%), and collagenase (0.4-0.8 mg/ml) for 15 to 20 min at a flow rate of 18 to 28 ml/min.

Microsomal Assays. Microsomal CYP3A4 and CYP2B6 activities were determined by measuring the formation rate of 6-hydroxytestosterone from testosterone and HBUP from bupropion, respectively (Faucette et al., 2000). Formation rates were determined in duplicate with microsomes prepared from individual treatment groups of hepatocyte cultures (n = 3). Preliminary experiments in pooled microsomes were conducted to identify microsomal protein amounts and incubation times resulting in linear rates of 6-hydroxytestosterone and HBUP formation.

HPLC Analysis. The HPLC system for quantitation of 6-hydroxytestosterone and cortisol (internal standard) consisted of a Hewlett Packard 1090L model system (Hewlett Packard, Palo Alto, CA) with UV detection set at 245 nm. 6-Hydroxytestosterone, and cortisol peaks were separated and resolved at 40°C on a Supelcosil LC-18, 150 × 4.6 mm, 5 µm column proceeded by a Pelliguard LC-18 2 cm guard column (Sigma-Aldrich). Mobile phases A (64% water, 35% methanol, 1% acetonitrile) and B (18% water, 80% methanol, 2% acetonitrile) were pumped over a gradient at a flow rate of 1 ml/min. Retention times for 6-hydroxytestosterone and cortisol were 4.3 and 5.5 min, respectively. The HPLC system for detection of HBUP and the internal standard triprolidine consisted of a Hewlett Packard 1100 liquid chromatograph (Hewlett Packard, Avondale, PA) connected to a Hewlett Packard model 1100 UV detector set at 214 nm. Peaks of interest were separated on a 5 µm Waters Symmetry 15 × 0.39 cm C₁₈ column (Millipore Corp., Bedford, MA). Mobile phases A (0.25% triethylamine, 15% acetonitrile, and 0.1% formic acid) and B (100% acetonitrile) were pumped at a flow rate of 1 ml/min using a gradient ranging from 0% B at 0 to 15.5 min, 10% B at 16 to 23 min, and 0% B at 23.5 to 30 min. The column temperature was maintained at 40°C. HBUP and triprolidine peaks were integrated using a Hewlett Packard Chemstation system. Retention times for HBUP and triprolidine were approximately 7.5 and 24.5 min, respectively.

Cotransfection Assays. HUH7 cells were seeded in 24-well plates (50,000/well) in high glucose DMEM supplemented with 10% charcoal/dextran-treated fetal bovine serum (HyClone Laboratories Inc., Logan, UT). Transfection mixes contained 100 ng of hPXR expression vector (pCMV-SPORT; Invitrogen, Carlsbad, CA), 100 ng of firefly luciferase reporter plasmid [(CYP3A4 ER₆)₂-GL3-Promoter Vector; Invitrogen], and 10 ng of Renilla luciferase reporter vector (pRL-TK Vector; Invitrogen) as internal control. Transfections were performed with Effectene (QIAGEN, Inc., Valencia, CA). Drug dilutions were prepared in phenol red-free medium supplemented with 10% charcoal-stripped, delipidated calf serum (Sigma-Aldrich). Cells were incubated for 24 h in the presence of drugs, and cell extracts were prepared in lysis buffer (Promega Corp., Madison, WI). Reporter activity was determined using the dual-luciferase reporter assay system, essentially according to the manufacturer's instructions (Promega).

Western Immunoblotting. Western immunoblotting was carried out as described previously by Parkinson and Gemzik (1991). Briefly, equal amounts of microsomal protein (10 µg for CYP3A4, 20 µg for CYP2B6) were loaded onto polyacrylamide gels. Proteins were transferred onto nylon membranes, and immunoblots were incubated with primary antibody overnight at room temperature. Levels of immunoreactive CYP3A4 protein were detected with antibodies against CYP3A4-specific peptide at a 1:1500 dilution. Levels of immunoreacture CYP2B6 proteins were detected with antibodies against CYP2B6-specific peptide at a 1:500 dilution as previously described (Faucette et al., 2000). Secondary labeling was achieved with alkaline phosphatase-conjugated goat affinity purified antibody to rabbit IgG (INC/Cappel, Aurora, OH), diluted 1:1000, and incubated for 3 h at room temperature. Blots were developed with 5-bromo-4-chloro-indolyl-phosphatase and nitroblue tetrazolium as substrate solution (Kirkegaard & Perry Laboratories Inc., Gaithersburg, MD).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Cyclophosphamide and Dexamethasone Concentration-Dependent Effects on CYP3A4 Activity. The concentration-dependent effects of CPA on CYP3A4 activity in 10 microsomal preparations of human hepatocytes treated with various concentrations of CPA (2 to 1500 µM) were examined in a series of preliminary experiments (data not shown). CYP3A4 activity, expressed as -fold induction over control 6-hydroxytestosterone formation, was increased by 1.5-fold at 10 µM CPA concentration, peaked at 3.5-fold at 250 µM, and gradually declined at concentrations greater than 250 µM. Significant variability was observed in the extent of induction, and CPA concentration at which maximal inductive effects occurred was observed. For the composite data, the concentration at which EC₅₀ was achieved was 40 µM, and the maximum effect was 3.5-fold induction over control. Microsomal protein concentration and lactate dehydrogenase leakage in hepatocyte cell culture treated with concentrations of CPA greater than 250 µM were not different from those treated with <250 µM (data not shown), suggesting that a cytotoxic effect of CPA on hepatocytes was not an explanation for the decrease in percent induction in CYP3A4 activity observed at higher CPA concentrations.

Effects of Cyclophosphamide (250 µM), Dexamethasone (10 µM) Alone and Cyclophosphamide (250 µM) with Dexamethasone (10 µM) on CYP3A4 Expression. Figure 1 illustrates the CYP3A4 activity in microsomes prepared from 10 human hepatocyte cultures treated with dimethyl sulfoxide (control), DEX (10 µM), CPA (250 µM), and DEX (10 µM) plus CPA (250 µM) for 72 h. The concentration of CPA (250 µM) chosen was that which produced maximal induction of CYP3A4 activity in preliminary experiments with CPA. The DEX concentration (10 µM) chosen represents a concentration approximately 5-fold higher than the maximum concentration observed following a 20-mg oral dose of DEX (Brophy et al., 1983) and a point on the upward curve of CYP3A4 induction observed with DEX. CPA concentrations <250 µM are commonly achieved in cancer patients receiving standard CPA doses (i.e., 1 g/m²), whereas >250 µM are observed in cancer patients receiving CPA as myeloablative therapy (Moore et al., 1988; Slattery et al., 1996; Yule et al., 1996; Ren et al., 1998; Busse et al., 1999; Yule et al., 1999b). Rifampin (10 µM) was used as a positive control in all hepatocyte preparations. Previous dose-response studies have revealed that maximal increases in CYP3A4 are attained at rifampin concentrations greater than 2 µM (Sahi et al., 2000).

View larger version (41K): [in this window] [in a new window]   Fig. 1.   Effect of CPA and DEX alone and in combination on the rate of 6-hydroxytestosterone formation CYP3A4 expression. Hepatocyte cultures were treated for 3 days with 0.1% dimethyl sulfoxide (control), 250 µM CPA, 10 µM DEX, or both CPA (250 µM) and DEX (10 µM), as indicated under Materials and Methods. Microsomes were subsequently prepared and analyzed for CYP3A4 enzyme activity based on the 6-hydroxylation of testosterone.

Cotransfection Assays. To determine whether the induction of CYP3A4 by CPA and DEX was mediated through activation of the PXR, the concentration-dependent effects of CPA (25-500 µM) alone and in combination with DEX (10 µM) were examined in HUH7 cells cotransfected with a hPXR expression vector and a reporter gene construct containing multiple copies of the CYP3A4 proximal PXRE ([ER₆]₂) (Lehmann et al., 1998). Figure 2 shows the -fold induction over control value for CPA (25-500 µM) with and without DEX (10 µM). Rifampin (10 µM) was included as a positive control. The high basal PXR activity is due to the fact that the ER₆ reporter plasmid that was used contains the Sv40 promoter, which is quite strong in the transfected cells. The results demonstrate that CPA produced a dose-dependent increase in PXR activation that was maximal at the highest CPA concentration studied (CPA 500 µM), and maximum activation observed was approximately 65% of rifampin activation. The addition of DEX to CPA resulted in only minor increases in PXR activation over CPA alone. PXR activation was approximately additive with respect to the effect of CPA and DEX alone. In separate experiments, CPA was more potent in activating PXR than other known inducers of CYP3A4 (DEX and phenytoin) when administered in equimolar amounts (data not shown). To our knowledge, this is the first report demonstrating that direct activation of PXR by CPA may be a possible molecular mechanism responsible for the autoinduction of CPA metabolism.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Effects of CPA on PXR activation in the absence and presence of DEX. HUH7 cells were cotransfected with expression plasmids for hPXR and the (CYP3A4 ER6)2-tk-luciferase reporter. Cells were treated for 24 h with each compound at final concentrations indicated, and cell extracts were subsequently assayed using the dual-luciferase reporter assay system as described under Materials and Methods. Data represent the mean ± S.E. from triplicate wells. *, p  0.05 compared with CPA-0 activity.

Concentration-Dependent Effects of Cyclophosphamide with and without Dexamethasone on CYP3A4 and CYP2B6 Activity and Protein Expression. To further characterize the apparent enhancement of CPA-mediated induction of CYP3A4 by DEX and to investigate potential inductive effects on CYP2B6, CPA alone and in combination with DEX (10 µM) was examined in three separate human hepatocyte preparations (HL092, HL093, HL100) over the range of CPA concentrations (25-750 µM) observed in patients receiving standard and high-dose CPA regimens. As observed in the prior experiments, CPA produced concentration-dependent increases in CYP3A4 activity that peaked at 250 µM and declined thereafter (Fig. 3A). DEX enhanced CPA induction of CYP3A4 activity, and this was particularly apparent at the low end of the CPA concentration range. CPA also produced concentration-dependent increases in CYP2B6 activity that peaked at 125 µM and were greater in magnitude than the effect of CPA on CYP3A4 activity (Fig. 3B). DEX enhanced the inductive effect of CPA on CYP2B6 activity and, again, this was particularly apparent at the low end of the CPA dose-response curve. DEX plus CPA resulted in greater -fold induction of CYP2B6 activity compared with CYP3A4. Corresponding increases in CYP3A4 and CYP2B6 immunoreactive protein as assessed by Western immunoblot and densitometric analysis of microsomes prepared from HL092 and HL100 confirmed CPA and CPA plus DEX induction of CYP3A4 and CYP2B6 activity (Figs. 4 and 5).

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Effect of CPA alone and in combination with DEX (10 µM) on CYP3A4 (3A) and CYP2B6 (3B) activity. Hepatocyte cultures were treated for 3 days with various concentrations of CPA (25-750 µM) alone or in combination with 10 µM DEX, as indicated under Materials and Methods. Microsomes were subsequently prepared and analyzed for CYP3A4 and CYP2B6 enzyme activities based on the 6-hydroxylation of testosterone and hydroxylation of bupropion, respectively. Data represent the mean ± S.E. from three separate preparations of hepatocytes (HL092, HL093, HL100). For comparison, dexamethasone (10 µM) alone produced a mean increase of 1.3 (S.D. = 0.11) and 1.1 (S.D. = 0.09) in CYP3A4 and CYP2B6 activity, respectively.

View larger version (40K): [in this window] [in a new window]   Fig. 4.   Effect of CPA alone and in combination with 10 µM DEX on the expression of CYP3A4 and CYP2B6 immunoreactive protein. Hepatocyte cultures from HL092 were treated for 3 days with solvent alone (control), 10 µM rifampicin (RIF), DEX (2, 10, and 50 µM), or various concentrations of CPA (25-750 µM) alone or in combination with 10 µM DEX. Microsomes were subsequently prepared and subjected to Western immunoblot and densitometric analysis as described under Materials and Methods.

View larger version (39K): [in this window] [in a new window]   Fig. 5.   Effect of CPA alone and in combination with 10 µM DEX on the expression of CYP3A4 and CYP2B6 immunoreactive protein. Hepatocyte cultures from HL100 were treated for 3 days with solvent alone (control), 10 µM rifampicin (RIF), DEX (10 and 50 µM), or various concentrations of CPA (25-750 µM) alone or in combination with 10 µM DEX. Microsomes were subsequently prepared and subjected to Western immunoblot and densitometric analysis as described under Materials and Methods.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Overall, these findings confirm and extend earlier reports showing CPA induction of CYP3A4 (Chang et al., 1997) and CYP2B6 (Gervot et al., 1999) in human hepatocyte to cultures. Specifically, CPA produced variable concentration-dependent changes in CYP3A4 activity that were inversely correlated with CYP3A4 baseline activity. CPA (250 µM) alone resulted in greater formation rates of 6-hydroxytestosterone than DEX (10 µM) alone in microsomes from human hepatocytes with baseline CYP3A4 activity <1500 pmol/min/mg of protein; however, CPA alone and DEX alone produced approximately equivalent increases in testosterone 6-hydroxylase activity in microsomes from hepatocytes with higher CYP3A4 baseline activity. Neither CPA nor DEX was as potent as the prototypical inducer rifampin at inducing CYP3A4 activity. At the concentrations tested, CPA produced greater increases in CYP2B6 than CYP3A4 activity in the three hepatocyte preparations studied. CPA appeared to reach maximum inductive effects on both CYP3A4 and CYP2B6 in concentrations in the range of 125 to 250 µM. Microsomal protein and lactate dehydrogenase activity remained relatively constant at high versus low CPA concentrations, indicating that a cytotoxic effect of higher concentrations of CPA in hepatocyte cell cultures was not the basis for this decline (data not shown). The importance of discordant effects of low and high CPA concentrations on CYP3A4 and CYP2B6 activity warrants further investigation as the concentration range investigated in cell culture was chosen to encompass the concentration range achieved in adult patients treated with standard and high-dose CPA regimens.

DEX enhanced the CYP3A4- and CYP2B6-inductive effect of CPA beyond that observed for CPA alone and DEX alone in microsomes prepared from cultured hepatocytes. However, in most cases, the increase in induction observed with CPA plus DEX was not additive. Increases in CYP3A4 activity with CPA and DEX were additive or synergistic in hepatocyte preparations with low baseline CYP3A4 activity. The amount of augmentation of enzyme activity was also concentration-dependent inasmuch as the greatest effects of DEX (10 µM) on both CYP3A4 and CYP2B6 activities were seen at the low end of the CPA range used for concentration-response experiments. DEX produced greater augmentation of the inductive effect of CPA on CYP2B6 compared with CYP3A4.

The results from the transient transfection assays show that CPA is capable of activating PXR, suggesting that this may be the predominant mechanism of the observed autoinduction exhibited by this drug in vivo. However, the dose response for PXR activation by CPA does not appear to be sufficient to account entirely for the extent of induction that is observed in primary hepatocyte cultures. This could be due to a number of factors including the involvement of other nuclear receptors or the formation of active metabolites of CPA that are more efficacious activators of PXR in primary cell culture systems. DEX enhanced the inductive effect of CPA far less in this model compared with human hepatocyte. Further research is required to fully understand the underlying mechanism(s) of CPA autoinduction in human liver.

The distinct dose-responses for CPA induction of 3A4 versus 2B6 may also be explained by the differential involvement of other nuclear receptors known to play a role in the regulation of these two genes, namely CAR and GR (Sueyoshi et al., 1999; Pascussi et al., 2000; Schuetz et al., 2000). Coregulation of CYP3A and CYP2B has been shown to occur by direct, but differential, binding of PXR and CAR to the same response elements in the CYP3A and CYP2B promoters (e.g., DR3/DR4) (Moore et al., 2000; Xie et al., 2000; Goodwin et al., 2001). Differential regulation of two separate genes can also occur if one gene is partially regulated by a repressor element that must first be overcome before up-regulation occurs. We would hypothesize that one of these, if not both, events is the predominant mechanism that determines the differences we observe. However, Pascussi et al. (2000) have shown that the glucocorticoid receptor can indirectly effect the regulation of these genes by modulating the levels of the receptors themselves. In addition, we and others have observed that the basal levels of 3A and 2B expression are differentially regulated by activation of the GR. For example, 3A basal expression is increased in a dose-dependent fashion by increasing levels of dexamethasone whereas basal levels of 2B6 are not. This, in effect, affects the overall -fold induction profile that is observed for each respective gene by inducers such as rifampicin. Thus, the regulation of these two genes is much more complicated than first meets the eye; the evidence suggests that PXR is a predominant factor in the regulation of both.

The importance of these findings to cancer patients receiving various doses of CPA is not clear. Although CYP3A4 activates CPA to its alkylating moiety, it has also recently been shown the P450 isoform responsible for its N-dechloroethylation and production of toxic byproducts (Roy et al., 1999). Therefore, induction of CYP3A4 may ultimately benefit or harm patients receiving this agent. Roy et al. (1999) demonstrated that CYP2B6 was the major CPA 4-hydroxylase isoform in a panel of 17 livers with an intrinsic clearance value 5 times that of CYP3A4. CYP2B6 expression is known to vary 20- to 250-fold in humans (Code et al., 1997; Ekins et al., 1998; Gervot et al., 1999; Stresser and Kupfer, 1999), and it is possible that patients with high CYP2B6 expression would be harmed by induction of CYP3A4 due to enhanced formation of toxic byproducts. A number of agents are available that inhibit the CYP3A4 pathway, and this strategy may ultimately prove useful in maximizing the therapeutic index of CPA. However, a large number of patients have low basal CYP2B6 activity, and the inductive effect of CPA, DEX, and other known enzyme inducers in patients is unknown. Patients with low CYP2B6 activity are therefore dependent on CYP3A4 and, to a lesser extent, CYP2C for activation of CPA to the 4-hydroxyCPA metabolite. In addition, it appears that treatment of patients receiving CPA with DEX would result in enhanced CPA induction of CYP2B6 and, to a lesser extent, CYP3A4 activity. This may favorably impact the metabolic disposition of CPA.