Epilepsia, 32(2):267-274. 1991 Raven Press, Ltd., New York 0 International League Against Epilepsy
Carbamazepine Dose Requirements During Stiripentol Therapy: Influence of Cytochrome P-450 Inhibition by Stiripentol Bradley M. Kerr, *Juan M. Martinez-Lage, *C. Viteri, ?Jacques Tor, $-A. Craig Eddy, and R e d H. Levy Departments of Pharmaceutics and Neurological Surgery, and $Department of Surgery, University of Washington, Seattle, Washington, U.S.A.; 'Department of Neurology, University of Navarra, Pamplona, Spain; and fLaboratoires Biocodex, Montrouge, France
Summary: The inhibitory effect of stiripentol (STP) on disposition of carbamazepine (CBZ) and carbamazepine10,ll-epoxide (CBZE) was quantitated to establish CBZ dosage reduction guidelines for future clinical add-on efficacy trials of STP. In seven epileptic patients, STP (1,50&3,000 mg/day for 2 weeks) inhibited CBZ clearance by 50 +. 16% (p = 0.001) and reduced the CBZEi CBZ plasma ratio by 45 +. 14% (p = 0.0005). The inhibitory effect was gradually manifested over a period of 7-10 days after initiation of STP therapy. In contrast to inhibition of CBZE formation, STP had no effect (p > 0.05) on elimination clearance or half-life (tli2)of CBZE
in six healthy volunteers. STP most likely exerts inhibitory effects through inhibition of cytochrome P-450. This hypothesis was confirmed in the present study by the finding that a therapeutic concentration of STP (7 pg/mL) inhibited l0,ll-epoxidation of CBZ in human liver microsomes by 40-50%. On the basis of results from this study, we propose that (a) CBZ dosage should be reduced in steps over a period of 7-10 days after initiation of STP, and (b) a CBZ dosage of 4.3 to 8.7 mg/kg/day will maintain therapeutic CBZ plasma levels of 5-10 pg/mL. Key Words: Carbamazepine-Carbamazepine epoxideCytochrome P-450-Stiripentol.
Stiripentol (STP); [(k)-l-(3,4-methylenedioxyphenyl)-4,4-dimethyl-l-penten-3-01]is a promising new anticonvulsant currently undergoing clinical trials in Europe and the United States. Initial studies in epileptic patients (Levy et al., 1984, 1985) showed that STP markedly reduces elimination clearances of several antiepileptic drugs (AEDs), including phenobarbital (PB), phenytoin (PHT), and carbamazepine (CBZ). These drug interactions are most likely attributable to an inhibition of cytochrome P-450 by stiripentol (Levy et al., 1984; Mesnil et al., 1988a,b). Drug interactions are known to hinder the usefulness of add-on trials for evaluating efficacy of AED candidates that inhibit cytochrome P-450, as shown previously with nafimidone (Treiman et al., 1985), verapamil (MacPhee et al., 1986), and diltiazem (Brodie and MacPhee, 1986). Similar problems with
drug interactions were anticipated in the design of clinical add-on trials of STP in patients receiving CBZ treatment (Rascol et al., 1989). The usefulness of add-on trials for evaluation of STP efficacy will be improved if the daily CBZ dose is reduced to minimize changes in plasma CBZ levels during STP treatment, but the manner in which CBZ dosage should be decreased has not been established. Future guidelines for adjusting CBZ dosage should be based on a detailed understanding of the inhibitory effect of STP. Therefore, the objectives of the present study were twofold: (a) to determine the extent and time course of the inhibitory effect of STP in epileptic patients receiving CBZ treatment; and (b) to elucidate the mechanism(s) by which STP alters plasma levels of CBZ and carbamazepine10,l I-epoxide (CBZE). METHODS
Received January 1990; revision accepted March 1990. Address correspondence and reprint requests to Dr. R . H. Levy at Department of Pharmaceutics, BG-20, University of Washington, Seattle, WA 98195, U.S.A.
Materials STP was a gift from Biocodex Laboratories (Paris, France). CBZ and CBZ-10,l l-transdihy267
B . M . KERR ET AL.
268
drodiol were gifts from Ciba-Geigy (Summit, NJ, U.S.A.). CBZE was synthesized from CBZ according to the procedure of Frigerio et al. (1972). Cyheptamide was purchased from Supelco (Bellefonte, PA, U.S.A.). NADPH (tetrasodium salt, type 1) was purchased from Sigma Chemical (St. Louis, MO, U.S.A.). High-performance liquid chromatography (HPLC) assays were performed with a Zorbax c8 column (DuPont, Wilmington, DE, U.S.A.). Ultraviolet detection was performed with a model 481 LC spectrophotometer (Waters Associates, Milford, MA, U.S.A.). STP/CBZ interaction in epileptic patients The effect of STP treatment on CBZ disposition was evaluated in seven adult epileptic patients after they had given informed consent. These clinical studies were performed at the University of Navarra, Pamplona, Spain. The CBZ monotherapy dosing regimen was unchanged for at least 2 weeks before STP was first administered. STP was initiated at a dose of 1,000-1,500 mg/day, and the final dose of 1,500-3,000 mgiday was attained in 10 days. The daily doses of STP and CBZ in the seven patients before and after 2 weeks of STP therapy are shown in Table 1. On the day before initiation of STP therapy, hourly blood samples were taken during a dosing interval of CBZ (control phase), with the first blood sample drawn just before CBZ administration. All subjects were instructed to empty their bladders immediately before CBZ intake, and urine was subsequently collected during the entire control phase dosing interval. After 2 weeks of STP treatment, hourly blood samples were again drawn and urine was collected during a CBZ dosing interval (STP treatment phase). Blood samples (8 ml) were anticoagulated with EDTA and centrifuged, and the plasma was collected and frozen at - 20°C until assayed for CBZ, CBZE, and STP. Urine volumes were measured, and aliquots were frozen at - 20°C TABLE 1. Daily doses of CBZ and STP ~
~~~
~
Control Phase
~~~
Stiripentol treatment phase
Patientlsexl weight (kg)
CBZ (mglday)
CBZ (mdday)
STP (mglday)
1lFl58 2lM165 3lFl60 4lFl66 5lFl50
800 800 1.200 1,400 800 1,000 1,200
800 800 600 1,000 400 600 800
1,500 1,500 3,000 3,000 2,000 2,000 2,000
6lF/50 7lFl60
CBZ, carbamazepine; STP, stiripentol. Epilepsia, Vol. 32, No. 2 , 1991
until assayed for CBZE and CBZ-l0,ll-transdihydrodiol. During the 2 weeks that elapsed between the two phases of the study, daily blood samples were drawn from two patients (patients 3 and4) for determination of trough plasma levels of CBZ, CBZE, and STP. CBZ and CBZE in plasma were simultaneously assayed by HPLC, with cyheptamide serving as internal standard. Plasma samples (100 p1) were extracted twice with n-hexane (2 X 2.5 mL) to remove STP selectively from the plasma, thereby avoiding chromatographic interference from STP. After the hexane was discarded, the compounds of interest were extracted from the plasma with 4 ml dichloromethane. Chromatographic separation was made at 50°C on a Zorbax c8 column ( 5 pm; 4.6 x 250 mm) with methanol/acetonitrile/water (42553, VOY vol/vol) as the mobile phase (flow rate 1.2 mumin). The compounds were detected by ultraviolet absorption at 210 nm. Retention times of CBZE, CBZ, and cyheptamide were 6.6, 11.4, and 16.2 min, respectively. CBZE in urine was assayed by a modified procedure in which urine samples were adjusted to pH 10 with 0.2 M carbonate buffer before extraction, with diethyl ether substituted for dichloromethane in the final step of the extraction. Previously published procedures were used for assay of CBZ-10,ll-transdihydrodiolin urine (Kerr et al., 1989) and STP in plasma (Lin and Levy, 1983). CBZ oral clearance (ClcBz) was estimated from the ratio of the daily dosing rate of CBZ to the average steady-state plasma concentration of CBZ determined during a single dosing interval. The molar plasma concentration ratio of CBZE to CBZ (CBZEICBZ) was calculated as the ratio of the areas under the plasma concentration-time curves measured over a single dosing interval. The fraction of CBZ metabolized to CBZE (fm) was calculated as the sum of the steady-state urinary excretion rates of CBZE and the 10,l l-transdihydrodiol metabolite (measured during a dosing interval) divided by the steady-state CBZ dosing rate. The formation clearance of CBZE (Cl,) was calculated by rnultiplication of ClCBZ and fm. The portion of ClCBZ devoted to elimination pathways other than formation of CBZE (CloTHER)was estimated as the difference between ClcBz and Cl,. Average STP plasma concentration during a dosing interval was calculated as the area under the plasma concentration-time plot divided by the time span of the dosing interval. STPKBZE interaction in normal volunteers The effect of STP on CBZE plasma levels in patients receiving CBZ is explained potentially as an
STIRIPENTOLICBZ INTERACTION
effect on epoxide formation andlor epoxide elimination. A method by which the effect on epoxide elimination clearance can be directly assessed is single-dose administration of CBZE (Kerr and Levy, 1989). In the present study, STP influence on disposition of an oral dose of CBZE was evaluated in six normal adult volunteers after they gave informed consent. CBZE was administered as a suspension in combination with antacid after an overnight fast, according to a previously described procedure (Kerr et al., 1989). Each volunteer ingested a 100-mg CBZE dose on two occasions, with a 1week interval allowed between the two doses. One CBZE dose served as a control, and the other CBZE dose was administered on the fourth day of a 7-day treatment regimen with STP 1,200 mg/day (600 mg twice daily). The order of the control and treatment phases was randomized. Blood samples (8 ml) were collected at times between 0 and 36 h after administration of CBZE. The blood samples were anticoagulated with EDTA, and the plasma was separated and frozen at -20°C until assayed for CBZE and STP. Plasma CBZE was assayed by means of a modified version of a published HPLC procedure (Kerr et al., 1989). The only modification was that plasma samples (250 p.L) were extracted twice with nhexane (2 x 2.5 ml) to remove STP selectively (thereby avoiding chromatographic interference) before extraction of CBZE from plasma with 5 ml dichloromethane. Assay of STP in plasma was performed as previously described (Lin and Levy, 1983). CBZE elimination half-life (tY2) was based on least-squares linear regression analysis of the log plasma concentration-time plot. Oral clearance of CBZE (ClcszE) was determined from the ratio of dose to area under the plasma concentration-time plot. Average STP plasma concentration during a dosing interval was calculated as the area under the plasma concentration-time plot divided by the time span of the dosing interval. Inhibition of CBZ metabolism by STP in microsomal preparations The effect of STP on cytochrome P-450-mediated conversion of CBZ to the 10,ll-epoxide metabolite was investigated in preparations of human liver microsomes. Samples of human liver tissue were obtained from organ transplant donors at Harborview Medical Center and University Hospital in Seattle, Washington. Microsomal incubations were performed in a 0.1 M phosphate buffer at a pH of 7.4. CBZ, STP, and NADPH were all dissolved in phosphate buffer without use of organic solvents before
269
addition to the microsomal incubations. Microsomes were preincubated with STP or blank buffer at 37°C for 90 s with NADPH added. After the preincubation period, CBZ was added to the incubations and the reaction was continued for 15 min at 37°C. The final volume of each incubation was I ml, protein concentration was 1 mg/ml, and the concentration of NADPH was 1 mM; concentration of CBZ was 14 pglml (preliminary studies showed no signs of enzyme saturation in the presence of CBZ concentrations S35 p.g/ml), and STP concentration was 0 or 7 p.g/ml. Enzymatic epoxidation of CBZ was terminated after 15 min by rapidly vortexing the 1-ml incubation with 4 ml dichloromethane. The internal standard cyheptamide (375 ng in 75 p.1 methanol) was then added to each sample. CBZE was extracted from the incubations and assayed by HPLC. Standards were prepared by adding appropriate amounts of phosphate buffer, fresh microsomal protein, dichloromethane, and internal standard to tubes containing 0-400 ng CBZE. After samples and standards were vortexed and centrifuged, the dichloromethane layer was transferred to a clean glass tube and evaporated to dryness. The residue was reconstituted with mobile phase, and chromatographic separation was made at room temperature on a Zorbax C, column (5 km; 4.6 x 250 mm) with acetonitrile/water (40:60, vol/ vol) as the mobile phase (flow rate 1.2 mlimin). The compounds were detected by ultraviolet absorption at 210 nm. Retention times of CBZE and cyheptamide were 4.7 and 8.6 min, respectively. RESULTS STP/CBZ interaction The daily dosing of CBZ was reduced in five of seven patients during STP therapy to avoid intoxication with CBZ (Table 1). After 2-week STP treatment, the average c1C.z in seven epileptic patients was reduced by an overall average of 50% (Table 2). The reduction of ClcBz was >50% in five patients (range 53-67%), but was only 27 and 29% in the other two patients. The mean CBZE/CBZ plasma concentration ratio was reduced by STP from 0.128 to 0.0696, an average decrease of -45% (Table 2). The extent of reduction in the CBZEiCBZ plasma ratio showed a strong tendency to parallel the extent of decrease in ClcBz; the 2 patients in whom ClCBZ was least inhibited (27 and 29% inhibition) also had the smallest reductions in the CBZE/CBZ plasma ratio (31 and 32% reduction), whereas 3 of the 5 patients in whom Clcsz was inhibited by 50-70% also had reductions in the CBZE/CBZ Epilepsia, Vol. 32, N o . 2 , 1991
B . M . KERR ET AL.
2 70
TABLE 2. Effect of STP on CBZ clearance (elcBz) and on plasma concentration ratio CBZEICBZ Clcsz (ml/min/kg) Patient
Control
1
0.930 1.43 1.59 1.37 0.924 1.31 1.17 1.25 0.252 p
2 3 4 5 6 7 Mean SD Paired t test
CBZE/CBZ
STP
=
Change (%)
Control
- 27 - 55 - 67 - 53 - 56 - 62 - 29 - 50 16
0.683 0.637 0.523 0.647 0.408 0.504 0.836 0.605 0.140 0.0011
STP
0.100 0.0686 0.129 0.0824 0.158 0.0613 0.128 0.0551 0.0967 0.0601 0.114 0.0446 0.170 0.115 0.128 0.0696 0.0278 0.0232 p = 0.0005
Average Plasma STP Change (%)
(pdml)
-31 - 36 - 61 - 57 - 38 - 61 - 32 - 45 14
1.45 1.16 6.63 2.02 6.06 10.5 4.76 4.65 3.40
CBZ clearance, ClCBz;plasma concentration ratio of CBZ epoxide to CBZ (CBZEKBZ); other abbreviations as in Table 1.
plasma ratio of -60% (Table 2). The effects of STP on plasma levels of CBZ and CBZE were similar to the effects noted previously for other inhibitors of cytochrome P-450, such as verapamil (MacPhee et al., 1986), propoxyphene (Dam et al., 1980), erythromycin (Barzaghi et al., 1987), and danazol (Kramer et al., 1986). Urinary excretion rates of the transdihydrodiol metabolite and unchanged CBZE indicated that the fraction of CBZ biotransformed to the 10 ,llepoxide metabolite (fm) was not significantly altered by STP (Table 3; control phase fm = 0.41 ? 0.14, STP phase fm = 0.39 ? 0.11). The pronounced inhibition of ClcBz in conjunction with no change in fm suggests that STP inhibits formation of CBZE (c1F) as well as alternative metabolic pathways of CBZ (Cl,,,,,). Indeed, both c1F and CloTHERwere reduced by -50% during treatment with STP (Table 3). The time-course studies showed that the ratio of CBZ dosing rate to trough plasma CBZ level (i.e., crude estimate of clearance under steady-state conditions) decreased 1-4 days after initiation of STP administration and became stabilized in 7-10 days
(Fig. 1A). The CBZE/CBZ plasma concentration ratio was decreased in similar fashion, requiring an equilibration period of 7-10 days (Fig. 1B). The CBZ doselplasma level ratio and the CBZE/CBZ plasma concentration ratio both decreased markedly when trough plasma concentrations of STP increased from 0 to 1 p,g/ml (Fig. 2), and maximal inhibition was observed at a somewhat higher concentration of STP. STP/CBZE interaction Treatment with STP had no significant effect on the tf or clearance of CBZE after single-dose administration to healthy volunteers (Table 4). The dose of STP in the CBZE interaction study (1,200 mg/day) was somewhat less than doses of STP that caused inhibition of CBZ clearance in the epileptic patients (1,500-3,000 mg/day). Similarly, the average plasma levels of STP in the CBZE single dose study (0.9-3.3 p,g/ml, Table 4) also tended to be less than STP plasma levels in the CBZ interaction study (1.2-10.5 p,g/ml, Table 2). However, STP is clearly a much more potent inhibitor of the metabolism of CBZ than of CBZE. Among the individuals
TABLE 3 . Effect of STP on f m , Cl, , and Cl,,, fm
C1, (mVmin/kg)
Cb-H,R (ml/minW
Patient
Control
STP
Control
STP
Control
STP
1
0.390 0.322 0.492 0.328 0.222 0.527 0.616 0.414 0.137
0.394 0.550 0.270 0.280 0.423 0.448 0.394 0.106
0.363 0.460 0.782 0.449 0.205 0.690 0.721 0.524 0.212
0.251 0.288 0.175 0.114 0.213 0.375 0.236 0.091
0.567 0.970 0.808 0.921 0.719 0.620 0.449 0.722 0.190
-a 0.386 0.235 0.472 0.294 0.291 0.461 0.356 0.098
2 3 4 5 6 7 Mean SD Paired t test
p
=
0.59
p = 0.0043
p = 0.0073
~~
fm, fraction of CBZ metabolized to CBZE; CI,, formation clearance of CBZE; ClOTHER, alternate CBZ clearance pathways; other abbreviations as in Table 1. a Incomplete urine collection. Epilepsia, Vol. 32, N o . 2 , 1991
2 71
STIRIPENTOLICBZ INTERACTION
1 1
.
3
.
140
j
120
100
n
P 0
(A)
0
120
Q)
x
x x
8o
.H
140
60 40
20
0
1
3
5
7
9
1 1 1 3 1 5
0
1
3
4
5
6
7
STP Plasma Level (pglmL)
Day
-
2
t i
0.02 0 1
3
5
7
9
1 1 1 3 1 5
Day FIG. 1. Time course of change in carbamazepine (CBZ) daily doselplasma level ratio (i-e., estimate of CBZ clearance) (A), and CBZ epoxide/CBZ plasma ratio (CBZE/CBZ)(B),after initiation of stiripentol therapy on day 1. Plasma drug concentrations (trough levels) were determined daily in patients 3 (0)and 4 (X).
in whom average STP plasma levels were in the range of 1-4 pg/ml, CBZ clearance was inhibited 27-55% (three epileptic patients, Table 2), whereas no consistent change was observed in clearance of CBZE (four volunteers, Table 4). The lack of effect of STP on the elimination clearance of CBZE indicates that the decreased CBZE/CBZ plasma ratio in epileptic patients during STP therapy (Table 2) results principally if not entirely from inhibition of the epoxide formation clearance. Inhibition of CBZ metabolism in microsomal preparations The 10,ll -epoxidation of CBZ by cytochrome P450 was significantly inhibited in the presence of a therapeutic concentration of STP (7 pg/ml) in four preparations of human liver microsomes (Table 5). The extent of inhibition in microsomes (-40-50% inhibition) was in excellent agreement with the extent of inhibition of CBZ metabolism in vivo in epileptic patients during STP therapy (Fig. 3). These
0.02 00
1
2
3
4
5
6
7
STP Plasma Level (pglmL) FIG. 2. Relationship between plasma concentration of stiripentol (STP)and carbamazepine (CBZ) daily dose/plasrna level ratio (i-e., estimate of CBZ clearance) (A), and carbarnazepine epoxide/CBZ plasma ratio (CBZE/CBZ) (B). Plasma drug concentrations (trough levels) were determined daily in patients 3 (0)and 4 (X).
results provide the first direct in vitro confirmation of the hypothesis (Levy et al., 1984; Moreland et al., 1986) that STP inhibits in vivo elimination of other AEDs by inhibiting cytochrome P-450 activity. DISCUSSION
The first evidence of the CBZ/STP drug interaction was noted during early clinical trials of STP, at which time STP markedly inhibited clearance of CBZ in a single patient (Levy et al., 1984). Results from the present study, as well as those of an earlier pilot investigation (Levy et al., 1985), confirm the pronounced in vivo inhibitory effects of STP. The mechanism by which STP inhibits CBZ clearance most likely involves formation of an inhibitory complex between the methylenedioxyphenyl moiety of stiripentol and cytochrome P-450(Levy et al., 1984; Mesnil et al., 1988a,b). In the present study, the ability of STP to inhibit cytochrome P-450 was confirmed both in vivo and in vitro by the following Epilepsia, Vol. 32, N o . 2 , 1991
272
B . M . KERR ET AL. TABLE 4. Effect of STP on Cl,, CICBzE(rnllminlkg)
Volunteerlsed weight (kg)
Control
I IF164
1.56 1.93 1.94 2.05 I .43 1.75 1.78 0.242
2lFl49 3lMl65 4lFl60 5lMl74 6lMl65
Mean SD Paired t test
p
=
and CBZEtII, CBZE ti/* (h)
STP
Change (%)
1.92 1.74 2.05 2.13 1.69 2.32 1.98 0.240
+ 23 -9.8 + 5.7 +3.9 + 18 + 33 + 12 15
0.122
Control
STP
5.54 6.35 6.60 5. I 6 6.10 5.59 5.89 0.549
p
Average plasma STP
=
Change (%)
(PLg/ml)
+ 12
6.21 5.93 5.66 4.89 6.10 5.1 1 5.65 0.541
1.36 3.27 0.88 I .65 0.88 I .30 1.56 0.89
- 6.6 - 14 - 5.2
0
- 8.6 - 3.7 9.0
0.327
ClcszE, oral clearance of CBZE; CBZE tllZ,half-life of CBZE; other abbreviations as in Table 1.
evidence: (a) The CBZE/CBZ plasma concentration ratio was substantially reduced during STP therapy; (b) the in vivo C1, of CBZE was inhibited by stiripentol; and (c) a therapeutic concentration of STP (7 pg/ml) inhibited the cytochrome P-450dependent conversion of CBZ to the 10,ll-epoxide in human liver microsomes. In addition to the effects on 10,ll-epoxide formation, STP also markedly inhibited alternative pathways of CBZ metabolism in vivo (CloTHER),which is consistent with inhibition of aromatic hydroxylation pathways. In contrast to the inhibitory effects toward oxidative metabolism, STP had no effect on in vivo elimination clearance of CBZE in healthy volunteers because elimination of CBZE is almost entirely mediated by microsomal epoxide hydrolase rather than cytochrome P-450 (Kerr et al., 1989). The extensive inhibition of CBZ clearance by STP clearly indicates that dosage adjustments are needed to avoid intoxication with CBZ after initiation of STP therapy. The average CBZ clearance of 0.605 ml/min/kg during the STP treatment phase indicates that a CBZ dose of 8.7 mg/kg/day should yield an average CBZ plasma level of 10 pg/ml. Therefore, therapeutic plasma levels of CBZ (5-10 pg/ml) during STP therapy should be achieved at CBZ dosages of 300-600 mg/day for an average size adult (70 kg), 200-400 mg/day for a small person (45
kg), and 400-800 mg/day for a large nonobese person (95 kg). Plasma levels of the pharmacologically active metabolite CBZE were sufficiently low in the present study for the epoxide to have no significant influence on CBZ dose-adjustment proposals. The time course of the STP/CBZ interaction in the present study showed that inhibition of CBZ clearance was detectable as early as the second or third day of STP therapy and that the inhibitory effect gradually became more pronounced in a 7- to 10-day period. To minimize changes in CBZ plasma levels, CBZ dose reductions should also be made gradually, with the initial dose reduction made on the third day of STP therapy and the final dose reduction made not later than 10 days after STP administration is started. The CBZ dosage should be reduced in steps if possible, because too rapid a dosage reduction may cause CBZ plasma levels to decrease temporarily to less than therapeutic levels
IN VITRO Mlcroromor
IN VlVO C B t Cloannco
IN VIVO EpoxldoICBZ
T
TABLE 5. Inhibition of biotransformation of CBZ" to CBZE by STP a in human liver microsames CBZE formation rateb (pmol x min-' mg protein-') Liver No. 102 103 105 108 a
Control 55.4 67.6 51.3 79.0
-t
1.5
2 0.6 -t -t
1.6 1.8
STP 30.8 2 1.5 34.0 2 1.9 31.1 ? 1.0 41.8 0.9
*
CBZ 14 pglml; STP = 7 pglml. Mean ? SD of quadruplicate incubations. Pooled t test, p < 0.001 in each liver.
Epilepsia, Vol. 32, N o . 2, 1991
Change
(%r
- 44% - 50% - 39%
- 47%
FIG. 3. Inhibitory effect of stiripentol (STP) on 1 0 , l l epoxidation of carbarnazepine (CBZ) in rnicrosomes(in vitro; STP 7 pglml) and effect on CBZ clearance and CBZ epoxidei CBZ plasma ratio in epileptic patients (in vivo; STP = 4.6 5 3.4 pglml). Plotted values are derived from data in Tables 2 and 5. The effect of stiripentol was statistically significant both in vitro and in vivo (p < 0.005).
2 73
STIRIPENTOLICBZ INTERACTION and possibly leave the patient unprotected against seizures. Complete discontinuation of STP from the treatment regimen is likely to result in a gradual reversal of the inhibitory effect, and a compensatory increase in the dosage of CBZ will probably be required; the time course of change in CBZ clearance after discontinuation of STP will be addressed in future studies. The inhibitory effect of STP may raise questions about the safety of STP polytherapy in future treatment of epilepsy. Clinically significant drug interactions involving enzyme induction, inhibition, or both are already common among the currently available AEDs (Pitlick, 1989), and inhibition of cytochrome P-450 by STP will further complicate this situation. Nevertheless, several benefits may be derived from intentional use of STP combination therapy; e.g., the ability of STP to decrease the CBZE/ CBZ plasma ratio (Table 2) indicates that neurologic side effects associated with CBZE (Schoeman et al., 1984) would be minimized during STP therapy. Perhaps more significant, the decreased CBZElCBZ plasma ratio suggests that STP may likewise decrease in vivo exposure to chemically reactive epoxide metabolites (arene oxides) of concurrently administered drugs. Arene oxide metabolites have been implicated as a cause of the infrequent but serious toxicities associated with AEDs, such as teratogenicity (Lindhout et al., I984), hepatitis (Spielberg et al., 1981), and blood dyscrasias (Gerson et al., 1983). The most notable AEDrelated toxicity attributed to an arene oxide metabolite is the teratogenic effect of phenytoin (fetal hydantoin syndrome) (Strickler et al., 1985). If STP itself is free of serious side effects, combination therapy with STP may reduce in vivo exposure to arene oxides and thereby protect against the severe toxicities of other AEDs. Additional benefits of STP polytherapy would probably include a substantially lower daily dose requirement and reduced frequency of administration for concomitant AEDs (Lockard and Levy, 1988; Levy and Kerr, 1989), but any benefit that might be associated with STP polytherapy will be achieved only if AED interactions with STP are carefully controlled. The inhibitory effect of STP was quantitated in some detail in the present study to develop approaches for managing AED interactions that might complicate the clinical efficacy trials of STP. Although the inhibitory effect of STP is pronounced, the present study documents that the time course and extent of inhibition show a surprising degree of intersubject consistency. These results indicate that the CBZlSTP interaction is highly predictable and reproducible and suggest that a standardized dose
adjustment procedure should successfully maintain CBZ plasma levels within therapeutic limits in most patients after initiation of STP therapy. Acknowledgment: We thank Dr. Milo Gibaldi, University of Washington, Seattle, WA, for helpful discussions and valuable suggestions concerning interpretations of study results, and Brian Bainbridge and Martha Moore for technical assistance. This work was supported by a research grant from Biocodex Laboratories, Montrouge, France, and by Grants No. PO1 NF17111 and T32 NS07289 from the National Institutes of Health.
REFERENCES Barzaghi N , Gatti G, Crema F, et al. Inhibition by erythromycin of the conversion of carbamazepine to its active 10,llepoxide metabolite. Er J Clin Pharmacol 1987;24:836-8. Brodie MJ, MacPhee GJ. Carbamazepine neurotoxicity precipitated by diltiazem. Er Med J 1986;292:1170-1. Dam M, Christensen JM, Brandt J, et al. Antiepileptic drugs: interaction with dextropropoxyphene. In: Johannessen SI, Morselli PL, Pippenger CE, Richens A, Schmidt D, Meinardi H, eds. Antiepileptic therapy: advances in drug monitoring. New York: Raven Press, 1980:29!%306. Frigerio A, Fanelli R, Biandrate P, et al. Mass spectrometric characterization of carbamazepine-l0,l 1-epoxide, a carbamazepine metabolite isolated from human urine. J Pharm Sci 1972;61:1144-7. Gerson GT, Fine DG, Spielberg SP, Sensenbrenner LL. Anticonvulsant-induced aplastic anemia: increased susceptibility to toxic drug metabolites in vitro. Blood 1983;61:889-93. Kerr BM, Levy RH. Carbamazepine epoxide. In: Levy R, Mattson R, Meldrum B, Penry JK, Dreifuss FE, eds. Antiepileptic drugs, 3rd ed. New York: Raven Press, 1989505-20. Kerr BM, Rettie AE, Eddy AC, et al. Inhibition of human liver microsomal epoxide hydrolase by valproate and valpromide: in v i t r o h vivo correlation. Clin Pharmacol Ther 1989;46:8293. Kramer G ,Theisohn M, von Unruh GE, Eichelbaum M. Carbamazepine-danazol drug interaction: its mechanism examined by a stable isotope technique. Ther Drug Monit 1986;8: 387-92. Levy RH, Loiseau P, Guyot M, et al. Stiripentol kinetics in epilepsy: nonlinearity and interactions. Ciin Pharmacol Ther 1984;36:661-9. Levy RH, Martinez-Lage JM, Tor J , et al. Stiripentol level-dose relationship and interaction with carbamazepine in epileptic patients. Epilepsia 1985;26:544-5. Levy RH, Kerr BM. Pharmacokinetics of old, new, and yetto-be-discovered antiepileptic drugs. Epiiepsia 1989;30 (SUPPIl):S35-S41. Lin H, Levy RH. Pharmacokinetic profile of a new anticonvulsant, stiripentol, in the rhesus monkey. Epilepsia 1983;24: 692-702. Lindhout D, Hoppener RJEA, Meinardi H. Teratogenicity of antiepileptic drug combinations with special emphasis on epoxidation (of carbamazepine). Epilepsia 1984;25:77-83. Lockard JS, Levy RH. Carbamazepine plus stiripentol: is polytherapy by design possible? Epilepsia 1988;29:476-81. MacPhee GJA, McInnes GT,Thompson GG,Brodie MJ. Verapamil potentiates carbamazepine neurotoxicity: a clinically important inhibitory interaction. Lancet 1986;1:70&3. Mesnil M, Testa B, Jenner P. In vitro inhibition by stiripentol of rat brain cytochrome P-450-mediated naphthalene hydroxyl18:1097-106. ation. Xenobiotica 1988~; Mesnil M, Testa B, Jenner P. Ex vivo inhibition of rat brain cytochrome P-450 activity by stiripentol. Eiochem Pharmacol 19886;37:3639-22. Epilepsia, Vol. 32, N o . 2, 1991
274
B . M . KERR ET AL.
Moreland TA, Astoin J, Lepage F, et al. The metabolic fate of stiripentol in man. Drug Metab Dispos 1986;14:654-62. Pitlick WH, ed. Antiepileptic drug interactions. New York: Demos Publications, 1989. Rascol 0, Squalli A, Montastruc JL, et al. A pilot study of stiripentol, a new anticonvulsant drug, in complex partial seizures uncontrolled by carbamazepine. Clin Neuropharmacol 1989;12:119-23. Schoeman JF, Elyas AA, Brett EM, Lascelles PT. Correlation between plasma carbamazepine- 10,l I-epoxide concentration and drug side-effects in children with epilepsy. Dev Med Child Neurol 1984;26:756-64. Spielberg SP, Gordon GB, Blake DA, et al. Predisposition to phenytoin hepatotoxicity assessed in vitro. N Eng J Med 1981;305~722-7. Strickler SM, Dansky LV, Miller MA, et al. Genetic predisposition to phenytoin-induced birth defects. Lancet 1985;2: 746-9. Treiman DM, Wilensky AJ, Ben-Menachem E, et al. Efficacy of nafimidone in the treatment of intractable partial seizures: report of a two-center study. Epilepsia 1985;26:607-11.
RJ?SUME L’effet inhibiteur du Stiripentol (STP) sur la disponibilitt de la carbamaztpine (CBZ) et du 10,l I-epoxide de carbamaztpine (CBZE) a t t t quantifiC dans le but d’Ctablir des recommandations pour la diminution des doses de CBZ dans les Ctudes cliniques futures d’adjonction du Stiripentol. Chez 7 patients tpileptiques, le STP (1500 B 3000 mg/j pendant deux semaines) a inhibt la clearance de CBZ de 50 ? 16% (p = 0.001) et a rtduit 14% (p = 0.0005). le rapport plamatique CBZE/CBZ de 45 L’effet inhibiteur s’est manifeste progressivement sur une pCriode de 7 ti 10jours aprts mise en route du STP. Contrairement a [’inhibition de la formation de CBZE, le STP n’a pas eu d’effet sur la clearance ou la demi-vie du CBZE chez 6 volontaires sains (p > 0.05). Le STP exerce trts vraisemblablement ses effets inhibiteurs par le biais d’une inhibition du cytochrome P-450. Cette
*
Epilepsia, Vol. 32, N o . 2,1991
hypothtse a Ctt confirmte dans ce travail par la constatation d’une inhibition par le STP a concentration thtrapeutique (7 mg/ ml) de la 10,11-epoxidation de la CBZ dans les microsomes hCpatiques humains, de 40 B 50%. Sur la base des resultats de cette ttude, les auteurs proposent que: 1”)la dose de CBZ soit rtduite par ttapes sur une ptriode de 7 B 10jours aprts mise en route du STP, et que: 2 7 la dose de CBZ de 4.3 a 8.7 mglkg par jour devrait sufire a maintenir des taux plasmatiques de CBZ thtrapeutiques, compris entre 5 et 10 mg/ml. (P. Genton, Marseille)
ZUSAMMENFASSUNG Die inhibitorische Wirkung von Stiripentol (STP) auf Carbamazepin (CBZ) und Carbamazepin-lO,1 I-Epoxyd (CBZE) wurde quantifiziert, um Richtlinien zur Carbamazepin-Dosisreduktion fur eine zukunftige klinische add-on-Wirkungsuntersuchung von Stiripentol zu erstellen. Bei sieben epileptischen Patienten hemmte Stiripentol (1,500-3,000 mg/Tag iiber 2 Wochen) die CBZ-Clearance um 50 + 16% (p = 0,001) und reduzierte die CBZE/CBZ-Plasma-Relation um 45 + 14% (p = 0.0005). Die inhibitorische Wirkung stellte sich schrittweise uber eine Periode von 7 bis 10 Tagen nach Beginn der STP-Therapie ein. Im Gegensatz zur Hemmung der CBZE-Bildung hatte STP keine Wirkung (p > 0.05) auf die Eliminations-Clearance oder Halbwertszeit von CBZE bei 6 gesunden Freiwilligen. STP entwickelt sehr wahrscheinlich seine inhibitorische Wirkung uber eine Hemmung von Zytochrom-P-450. Diese Hypothese wurde in der vorliegenden Studie durch den Befund bestatigt, daR eine therapeutische Konzentration von STP (7 pg/ml) die 10,llEpoxydation von CBZ in menschlichen Lebermikrosomen um 40 bis 50% inhibierte. Auf der Grundlage dieser Ergebnisse wird vorgeschlagen, daR 1. die CBZ-Dosis nach Beginn einer STPTherapie stufenweise uber 7 bis 10 Tage reduziert wird und 2. eine CBZ-Dosis von 4.3 bis 8.7 mg/kg/Tag fur therapeutische CBZ-Plasmaspiegel von 5 bis 10 mglml ausreichend ist. (C. Benninger, Heidelberg)
Carbamazepine Dose Requirements During Stiripentol Therapy: Influence of Cytochrome P-450 Inhibition by Stiripentol Bradley M. Kerr, *Juan M. Martinez-Lage, *C. Viteri, ?Jacques Tor, $-A. Craig Eddy, and R e d H. Levy Departments of Pharmaceutics and Neurological Surgery, and $Department of Surgery, University of Washington, Seattle, Washington, U.S.A.; 'Department of Neurology, University of Navarra, Pamplona, Spain; and fLaboratoires Biocodex, Montrouge, France
Summary: The inhibitory effect of stiripentol (STP) on disposition of carbamazepine (CBZ) and carbamazepine10,ll-epoxide (CBZE) was quantitated to establish CBZ dosage reduction guidelines for future clinical add-on efficacy trials of STP. In seven epileptic patients, STP (1,50&3,000 mg/day for 2 weeks) inhibited CBZ clearance by 50 +. 16% (p = 0.001) and reduced the CBZEi CBZ plasma ratio by 45 +. 14% (p = 0.0005). The inhibitory effect was gradually manifested over a period of 7-10 days after initiation of STP therapy. In contrast to inhibition of CBZE formation, STP had no effect (p > 0.05) on elimination clearance or half-life (tli2)of CBZE
in six healthy volunteers. STP most likely exerts inhibitory effects through inhibition of cytochrome P-450. This hypothesis was confirmed in the present study by the finding that a therapeutic concentration of STP (7 pg/mL) inhibited l0,ll-epoxidation of CBZ in human liver microsomes by 40-50%. On the basis of results from this study, we propose that (a) CBZ dosage should be reduced in steps over a period of 7-10 days after initiation of STP, and (b) a CBZ dosage of 4.3 to 8.7 mg/kg/day will maintain therapeutic CBZ plasma levels of 5-10 pg/mL. Key Words: Carbamazepine-Carbamazepine epoxideCytochrome P-450-Stiripentol.
Stiripentol (STP); [(k)-l-(3,4-methylenedioxyphenyl)-4,4-dimethyl-l-penten-3-01]is a promising new anticonvulsant currently undergoing clinical trials in Europe and the United States. Initial studies in epileptic patients (Levy et al., 1984, 1985) showed that STP markedly reduces elimination clearances of several antiepileptic drugs (AEDs), including phenobarbital (PB), phenytoin (PHT), and carbamazepine (CBZ). These drug interactions are most likely attributable to an inhibition of cytochrome P-450 by stiripentol (Levy et al., 1984; Mesnil et al., 1988a,b). Drug interactions are known to hinder the usefulness of add-on trials for evaluating efficacy of AED candidates that inhibit cytochrome P-450, as shown previously with nafimidone (Treiman et al., 1985), verapamil (MacPhee et al., 1986), and diltiazem (Brodie and MacPhee, 1986). Similar problems with
drug interactions were anticipated in the design of clinical add-on trials of STP in patients receiving CBZ treatment (Rascol et al., 1989). The usefulness of add-on trials for evaluation of STP efficacy will be improved if the daily CBZ dose is reduced to minimize changes in plasma CBZ levels during STP treatment, but the manner in which CBZ dosage should be decreased has not been established. Future guidelines for adjusting CBZ dosage should be based on a detailed understanding of the inhibitory effect of STP. Therefore, the objectives of the present study were twofold: (a) to determine the extent and time course of the inhibitory effect of STP in epileptic patients receiving CBZ treatment; and (b) to elucidate the mechanism(s) by which STP alters plasma levels of CBZ and carbamazepine10,l I-epoxide (CBZE). METHODS
Received January 1990; revision accepted March 1990. Address correspondence and reprint requests to Dr. R . H. Levy at Department of Pharmaceutics, BG-20, University of Washington, Seattle, WA 98195, U.S.A.
Materials STP was a gift from Biocodex Laboratories (Paris, France). CBZ and CBZ-10,l l-transdihy267
B . M . KERR ET AL.
268
drodiol were gifts from Ciba-Geigy (Summit, NJ, U.S.A.). CBZE was synthesized from CBZ according to the procedure of Frigerio et al. (1972). Cyheptamide was purchased from Supelco (Bellefonte, PA, U.S.A.). NADPH (tetrasodium salt, type 1) was purchased from Sigma Chemical (St. Louis, MO, U.S.A.). High-performance liquid chromatography (HPLC) assays were performed with a Zorbax c8 column (DuPont, Wilmington, DE, U.S.A.). Ultraviolet detection was performed with a model 481 LC spectrophotometer (Waters Associates, Milford, MA, U.S.A.). STP/CBZ interaction in epileptic patients The effect of STP treatment on CBZ disposition was evaluated in seven adult epileptic patients after they had given informed consent. These clinical studies were performed at the University of Navarra, Pamplona, Spain. The CBZ monotherapy dosing regimen was unchanged for at least 2 weeks before STP was first administered. STP was initiated at a dose of 1,000-1,500 mg/day, and the final dose of 1,500-3,000 mgiday was attained in 10 days. The daily doses of STP and CBZ in the seven patients before and after 2 weeks of STP therapy are shown in Table 1. On the day before initiation of STP therapy, hourly blood samples were taken during a dosing interval of CBZ (control phase), with the first blood sample drawn just before CBZ administration. All subjects were instructed to empty their bladders immediately before CBZ intake, and urine was subsequently collected during the entire control phase dosing interval. After 2 weeks of STP treatment, hourly blood samples were again drawn and urine was collected during a CBZ dosing interval (STP treatment phase). Blood samples (8 ml) were anticoagulated with EDTA and centrifuged, and the plasma was collected and frozen at - 20°C until assayed for CBZ, CBZE, and STP. Urine volumes were measured, and aliquots were frozen at - 20°C TABLE 1. Daily doses of CBZ and STP ~
~~~
~
Control Phase
~~~
Stiripentol treatment phase
Patientlsexl weight (kg)
CBZ (mglday)
CBZ (mdday)
STP (mglday)
1lFl58 2lM165 3lFl60 4lFl66 5lFl50
800 800 1.200 1,400 800 1,000 1,200
800 800 600 1,000 400 600 800
1,500 1,500 3,000 3,000 2,000 2,000 2,000
6lF/50 7lFl60
CBZ, carbamazepine; STP, stiripentol. Epilepsia, Vol. 32, No. 2 , 1991
until assayed for CBZE and CBZ-l0,ll-transdihydrodiol. During the 2 weeks that elapsed between the two phases of the study, daily blood samples were drawn from two patients (patients 3 and4) for determination of trough plasma levels of CBZ, CBZE, and STP. CBZ and CBZE in plasma were simultaneously assayed by HPLC, with cyheptamide serving as internal standard. Plasma samples (100 p1) were extracted twice with n-hexane (2 X 2.5 mL) to remove STP selectively from the plasma, thereby avoiding chromatographic interference from STP. After the hexane was discarded, the compounds of interest were extracted from the plasma with 4 ml dichloromethane. Chromatographic separation was made at 50°C on a Zorbax c8 column ( 5 pm; 4.6 x 250 mm) with methanol/acetonitrile/water (42553, VOY vol/vol) as the mobile phase (flow rate 1.2 mumin). The compounds were detected by ultraviolet absorption at 210 nm. Retention times of CBZE, CBZ, and cyheptamide were 6.6, 11.4, and 16.2 min, respectively. CBZE in urine was assayed by a modified procedure in which urine samples were adjusted to pH 10 with 0.2 M carbonate buffer before extraction, with diethyl ether substituted for dichloromethane in the final step of the extraction. Previously published procedures were used for assay of CBZ-10,ll-transdihydrodiolin urine (Kerr et al., 1989) and STP in plasma (Lin and Levy, 1983). CBZ oral clearance (ClcBz) was estimated from the ratio of the daily dosing rate of CBZ to the average steady-state plasma concentration of CBZ determined during a single dosing interval. The molar plasma concentration ratio of CBZE to CBZ (CBZEICBZ) was calculated as the ratio of the areas under the plasma concentration-time curves measured over a single dosing interval. The fraction of CBZ metabolized to CBZE (fm) was calculated as the sum of the steady-state urinary excretion rates of CBZE and the 10,l l-transdihydrodiol metabolite (measured during a dosing interval) divided by the steady-state CBZ dosing rate. The formation clearance of CBZE (Cl,) was calculated by rnultiplication of ClCBZ and fm. The portion of ClCBZ devoted to elimination pathways other than formation of CBZE (CloTHER)was estimated as the difference between ClcBz and Cl,. Average STP plasma concentration during a dosing interval was calculated as the area under the plasma concentration-time plot divided by the time span of the dosing interval. STPKBZE interaction in normal volunteers The effect of STP on CBZE plasma levels in patients receiving CBZ is explained potentially as an
STIRIPENTOLICBZ INTERACTION
effect on epoxide formation andlor epoxide elimination. A method by which the effect on epoxide elimination clearance can be directly assessed is single-dose administration of CBZE (Kerr and Levy, 1989). In the present study, STP influence on disposition of an oral dose of CBZE was evaluated in six normal adult volunteers after they gave informed consent. CBZE was administered as a suspension in combination with antacid after an overnight fast, according to a previously described procedure (Kerr et al., 1989). Each volunteer ingested a 100-mg CBZE dose on two occasions, with a 1week interval allowed between the two doses. One CBZE dose served as a control, and the other CBZE dose was administered on the fourth day of a 7-day treatment regimen with STP 1,200 mg/day (600 mg twice daily). The order of the control and treatment phases was randomized. Blood samples (8 ml) were collected at times between 0 and 36 h after administration of CBZE. The blood samples were anticoagulated with EDTA, and the plasma was separated and frozen at -20°C until assayed for CBZE and STP. Plasma CBZE was assayed by means of a modified version of a published HPLC procedure (Kerr et al., 1989). The only modification was that plasma samples (250 p.L) were extracted twice with nhexane (2 x 2.5 ml) to remove STP selectively (thereby avoiding chromatographic interference) before extraction of CBZE from plasma with 5 ml dichloromethane. Assay of STP in plasma was performed as previously described (Lin and Levy, 1983). CBZE elimination half-life (tY2) was based on least-squares linear regression analysis of the log plasma concentration-time plot. Oral clearance of CBZE (ClcszE) was determined from the ratio of dose to area under the plasma concentration-time plot. Average STP plasma concentration during a dosing interval was calculated as the area under the plasma concentration-time plot divided by the time span of the dosing interval. Inhibition of CBZ metabolism by STP in microsomal preparations The effect of STP on cytochrome P-450-mediated conversion of CBZ to the 10,ll-epoxide metabolite was investigated in preparations of human liver microsomes. Samples of human liver tissue were obtained from organ transplant donors at Harborview Medical Center and University Hospital in Seattle, Washington. Microsomal incubations were performed in a 0.1 M phosphate buffer at a pH of 7.4. CBZ, STP, and NADPH were all dissolved in phosphate buffer without use of organic solvents before
269
addition to the microsomal incubations. Microsomes were preincubated with STP or blank buffer at 37°C for 90 s with NADPH added. After the preincubation period, CBZ was added to the incubations and the reaction was continued for 15 min at 37°C. The final volume of each incubation was I ml, protein concentration was 1 mg/ml, and the concentration of NADPH was 1 mM; concentration of CBZ was 14 pglml (preliminary studies showed no signs of enzyme saturation in the presence of CBZ concentrations S35 p.g/ml), and STP concentration was 0 or 7 p.g/ml. Enzymatic epoxidation of CBZ was terminated after 15 min by rapidly vortexing the 1-ml incubation with 4 ml dichloromethane. The internal standard cyheptamide (375 ng in 75 p.1 methanol) was then added to each sample. CBZE was extracted from the incubations and assayed by HPLC. Standards were prepared by adding appropriate amounts of phosphate buffer, fresh microsomal protein, dichloromethane, and internal standard to tubes containing 0-400 ng CBZE. After samples and standards were vortexed and centrifuged, the dichloromethane layer was transferred to a clean glass tube and evaporated to dryness. The residue was reconstituted with mobile phase, and chromatographic separation was made at room temperature on a Zorbax C, column (5 km; 4.6 x 250 mm) with acetonitrile/water (40:60, vol/ vol) as the mobile phase (flow rate 1.2 mlimin). The compounds were detected by ultraviolet absorption at 210 nm. Retention times of CBZE and cyheptamide were 4.7 and 8.6 min, respectively. RESULTS STP/CBZ interaction The daily dosing of CBZ was reduced in five of seven patients during STP therapy to avoid intoxication with CBZ (Table 1). After 2-week STP treatment, the average c1C.z in seven epileptic patients was reduced by an overall average of 50% (Table 2). The reduction of ClcBz was >50% in five patients (range 53-67%), but was only 27 and 29% in the other two patients. The mean CBZE/CBZ plasma concentration ratio was reduced by STP from 0.128 to 0.0696, an average decrease of -45% (Table 2). The extent of reduction in the CBZEiCBZ plasma ratio showed a strong tendency to parallel the extent of decrease in ClcBz; the 2 patients in whom ClCBZ was least inhibited (27 and 29% inhibition) also had the smallest reductions in the CBZE/CBZ plasma ratio (31 and 32% reduction), whereas 3 of the 5 patients in whom Clcsz was inhibited by 50-70% also had reductions in the CBZE/CBZ Epilepsia, Vol. 32, N o . 2 , 1991
B . M . KERR ET AL.
2 70
TABLE 2. Effect of STP on CBZ clearance (elcBz) and on plasma concentration ratio CBZEICBZ Clcsz (ml/min/kg) Patient
Control
1
0.930 1.43 1.59 1.37 0.924 1.31 1.17 1.25 0.252 p
2 3 4 5 6 7 Mean SD Paired t test
CBZE/CBZ
STP
=
Change (%)
Control
- 27 - 55 - 67 - 53 - 56 - 62 - 29 - 50 16
0.683 0.637 0.523 0.647 0.408 0.504 0.836 0.605 0.140 0.0011
STP
0.100 0.0686 0.129 0.0824 0.158 0.0613 0.128 0.0551 0.0967 0.0601 0.114 0.0446 0.170 0.115 0.128 0.0696 0.0278 0.0232 p = 0.0005
Average Plasma STP Change (%)
(pdml)
-31 - 36 - 61 - 57 - 38 - 61 - 32 - 45 14
1.45 1.16 6.63 2.02 6.06 10.5 4.76 4.65 3.40
CBZ clearance, ClCBz;plasma concentration ratio of CBZ epoxide to CBZ (CBZEKBZ); other abbreviations as in Table 1.
plasma ratio of -60% (Table 2). The effects of STP on plasma levels of CBZ and CBZE were similar to the effects noted previously for other inhibitors of cytochrome P-450, such as verapamil (MacPhee et al., 1986), propoxyphene (Dam et al., 1980), erythromycin (Barzaghi et al., 1987), and danazol (Kramer et al., 1986). Urinary excretion rates of the transdihydrodiol metabolite and unchanged CBZE indicated that the fraction of CBZ biotransformed to the 10 ,llepoxide metabolite (fm) was not significantly altered by STP (Table 3; control phase fm = 0.41 ? 0.14, STP phase fm = 0.39 ? 0.11). The pronounced inhibition of ClcBz in conjunction with no change in fm suggests that STP inhibits formation of CBZE (c1F) as well as alternative metabolic pathways of CBZ (Cl,,,,,). Indeed, both c1F and CloTHERwere reduced by -50% during treatment with STP (Table 3). The time-course studies showed that the ratio of CBZ dosing rate to trough plasma CBZ level (i.e., crude estimate of clearance under steady-state conditions) decreased 1-4 days after initiation of STP administration and became stabilized in 7-10 days
(Fig. 1A). The CBZE/CBZ plasma concentration ratio was decreased in similar fashion, requiring an equilibration period of 7-10 days (Fig. 1B). The CBZ doselplasma level ratio and the CBZE/CBZ plasma concentration ratio both decreased markedly when trough plasma concentrations of STP increased from 0 to 1 p,g/ml (Fig. 2), and maximal inhibition was observed at a somewhat higher concentration of STP. STP/CBZE interaction Treatment with STP had no significant effect on the tf or clearance of CBZE after single-dose administration to healthy volunteers (Table 4). The dose of STP in the CBZE interaction study (1,200 mg/day) was somewhat less than doses of STP that caused inhibition of CBZ clearance in the epileptic patients (1,500-3,000 mg/day). Similarly, the average plasma levels of STP in the CBZE single dose study (0.9-3.3 p,g/ml, Table 4) also tended to be less than STP plasma levels in the CBZ interaction study (1.2-10.5 p,g/ml, Table 2). However, STP is clearly a much more potent inhibitor of the metabolism of CBZ than of CBZE. Among the individuals
TABLE 3 . Effect of STP on f m , Cl, , and Cl,,, fm
C1, (mVmin/kg)
Cb-H,R (ml/minW
Patient
Control
STP
Control
STP
Control
STP
1
0.390 0.322 0.492 0.328 0.222 0.527 0.616 0.414 0.137
0.394 0.550 0.270 0.280 0.423 0.448 0.394 0.106
0.363 0.460 0.782 0.449 0.205 0.690 0.721 0.524 0.212
0.251 0.288 0.175 0.114 0.213 0.375 0.236 0.091
0.567 0.970 0.808 0.921 0.719 0.620 0.449 0.722 0.190
-a 0.386 0.235 0.472 0.294 0.291 0.461 0.356 0.098
2 3 4 5 6 7 Mean SD Paired t test
p
=
0.59
p = 0.0043
p = 0.0073
~~
fm, fraction of CBZ metabolized to CBZE; CI,, formation clearance of CBZE; ClOTHER, alternate CBZ clearance pathways; other abbreviations as in Table 1. a Incomplete urine collection. Epilepsia, Vol. 32, N o . 2 , 1991
2 71
STIRIPENTOLICBZ INTERACTION
1 1
.
3
.
140
j
120
100
n
P 0
(A)
0
120
Q)
x
x x
8o
.H
140
60 40
20
0
1
3
5
7
9
1 1 1 3 1 5
0
1
3
4
5
6
7
STP Plasma Level (pglmL)
Day
-
2
t i
0.02 0 1
3
5
7
9
1 1 1 3 1 5
Day FIG. 1. Time course of change in carbamazepine (CBZ) daily doselplasma level ratio (i-e., estimate of CBZ clearance) (A), and CBZ epoxide/CBZ plasma ratio (CBZE/CBZ)(B),after initiation of stiripentol therapy on day 1. Plasma drug concentrations (trough levels) were determined daily in patients 3 (0)and 4 (X).
in whom average STP plasma levels were in the range of 1-4 pg/ml, CBZ clearance was inhibited 27-55% (three epileptic patients, Table 2), whereas no consistent change was observed in clearance of CBZE (four volunteers, Table 4). The lack of effect of STP on the elimination clearance of CBZE indicates that the decreased CBZE/CBZ plasma ratio in epileptic patients during STP therapy (Table 2) results principally if not entirely from inhibition of the epoxide formation clearance. Inhibition of CBZ metabolism in microsomal preparations The 10,ll -epoxidation of CBZ by cytochrome P450 was significantly inhibited in the presence of a therapeutic concentration of STP (7 pg/ml) in four preparations of human liver microsomes (Table 5). The extent of inhibition in microsomes (-40-50% inhibition) was in excellent agreement with the extent of inhibition of CBZ metabolism in vivo in epileptic patients during STP therapy (Fig. 3). These
0.02 00
1
2
3
4
5
6
7
STP Plasma Level (pglmL) FIG. 2. Relationship between plasma concentration of stiripentol (STP)and carbamazepine (CBZ) daily dose/plasrna level ratio (i-e., estimate of CBZ clearance) (A), and carbarnazepine epoxide/CBZ plasma ratio (CBZE/CBZ) (B). Plasma drug concentrations (trough levels) were determined daily in patients 3 (0)and 4 (X).
results provide the first direct in vitro confirmation of the hypothesis (Levy et al., 1984; Moreland et al., 1986) that STP inhibits in vivo elimination of other AEDs by inhibiting cytochrome P-450 activity. DISCUSSION
The first evidence of the CBZ/STP drug interaction was noted during early clinical trials of STP, at which time STP markedly inhibited clearance of CBZ in a single patient (Levy et al., 1984). Results from the present study, as well as those of an earlier pilot investigation (Levy et al., 1985), confirm the pronounced in vivo inhibitory effects of STP. The mechanism by which STP inhibits CBZ clearance most likely involves formation of an inhibitory complex between the methylenedioxyphenyl moiety of stiripentol and cytochrome P-450(Levy et al., 1984; Mesnil et al., 1988a,b). In the present study, the ability of STP to inhibit cytochrome P-450 was confirmed both in vivo and in vitro by the following Epilepsia, Vol. 32, N o . 2 , 1991
272
B . M . KERR ET AL. TABLE 4. Effect of STP on Cl,, CICBzE(rnllminlkg)
Volunteerlsed weight (kg)
Control
I IF164
1.56 1.93 1.94 2.05 I .43 1.75 1.78 0.242
2lFl49 3lMl65 4lFl60 5lMl74 6lMl65
Mean SD Paired t test
p
=
and CBZEtII, CBZE ti/* (h)
STP
Change (%)
1.92 1.74 2.05 2.13 1.69 2.32 1.98 0.240
+ 23 -9.8 + 5.7 +3.9 + 18 + 33 + 12 15
0.122
Control
STP
5.54 6.35 6.60 5. I 6 6.10 5.59 5.89 0.549
p
Average plasma STP
=
Change (%)
(PLg/ml)
+ 12
6.21 5.93 5.66 4.89 6.10 5.1 1 5.65 0.541
1.36 3.27 0.88 I .65 0.88 I .30 1.56 0.89
- 6.6 - 14 - 5.2
0
- 8.6 - 3.7 9.0
0.327
ClcszE, oral clearance of CBZE; CBZE tllZ,half-life of CBZE; other abbreviations as in Table 1.
evidence: (a) The CBZE/CBZ plasma concentration ratio was substantially reduced during STP therapy; (b) the in vivo C1, of CBZE was inhibited by stiripentol; and (c) a therapeutic concentration of STP (7 pg/ml) inhibited the cytochrome P-450dependent conversion of CBZ to the 10,ll-epoxide in human liver microsomes. In addition to the effects on 10,ll-epoxide formation, STP also markedly inhibited alternative pathways of CBZ metabolism in vivo (CloTHER),which is consistent with inhibition of aromatic hydroxylation pathways. In contrast to the inhibitory effects toward oxidative metabolism, STP had no effect on in vivo elimination clearance of CBZE in healthy volunteers because elimination of CBZE is almost entirely mediated by microsomal epoxide hydrolase rather than cytochrome P-450 (Kerr et al., 1989). The extensive inhibition of CBZ clearance by STP clearly indicates that dosage adjustments are needed to avoid intoxication with CBZ after initiation of STP therapy. The average CBZ clearance of 0.605 ml/min/kg during the STP treatment phase indicates that a CBZ dose of 8.7 mg/kg/day should yield an average CBZ plasma level of 10 pg/ml. Therefore, therapeutic plasma levels of CBZ (5-10 pg/ml) during STP therapy should be achieved at CBZ dosages of 300-600 mg/day for an average size adult (70 kg), 200-400 mg/day for a small person (45
kg), and 400-800 mg/day for a large nonobese person (95 kg). Plasma levels of the pharmacologically active metabolite CBZE were sufficiently low in the present study for the epoxide to have no significant influence on CBZ dose-adjustment proposals. The time course of the STP/CBZ interaction in the present study showed that inhibition of CBZ clearance was detectable as early as the second or third day of STP therapy and that the inhibitory effect gradually became more pronounced in a 7- to 10-day period. To minimize changes in CBZ plasma levels, CBZ dose reductions should also be made gradually, with the initial dose reduction made on the third day of STP therapy and the final dose reduction made not later than 10 days after STP administration is started. The CBZ dosage should be reduced in steps if possible, because too rapid a dosage reduction may cause CBZ plasma levels to decrease temporarily to less than therapeutic levels
IN VITRO Mlcroromor
IN VlVO C B t Cloannco
IN VIVO EpoxldoICBZ
T
TABLE 5. Inhibition of biotransformation of CBZ" to CBZE by STP a in human liver microsames CBZE formation rateb (pmol x min-' mg protein-') Liver No. 102 103 105 108 a
Control 55.4 67.6 51.3 79.0
-t
1.5
2 0.6 -t -t
1.6 1.8
STP 30.8 2 1.5 34.0 2 1.9 31.1 ? 1.0 41.8 0.9
*
CBZ 14 pglml; STP = 7 pglml. Mean ? SD of quadruplicate incubations. Pooled t test, p < 0.001 in each liver.
Epilepsia, Vol. 32, N o . 2, 1991
Change
(%r
- 44% - 50% - 39%
- 47%
FIG. 3. Inhibitory effect of stiripentol (STP) on 1 0 , l l epoxidation of carbarnazepine (CBZ) in rnicrosomes(in vitro; STP 7 pglml) and effect on CBZ clearance and CBZ epoxidei CBZ plasma ratio in epileptic patients (in vivo; STP = 4.6 5 3.4 pglml). Plotted values are derived from data in Tables 2 and 5. The effect of stiripentol was statistically significant both in vitro and in vivo (p < 0.005).
2 73
STIRIPENTOLICBZ INTERACTION and possibly leave the patient unprotected against seizures. Complete discontinuation of STP from the treatment regimen is likely to result in a gradual reversal of the inhibitory effect, and a compensatory increase in the dosage of CBZ will probably be required; the time course of change in CBZ clearance after discontinuation of STP will be addressed in future studies. The inhibitory effect of STP may raise questions about the safety of STP polytherapy in future treatment of epilepsy. Clinically significant drug interactions involving enzyme induction, inhibition, or both are already common among the currently available AEDs (Pitlick, 1989), and inhibition of cytochrome P-450 by STP will further complicate this situation. Nevertheless, several benefits may be derived from intentional use of STP combination therapy; e.g., the ability of STP to decrease the CBZE/ CBZ plasma ratio (Table 2) indicates that neurologic side effects associated with CBZE (Schoeman et al., 1984) would be minimized during STP therapy. Perhaps more significant, the decreased CBZElCBZ plasma ratio suggests that STP may likewise decrease in vivo exposure to chemically reactive epoxide metabolites (arene oxides) of concurrently administered drugs. Arene oxide metabolites have been implicated as a cause of the infrequent but serious toxicities associated with AEDs, such as teratogenicity (Lindhout et al., I984), hepatitis (Spielberg et al., 1981), and blood dyscrasias (Gerson et al., 1983). The most notable AEDrelated toxicity attributed to an arene oxide metabolite is the teratogenic effect of phenytoin (fetal hydantoin syndrome) (Strickler et al., 1985). If STP itself is free of serious side effects, combination therapy with STP may reduce in vivo exposure to arene oxides and thereby protect against the severe toxicities of other AEDs. Additional benefits of STP polytherapy would probably include a substantially lower daily dose requirement and reduced frequency of administration for concomitant AEDs (Lockard and Levy, 1988; Levy and Kerr, 1989), but any benefit that might be associated with STP polytherapy will be achieved only if AED interactions with STP are carefully controlled. The inhibitory effect of STP was quantitated in some detail in the present study to develop approaches for managing AED interactions that might complicate the clinical efficacy trials of STP. Although the inhibitory effect of STP is pronounced, the present study documents that the time course and extent of inhibition show a surprising degree of intersubject consistency. These results indicate that the CBZlSTP interaction is highly predictable and reproducible and suggest that a standardized dose
adjustment procedure should successfully maintain CBZ plasma levels within therapeutic limits in most patients after initiation of STP therapy. Acknowledgment: We thank Dr. Milo Gibaldi, University of Washington, Seattle, WA, for helpful discussions and valuable suggestions concerning interpretations of study results, and Brian Bainbridge and Martha Moore for technical assistance. This work was supported by a research grant from Biocodex Laboratories, Montrouge, France, and by Grants No. PO1 NF17111 and T32 NS07289 from the National Institutes of Health.
REFERENCES Barzaghi N , Gatti G, Crema F, et al. Inhibition by erythromycin of the conversion of carbamazepine to its active 10,llepoxide metabolite. Er J Clin Pharmacol 1987;24:836-8. Brodie MJ, MacPhee GJ. Carbamazepine neurotoxicity precipitated by diltiazem. Er Med J 1986;292:1170-1. Dam M, Christensen JM, Brandt J, et al. Antiepileptic drugs: interaction with dextropropoxyphene. In: Johannessen SI, Morselli PL, Pippenger CE, Richens A, Schmidt D, Meinardi H, eds. Antiepileptic therapy: advances in drug monitoring. New York: Raven Press, 1980:29!%306. Frigerio A, Fanelli R, Biandrate P, et al. Mass spectrometric characterization of carbamazepine-l0,l 1-epoxide, a carbamazepine metabolite isolated from human urine. J Pharm Sci 1972;61:1144-7. Gerson GT, Fine DG, Spielberg SP, Sensenbrenner LL. Anticonvulsant-induced aplastic anemia: increased susceptibility to toxic drug metabolites in vitro. Blood 1983;61:889-93. Kerr BM, Levy RH. Carbamazepine epoxide. In: Levy R, Mattson R, Meldrum B, Penry JK, Dreifuss FE, eds. Antiepileptic drugs, 3rd ed. New York: Raven Press, 1989505-20. Kerr BM, Rettie AE, Eddy AC, et al. Inhibition of human liver microsomal epoxide hydrolase by valproate and valpromide: in v i t r o h vivo correlation. Clin Pharmacol Ther 1989;46:8293. Kramer G ,Theisohn M, von Unruh GE, Eichelbaum M. Carbamazepine-danazol drug interaction: its mechanism examined by a stable isotope technique. Ther Drug Monit 1986;8: 387-92. Levy RH, Loiseau P, Guyot M, et al. Stiripentol kinetics in epilepsy: nonlinearity and interactions. Ciin Pharmacol Ther 1984;36:661-9. Levy RH, Martinez-Lage JM, Tor J , et al. Stiripentol level-dose relationship and interaction with carbamazepine in epileptic patients. Epilepsia 1985;26:544-5. Levy RH, Kerr BM. Pharmacokinetics of old, new, and yetto-be-discovered antiepileptic drugs. Epiiepsia 1989;30 (SUPPIl):S35-S41. Lin H, Levy RH. Pharmacokinetic profile of a new anticonvulsant, stiripentol, in the rhesus monkey. Epilepsia 1983;24: 692-702. Lindhout D, Hoppener RJEA, Meinardi H. Teratogenicity of antiepileptic drug combinations with special emphasis on epoxidation (of carbamazepine). Epilepsia 1984;25:77-83. Lockard JS, Levy RH. Carbamazepine plus stiripentol: is polytherapy by design possible? Epilepsia 1988;29:476-81. MacPhee GJA, McInnes GT,Thompson GG,Brodie MJ. Verapamil potentiates carbamazepine neurotoxicity: a clinically important inhibitory interaction. Lancet 1986;1:70&3. Mesnil M, Testa B, Jenner P. In vitro inhibition by stiripentol of rat brain cytochrome P-450-mediated naphthalene hydroxyl18:1097-106. ation. Xenobiotica 1988~; Mesnil M, Testa B, Jenner P. Ex vivo inhibition of rat brain cytochrome P-450 activity by stiripentol. Eiochem Pharmacol 19886;37:3639-22. Epilepsia, Vol. 32, N o . 2, 1991
274
B . M . KERR ET AL.
Moreland TA, Astoin J, Lepage F, et al. The metabolic fate of stiripentol in man. Drug Metab Dispos 1986;14:654-62. Pitlick WH, ed. Antiepileptic drug interactions. New York: Demos Publications, 1989. Rascol 0, Squalli A, Montastruc JL, et al. A pilot study of stiripentol, a new anticonvulsant drug, in complex partial seizures uncontrolled by carbamazepine. Clin Neuropharmacol 1989;12:119-23. Schoeman JF, Elyas AA, Brett EM, Lascelles PT. Correlation between plasma carbamazepine- 10,l I-epoxide concentration and drug side-effects in children with epilepsy. Dev Med Child Neurol 1984;26:756-64. Spielberg SP, Gordon GB, Blake DA, et al. Predisposition to phenytoin hepatotoxicity assessed in vitro. N Eng J Med 1981;305~722-7. Strickler SM, Dansky LV, Miller MA, et al. Genetic predisposition to phenytoin-induced birth defects. Lancet 1985;2: 746-9. Treiman DM, Wilensky AJ, Ben-Menachem E, et al. Efficacy of nafimidone in the treatment of intractable partial seizures: report of a two-center study. Epilepsia 1985;26:607-11.
RJ?SUME L’effet inhibiteur du Stiripentol (STP) sur la disponibilitt de la carbamaztpine (CBZ) et du 10,l I-epoxide de carbamaztpine (CBZE) a t t t quantifiC dans le but d’Ctablir des recommandations pour la diminution des doses de CBZ dans les Ctudes cliniques futures d’adjonction du Stiripentol. Chez 7 patients tpileptiques, le STP (1500 B 3000 mg/j pendant deux semaines) a inhibt la clearance de CBZ de 50 ? 16% (p = 0.001) et a rtduit 14% (p = 0.0005). le rapport plamatique CBZE/CBZ de 45 L’effet inhibiteur s’est manifeste progressivement sur une pCriode de 7 ti 10jours aprts mise en route du STP. Contrairement a [’inhibition de la formation de CBZE, le STP n’a pas eu d’effet sur la clearance ou la demi-vie du CBZE chez 6 volontaires sains (p > 0.05). Le STP exerce trts vraisemblablement ses effets inhibiteurs par le biais d’une inhibition du cytochrome P-450. Cette
*
Epilepsia, Vol. 32, N o . 2,1991
hypothtse a Ctt confirmte dans ce travail par la constatation d’une inhibition par le STP a concentration thtrapeutique (7 mg/ ml) de la 10,11-epoxidation de la CBZ dans les microsomes hCpatiques humains, de 40 B 50%. Sur la base des resultats de cette ttude, les auteurs proposent que: 1”)la dose de CBZ soit rtduite par ttapes sur une ptriode de 7 B 10jours aprts mise en route du STP, et que: 2 7 la dose de CBZ de 4.3 a 8.7 mglkg par jour devrait sufire a maintenir des taux plasmatiques de CBZ thtrapeutiques, compris entre 5 et 10 mg/ml. (P. Genton, Marseille)
ZUSAMMENFASSUNG Die inhibitorische Wirkung von Stiripentol (STP) auf Carbamazepin (CBZ) und Carbamazepin-lO,1 I-Epoxyd (CBZE) wurde quantifiziert, um Richtlinien zur Carbamazepin-Dosisreduktion fur eine zukunftige klinische add-on-Wirkungsuntersuchung von Stiripentol zu erstellen. Bei sieben epileptischen Patienten hemmte Stiripentol (1,500-3,000 mg/Tag iiber 2 Wochen) die CBZ-Clearance um 50 + 16% (p = 0,001) und reduzierte die CBZE/CBZ-Plasma-Relation um 45 + 14% (p = 0.0005). Die inhibitorische Wirkung stellte sich schrittweise uber eine Periode von 7 bis 10 Tagen nach Beginn der STP-Therapie ein. Im Gegensatz zur Hemmung der CBZE-Bildung hatte STP keine Wirkung (p > 0.05) auf die Eliminations-Clearance oder Halbwertszeit von CBZE bei 6 gesunden Freiwilligen. STP entwickelt sehr wahrscheinlich seine inhibitorische Wirkung uber eine Hemmung von Zytochrom-P-450. Diese Hypothese wurde in der vorliegenden Studie durch den Befund bestatigt, daR eine therapeutische Konzentration von STP (7 pg/ml) die 10,llEpoxydation von CBZ in menschlichen Lebermikrosomen um 40 bis 50% inhibierte. Auf der Grundlage dieser Ergebnisse wird vorgeschlagen, daR 1. die CBZ-Dosis nach Beginn einer STPTherapie stufenweise uber 7 bis 10 Tage reduziert wird und 2. eine CBZ-Dosis von 4.3 bis 8.7 mg/kg/Tag fur therapeutische CBZ-Plasmaspiegel von 5 bis 10 mglml ausreichend ist. (C. Benninger, Heidelberg)
