BIOPHARMACEUTICS & DRUG DISPOSITION, VOL. 13, 337-344 (1992)
PHARMACOKINETICS OF IBUPROFEN ENANTIOMERS AFTER SINGLE AND REPEATED DOSES IN MAN J. OLIARY., M. TOD**, P. NICOLAS*, 0. PETITJEAN* AND G . CAILLfi'
*Ddpartement de Pharmacologie hospitaliPre, HGpital Avicenne, 125 route de Stalingrad, 93000 Bobigny, France tD&partementde Pharmacologie. Facultd de Mddeeine, 2900 Bvd E. Montpetit. CP 6128. Montreal, Canada
ABSTRACT The pharmacokinetic parameters of ibuprofen enantiomers after a single 600 mg dose and repeated 3 x 400 mg doses of Nurofen@ were determined in 12 healthy volunteers. Terminal half-lives were similar for both enantiomers, but plasma levels of S-ibuprofen were higher than those of R-ibuprofen, due to the chiral inversion and differences in distribution and metabolism. Comparison of maximal concentrations and areas under the concentration vs time curves between the first and last doses for each enantiomer indicated linear pharmacokinetics with no time-dependency. A large interindividual variability in the ratio of 9 to R-ibuprofen average concentrations at steady-state was observed (mean f SD 1 89 k 0 - 89) and probably accounts for the known lack of correlation between racemic ibuprofen concentrations and therapeutic efficacy.
-
KEY WORDS
Ibuprofen Enantiomers Pharmacokinetics
INTRODUCTION Ibuprofen, a 2-arylpropionic acid, is administered as the racemic drug (R, Sibuprofen). Recent review article~l-~ have emphasized that consideration of the stereochemical aspects of its metabolic fate could provide notable insights into its concentration-effect relationship and safety evaluation. In particular, it has been shown that only the S (+) -enantiomer inhibited prostaglandin synthesis,' that 63 f 6 per cent of an administered dose of R (-) -ibuprofen was specifically inverted into the S(+)enantiometer,6 and that in the rat, only the R ( + ) enantiometer could form 'hybrid' triglycerides.' This latter point could be of therapeutic relevance since the hybrid triglycerides may accumulate in the brain and be responsible, in part, for ibuprofen's central nervous system toxicity. *Addresseefor correspondence.
0142-2782/92/050337-08$09.00 01992 by John Wiley & Sons, Ltd.
Received 27 March 1991 Revised 9 December I991 Accepted 13 December 1991
338
J. OLIARY ET AL.
Conventional studies with racemic ibuprofen showed no correlation between the area under the racemic concentration curve (AUC) and clinical response in rheumatoid arthritis,* and this may be attributed to inter-individualvariation of the steady-state concentration ratio of S- to R-ibuprofen. Furthermore, ibuprofen exhibits nonlinear kinetics, inasmuch as the total plasma AUC increases less than proportionately with the given dose, because of non-linear plasma protein binding.9 Thus, the steady-state concentration of the active Senantiomer is difficult to predict, and the aim of this study was to evaluate the plasma levels of ibuprofen enantiomers after administration of repeated oral doses compared to a single dose. METHODS
Subjects Twelve healthy male volunteers participated in talestudy after they had given informed consent. Their mean age f SD was 23.6 f2 - 7 years (range 19 to 27) and mean weight f SD was 70.6 f7 - 2kg (range 58 to 85). The volunteers were examined using standard methods of clinical chemistry and hematology before and after the study. None of the participants had received any form of medication during the 2 weeks prior to the experiments. The study was approved by the local Ethical Committee in accordance with the Declaration of Helsinki.
Experimental design Two different experimental protocols were investigated in this study. In both cases, the drug was given to each subject with 150ml of water and no food or fluids were ingested over the next 4 h. Blood samples (6 ml) were drawn onto ethylenediaminetetra-acetic acid (EDTA) by venipuncture.
Single dose. A single oral dose of 600 mg (three 200 mg tablets of NurofenO ) of racemic ibuprofen was given to each fasted subject in the morning. A pre-dose blood sample served as an analytical blank. Subsequent samples weredrawn0.33, 0.67, 1, 2, 3, 4, 5 , 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, and 24 h after administration. Multiple dose. Racernic ibuprofen (400 mg; two 200 rng tablets of NurofenO ) was given every 8 h for 5 days. Samples were drawn at 0.33, 0-67, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 24, 30, 36, and 48h after the last administration. Blood samples were centrifuged at 2000 x g. Plasma was separated, aliquoted in polypropylene tubes and stored at - 30" until analyzed.
IBUPROFEN ENANTIOMERS
339
Drug assay Plasma levels of ibuprofen were measured using the high pressure liquid chromatography (HPLC) method of Avgerinos and Hutt'O with slight modifications. To 1 ml plasma samples, 5 pg of 2-4-dichlorophenoxyaceticacid were added, followed by 1 ml of 1 M HC1 and 5 ml of benzene. Samples were mixed slowly for 10 min and then centrifuged. The organic layer was collected in glass tubes. After evaporation under nitrogen gas, the drug residue was dissolved in 100 pl of 1-hydroxybenzotriazole, 1-(3-dimethylarninopropyl)-3-ethyl carbodiimide hydrochloride, and S ( + ) (naphthene-1) ethyl-1-amine. These derivatization agents were in solution at 0.1 per cent (m/v) in dichloromethane. After 2 h, the solvent was evaporated and the drug residue was dissolved in 300 p1 of the mobile phase (50 per cent acetonitrile in 0-015 M H3P04, final pH 3 5 ) before injection. A Hychrom C8 Spherisorb 100x 4.6 mm column and a spectrophotometric detector were used for analysis. At a flow rate of 2.5 ml min- l , the internal standard, S (+) -ibuprofen and R ( - ) -ibuprofen eluted, respectively, at 4.5, 7.2, 8.2min. The within-day coefficient of variation was 6 - 1 per cent at 5 mg 1- and the recovery was about 95 per cent. The limit of quantification was 1 mgl-l.
Data analysis For each subject and each enantiomer, the concentrationvs time data were fitted to a biexponential (R-ibuprofen: subjects 1, 3 to 9;S-ibuprofen: subjects 1 to 11) or a triexponential (R-ibuprofen: subjects 2, 10 to 12; S-ibuprofen: subject 12) model with a lag time, using the SIPHAR program (Simed, Crheil, 94008, France). Nonlinear regression analysis was performed using extended least square. Choice between alternative models was made on the basis of the Schwartz criterion]' S (S = - Log likelihood + 0.5 log N, where Nis the number of data points), the standard error of the estimated parameters, and the ratio between experimental and extrapolated to infinity AUC. Several parameters were considered for the pharmacokinetic analysis: the maximal concentration C,,, the time to reach maximal concentration tmaxand the half-life of the elimination phase t%. The experimental area under the concentration vs time curve (AUC,,,) was calculated by the log-trapezoidal rule up to the last sampling time. The total where C, is the area (AUC,) was obtained by adding CJk, to AUC,,, concentration of the last sample and k, the slope of the terminal phase. Clearance was calculated as CL/F=D/AUC, where F is the (unknown) bioavailability and D the dose. The volume of distribution ( V , was determined as: Vd/F=D/k, AUC,. The average concentration over 24 h was calculated by the ratio AUCeX,/24. For the multiple dose study, AUC, (e.g. the area under curve from 0 to 8 h after the last dose) was determined using the log-trapezoidal rule. The average
1* 2 2 0-66 1 5* 3* 6* 1 2 1* 1 2.1 1.72
(h) S
15.2 pco.001
50.5
36.0 47.7 56.9 69-7 62.5 36.7 65.2 55.2 53.1 65.0 19.0 38.5
R
AUC, 47.9 103.1 87.9 89.5 70.4 45.8 82.8 102.3 72.1 111.4 59.0 75.2 78.9 6.1
S
(h.mg1-')
*Double peak: in this table only the main peak was taken into account.
1* 4* 1 0.66 1* 1 1.7 1.42 NS
5*
0.66* 2 2 0-66 1
R
S
R
15.4 21.0 14.3 17.8 25.7 7.1 11.0 14.7 13.0 22.2 12.5 20.0 16-2 5-28
fmax
(mg1-9
c ,
14.8 17.3 11-4 18.3 5 25.9 6 4.1 7 10.0 8 10.0 13-4 9 19.2 10 11 8.8 12 14.8 14.0 Mean SD 5.76 Significance p c o . 0 1
1 2 3 4
Subjects
-
19.0 23.5 20.1 9.5 11.8 79.8 33.3 38.8 18-5 18.1 94.5 28.2 32.9 26.8
R
Vd/F
5.5
2.8 2.5 2.3 3.3 2.4 3.1 1.20
5.0
2.0 2.5 4-8 2.5 1.9 4.6
S
(h)
4.9 2.3 2.7 4.1 2.5 3.2 1-65 NS
1.6 2.6 2.6 1.5 1.7 6.8
R
t,
2-6 1.2 2.6 1.9 2-1 2.7 0.7 1.6 2.0 0.66 pco.001
4.80 8.17 4-60 5.44 5.63 4.62 15.80 7-79 6.75 3.19
-
1.5 2.0 2.1 2.9
R
C,,
8-34 6.28 5-27 4.30
CL/F R
2.9 1.6 3.2 4.1 2.9 4.6 2.4 3.1 3.2 0.91
2.0 4.3 3.5 3.7
S
(mg1-I)
Table 1. Pharmacokinetic parameters of ibuprofen enantiomers after a single 600 mg oral dose of a racemic formulation
?
31
.e
E
P
?
P 0
w
IBUPROFEN ENANTIOMERS
341
steady-state concentration Cavwas calculated as AUCJ7, where 7 is the interval of administration, e.g. 8 h. The predicted maximal concentration at steady-state, C,, cdc., was calculated assuming linear kinetics, as follows:
C,,
cdc, = 0-66
C,,/l-
e-k
where C,, and k, are the parameters determined after the single dose administration. The factor 0.66 corrects for the doses that are different (300 mg 'vs 200 mg) between the two sequences. Clearance at the steady state was calculated as CL/F=D/AUC, and Vd determined from the last dose was Vd/F=D/ke AUC,
-
Statistical analysis Pharmacokinetic parameters of R- and S-ibuprofen were compared using the Student's t-test for paired data except for t, (Wilcoxon t-test). All results are expressed as means fSD. RESULTS Pharmacokinetic parameters of R- and S-ibuprofen after a 300 mg single dose of each enantiomer are shown in Table 1. The mean fSD C,, for R- and Sibuprofen are significantly different: 14.0k5.8 vs 1 6 . 2 f 5 - 3 m g * I-' (p
PHARMACOKINETICS OF IBUPROFEN ENANTIOMERS AFTER SINGLE AND REPEATED DOSES IN MAN J. OLIARY., M. TOD**, P. NICOLAS*, 0. PETITJEAN* AND G . CAILLfi'
*Ddpartement de Pharmacologie hospitaliPre, HGpital Avicenne, 125 route de Stalingrad, 93000 Bobigny, France tD&partementde Pharmacologie. Facultd de Mddeeine, 2900 Bvd E. Montpetit. CP 6128. Montreal, Canada
ABSTRACT The pharmacokinetic parameters of ibuprofen enantiomers after a single 600 mg dose and repeated 3 x 400 mg doses of Nurofen@ were determined in 12 healthy volunteers. Terminal half-lives were similar for both enantiomers, but plasma levels of S-ibuprofen were higher than those of R-ibuprofen, due to the chiral inversion and differences in distribution and metabolism. Comparison of maximal concentrations and areas under the concentration vs time curves between the first and last doses for each enantiomer indicated linear pharmacokinetics with no time-dependency. A large interindividual variability in the ratio of 9 to R-ibuprofen average concentrations at steady-state was observed (mean f SD 1 89 k 0 - 89) and probably accounts for the known lack of correlation between racemic ibuprofen concentrations and therapeutic efficacy.
-
KEY WORDS
Ibuprofen Enantiomers Pharmacokinetics
INTRODUCTION Ibuprofen, a 2-arylpropionic acid, is administered as the racemic drug (R, Sibuprofen). Recent review article~l-~ have emphasized that consideration of the stereochemical aspects of its metabolic fate could provide notable insights into its concentration-effect relationship and safety evaluation. In particular, it has been shown that only the S (+) -enantiomer inhibited prostaglandin synthesis,' that 63 f 6 per cent of an administered dose of R (-) -ibuprofen was specifically inverted into the S(+)enantiometer,6 and that in the rat, only the R ( + ) enantiometer could form 'hybrid' triglycerides.' This latter point could be of therapeutic relevance since the hybrid triglycerides may accumulate in the brain and be responsible, in part, for ibuprofen's central nervous system toxicity. *Addresseefor correspondence.
0142-2782/92/050337-08$09.00 01992 by John Wiley & Sons, Ltd.
Received 27 March 1991 Revised 9 December I991 Accepted 13 December 1991
338
J. OLIARY ET AL.
Conventional studies with racemic ibuprofen showed no correlation between the area under the racemic concentration curve (AUC) and clinical response in rheumatoid arthritis,* and this may be attributed to inter-individualvariation of the steady-state concentration ratio of S- to R-ibuprofen. Furthermore, ibuprofen exhibits nonlinear kinetics, inasmuch as the total plasma AUC increases less than proportionately with the given dose, because of non-linear plasma protein binding.9 Thus, the steady-state concentration of the active Senantiomer is difficult to predict, and the aim of this study was to evaluate the plasma levels of ibuprofen enantiomers after administration of repeated oral doses compared to a single dose. METHODS
Subjects Twelve healthy male volunteers participated in talestudy after they had given informed consent. Their mean age f SD was 23.6 f2 - 7 years (range 19 to 27) and mean weight f SD was 70.6 f7 - 2kg (range 58 to 85). The volunteers were examined using standard methods of clinical chemistry and hematology before and after the study. None of the participants had received any form of medication during the 2 weeks prior to the experiments. The study was approved by the local Ethical Committee in accordance with the Declaration of Helsinki.
Experimental design Two different experimental protocols were investigated in this study. In both cases, the drug was given to each subject with 150ml of water and no food or fluids were ingested over the next 4 h. Blood samples (6 ml) were drawn onto ethylenediaminetetra-acetic acid (EDTA) by venipuncture.
Single dose. A single oral dose of 600 mg (three 200 mg tablets of NurofenO ) of racemic ibuprofen was given to each fasted subject in the morning. A pre-dose blood sample served as an analytical blank. Subsequent samples weredrawn0.33, 0.67, 1, 2, 3, 4, 5 , 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, and 24 h after administration. Multiple dose. Racernic ibuprofen (400 mg; two 200 rng tablets of NurofenO ) was given every 8 h for 5 days. Samples were drawn at 0.33, 0-67, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 24, 30, 36, and 48h after the last administration. Blood samples were centrifuged at 2000 x g. Plasma was separated, aliquoted in polypropylene tubes and stored at - 30" until analyzed.
IBUPROFEN ENANTIOMERS
339
Drug assay Plasma levels of ibuprofen were measured using the high pressure liquid chromatography (HPLC) method of Avgerinos and Hutt'O with slight modifications. To 1 ml plasma samples, 5 pg of 2-4-dichlorophenoxyaceticacid were added, followed by 1 ml of 1 M HC1 and 5 ml of benzene. Samples were mixed slowly for 10 min and then centrifuged. The organic layer was collected in glass tubes. After evaporation under nitrogen gas, the drug residue was dissolved in 100 pl of 1-hydroxybenzotriazole, 1-(3-dimethylarninopropyl)-3-ethyl carbodiimide hydrochloride, and S ( + ) (naphthene-1) ethyl-1-amine. These derivatization agents were in solution at 0.1 per cent (m/v) in dichloromethane. After 2 h, the solvent was evaporated and the drug residue was dissolved in 300 p1 of the mobile phase (50 per cent acetonitrile in 0-015 M H3P04, final pH 3 5 ) before injection. A Hychrom C8 Spherisorb 100x 4.6 mm column and a spectrophotometric detector were used for analysis. At a flow rate of 2.5 ml min- l , the internal standard, S (+) -ibuprofen and R ( - ) -ibuprofen eluted, respectively, at 4.5, 7.2, 8.2min. The within-day coefficient of variation was 6 - 1 per cent at 5 mg 1- and the recovery was about 95 per cent. The limit of quantification was 1 mgl-l.
Data analysis For each subject and each enantiomer, the concentrationvs time data were fitted to a biexponential (R-ibuprofen: subjects 1, 3 to 9;S-ibuprofen: subjects 1 to 11) or a triexponential (R-ibuprofen: subjects 2, 10 to 12; S-ibuprofen: subject 12) model with a lag time, using the SIPHAR program (Simed, Crheil, 94008, France). Nonlinear regression analysis was performed using extended least square. Choice between alternative models was made on the basis of the Schwartz criterion]' S (S = - Log likelihood + 0.5 log N, where Nis the number of data points), the standard error of the estimated parameters, and the ratio between experimental and extrapolated to infinity AUC. Several parameters were considered for the pharmacokinetic analysis: the maximal concentration C,,, the time to reach maximal concentration tmaxand the half-life of the elimination phase t%. The experimental area under the concentration vs time curve (AUC,,,) was calculated by the log-trapezoidal rule up to the last sampling time. The total where C, is the area (AUC,) was obtained by adding CJk, to AUC,,, concentration of the last sample and k, the slope of the terminal phase. Clearance was calculated as CL/F=D/AUC, where F is the (unknown) bioavailability and D the dose. The volume of distribution ( V , was determined as: Vd/F=D/k, AUC,. The average concentration over 24 h was calculated by the ratio AUCeX,/24. For the multiple dose study, AUC, (e.g. the area under curve from 0 to 8 h after the last dose) was determined using the log-trapezoidal rule. The average
1* 2 2 0-66 1 5* 3* 6* 1 2 1* 1 2.1 1.72
(h) S
15.2 pco.001
50.5
36.0 47.7 56.9 69-7 62.5 36.7 65.2 55.2 53.1 65.0 19.0 38.5
R
AUC, 47.9 103.1 87.9 89.5 70.4 45.8 82.8 102.3 72.1 111.4 59.0 75.2 78.9 6.1
S
(h.mg1-')
*Double peak: in this table only the main peak was taken into account.
1* 4* 1 0.66 1* 1 1.7 1.42 NS
5*
0.66* 2 2 0-66 1
R
S
R
15.4 21.0 14.3 17.8 25.7 7.1 11.0 14.7 13.0 22.2 12.5 20.0 16-2 5-28
fmax
(mg1-9
c ,
14.8 17.3 11-4 18.3 5 25.9 6 4.1 7 10.0 8 10.0 13-4 9 19.2 10 11 8.8 12 14.8 14.0 Mean SD 5.76 Significance p c o . 0 1
1 2 3 4
Subjects
-
19.0 23.5 20.1 9.5 11.8 79.8 33.3 38.8 18-5 18.1 94.5 28.2 32.9 26.8
R
Vd/F
5.5
2.8 2.5 2.3 3.3 2.4 3.1 1.20
5.0
2.0 2.5 4-8 2.5 1.9 4.6
S
(h)
4.9 2.3 2.7 4.1 2.5 3.2 1-65 NS
1.6 2.6 2.6 1.5 1.7 6.8
R
t,
2-6 1.2 2.6 1.9 2-1 2.7 0.7 1.6 2.0 0.66 pco.001
4.80 8.17 4-60 5.44 5.63 4.62 15.80 7-79 6.75 3.19
-
1.5 2.0 2.1 2.9
R
C,,
8-34 6.28 5-27 4.30
CL/F R
2.9 1.6 3.2 4.1 2.9 4.6 2.4 3.1 3.2 0.91
2.0 4.3 3.5 3.7
S
(mg1-I)
Table 1. Pharmacokinetic parameters of ibuprofen enantiomers after a single 600 mg oral dose of a racemic formulation
?
31
.e
E
P
?
P 0
w
IBUPROFEN ENANTIOMERS
341
steady-state concentration Cavwas calculated as AUCJ7, where 7 is the interval of administration, e.g. 8 h. The predicted maximal concentration at steady-state, C,, cdc., was calculated assuming linear kinetics, as follows:
C,,
cdc, = 0-66
C,,/l-
e-k
where C,, and k, are the parameters determined after the single dose administration. The factor 0.66 corrects for the doses that are different (300 mg 'vs 200 mg) between the two sequences. Clearance at the steady state was calculated as CL/F=D/AUC, and Vd determined from the last dose was Vd/F=D/ke AUC,
-
Statistical analysis Pharmacokinetic parameters of R- and S-ibuprofen were compared using the Student's t-test for paired data except for t, (Wilcoxon t-test). All results are expressed as means fSD. RESULTS Pharmacokinetic parameters of R- and S-ibuprofen after a 300 mg single dose of each enantiomer are shown in Table 1. The mean fSD C,, for R- and Sibuprofen are significantly different: 14.0k5.8 vs 1 6 . 2 f 5 - 3 m g * I-' (p
