- Email: [email protected]
Gabapentin bioavailability: effect of dose and frequency of administration in adult patients with epilepsy
Epilepsy Research 31 (1998) 91 – 99
Gabapentin bioavailability: effect of dose and frequency of administration in adult patients with epilepsy Barry E. Gidal a,b,*, John DeCerce c, Howard N. Bockbrader d, Jose Gonzalez c, Sarah Kruger b, Michael E. Pitterle a, Paul Rutecki b, R. Eugene Ramsay c b
a School of Pharmacy, Uni6ersity of Wisconsin, 425 N. Charter Street, Madison, WI 53706, USA Department of Neurology, Uni6ersity of Wisconsin, 425 N. Charter Street, Madison, WI 53706, USA c Uni6ersity of Miami, International Center for Epilepsy, Miami, FL, USA d Warner-Lambert Co., Parke-Da6is Pharmaceutical Research Di6ision, Ann Arbor, MI, USA
Received 2 December 1997; received in revised form 10 March 1998; accepted 11 March 1998
Abstract Gabapentin (GBP) is a non-metabolized antiepileptic drug that is eliminated by renal excretion and displays saturable, dose dependent absorption. The recommended dosing schedule for GBP is t.i.d. At large daily doses, oral bioavailability (F) may be improved by giving the daily dose more frequently. Objecti6e: To evaluate whether switching GBP dosage regimen from t.i.d. to q.i.d. results in increased oral bioavailability. Methods: This study consisted of two parts; a computer simulated pharmacokinetic model and a clinical pharmacokinetic study in nine adult epileptic patients receiving 3600 mg/day and 11 receiving 4800 mg/day. All patients were evaluated during both t.i.d. and q.i.d. regimens. F were determined by calculation of percent of dose excreted unchanged using steady-state 24-h urine collections and were compared using a paired t-test. Results: At 3600 mg/day, mean F following t.i.d. and q.i.d. dosing were 38.7 922.1% and 40.0 918.9%, respectively (P= 0.738). At 4800 mg/day, mean F following t.i.d. and q.i.d. dosing were 29.2 916.2% and 35.6 917.6%, respectively (P=0.006). Discussion: Good agreement was observed between values from this study and predicted values based on the pharmacokinetic model. Improved GBP F at doses of 3600 mg/day was not achieved with more frequent drug administration, and thus is not warranted. At 4800 mg/day, a 22% increase in F was observed with more frequent drug dosing. Conclusion: GBP F may be significantly increased by q.i.d. versus t.i.d. dosing, depending upon dose level. This increase in F however must be balanced against the inconvenience of more frequent dosing. Therapeutic drug level monitoring may aid in the evaluation of such pharmacokinetic maneuvers. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Gabapentin; Bioavailability; Epilepsy
* Corresponding author. Tel.: +1 608 2623280; fax: +1 608 2655421; e-mail: [email protected] 0920-1211/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S0920-1211(98)00020-5
92
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
1. Introduction
2. Aims and objectives
Gabapentin is an antiepileptic medication approved in the USA for adjunct therapy of adult patients with partial seizures, with and without secondary generalization. In addition to well established efficacy and a modest toxicity profile, gabapentin has a unique pharmacokinetic profile. Gabapentin is excreted unchanged in the urine, with its clearance correlating with creatinine clearance (McLean, 1994). Gabapentin has a elimination half-life of approximately 6 h in patients with normal renal function, does not bind to plasma proteins, and does not appear to interact pharmacokinetically with other antiepileptic medications. Of clinical interest, gabapentin displays dose dependent absorption, with systemic bioavailability decreasing with increasing doses. The mechanism underlying this phenomenon appears to involve an active and saturable transport mechanism (Stewart et al., 1993). In clinical trials, plasma gabapentin concentrations increase with increasing doses but the increases were not dose proportional (linear). Bioavailability following single doses in healthy subjects is approximately 57% at 300 mg and 42% at 600 mg (McLean, 1994). Currently, there is no recognized maximal daily dose of gabapentin and many patients are receiving more than 1800 mg/day. Given this well documented property of saturable absorption, it is reasonable to hypothesize that the current recommendation of a three times per day (t.i.d.) administration schedule may not be optimal. At higher doses (\ 1800 mg/day), a four times per day (q.i.d.) schedule may result in greater systemic bioavailability. This may not only have clinical implications, in that gabapentin efficacy appears to increase with increasing daily dose, but economic consequences as well. At a given daily dose, if systemic bioavailability is increased, the ‘effective’ systemic dose or exposure to gabapentin would in essence, be increased with no increase in medication expense.
The aim of this study was to evaluate the overall pharmacokinetic impact of modifications in gabapentin oral administration schedules. This study was conducted in two parts including the application of a computer-modeled pharmacokinetic simulation and a clinical pharmacokinetic study in adult epileptic patients. The first component investigates predicted steady-state plasma gabapentin concentrations and absorption parameters that would be expected from t.i.d. and q.i.d. drug administration regimens. The second component is a clinical study to test the hypothesis that increasing the frequency of gabapentin administration results in a significant increase in gabapentin bioavailability. The clinical study evaluated the impact of increasing the gabapentin dosage regimens from t.i.d. to q.i.d. in epileptic patients currently maintained on large yet clinically relevant dosages of 3600 or 4800 mg/day.
3. Methods
3.1. Part I. Simulated pharmacokinetic model — predicted bioa6ailability and steady-state plasma gabapentin concentrations A pharmacokinetic model was developed to predict steady-state plasma gabapentin concentrations. Following oral administration, plasma gabapentin concentrations are adequately described by a one compartment open model with first order absorption and elimination rate processes (Vollmer et al., 1989). Eq. (1) (Gilbaldi and Perrier, 1975) describes steady-state plasma gabapentin concentrations (C) as a function of time (t) for a one compartment open model with first order absorption and elimination where ka is the absorption rate constant, kel is the elimination rate constant, F is the fraction of drug absorbed, Vd is the apparent distribution volume, and t is the dosing interval. C=
kaFDose(e − kel t − e − ka t) Vd(ka − kel)(1− e kelt)(1− e kat)
(1)
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
Estimates for ka, kel, and Vd were obtained from a non-linear least-squares regression analysis (Heinzel et al., 1993) of mean plasma gabapentin concentration-time data obtained from a 300-mg single dose bioavailability study in healthy volunteers (data on file, Study 945-055-0, Parke-Davis Pharmaceutical Research, Division of WarnerLambert). The values for ka, kel, and Vd were 0.778 h − 1, 0.139 h − 1, and 45.4 l, respectively. Because gabapentin bioavailability decreases with increasing dose, a non-linear function (Eq. (2)) was used to describe the relationship between gabapentin bioavailability and daily dose, and relates gabapentin bioavailability (F) to daily dose over the dosage range of 300 – 4800 mg/day (Bockbrader et al., 1996). The fraction of drug absorbed in Eq. (2) is defined as the ratio of Ae, the amount of gabapentin excreted unchanged in the urine over 24 h (for gabapentin this value is equal to the dose absorbed), to daily dose (Dose). The fraction absorbed varies with the maximum amount of drug absorbed per day (Dmax) and inversely to the sum of daily dose and the dose where absorption is half-saturated (D50). F =Ae/Dose=Dmax/(D50 +Dose)
(2)
Best estimates for Dmax and D50 were obtained from a non-linear least-squares regression analysis (PCNONLIN) of mean amount of gabapentin excreted in urine over 24 h and daily dose from three clinical pharmacokinetic studies in healthy volunteers (Bockbrader, 1995).
3.2. Part II. Clinical study 3.2.1. Patients Eligible study subjects included all adult ( \ 18 years) men and women patients with epilepsy who were currently receiving gabapentin as either adjunctive or monotherapy. Eligible patients must have been receiving gabapentin at a stable dosage of either 3600 or 4800 mg/day for at least 1 month prior to evaluation. Patients with known or suspected histories of non-compliance, renal or gastrointestinal disorders were not eligible for study participation. This study was approved by the University of Wisconsin and University of Miami Institutional Review Boards, and informed
93
written consent was obtained prior to study enrollment.
3.2.2. Study design The clinical study was conducted in three phases, with Phase I being the baseline medication standardization period, Phase II being the first pharmacokinetic assessment period, and Phase III being the second pharmacokinetics assessment period. Total study duration was 9 days. 3.2.3. Pharmacokinetic analysis Since the only route for gabapentin elimination is renal excretion, the amount of drug present in the urine is equal to the amount absorbed. Total amount of drug excreted in urine (Ae), was calculated as the product of measured urinary gabapentin concentration and total urine volume. The fraction of drug recovered in urine (and therefore bioavailability (F)), was calculated as the quotient of Ae and individual daily gabapentin dose, and expressed as a percentage (Eq. (3)). F(%)= Ae/Dose×100
(3)
3.2.4. Study procedures 3.2.4.1. Phase I. Baseline assessment period. Eligible patients were instructed to standardize their current prescribed gabapentin regimen in the following fashion: patients currently taking gabapentin on a t.i.d. schedule, were asked to take their medication at 08:00, 16:00 and 22:00 h, while patients currently taking gabapentin on a q.i.d. schedule were instructed to take their medication at 08:00, 12:00, 18:00 and 22:00 h. In order to facilitate compliance, medication was pre-loaded in a medication pill-box container, and patients were instructed on its use. Medication administration was standardized in this manner for 3 days prior to commencement of the first pharmacokinetic assessment period, in order to assure that steady-state conditions were achieved. 3.2.4.2. Phase II. First pharmacokinetic assessment period. Following completion of the baseline standardization period, patients were asked to report to the epilepsy clinic in the early morning, prior to
94
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
taking their first morning dose of gabapentin. Medication compliance was assessed via patient interview and pill counts. Patients were then instructed to completely void, and were given both written and verbal instructions regarding 24-h urine collections. Patients were allowed to take their first morning dose of gabapentin, and instructed to continue taking gabapentin as during the baseline period. The timed urine collection was conducted over 24 h.
3.2.4.3. Phase III. Second pharmacokinetic assessment period. Following completion of Phase II, patients were given written and verbal instructions as to their new medication administration schedule (t.i.d. to q.i.d.) which they were to follow for the next 72 h. Medication again was provided in pre-loaded pillboxes. An identical program of urine collection was employed during this final study phase. Following the completion of Phase III, all patients were instructed to resume their original prescribed medication regimen.
4. Laboratory analysis Urine gabapentin concentrations were determined using a modified HPLC technique as described previously (Lensmeyer et al., 1995). Briefly, gabapentin is extracted from urine with an octyldecyl (C-18) solid phase adsorbent column. Analytical detection was enhanced using derivatization with trinitrobenzene, and then concentrated on a thin solid phase C-18 membrane. Following elution from the membrane, the derivatives were injected directly onto a Ultrasphere C-18 HPLC column with UV detection at 340 nm. This analytical method is a unique process that uses two solid phase extractions to achieve selectivity. The first extraction isolates gabapentin from potentially interfering compounds such as proteins and amino acids. The second extraction separates the derivatized gabapentin from the derivatizing reagent and from compounds that may have also been derivatized. Between-run coefficient of variation was between 2.3 and 2.9%.
5. Statistical analysis Patients in each gabapentin dosage group (3600 and 4800 mg/day) served as their own control. Fraction of gabapentin dose excreted in urine was compared using Student’s t-test for paired samples. All data is presented as mean values9 1 S.D. Interpatient variability in bioavailability was also expressed as a coefficient of variation (CV). Statistical significance was assigned at PB 0.05. A 20% increase in bioavailability was considered clinically significant.
6. Results
6.1. Part I. Predicted data Fig. 1 illustrates the non-linear relationship between amount of drug absorbed and oral gabapentin dose. The predicted line is based on a non-linear least squares regression analysis of amount absorbed and daily dose assuming a saturable absorption process. The best estimates for the model fit is a maximum absorption rate of 2720 mg/day and a daily dose where the absorption process is half saturated at 4080 mg. Fig. 2A illustrates the predicted steady-state plasma gabapentin concentration-time profiles following daily doses of 1200, 2400, 3600, and 4800 mg/day given in equally divided doses q8h. The predicted plasma gabapentin concentrationtime profiles are in excellent agreement with steady-state data obtained in a multiple-dose dose-proportionality study in healthy volunteers (Vollmer et al., 1989). Fig. 2B illustrates the predicted steady-state plasma gabapentin concentration-time profiles for the same daily doses used in Fig. 2A except the drug is given in equally divided doses q6h. The plasma gabapentin concentrations and absorption parameters predicted for the q8h and q6h dosing schedules are given in Table 1. The steady-state plasma gabapentin concentrations and amount absorbed are similar at 1200 and 2400 mg/day for both the q6h and q8h administration. At doses above 2400 mg/day, the effect of more frequent dosing administration becomes apparent.
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
95
Fig. 1. Observed (solid circles) and predicted (solid line) amount of gabapentin absorbed following a single oral dose of gabapentin to healthy volunteers.
6.2. Part II. Clinical study A total of twenty adult patients completed this study. There were nine patients (two women, seven men), with a mean age of 38.7 910 years, Table 1 Predicted pharmacokinetic values for q8h and q6h dosing Daily gabapentin dose 1200
2400
3600
4800
Given q6h Cmax Cavg Cmin Ae F
4.8 4 3.2 655 55
8.2 6.8 5.5 1110 46
10.6 8.9 7.2 1444 40
12.5 10.5 8.4 1700 35
Given q8h Cmax Cavg Cmin Ae F
4.9 3.9 2.6 618 52
8.1 6.2 4.3 1010 42
10.2 7.8 5.4 1280 35
11.8 9 6.3 1470 31
and weight of 80.49 20.9 kg in the 3600-mg/day group, while 11 patients (six women, five men), with a mean age of 32.69 10.1 years and weight of 74.2 914 kg comprised the 4800-mg/day group. Concomitant antiepileptic drugs were allowed in this study and included; carbamazepine (n= 10), phenytoin (n= 4), valproic acid (n =2) and lamotrigine (n= 1). At 3600 mg/day, the mean fraction of drug recovered unchanged in urine (i.e. bioavailability, F) following a three times per day (t.i.d.) dosage administration schedule was 38.69 22.1%. After conversion to a four times per day (q.i.d.) schedule, bioavailability was 40.09 18.9%, a difference that was not significantly different (P = 0.738). Bioavailability CV and individual patient data are displayed in Table 2. For patients receiving the higher dosage of 4800 mg/day, bioavailability while receiving gabapentin on a t.i.d. basis was 29.29 16.2%. Following conversion to the q.i.d. administration schedule, bioavailability was noted to significantly increase to 35.79 17.6% (P= 0.006). Bioavailability CV
96
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
Fig. 2. (A) Observed and predicted steady-state plasma gabapentin concentrations following administration of 1200 mg/day (solid line, open circle), 2400 mg/day (short dashed line, triangle pointing downwards), 3600 mg/day (long dashed line, square), and 4800 mg/day (medium dashed line, triangle pointing upwards) given in equally divided doses every 8 h. (B) Predicted steady-state plasma gabapentin concentrations following administration of 1200 mg/day (solid line), 2400 mg/day (long dashed line), 3600 mg/day (medium dashed line), and 4800 mg/day (short dashed line) given in equally divided doses every 6 h.
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
97
Table 2 Patient demographics and gabapentin bioavailability, 3600 mg/day Patient no.
Gender
Age (years)
Weight (kg)
F (%) t.i.d.
F (%) q.i.d.
1 2 3 4 5 6 7 8 9
M F M M M M M M F
35 61 36 43 42 37 40 28 27
81.8 56.4 91.8 70 87.2 73 70.9 128 64.5
33.4 71.2 15.6 54.3 72.5 29 19 18.8 34.1
40.9 69.3 20.2 38.1 65.9 36.3 10.5 44 35.2
80.4 20.9
38.7 22.1 57
40* 18.9 47
Mean S.D. CV (%)
38.7 10
*Difference not statistically significant, P = 0.738.
and individual patient data are displayed in Table 3.
7. Adverse effects Although this study was not designed to evaluate clinical outcomes, no patient reported any change in either seizure frequency or perception of adverse effects during this study period.
8. Discussion Gabapentin is a hydrophilic, zwitterionic compound, whose absorption most likely includes a carrier-mediated process (L-amino acid transport system), which seems to be responsible for the dose dependent, saturable absorption pattern of this drug (Stewart et al., 1993). In addition, it has been postulated that a non-saturable, paracellular pathway may also exist (Stevenson et al., 1997a). Gabapenin is poorly absorbed from the colon in both animals and humans, and absorption therefore is probably limited to the small intestine (Stevenson et al., 1997a,b). Good agreement exists between the predicted and observed gabapentin bioavailability at 3600 and 4800 mg/day following either t.i.d. or q.i.d. dosing. The biggest difference between predicted
and observed values occurred at 3600 mg/day given t.i.d. with a model predicted bioavailability of 35%, versus 38.7% observed in the clinical study. Our data demonstrate that increasing the frequency of drug administration may result in significant increases in gabapentin availability, with the magnitude of effect dependant upon drug dosage level. At daily doses of 3600–4800 mg/ day, the effect of more frequent gabapentin administration becomes more apparent. The fluctuation between Cmax and Cmin is less and Cmin values are higher for q6h versus q8h administration. Based upon both model predicted and observed, the amount absorbed per day is somewhat higher for the q6h administration. At 3600 mg/ day, approximately 50–160 mg more drug is absorbed with the q6h as compared to the q8h administration. These data suggest therefore that at 3600 mg/ day, the difference in gabapentin bioavailability between q8h and q6h dosing regimens is not clinically significant. Based upon both the predicted and clinical data, gabapentin bioavailability may be expected to increase by only 3–14% in the average patient taking 3600 mg/day. Examination of individual patient data however (Table 2) shows that 4/9 patients did display clinically significant (\ 20%) increases in bioavailability. Nevertheless, more frequent administration schedules
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
98
Table 3 Patient demographics and gabapentin bioavailability, 4800 mg/day Patient no.
Gender
Age (years)
Weight (kg)
F (%) t.i.d.
F (%) q.i.d.
1 2 3 4 5 6 7 8 9 10 11
F F M F F M F F M M M
24 45 28 45 22 39 21 37 21 31 46
72.3 79.1 82.3 57.3 57.7 76.4 94.5 73.2 50 85 88.6
19.8 15.5 31.6 37.9 11.2 21.8 43.2 61.6 20.5 12.1 46.2
29.8 24.1 30.9 48.4 11.2 24.0 54.3 74.2 30.9 23.7 40.8
32.6 10.1
74.2 14
29.2 16.2 55
35.6* 17.6 49
Mean S.D. CV (%)
*Difference statistically significant, P= 0.006.
would not appear to be pharmacokinetically beneficial at these (and presumably lower) daily dosages in the average patient. At daily dosages of 4800 mg however, between 230 and 312 mg more gabapentin is predicted to be absorbed following q6h administration as compared to q8h. When doses of 4800 mg/day are given q6h, our data would predict an increase in gabapentin bioavailability of 13 – 22%, relative to a q8h administration schedule. Thus, one would expect higher steady-state Cavg and Cmin plasma gabapentin concentrations and better bioavailability with more frequent drug administration. Although the primary objective of this study was to evaluate the effect of drug administration frequency on bioavailability, both our simulated, and clinical pharmacokinetic data do provide some other interesting observations. When viewed in terms of absolute increases in drug absorption, a difference in gabapentin daily dosage between 3600 and 4800 mg yields only modest additional absorbed drug. We observed that a 33% dosage increase yields on average only approximately 15% additional systemically available drug. These observations are clearly consistent with saturable drug absorption. Finally, our data also suggest that substantial intersubject variability exists for gabapentin bioavailability, with CVs ranging between 47 and
57%. This observation is in agreement with the observations of Hellriegel et al. (1996) who have shown that the lower a given drug’s bioavailability, the greater the intersubject variability in bioavailability. This may be clinically relevant in the following sense: although, on average, patients receiving gabapentin dosages of 3600 mg/day or less may not see an increase in absorption with more frequent drug administration, individual patients might. Conceivably, therapeutic drug concentration monitoring may be clinically useful in determining the pharmacokinetic outcome of drug regimen modifications in individual patients receiving gabapentin. Although statistically significant increases in gabapentin bioavailability may be achieved by more frequent drug administration, these potential gains are modest, and may come at a price of diminished medication compliance. Indeed, Cramer et al. (1989), have reported that patients with epilepsy are compliant with prescribed regimens at a rate of 77% when taking medications three times daily, versus a compliance rate of 39% when taking medications four times daily. Given these observations, any benefit that may result from increased systemic drug availability may, in some patients, be counterbalanced by virtue of complicating the medication administration schedule. Clinicians must therefore be vigilant in
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
looking for medication non-compliance when schedules that require more frequent dosing are employed. In summary, data derived from a clinical pharmacokinetic study in adult patients with epilepsy as well as that from a simulated kinetic model suggest that at large daily doses (4800 mg/day), alteration of drug administration frequency may modestly enhance gabapentin systemic availability, resulting in increased serum concentrations. At lower daily doses, there appears to be no significant advantage to this manipulation.
Acknowledgements Funding and gabapentin study medication was provided by Parke-Davis.
References Bockbrader, H.N., 1995. Clinical pharmacokinetics of gabapentin. Drugs Today 31, 613–619. Bockbrader, H.N., Breslin, E.M., Underwood, B.A., Posvar, E.L., Sedman, A.J., 1996. Multiple-dose, dose-proportionality study of Neurontin (gabapentin) in health volunteers. Epilepsia 37 (5), 159.
.
99
Cramer, J.A., Mattson, R.H., Prevey, M.L., et al., 1989. How often is medication taken as prescribed? A novel assessment technique. J. Am. Med. Assoc. 261, 3273 – 3277. Gilbaldi, M., Perrier, D. (Eds.), 1975. Pharmacokinetics, vol. 1. Marcel Decker, New York. Heinzel, G., Woloszczak, R., Thomann, P., 1993. Topfit version 2.0: Pharmacokinetic and Pharmacodynamic Data Analysis System for the PC. Fischer, Stuttgart. Hellriegel, E.T., Bjornsson, T.D., Hauck, W.W., 1996. Interpatient variability in bioavailability is related to the extent of absorption: implications for bioavailability and bioequivalence studies. Clin. Pharmacol. Ther. 60, 601 – 607. Lensmeyer, G., Kempf, T., Gidal, B., Weibe, D., 1995. Optimized method for determination of gabapentin in serum by HPL. Ther. Drug Monit. 17, 251 – 258. McLean, M.J., 1994. Clinical pharmacokinetics of gabapentin. Neurology 44, S17 – S22. Stevenson, C.M., Radulovic, L.L., Bockbrader, H.N., Fleisher, D., 1997a. Contrasting nutrient effects on the plasma levels of an amino acid like antiepileptic agent from the jejunal administration in dogs. J. Pharm. Sci. 86, 953 – 957. Stevenson, C.M., Kim, J., Fleisher, D., 1997b. Colonic absorption of antiepileptic agents. Epilepsia 38, 63 – 67. Stewart, B.H., Kugler, A.R., Thompson, P.R., Bockbrader, H.N., 1993. A saturable transport mechanism in the intestinal absorption of gabapentin is the underlying cause of the lack of proportionality between increasing dose and drug levels in plasma. Pharm. Res. 10, 276 – 282. Vollmer, K.O., Anhut, H., Thomann, P., Wagner, F., Jaehnchen, D., 1989. Pharmacokinetic model and absolute bioavailability of the new anticonvulsant gabapentin. Adv. Epileptol. 17, 209 – 211.
.
Gabapentin bioavailability: effect of dose and frequency of administration in adult patients with epilepsy Barry E. Gidal a,b,*, John DeCerce c, Howard N. Bockbrader d, Jose Gonzalez c, Sarah Kruger b, Michael E. Pitterle a, Paul Rutecki b, R. Eugene Ramsay c b
a School of Pharmacy, Uni6ersity of Wisconsin, 425 N. Charter Street, Madison, WI 53706, USA Department of Neurology, Uni6ersity of Wisconsin, 425 N. Charter Street, Madison, WI 53706, USA c Uni6ersity of Miami, International Center for Epilepsy, Miami, FL, USA d Warner-Lambert Co., Parke-Da6is Pharmaceutical Research Di6ision, Ann Arbor, MI, USA
Received 2 December 1997; received in revised form 10 March 1998; accepted 11 March 1998
Abstract Gabapentin (GBP) is a non-metabolized antiepileptic drug that is eliminated by renal excretion and displays saturable, dose dependent absorption. The recommended dosing schedule for GBP is t.i.d. At large daily doses, oral bioavailability (F) may be improved by giving the daily dose more frequently. Objecti6e: To evaluate whether switching GBP dosage regimen from t.i.d. to q.i.d. results in increased oral bioavailability. Methods: This study consisted of two parts; a computer simulated pharmacokinetic model and a clinical pharmacokinetic study in nine adult epileptic patients receiving 3600 mg/day and 11 receiving 4800 mg/day. All patients were evaluated during both t.i.d. and q.i.d. regimens. F were determined by calculation of percent of dose excreted unchanged using steady-state 24-h urine collections and were compared using a paired t-test. Results: At 3600 mg/day, mean F following t.i.d. and q.i.d. dosing were 38.7 922.1% and 40.0 918.9%, respectively (P= 0.738). At 4800 mg/day, mean F following t.i.d. and q.i.d. dosing were 29.2 916.2% and 35.6 917.6%, respectively (P=0.006). Discussion: Good agreement was observed between values from this study and predicted values based on the pharmacokinetic model. Improved GBP F at doses of 3600 mg/day was not achieved with more frequent drug administration, and thus is not warranted. At 4800 mg/day, a 22% increase in F was observed with more frequent drug dosing. Conclusion: GBP F may be significantly increased by q.i.d. versus t.i.d. dosing, depending upon dose level. This increase in F however must be balanced against the inconvenience of more frequent dosing. Therapeutic drug level monitoring may aid in the evaluation of such pharmacokinetic maneuvers. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Gabapentin; Bioavailability; Epilepsy
* Corresponding author. Tel.: +1 608 2623280; fax: +1 608 2655421; e-mail: [email protected] 0920-1211/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S0920-1211(98)00020-5
92
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
1. Introduction
2. Aims and objectives
Gabapentin is an antiepileptic medication approved in the USA for adjunct therapy of adult patients with partial seizures, with and without secondary generalization. In addition to well established efficacy and a modest toxicity profile, gabapentin has a unique pharmacokinetic profile. Gabapentin is excreted unchanged in the urine, with its clearance correlating with creatinine clearance (McLean, 1994). Gabapentin has a elimination half-life of approximately 6 h in patients with normal renal function, does not bind to plasma proteins, and does not appear to interact pharmacokinetically with other antiepileptic medications. Of clinical interest, gabapentin displays dose dependent absorption, with systemic bioavailability decreasing with increasing doses. The mechanism underlying this phenomenon appears to involve an active and saturable transport mechanism (Stewart et al., 1993). In clinical trials, plasma gabapentin concentrations increase with increasing doses but the increases were not dose proportional (linear). Bioavailability following single doses in healthy subjects is approximately 57% at 300 mg and 42% at 600 mg (McLean, 1994). Currently, there is no recognized maximal daily dose of gabapentin and many patients are receiving more than 1800 mg/day. Given this well documented property of saturable absorption, it is reasonable to hypothesize that the current recommendation of a three times per day (t.i.d.) administration schedule may not be optimal. At higher doses (\ 1800 mg/day), a four times per day (q.i.d.) schedule may result in greater systemic bioavailability. This may not only have clinical implications, in that gabapentin efficacy appears to increase with increasing daily dose, but economic consequences as well. At a given daily dose, if systemic bioavailability is increased, the ‘effective’ systemic dose or exposure to gabapentin would in essence, be increased with no increase in medication expense.
The aim of this study was to evaluate the overall pharmacokinetic impact of modifications in gabapentin oral administration schedules. This study was conducted in two parts including the application of a computer-modeled pharmacokinetic simulation and a clinical pharmacokinetic study in adult epileptic patients. The first component investigates predicted steady-state plasma gabapentin concentrations and absorption parameters that would be expected from t.i.d. and q.i.d. drug administration regimens. The second component is a clinical study to test the hypothesis that increasing the frequency of gabapentin administration results in a significant increase in gabapentin bioavailability. The clinical study evaluated the impact of increasing the gabapentin dosage regimens from t.i.d. to q.i.d. in epileptic patients currently maintained on large yet clinically relevant dosages of 3600 or 4800 mg/day.
3. Methods
3.1. Part I. Simulated pharmacokinetic model — predicted bioa6ailability and steady-state plasma gabapentin concentrations A pharmacokinetic model was developed to predict steady-state plasma gabapentin concentrations. Following oral administration, plasma gabapentin concentrations are adequately described by a one compartment open model with first order absorption and elimination rate processes (Vollmer et al., 1989). Eq. (1) (Gilbaldi and Perrier, 1975) describes steady-state plasma gabapentin concentrations (C) as a function of time (t) for a one compartment open model with first order absorption and elimination where ka is the absorption rate constant, kel is the elimination rate constant, F is the fraction of drug absorbed, Vd is the apparent distribution volume, and t is the dosing interval. C=
kaFDose(e − kel t − e − ka t) Vd(ka − kel)(1− e kelt)(1− e kat)
(1)
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
Estimates for ka, kel, and Vd were obtained from a non-linear least-squares regression analysis (Heinzel et al., 1993) of mean plasma gabapentin concentration-time data obtained from a 300-mg single dose bioavailability study in healthy volunteers (data on file, Study 945-055-0, Parke-Davis Pharmaceutical Research, Division of WarnerLambert). The values for ka, kel, and Vd were 0.778 h − 1, 0.139 h − 1, and 45.4 l, respectively. Because gabapentin bioavailability decreases with increasing dose, a non-linear function (Eq. (2)) was used to describe the relationship between gabapentin bioavailability and daily dose, and relates gabapentin bioavailability (F) to daily dose over the dosage range of 300 – 4800 mg/day (Bockbrader et al., 1996). The fraction of drug absorbed in Eq. (2) is defined as the ratio of Ae, the amount of gabapentin excreted unchanged in the urine over 24 h (for gabapentin this value is equal to the dose absorbed), to daily dose (Dose). The fraction absorbed varies with the maximum amount of drug absorbed per day (Dmax) and inversely to the sum of daily dose and the dose where absorption is half-saturated (D50). F =Ae/Dose=Dmax/(D50 +Dose)
(2)
Best estimates for Dmax and D50 were obtained from a non-linear least-squares regression analysis (PCNONLIN) of mean amount of gabapentin excreted in urine over 24 h and daily dose from three clinical pharmacokinetic studies in healthy volunteers (Bockbrader, 1995).
3.2. Part II. Clinical study 3.2.1. Patients Eligible study subjects included all adult ( \ 18 years) men and women patients with epilepsy who were currently receiving gabapentin as either adjunctive or monotherapy. Eligible patients must have been receiving gabapentin at a stable dosage of either 3600 or 4800 mg/day for at least 1 month prior to evaluation. Patients with known or suspected histories of non-compliance, renal or gastrointestinal disorders were not eligible for study participation. This study was approved by the University of Wisconsin and University of Miami Institutional Review Boards, and informed
93
written consent was obtained prior to study enrollment.
3.2.2. Study design The clinical study was conducted in three phases, with Phase I being the baseline medication standardization period, Phase II being the first pharmacokinetic assessment period, and Phase III being the second pharmacokinetics assessment period. Total study duration was 9 days. 3.2.3. Pharmacokinetic analysis Since the only route for gabapentin elimination is renal excretion, the amount of drug present in the urine is equal to the amount absorbed. Total amount of drug excreted in urine (Ae), was calculated as the product of measured urinary gabapentin concentration and total urine volume. The fraction of drug recovered in urine (and therefore bioavailability (F)), was calculated as the quotient of Ae and individual daily gabapentin dose, and expressed as a percentage (Eq. (3)). F(%)= Ae/Dose×100
(3)
3.2.4. Study procedures 3.2.4.1. Phase I. Baseline assessment period. Eligible patients were instructed to standardize their current prescribed gabapentin regimen in the following fashion: patients currently taking gabapentin on a t.i.d. schedule, were asked to take their medication at 08:00, 16:00 and 22:00 h, while patients currently taking gabapentin on a q.i.d. schedule were instructed to take their medication at 08:00, 12:00, 18:00 and 22:00 h. In order to facilitate compliance, medication was pre-loaded in a medication pill-box container, and patients were instructed on its use. Medication administration was standardized in this manner for 3 days prior to commencement of the first pharmacokinetic assessment period, in order to assure that steady-state conditions were achieved. 3.2.4.2. Phase II. First pharmacokinetic assessment period. Following completion of the baseline standardization period, patients were asked to report to the epilepsy clinic in the early morning, prior to
94
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
taking their first morning dose of gabapentin. Medication compliance was assessed via patient interview and pill counts. Patients were then instructed to completely void, and were given both written and verbal instructions regarding 24-h urine collections. Patients were allowed to take their first morning dose of gabapentin, and instructed to continue taking gabapentin as during the baseline period. The timed urine collection was conducted over 24 h.
3.2.4.3. Phase III. Second pharmacokinetic assessment period. Following completion of Phase II, patients were given written and verbal instructions as to their new medication administration schedule (t.i.d. to q.i.d.) which they were to follow for the next 72 h. Medication again was provided in pre-loaded pillboxes. An identical program of urine collection was employed during this final study phase. Following the completion of Phase III, all patients were instructed to resume their original prescribed medication regimen.
4. Laboratory analysis Urine gabapentin concentrations were determined using a modified HPLC technique as described previously (Lensmeyer et al., 1995). Briefly, gabapentin is extracted from urine with an octyldecyl (C-18) solid phase adsorbent column. Analytical detection was enhanced using derivatization with trinitrobenzene, and then concentrated on a thin solid phase C-18 membrane. Following elution from the membrane, the derivatives were injected directly onto a Ultrasphere C-18 HPLC column with UV detection at 340 nm. This analytical method is a unique process that uses two solid phase extractions to achieve selectivity. The first extraction isolates gabapentin from potentially interfering compounds such as proteins and amino acids. The second extraction separates the derivatized gabapentin from the derivatizing reagent and from compounds that may have also been derivatized. Between-run coefficient of variation was between 2.3 and 2.9%.
5. Statistical analysis Patients in each gabapentin dosage group (3600 and 4800 mg/day) served as their own control. Fraction of gabapentin dose excreted in urine was compared using Student’s t-test for paired samples. All data is presented as mean values9 1 S.D. Interpatient variability in bioavailability was also expressed as a coefficient of variation (CV). Statistical significance was assigned at PB 0.05. A 20% increase in bioavailability was considered clinically significant.
6. Results
6.1. Part I. Predicted data Fig. 1 illustrates the non-linear relationship between amount of drug absorbed and oral gabapentin dose. The predicted line is based on a non-linear least squares regression analysis of amount absorbed and daily dose assuming a saturable absorption process. The best estimates for the model fit is a maximum absorption rate of 2720 mg/day and a daily dose where the absorption process is half saturated at 4080 mg. Fig. 2A illustrates the predicted steady-state plasma gabapentin concentration-time profiles following daily doses of 1200, 2400, 3600, and 4800 mg/day given in equally divided doses q8h. The predicted plasma gabapentin concentrationtime profiles are in excellent agreement with steady-state data obtained in a multiple-dose dose-proportionality study in healthy volunteers (Vollmer et al., 1989). Fig. 2B illustrates the predicted steady-state plasma gabapentin concentration-time profiles for the same daily doses used in Fig. 2A except the drug is given in equally divided doses q6h. The plasma gabapentin concentrations and absorption parameters predicted for the q8h and q6h dosing schedules are given in Table 1. The steady-state plasma gabapentin concentrations and amount absorbed are similar at 1200 and 2400 mg/day for both the q6h and q8h administration. At doses above 2400 mg/day, the effect of more frequent dosing administration becomes apparent.
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
95
Fig. 1. Observed (solid circles) and predicted (solid line) amount of gabapentin absorbed following a single oral dose of gabapentin to healthy volunteers.
6.2. Part II. Clinical study A total of twenty adult patients completed this study. There were nine patients (two women, seven men), with a mean age of 38.7 910 years, Table 1 Predicted pharmacokinetic values for q8h and q6h dosing Daily gabapentin dose 1200
2400
3600
4800
Given q6h Cmax Cavg Cmin Ae F
4.8 4 3.2 655 55
8.2 6.8 5.5 1110 46
10.6 8.9 7.2 1444 40
12.5 10.5 8.4 1700 35
Given q8h Cmax Cavg Cmin Ae F
4.9 3.9 2.6 618 52
8.1 6.2 4.3 1010 42
10.2 7.8 5.4 1280 35
11.8 9 6.3 1470 31
and weight of 80.49 20.9 kg in the 3600-mg/day group, while 11 patients (six women, five men), with a mean age of 32.69 10.1 years and weight of 74.2 914 kg comprised the 4800-mg/day group. Concomitant antiepileptic drugs were allowed in this study and included; carbamazepine (n= 10), phenytoin (n= 4), valproic acid (n =2) and lamotrigine (n= 1). At 3600 mg/day, the mean fraction of drug recovered unchanged in urine (i.e. bioavailability, F) following a three times per day (t.i.d.) dosage administration schedule was 38.69 22.1%. After conversion to a four times per day (q.i.d.) schedule, bioavailability was 40.09 18.9%, a difference that was not significantly different (P = 0.738). Bioavailability CV and individual patient data are displayed in Table 2. For patients receiving the higher dosage of 4800 mg/day, bioavailability while receiving gabapentin on a t.i.d. basis was 29.29 16.2%. Following conversion to the q.i.d. administration schedule, bioavailability was noted to significantly increase to 35.79 17.6% (P= 0.006). Bioavailability CV
96
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
Fig. 2. (A) Observed and predicted steady-state plasma gabapentin concentrations following administration of 1200 mg/day (solid line, open circle), 2400 mg/day (short dashed line, triangle pointing downwards), 3600 mg/day (long dashed line, square), and 4800 mg/day (medium dashed line, triangle pointing upwards) given in equally divided doses every 8 h. (B) Predicted steady-state plasma gabapentin concentrations following administration of 1200 mg/day (solid line), 2400 mg/day (long dashed line), 3600 mg/day (medium dashed line), and 4800 mg/day (short dashed line) given in equally divided doses every 6 h.
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
97
Table 2 Patient demographics and gabapentin bioavailability, 3600 mg/day Patient no.
Gender
Age (years)
Weight (kg)
F (%) t.i.d.
F (%) q.i.d.
1 2 3 4 5 6 7 8 9
M F M M M M M M F
35 61 36 43 42 37 40 28 27
81.8 56.4 91.8 70 87.2 73 70.9 128 64.5
33.4 71.2 15.6 54.3 72.5 29 19 18.8 34.1
40.9 69.3 20.2 38.1 65.9 36.3 10.5 44 35.2
80.4 20.9
38.7 22.1 57
40* 18.9 47
Mean S.D. CV (%)
38.7 10
*Difference not statistically significant, P = 0.738.
and individual patient data are displayed in Table 3.
7. Adverse effects Although this study was not designed to evaluate clinical outcomes, no patient reported any change in either seizure frequency or perception of adverse effects during this study period.
8. Discussion Gabapentin is a hydrophilic, zwitterionic compound, whose absorption most likely includes a carrier-mediated process (L-amino acid transport system), which seems to be responsible for the dose dependent, saturable absorption pattern of this drug (Stewart et al., 1993). In addition, it has been postulated that a non-saturable, paracellular pathway may also exist (Stevenson et al., 1997a). Gabapenin is poorly absorbed from the colon in both animals and humans, and absorption therefore is probably limited to the small intestine (Stevenson et al., 1997a,b). Good agreement exists between the predicted and observed gabapentin bioavailability at 3600 and 4800 mg/day following either t.i.d. or q.i.d. dosing. The biggest difference between predicted
and observed values occurred at 3600 mg/day given t.i.d. with a model predicted bioavailability of 35%, versus 38.7% observed in the clinical study. Our data demonstrate that increasing the frequency of drug administration may result in significant increases in gabapentin availability, with the magnitude of effect dependant upon drug dosage level. At daily doses of 3600–4800 mg/ day, the effect of more frequent gabapentin administration becomes more apparent. The fluctuation between Cmax and Cmin is less and Cmin values are higher for q6h versus q8h administration. Based upon both model predicted and observed, the amount absorbed per day is somewhat higher for the q6h administration. At 3600 mg/ day, approximately 50–160 mg more drug is absorbed with the q6h as compared to the q8h administration. These data suggest therefore that at 3600 mg/ day, the difference in gabapentin bioavailability between q8h and q6h dosing regimens is not clinically significant. Based upon both the predicted and clinical data, gabapentin bioavailability may be expected to increase by only 3–14% in the average patient taking 3600 mg/day. Examination of individual patient data however (Table 2) shows that 4/9 patients did display clinically significant (\ 20%) increases in bioavailability. Nevertheless, more frequent administration schedules
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
98
Table 3 Patient demographics and gabapentin bioavailability, 4800 mg/day Patient no.
Gender
Age (years)
Weight (kg)
F (%) t.i.d.
F (%) q.i.d.
1 2 3 4 5 6 7 8 9 10 11
F F M F F M F F M M M
24 45 28 45 22 39 21 37 21 31 46
72.3 79.1 82.3 57.3 57.7 76.4 94.5 73.2 50 85 88.6
19.8 15.5 31.6 37.9 11.2 21.8 43.2 61.6 20.5 12.1 46.2
29.8 24.1 30.9 48.4 11.2 24.0 54.3 74.2 30.9 23.7 40.8
32.6 10.1
74.2 14
29.2 16.2 55
35.6* 17.6 49
Mean S.D. CV (%)
*Difference statistically significant, P= 0.006.
would not appear to be pharmacokinetically beneficial at these (and presumably lower) daily dosages in the average patient. At daily dosages of 4800 mg however, between 230 and 312 mg more gabapentin is predicted to be absorbed following q6h administration as compared to q8h. When doses of 4800 mg/day are given q6h, our data would predict an increase in gabapentin bioavailability of 13 – 22%, relative to a q8h administration schedule. Thus, one would expect higher steady-state Cavg and Cmin plasma gabapentin concentrations and better bioavailability with more frequent drug administration. Although the primary objective of this study was to evaluate the effect of drug administration frequency on bioavailability, both our simulated, and clinical pharmacokinetic data do provide some other interesting observations. When viewed in terms of absolute increases in drug absorption, a difference in gabapentin daily dosage between 3600 and 4800 mg yields only modest additional absorbed drug. We observed that a 33% dosage increase yields on average only approximately 15% additional systemically available drug. These observations are clearly consistent with saturable drug absorption. Finally, our data also suggest that substantial intersubject variability exists for gabapentin bioavailability, with CVs ranging between 47 and
57%. This observation is in agreement with the observations of Hellriegel et al. (1996) who have shown that the lower a given drug’s bioavailability, the greater the intersubject variability in bioavailability. This may be clinically relevant in the following sense: although, on average, patients receiving gabapentin dosages of 3600 mg/day or less may not see an increase in absorption with more frequent drug administration, individual patients might. Conceivably, therapeutic drug concentration monitoring may be clinically useful in determining the pharmacokinetic outcome of drug regimen modifications in individual patients receiving gabapentin. Although statistically significant increases in gabapentin bioavailability may be achieved by more frequent drug administration, these potential gains are modest, and may come at a price of diminished medication compliance. Indeed, Cramer et al. (1989), have reported that patients with epilepsy are compliant with prescribed regimens at a rate of 77% when taking medications three times daily, versus a compliance rate of 39% when taking medications four times daily. Given these observations, any benefit that may result from increased systemic drug availability may, in some patients, be counterbalanced by virtue of complicating the medication administration schedule. Clinicians must therefore be vigilant in
B.E. Gidal et al. / Epilepsy Research 31 (1998) 91–99
looking for medication non-compliance when schedules that require more frequent dosing are employed. In summary, data derived from a clinical pharmacokinetic study in adult patients with epilepsy as well as that from a simulated kinetic model suggest that at large daily doses (4800 mg/day), alteration of drug administration frequency may modestly enhance gabapentin systemic availability, resulting in increased serum concentrations. At lower daily doses, there appears to be no significant advantage to this manipulation.
Acknowledgements Funding and gabapentin study medication was provided by Parke-Davis.
References Bockbrader, H.N., 1995. Clinical pharmacokinetics of gabapentin. Drugs Today 31, 613–619. Bockbrader, H.N., Breslin, E.M., Underwood, B.A., Posvar, E.L., Sedman, A.J., 1996. Multiple-dose, dose-proportionality study of Neurontin (gabapentin) in health volunteers. Epilepsia 37 (5), 159.
.
99
Cramer, J.A., Mattson, R.H., Prevey, M.L., et al., 1989. How often is medication taken as prescribed? A novel assessment technique. J. Am. Med. Assoc. 261, 3273 – 3277. Gilbaldi, M., Perrier, D. (Eds.), 1975. Pharmacokinetics, vol. 1. Marcel Decker, New York. Heinzel, G., Woloszczak, R., Thomann, P., 1993. Topfit version 2.0: Pharmacokinetic and Pharmacodynamic Data Analysis System for the PC. Fischer, Stuttgart. Hellriegel, E.T., Bjornsson, T.D., Hauck, W.W., 1996. Interpatient variability in bioavailability is related to the extent of absorption: implications for bioavailability and bioequivalence studies. Clin. Pharmacol. Ther. 60, 601 – 607. Lensmeyer, G., Kempf, T., Gidal, B., Weibe, D., 1995. Optimized method for determination of gabapentin in serum by HPL. Ther. Drug Monit. 17, 251 – 258. McLean, M.J., 1994. Clinical pharmacokinetics of gabapentin. Neurology 44, S17 – S22. Stevenson, C.M., Radulovic, L.L., Bockbrader, H.N., Fleisher, D., 1997a. Contrasting nutrient effects on the plasma levels of an amino acid like antiepileptic agent from the jejunal administration in dogs. J. Pharm. Sci. 86, 953 – 957. Stevenson, C.M., Kim, J., Fleisher, D., 1997b. Colonic absorption of antiepileptic agents. Epilepsia 38, 63 – 67. Stewart, B.H., Kugler, A.R., Thompson, P.R., Bockbrader, H.N., 1993. A saturable transport mechanism in the intestinal absorption of gabapentin is the underlying cause of the lack of proportionality between increasing dose and drug levels in plasma. Pharm. Res. 10, 276 – 282. Vollmer, K.O., Anhut, H., Thomann, P., Wagner, F., Jaehnchen, D., 1989. Pharmacokinetic model and absolute bioavailability of the new anticonvulsant gabapentin. Adv. Epileptol. 17, 209 – 211.
.