Gabapentin and Pregabalin

Michael J. McLean

Barry E. Gidal

Gabapentin (1-[aminomethyl]cyclohexaneacetic acid), or (3-cyclohexyl γ-aminobutyric acid [GABA]; see Fig. 59.1) resulted from bonding cyclohexane to the 3-position of the GABA backbone to produce a structural analogue of baclofen. Although designed as a spasmolytic agent, gabapentin was developed to treat epilepsy (1, 2, 3). Initially, gabapentin was approved by the U.S. Food and Drug Administration (FDA) at the end of 1993 as an adjunctive agent for the treatment of complex partial seizures with or without secondary generalization in patients over 12 years of age. It was approved for children 3 to 12 years of age in 2001. The agent has been approved as initial monotherapy in about 40 countries outside the United States. In 2002, gabapentin was also approved for the treatment of postherpetic neuralgia.

Broad utility of gabapentin in open use led to the development of other 3-substituted GABA analogues—a group of compounds that have been dubbed “gabapentinoids.” One of these, pregabalin (3-[aminomethyl]-5-methyl-,[3S]-hexanoic acid), or (3-isobutyl-GABA; see Fig. 59.1), has been shown to be superior to placebo as add-on therapy for patients with refractory partial and secondarily generalized seizures in three pivotal trials. Anxiolytic and analgesic properties of pregabalin have also been demonstrated in placebo-controlled trials. The new drug application (NDA) containing data supporting several indications for pregabalin was submitted to the FDA in November 2003. As of July 2004, pregabalin had approval from the European Commission to market pregabalin as adjunctive treatment for partial seizures, as well as for the treatment of neuropathic pain.

CHEMISTRY

Gabapentin

Gabapentin is an amorphous crystalline substance with a molecular weight of 171.24. It is freely soluble in water (1,4). The structure of gabapentin combines an inhibitory amino acid, GABA, and a cyclohexane ring that result in a zwitterion at physiologic pH (1,4). The agent is actively transported between body compartments by the L-system amino acid transporter, which recognizes such naturally occurring, bulky, neutral amino acids as L-leucine, L-isoleucine, L-valine, and L-phenylalanine. The same carrier is presumed to mediate transport across the gut wall, the blood-brain barrier, and cell membranes (5,6). Gabapentin concentrations can be measured in protein-free plasma samples by high-performance liquid chromatography (7,8) and gas chromatography (9). Blood level assays are commercially available. Gabapentin degrades slowly to a lactam in solution as a function of pH, temperature, and buffer concentration (10). The lactam is formed during synthesis of gabapentin, and has both proconvulsant (11) and neuroprotective (12) properties in laboratory models. Presumably, the lactam does not accumulate in sufficient quantity to be clinically significant. The current proprietary synthetic methods result in a low-lactam product (13).

Pregabalin

Pregabalin is also a water-soluble compound, with a molecular weight of 159.23. It is several times more potent than gabapentin on a mg/kg basis in various animal seizure models (14). Bioavailability is 90% or more. This suggests that the L-system amino acid transporter concentrated in the duodenum is not saturated by useful doses of pregabalin (14). Additional uptake mechanisms for pregabalin may exist. Serum concentrations have been determined with high-performance liquid chromatography-ultraviolet (UV) methods (15). This suggests that commercial laboratories will offer pregabalin plasma level determinations once the agent has been approved for marketing.

Figure 59.1 Chemical structure of gabapentin and pregabalin.

MECHANISMS OF ACTION

Multiple actions of gabapentin and pregabalin have been reported in animal and cell models (14,16,17), at least some of which occurred at therapeutically relevant doses or concentrations. The actions of the two agents were similar in studies of in vitro and animal models when compared via similar technique(s) (Table 59.1). Effects of both gabapentinoids have been noted on calcium and potassium channels, as well as on neurotransmitters, including GABA, glutamate, and serotonin.

TABLE 59.1 EFFECTS OF GABAPENTIN AND PREGABALIN IN VITRO AND IN ANIMAL MODELS

	Gabapentin	Pregabalin
*In vitro models*
Binding site	α₂δ subunit of voltage-sensitive calcium channels	α₂δ subunit of voltage-sensitive calcium channels
Voltage-sensitive Ca⁺⁺ channels	Current through multiple types reduced	Current through multiple types reduced
Ca voltage-sensitive Na⁺ channels	No direct effect	—
Voltage-sensitive K⁺ channels	K_ATP current increased	K_ATP current increased
GABA	Increased concentration by MRS Increased GAD activity Increased GABA release Increased transporter expression	Inferred from animal models of decreased GABA synthesis or blockade
Glutamate	Decreased concentration	Decreased synthesis
Serotonin	Increased plasma levels associated with ↑ stage 3,4 sleep	Increase inferred from ↑ stage 3,4 sleep
Other neurotransmitters	Decreased release (↓Ca⁺⁺ and ↑K⁺ currents?)	Decreased release (↓Ca⁺⁺ and ↓K⁺ currents?)
Enzymes	Multiple effects to increase GABA, decrease glutamate; effects on PKA and PKC	—
Animal models
Maximal electroshock model	Effective	More potent than GBP
Models of reduced GABA synthesis or blockade	Partially effective	Partially effective
Genetic absence model (GAERS)	Ineffective	Ineffective
Abbreviation: ATP, adenosine triphosphate; GABA, γ-aminobutyric acid; GAD, glutamic acid decarboxylase; GAERS, genetic absence rats from Strasbourg; MRS, magnetic resonance spectroscopy; PKA, protein kinase A; PKC, protein kinase C.

Gabapentin is generally well tolerated, and pregabalin was well tolerated in clinical trials. This suggests that the sum of their actions is most likely modulatory, rather than highly potent. If the effects were highly potent, adverse events would limit clinical utility. Some actions may be
subtle under various clinical circumstances, with their relative clinical impact difficult to weigh. Thus, the mechanism(s) underlying clinical efficacy have not been determined with certainty.

PHARMACOKINETICS

The pharmacokinetic properties of gabapentin are generally favorable. However, dose-dependent bioavailability and interindividual variability in uptake require optimization on an individual basis. It is conceivable that some patients may fail to achieve desired seizure control because of inadequate absorption. This may have confounded the outcome of dose-controlled studies of gabapentin monotherapy (Vide Infra). Oral bioavailability of pregabalin, on the other hand, approaches 100% (14). This could contribute to the greater efficacy of pregabalin in clinical trials. Table 59.2 summarizes some important pharmacokinetic properties of these two agents.

Absorption

Gabapentin

Gabapentin is absorbed primarily in the small intestine, presumably because the L-amino acid transporter is concentrated there (18). Absorption of gabapentin in the colon is poor in both animals and humans (19,20).

Bioavailability of gabapentin is limited and dose-dependent. The plasma level after a single 300-mg oral dose, either as a capsule or as solution, was about 60% that of an intravenous (IV) formulation (21). On a multi-dose regimen of 1600 mg tid, bioavailability decreased to about 35% (22).

TABLE 59.2 PHARMACOKINETIC PROFILES OF GABAPENTIN AND PREGABALIN, WITH SIGNIFICANT DIFFERENCES LIMITED TO ABSORPTION

Absorption	Mediated by L-amino acid transporter Gabapentin: Dose-limiting bioavailability Increased by naproxen and morphine Decreased by hydrocodone Pregabalin: >90% bioavailability
Distribution	Water soluble Not extensively bound to plasma proteins
Metabolism	Not metabolized by liver No induction of hepatic enzymes No autoinduction No inhibition of hepatic enzymes
Elimination	Excreted intact in urine Excretion proportional to creatinine clearance Gabapentin: t_½ 4 to 22 hours; average, 5 to 7 hours Pregabalin: t_½ 5 to 7 hours
Interactions	No effect on other AEDs or oral contraceptives No effect of gabapentin on levels of other AEDs None with hepatic enzymes None with protein-binding sites
Abbreviation: AEDs, antiepileptic drugs; t_½, half-life;.

Mean maximal plasma levels (C_max) were 2.7 to 2.9 μg/mL within 2 to 3 hours after oral administration of a single 300-mg capsule to healthy volunteers, and peak steady-state plasma concentrations averaged 4 μg/mL following oral administration of 300 mg every 8 hours (22, 23, 24, 25).

A nonlinear relationship between dose and plasma levels may result from saturable absorption of gabapentin from the intestine (5). In phase 1 pharmacokinetic studies, plasma concentrations of gabapentin increased in proportion to the dose, up to 1800 mg per day. At doses from 1800 to 4800 mg per day (600 to 1600 mg q8h), plasma levels continued to rise, but less than expected (26). A non-linear increase in plasma levels was also noted in the data from some clinical trials (27,28). Data from pharmacokinetic studies conducted in both volunteers and patients suggest that substantial interpatient variability exists for gabapentin absorption. Plasma concentrations from the open studies of patients taking 2700 to 6000 mg per day were linearly related to dose, but interindividual variability was pronounced (29,30). While apparent variability of uptake between individuals can be substantial, far less variability is noted within subjects (31).

Administration with food or typical enteral nutritional formulations does not impair absorption of gabapentin (22,31). In that gabapentin is absorbed by the L-system amino acid transporter, it may be speculated that concomitant administration with high-protein meals may interfere with absorption of the agent. Surprisingly, high-protein meals and meals rich in neutral amino acids have been reported to enhance mean plasma levels by 36% and the area under the curve (AUC) by 12% (32,33). The physiologic basis for this effect is undetermined. Enhanced amino acid transport or increased paracellular absorption could contribute. Another group of investigators also confirmed the lack of impairment in gabapentin absorption following a high-protein meal, although they did not demonstrate significantly increased oral absorption (34).

Pharmacokinetic studies suggest that patients should receive four doses per day when the total daily dose is ≥3600 mg per day (35). In one study of 36 healthy volunteers, neither C_max nor time to maximal serum concentration (T_max) were influenced by subject age, implying that absorption does not significantly change with aging (36).

Pregabalin

The absorption of pregabalin was linearly related to the dose in studies of single and multiple oral doses in patients (37). T_max was about 1 hour with oral bioavailability 90 percent or more (37), compared with the limited, dose-dependent uptake of gabapentin. This suggests that pregabalin absorption is not limited by a saturable process and may involve different or multiple absorption mechanisms. Pregabalin uptake was sodium-dependent and involved multiple amino
acid carriers (b^0,+, B⁰, and B^0,+) in brush-border membrane vesicles prepared from duodenum, jejunum, and ileum of rats and rabbits (38). In the same model, gabapentin absorption was mediated by a sodium-independent transporter (b^0,+) and was greatest in the duodenum and ileum (38). Other mechanisms for pregabalin absorption have not been ruled out, but the short elimination half-life (t_½) suggests that absorption does not take place throughout the intestine.

Distribution

Gabapentinoids do not bind significantly to plasma proteins (14,39,40). Pooled data from several studies of gabapentin yield a mean volume of distribution (Vd) of 60.9 L, or 0.65 to 1.04 L/kg (22,40).

Gabapentin

Gabapentin crosses the human blood-brain barrier and is distributed to the central nervous system (CNS). Ratios of cerebrospinal fluid (CSF) to plasma concentration were 0.1 at 6 hours and 0.2 at 24 hours after a single 1,200-mg oral dose of gabapentin. After 3 months of treatment with gabapentin 900 or 1200 mg per day, concentrations in CSF varied from 6% to 34% of those in plasma (41, 42, 43). Two clearance mechanisms—passive diffusion and active transport—appear to limit accumulation (44, 45, 46).

The CNS to plasma partition ratio was 0.8:1.0 between 1 and 8 hours after a single IV dose of gabapentin (47). Assuming a partition ratio of 0.8 (47) and C_max ranging from 2 to 25 μg/mL at steady state, brain tissue concentrations of 1.6 to 20 μg/g are achievable.

Pregabalin

Pregabalin is not metabolized in humans, is not bound significantly to plasma proteins, and enters the brain readily (14). As in the case of gabapentin, anticonvulsant efficacy appeared with a delay after entry of pregabalin into the brain, as measured by microdialysis, and persisted to some extent as interstitial brain concentrations fell (48). Efficacy was not strictly proportional to the concentration of pregabalin in the brain, and could have been caused by delayed (e.g., biochemical) action of the agent (48).

Elimination

Gabapentin

The absorbed fraction of gabapentin is excreted unchanged in the urine (1,22). There is no evidence that gabapentin is metabolized in humans (1,22). Repeated dosing does not affect the elimination of gabapentin (39,40).

The elimination t_½ of gabapentin was originally estimated to be 7 to 9 hours (40,46); however, more recent data indicate a broader range of elimination t_½s from 4 to 22 hours (49). Whereas the oral absorption of gabapentin appears to be nonlinear, the renal clearance of the agent has a linear relationship to creatinine clearance (ClCr) and glomerular filtration rate in both adults and children (22,49, 50, 51). Oral clearance of gabapentin appears similar between genders (36). When normalized for body weight, gabapentin oral clearance is more variable in young children (younger than 5 years of age) as compared with older children. On an mg/kg basis, younger children appear to require doses approximately 33% larger than those of older children (51).

Age- and disease-related decreases in renal function substantially reduce elimination (36,49). It is therefore reasonable to expect longer elimination t_½s and higher relative steady-state plasma concentrations in elderly patients compared with younger individuals. Gabapentin apparent oral clearance has been reported to range from approximately 225 mL per minute in individuals younger than 30 years of age to approximately 125 mL per minute in those older than 70 years of age. Dosage guidelines based on renal function have been generated from pharmacokinetic studies (49).

Pregabalin

Pregabalin is also excreted intact in the urine in proportion to ClCr (52). The elimination t_½ was approximately 9 hours for ClCr >60 mL per minute, 25 hours for ClCr 15 to 30 mL per minute, and 55 hours for hemodialysis patients (52). Renal function was the only factor that altered pregabalin pharmacokinetics; age was not an independent factor (53).

Interactions with Other Medications

Gabapentin

There is no evidence that gabapentin either induces or inhibits hepatic microsomal enzymes involved in the metabolism of other agents (40). Gabapentin does not appear to alter the metabolism of antiepileptic drugs (AEDs; carbamazepine or its epoxide, phenobarbital, phenytoin, or valproate), with the exception of felbamate (26). Hussein and associates reported that coad-ministration of gabapentin was associated with a 50% extension in the elimination t_½ of felbamate in 11 patients, presumably by a renal interaction (54). The older AEDs do not affect the pharmacokinetics of gabapentin. Similarly, no clinically significant interactions were noted with antacids (55), oral contraceptives (56), or lithium (57).

Pregabalin

Steady-state plasma levels of pregabalin were not affected significantly by carbamazepine, lamotrigine, phenobarbital, phenytoin, tiagabine, topiramate, or valproate (58). The addition of pregabalin had no effect on trough plasma levels, C_max, AUC, or t_½ of carbamazepine, lamotrigine, phenytoin, or valproate administered as monotherapy in patients with partial epilepsy (58).

Concentration-Effect Relationship

Gabapentin

The therapeutic range of gabapentin concentrations in plasma is not completely characterized. In healthy volunteers, oral dosing with 100 mg every 8 hours resulted in a mean steady-state peak plasma level of 1.91 μg/mL (22). Plasma levels ≥2μg/mL were associated with significant clinical improvement in controlled studies (59,60). Well-tolerated, early-morning trough plasma levels exceeded 20 μg/mL in some patients receiving 4800 mg per day (50). Wilson and colleagues noted improved clinical response in a group of patients with refractory partial seizures, with gabapentin serum concentrations ranging between 6 and 20 μg/mL (30). Other studies, however, have been unable to establish a significant concentration-effect relationship (61). Nonetheless, clinical data suggest that initial plasma target concentrations between 2 and 20 μg/mL may be associated with improved clinical response. An upper limit or maximal gabapentin plasma concentration has not been identified. It is important to recognize that adequate concentration-controlled studies with gabapentin have not been conducted. Therefore, these proposed concentrations should serve as a guide, not as a conclusively established therapeutic target.

TABLE 59.3 PIVOTAL TRIALS OF GABAPENTIN (GBP) AS ADD-ON THERAPY FOR PATIENTS WITH REFRACTORY PARTIAL AND SECONDARILY GENERALIZED TONIC-CLONIC SEIZURES

Study	Number of Subjects	Doses	Mean RRatio^*	Responder Rate; % (P Value)
UK Gabapentin Study Group (1990)	127	GBP 1,200 mg/day vs. placebo	GBP: -0.192 Placebo: -0.060 (p = 0.0056)	GBP: 23% Placebo: 9% (p = 0.049)
US Gabapentin Study Group No. 5 (1993)	306	GBP 600, 1,200, and 1,800 mg/day vs. placebo	1,800 mg/day: -0.233 (p <0.001) 1,200 mg/day: -0.118 (p <0.023) 600 mg/day: -0.151 (p <0.007) Placebo: -0.025	1,800 mg/day: 1,200 mg/day: 600 mg/day: Placebo:	26.4% (p <0.007) 17.6% (p <0.080) 18.4% (p <0.138) 8.4%
International Gabapentin Study Group (Anhut et al., 1994)	272	GBP 900 and 1,200 mg/day vs. placebo	1,200 mg/day: -0.157 (p = 0.0055) 900 mg/day: -0.136 (p = 0.0046) Placebo: -0.025	1,200 mg/day: 900 mg/day: Placebo:	28.0% (p = 0.008) 22.9% (p = 0.0046) 10.1%
The three parallel-group studies were 12 weeks in duration, multicenter, randomized, double-blind, and placebo-controlled. Study medications were given in three divided doses. Significance is expressed as p values versus placebo.
From refs 62, 63, 64.
Abbreviation: RRatio, response ratio.