PRECAUTIONS SECTION

PRECAUTIONS

General

Use in Non-Induced Patients: Virtually all experience with GABITRIL has been obtained in patients with epilepsy receiving at least one concomitant enzyme-inducing antiepilepsy drug (AED), which lowers the plasma levels of tiagabine. Use in non-induced patients requires lower doses of GABITRIL. These patients may also require a slower titration of GABITRIL compared to that of induced patients (seeDOSAGE AND ADMINISTRATION). Patients taking a combination of inducing and non-inducing agents (e.g., carbamazepine and valproate) should be considered to be induced. Patients not receiving hepatic enzyme-inducing agents are referred to as non-induced patients.

Generalized Weakness: Moderately severe to incapacitating generalized weakness has been reported following administration of GABITRIL in 28 of 2531 (approximately 1%) patients with epilepsy. The weakness resolved in all cases after a reduction in dose or discontinuation of GABITRIL.

Binding in the Eye and Other Melanin-Containing Tissues: When dogs received a single dose of radiolabeled tiagabine, there was evidence of residual binding in the retina and uvea after 3 weeks (the latest time point measured). Although not directly measured, melanin binding is suggested. The ability of available tests to detect potentially adverse consequences, if any, of the binding of tiagabine to melanin-containing tissue is unknown and there was no systematic monitoring for relevant ophthalmological changes during the clinical development of GABITRIL. However, long-term (up to one year) toxicological studies of tiagabine in dogs showed no treatment-related ophthalmoscopic changes and macro- and microscopic examinations of the eye were unremarkable. Accordingly, although there are no specific recommendations for periodic ophthalmologic monitoring, prescribers should be aware of the possibility of long-term ophthalmologic effects.

**Use in Hepatically-Impaired Patients:**Because the clearance of tiagabine is reduced in patients with liver disease, dosage reduction may be necessary in these patients.

Serious Rash: Four patients treated with tiagabine during the product’s premarketing clinical testing developed what were considered to be serious rashes. In two patients, the rash was described as maculopapular; in one it was described as vesiculobullous; and in the 4th case, a diagnosis of Stevens Johnson syndrome was made. In none of the 4 cases is it certain that tiagabine was the primary, or even a contributory, cause of the rash. Nevertheless, drug associated rash can, if extensive and serious, cause irreversible morbidity, even death.

Information for Patients

Patients should be informed of the availability of a Medication Guide, and they should be instructed to read it prior to taking GABITRIL. The complete text of the Medication Guide is provided at the end of this labeling.

Suicidal Thinking and Behavior: Patients, their caregivers, and families should be counseled that AEDs, including GABITRIL, may increase the risk of suicidal thoughts and behavior and should be advised of the need to be alert for the emergence or worsening of symptoms of depression, any unusual changes in mood or behavior, or the emergence of suicidal thoughts, behavior, or thoughts about self-harm. Behaviors of concern should be reported immediately to healthcare providers.

Patients should be advised that GABITRIL may cause dizziness, somnolence, and other symptoms and signs of CNS depression. Accordingly, patients should be advised neither to drive nor to operate other complex machinery until they have gained sufficient experience on GABITRIL to gauge whether or not it affects their mental and/or motor performance adversely. Because of the possible additive depressive effects, caution should also be used when patients are taking other CNS depressants in combination with GABITRIL.

Because teratogenic effects were seen in the offspring of rats exposed to maternally toxic doses of tiagabine and because experience in humans is limited, patients should be advised to notify their physicians if they become pregnant or intend to become pregnant during therapy.

Because of the possibility that tiagabine may be excreted in breast milk, patients should be advised to notify those providing care to themselves and their children if they intend to breastfeed or are breastfeeding an infant.

Patients should be encouraged to enroll in the North American Antiepileptic Drug (NAAED) Pregnancy Registry if they become pregnant. This registry is collecting information about the safety of antiepileptic drugs during pregnancy. To enroll, patients can call the toll free number 1-888-233-2334 (seePRECAUTIONS, Pregnancy).

Laboratory Tests

Therapeutic Monitoring of Plasma Concentrations of Tiagabine: A therapeutic range for tiagabine plasma concentrations has not been established. In controlled trials, trough plasma concentrations observed among patients randomized to doses of tiagabine that were statistically significantly more effective than placebo ranged from <1 ng/mL to 234 ng/mL (median, 10th and 90th percentiles are 23.7 ng/mL, 5.4 ng/mL, and 69.8 ng/mL, respectively). Because of the potential for pharmacokinetic interactions between GABITRIL and drugs that induce or inhibit hepatic metabolizing enzymes, it may be useful to obtain plasma levels of tiagabine before and after changes are made in the therapeutic regimen.

**Clinical Chemistry and Hematology:**During the development of GABITRIL, no systematic abnormalities on routine laboratory testing were noted. Therefore, no specific guidance is offered regarding routine monitoring; the practitioner retains responsibility for determining how best to monitor the patient in his/her care.

EEG: Patients with a history of spike and wave discharges on EEG have been reported to have exacerbations of their EEG abnormalities associated with cognitive/neuropsychiatric events. This raises the possibility that these clinical events may, in some cases, be a manifestation of underlying seizure activity (seeWARNINGS, Cognitive/Neuropsychiatric Adverse Events). In the documented cases of spike and wave discharges on EEG with cognitive/neuropsychiatric events, patients usually continued tiagabine, but required dosage adjustment.

Drug Interactions

In evaluating the potential for interactions among co-administered antiepilepsy drugs (AEDs), whether or not an AED induces or does not induce metabolic enzymes is an important consideration. Carbamazepine, phenytoin, primidone, and phenobarbital are generally classified as enzyme inducers; valproate and gabapentin are not. GABITRIL is considered to be a non-enzyme inducing AED (seePRECAUTIONS, General, Use in Non-Induced Patients).

The drug interaction data described in this section were obtained from studies involving either healthy subjects or patients with epilepsy.

Effects of GABITRIL on other Antiepilepsy Drugs (AEDs):

Phenytoin: Tiagabine had no effect on the steady-state plasma concentrations of phenytoin in patients with epilepsy.

Carbamazepine: Tiagabine had no effect on the steady-state plasma concentrations of carbamazepine or its epoxide metabolite in patients with epilepsy.

Valproate: Tiagabine causes a slight decrease (about 10%) in steady-state valproate concentrations.

Phenobarbital or Primidone: No formal pharmacokinetic studies have been performed examining the addition of tiagabine to regimens containing phenobarbital or primidone. The addition of tiagabine in a limited number of patients in three well-controlled studies caused no systematic changes in phenobarbital or primidone concentrations when compared to placebo.

Effects of other Antiepilepsy Drugs (AEDs) on GABITRIL:

Carbamazepine: Population pharmacokinetic analyses indicate that tiagabine clearance is 60% greater in patients taking carbamazepine with or without other enzyme-inducing AEDs.

Phenytoin: Population pharmacokinetic analyses indicate that tiagabine clearance is 60% greater in patients taking phenytoin with or without other enzyme-inducing AEDs.

Phenobarbital (Primidone): Population pharmacokinetic analyses indicate that tiagabine clearance is 60% greater in patients taking phenobarbital (primidone) with or without other enzyme-inducing AEDs.

Valproate: The addition of tiagabine to patients taking valproate chronically had no effect on tiagabine pharmacokinetics, but valproate significantly decreased tiagabine binding in vitro from 96.3 to 94.8%, which resulted in an increase of approximately 40% in the free tiagabine concentration. The clinical relevance of this in vitro finding is unknown.

Interaction of GABITRIL with Other Drugs:

Cimetidine: Co-administration of cimetidine (800 mg/day) to patients taking tiagabine chronically had no effect on tiagabine pharmacokinetics.

Theophylline: A single 10 mg dose of tiagabine did not affect the pharmacokinetics of theophylline at steady state.

Warfarin: No significant differences were observed in the steady-state pharmacokinetics of R-warfarin or S-warfarin with the addition of tiagabine given as a single dose. Prothrombin times were not affected by tiagabine.

Digoxin: Concomitant administration of tiagabine did not affect the steady- state pharmacokinetics of digoxin or the mean daily trough serum level of digoxin.

Ethanol or Triazolam: No significant differences were observed in the pharmacokinetics of triazolam (0.125 mg) and tiagabine (10 mg) when given together as a single dose. The pharmacokinetics of ethanol were not affected by multiple-dose administration of tiagabine. Tiagabine has shown no clinically important potentiation of the pharmacodynamic effects of triazolam or alcohol. Because of the possible additive effects of drugs that may depress the nervous system, ethanol or triazolam should be used cautiously in combination with tiagabine.

Oral Contraceptives: Multiple dose administration of tiagabine (8 mg/day monotherapy) did not alter the pharmacokinetics of oral contraceptives in healthy women of child-bearing age.

Antipyrine: Antipyrine pharmacokinetics were not significantly different before and after tiagabine multiple-dose regimens. This indicates that tiagabine does not cause induction or inhibition of the hepatic microsomal enzyme systems responsible for the metabolism of antipyrine.

St. John’s wort: Concomitant use of St. John’s wort may enhance the metabolism of tiagabine.

Interaction of GABITRIL with Highly Protein Bound Drugs:

In vitro data showed that tiagabine is 96% bound to human plasma protein and therefore has the potential to interact with other highly protein bound compounds. Such an interaction can potentially lead to higher free fractions of either tiagabine or the competing drug.

Carcinogenesis, Mutagenesis, Impairment of Fertility

Carcinogenesis: In rats, a study of the potential carcinogenicity associated with tiagabine HCl administration showed that 200 mg/kg/day (plasma exposure [AUC] 36 to 100 times that at the maximum recommended human dosage [MRHD] of 56 mg/day) for 2 years resulted in small, but statistically significant increases in the incidences of hepatocellular adenomas in females and Leydig cell tumors of the testis in males. The significance of these findings relative to the use of GABITRIL in humans is unknown. The no effect dosage for induction of tumors in this study was 100 mg/kg/day (17 to 50 times the exposure at the MRHD). No statistically significant increases in tumor formation were noted in mice at dosages up to 250 mg/kg/day (20 times the MRHD on a mg/m2 basis).

Mutagenesis: Tiagabine produced an increase in structural chromosome aberration frequency in human lymphocytes in vitro in the absence of metabolic activation. No increase in chromosomal aberration frequencies was demonstrated in this assay in the presence of metabolic activation. No evidence of genetic toxicity was found in the in vitro bacterial gene mutation assays, the in vitro HGPRT forward mutation assay in Chinese hamster lung cells, the in vivo mouse micronucleus test, or an unscheduled DNA synthesis assay.

Impairment of Fertility: Studies of male and female rats administered dosages of tiagabine HCl prior to and during mating, gestation, and lactation have shown no impairment of fertility at doses up to 100 mg/kg/day. This dose represents approximately 16 times the maximum recommended human dose (MRHD) of 56 mg/day, based on body surface area (mg/m2). Lowered maternal weight gain and decreased viability and growth in the rat pups were found at 100 mg/kg, but not at 20 mg/kg/day (3 times the MRHD on a mg/m2 basis).

Pregnancy:

Tiagabine has been shown to have adverse effects on embryo-fetal development, including teratogenic effects, when administered to pregnant rats and rabbits at doses greater than the human therapeutic dose.

An increased incidence of malformed fetuses (various craniofacial, appendicular, and visceral defects) and decreased fetal weights were observed following oral administration of 100 mg/kg/day to pregnant rats during the period of organogenesis. This dose is approximately 16 times the maximum recommended human dose (MRHD) of 56 mg/day, based on body surface area (mg/m2). Maternal toxicity (transient weight loss/reduced maternal weight gain during gestation) was associated with this dose, but there is no evidence to suggest that the teratogenic effects were secondary to the maternal effects. No adverse maternal or embryo-fetal effects were seen at a dose of 20 mg/kg/day (3 times the MRHD on a mg/m2 basis).

Decreased maternal weight gain, increased resorption of embryos and increased incidences of fetal variations, but not malformations, were observed when pregnant rabbits were given 25 mg/kg/day (8 times the MRHD on a mg/m2 basis) during organogenesis. The no effect level for maternal and embryo-fetal toxicity in rabbits was 5 mg/kg/day (equivalent to the MRHD on a mg/m2 basis).

When female rats were given tiagabine 100 mg/kg/day during late gestation and throughout parturition and lactation, decreased maternal weight gain during gestation, an increase in stillbirths, and decreased postnatal offspring viability and growth were found. There are no adequate and well-controlled studies in pregnant women. Tiagabine should be used during pregnancy only if clearly needed.

To provide additional information regarding the effects of in utero exposure to GABITRIL, physicians are advised to recommend that pregnant patients taking GABITRIL enroll in the NAAED Pregnancy Registry. This can be done by calling the toll free number 1-888-233-2334, and must be done by patients themselves. Information on the registry can also be found at the website http://www.aedpregnancyregistry.org/.

Use in Nursing Mothers:

Studies in rats have shown that tiagabine HCl and/or its metabolites are excreted in the milk of that species. Levels of excretion of tiagabine and/or its metabolites in human milk have not been determined and effects on the nursing infant are unknown. GABITRIL should be used in women who are nursing only if the benefits clearly outweigh the risks.

Pediatric Use:

Safety and effectiveness in pediatric patients below the age of 12 have not been established. The pharmacokinetics of tiagabine were evaluated in pediatric patients age 3 to 10 years (seeCLINICAL PHARMACOLOGY, Special Populations, Pediatric).

Geriatric Use:

Because few patients over the age of 65 (approximately 20) were exposed to GABITRIL during its clinical evaluation, no specific statements about the safety or effectiveness of GABITRIL in this age group could be made.

---

DRUG ABUSE AND DEPENDENCE SECTION

DRUG ABUSE AND DEPENDENCE

The abuse and dependence potential of GABITRIL have not been evaluated in human studies.

---

ADVERSE REACTIONS SECTION

ADVERSE REACTIONS

The most commonly observed adverse events in placebo-controlled, parallel- group, add-on epilepsy trials associated with the use of GABITRIL in combination with other antiepilepsy drugs not seen at an equivalent frequency among placebo-treated patients were dizziness/light-headedness, asthenia/lack of energy, somnolence, nausea, nervousness/irritability, tremor, abdominal pain, and thinking abnormal/difficulty with concentration or attention.

Approximately 21% of the 2531 patients who received GABITRIL in clinical trials of epilepsy discontinued treatment because of an adverse event. The adverse events most commonly associated with discontinuation were dizziness (1.7%), somnolence (1.6%), depression (1.3%), confusion (1.1%), and asthenia (1.1%).

In Studies 1 and 2 (U.S. studies), the double-blind, placebo-controlled, parallel-group, add-on studies, the proportion of patients who discontinued treatment because of adverse events was 11% for the group treated with GABITRIL and 6% for the placebo group. The most common adverse events considered the primary reason for discontinuation were confusion (1.2%), somnolence (1.0%), and ataxia (1.0%).

Adverse Event Incidence in Controlled Clinical Trials: Table 5 lists treatment-emergent signs and symptoms that occurred in at least 1% of patients treated with GABITRIL for epilepsy participating in parallel-group, placebo- controlled trials and were numerically more common in the GABITRIL group. In these studies, either GABITRIL or placebo was added to the patient’s current antiepilepsy drug therapy. Adverse events were usually mild or moderate in intensity.

The prescriber should be aware that these figures, obtained when GABITRIL was added to concurrent antiepilepsy drug therapy, cannot be used to predict the frequency of adverse events in the course of usual medical practice when patient characteristics and other factors may differ from those prevailing during clinical studies. Similarly, the cited frequencies cannot be directly compared with figures obtained from other clinical investigations involving different treatments, uses, or investigators. An inspection of these frequencies, however, does provide the prescribing physician with one basis to estimate the relative contribution of drug and non-drug factors to the adverse event incidences in the population studied.

Table 5: Treatment-Emergent Adverse Event1 Incidence in Parallel- Group, Placebo-Controlled, Add-On Trials (events in at least 1% of patients treated with GABITRIL and numerically more frequent than in the placebo group)

1 Patients in these add-on studies were receiving one to three concomitant enzyme-inducing antiepilepsy drugs in addition to GABITRIL or placebo. Patients may have reported multiple adverse experiences; thus, patients may be included in more than one category.

• COSTART term substituted with a more clinically descriptive term.

Other events reported by 1% or more of patients treated with GABITRIL but equally or more frequent in the placebo group were: accidental injury, chest pain, constipation, flu syndrome, rhinitis, anorexia, back pain, dry mouth, flatulence, ecchymosis, twitching, fever, amblyopia, conjunctivitis, urinary tract infection, urinary frequency, infection, dyspepsia, gastroenteritis, nausea and vomiting, myalgia, diplopia, headache, anxiety, acne, sinusitis, and incoordination.

Study 1 was a dose-response study including doses of 32 mg and 56 mg. Table 6 shows adverse events reported at a rate of ≥ 5% in at least one GABITRIL group and more frequent than in the placebo group. Among these events, depression, tremor, nervousness, difficulty with concentration/attention, and perhaps asthenia exhibited a positive relationship to dose.

Table 6: Treatment-Emergent Adverse Event Incidence in Study 1† (events in at least 5% of patients treated with GABITRIL 32 or 56 mg and numerically more frequent than in the placebo group)

† Patients in this study were receiving one to three concomitant enzyme- inducing antiepilepsy drugs in addition to GABITRIL or placebo. Patients may have reported multiple adverse experiences; thus, patients may be included in more than one category.

• COSTART term substituted with a more clinically descriptive term.

The effects of GABITRIL in relation to those of placebo on the incidence of adverse events and the types of adverse events reported were independent of age, weight, and gender. Because only 10% of patients were non-Caucasian in parallel-group, placebo-controlled trials, there is insufficient data to support a statement regarding the distribution of adverse experience reports by race.

Other Adverse Events Observed During All Clinical Trials: GABITRIL has been administered to 2531 patients during all phase 2/3 clinical trials, only some of which were placebo-controlled. During these trials, all adverse events were recorded by the clinical investigators using terminology of their own choosing. To provide a meaningful estimate of the proportion of individuals having adverse events, similar types of events were grouped into a smaller number of standardized categories using modified COSTART dictionary terminology. These categories are used in the listing below. The frequencies presented represent the proportion of the 2531 patients exposed to GABITRIL who experienced events of the type cited on at least one occasion while receiving GABITRIL. All reported events are included except those already listed above, events seen only three times or fewer (unless potentially important), events very unlikely to be drug-related, and those too general to be informative. Events are included without regard to determination of a causal relationship to tiagabine.

Events are further classified within body system categories and enumerated in order of decreasing frequency using the following definitions: frequent adverse events are defined as those occurring in at least 1/100 patients; infrequent adverse events are those occurring in 1/100 to 1/1000 patients; rare events are those occurring in fewer than 1/1000 patients.

Body as a Whole: Frequent: Allergic reaction, chest pain, chills, cyst, neck pain, and malaise. Infrequent: Abscess, cellulitis, facial edema, halitosis, hernia, neck rigidity, neoplasm, pelvic pain, photosensitivity reaction, sepsis, sudden death, and suicide attempt.

Cardiovascular System: Frequent: Hypertension, palpitation, syncope, and tachycardia. Infrequent: Angina pectoris, cerebral ischemia, electrocardiogram abnormal, hemorrhage, hypotension, myocardial infarct, pallor, peripheral vascular disorder, phlebitis, postural hypotension, and thrombophlebitis.

Digestive System: Frequent: Gingivitis and stomatitis. Infrequent: Abnormal stools, cholecystitis, cholelithiasis, dysphagia, eructation, esophagitis, fecal incontinence, gastritis, gastrointestinal hemorrhage, glossitis, gum hyperplasia, hepatomegaly, increased salivation, liver function tests abnormal, melena, periodontal abscess, rectal hemorrhage, thirst, tooth caries, and ulcerative stomatitis.

Endocrine System: Infrequent: Goiter and hypothyroidism.

Hemic and Lymphatic System: Frequent: Lymphadenopathy. Infrequent: Anemia, erythrocytes abnormal, leukopenia, petechia, and thrombocytopenia.

Metabolic and Nutritional: Frequent: Edema, peripheral edema, weight gain, and weight loss. Infrequent: Dehydration, hypercholesteremia, hyperglycemia, hyperlipemia, hypoglycemia, hypokalemia, and hyponatremia.

Musculoskeletal System: Frequent: Arthralgia. Infrequent: Arthritis, arthrosis, bursitis, generalized spasm, and tendinous contracture.

Nervous System: Frequent: Depersonalization, dysarthria, euphoria, hallucination, hyperkinesia, hypertonia, hypesthesia, hypokinesia, hypotonia, migraine, myoclonus, paranoid reaction, personality disorder, reflexes decreased, stupor, twitching, and vertigo. Infrequent: Abnormal dreams, apathy, choreoathetosis, circumoral paresthesia, CNS neoplasm, coma, delusions, dry mouth, dystonia, encephalopathy, hemiplegia, leg cramps, libido increased, libido decreased, movement disorder, neuritis, neurosis, paralysis, peripheral neuritis, psychosis, reflexes increased, and urinary retention.

Respiratory System: Frequent: Bronchitis, dyspnea, epistaxis, and pneumonia. Infrequent: Apnea, asthma, hemoptysis, hiccups, hyperventilation, laryngitis, respiratory disorder, and voice alteration.

Skin and Appendages: Frequent: Alopecia, dry skin, and sweating. Infrequent: Contact dermatitis, eczema, exfoliative dermatitis, furunculosis, herpes simplex, herpes zoster, hirsutism, maculopapular rash, psoriasis, skin benign neoplasm, skin carcinoma, skin discolorations, skin nodules, skin ulcer, subcutaneous nodule, urticaria, and vesiculobullous rash.

Special Senses: Frequent: Abnormal vision, ear pain, otitis media, and tinnitus. Infrequent: Blepharitis, blindness, deafness, eye pain, hyperacusis, keratoconjunctivitis, otitis externa, parosmia, photophobia, taste loss, taste perversion, and visual field defect.

Urogenital System: Frequent: Dysmenorrhea, dysuria, metrorrhagia, urinary incontinence, and vaginitis. Infrequent: Abortion, amenorrhea, breast enlargement, breast pain, cystitis, fibrocystic breast, hematuria, impotence, kidney failure, menorrhagia, nocturia, papanicolaou smear suspicious, polyuria, pyelonephritis, salpingitis, urethritis, urinary urgency, and vaginal hemorrhage.

Postmarketing Reports

The following adverse reactions have been identified during postapproval use of GABITRIL. Because these reactions are reported voluntarily from a population of uncertain size, it is not always possible to reliably estimate their frequency or establish a causal relationship to drug exposure.

**Skin and subcutaneous tissue disorders:**bullous dermatitis

Eye disorders: vision blurred

---