Jun 28, 2001 - ¶Keith Grainger, **Christine Kilpatrick, â â Roderick McKenzie, â¡â¡Daniel ... at Ladhope Chambers, 131 Wickham Terrace, Brisbane, QLD, 4000,.
Epilepsia, 42(10):1335–1339, 2001 Blackwell Science, Inc. © International League Against Epilepsy
AUStralian Study of Titration to Effect Profile of Safety (AUS-STEPS): High-Dose Gabapentin (Neurontin) in Partial Seizures *Roy Beran, †Samuel Berkovic, ‡Andrew Black, §Gytis Danta, 㛳John Dunne, ‡Joseph Frasca, ¶Keith Grainger, **Christine Kilpatrick, ††Roderick McKenzie, ‡‡Daniel McLaughlin, §§Graham Schapel, and 㛳㛳Ernest Somerville *Strategic Health Evaluators, Chatswood, and the Department of Neurology, Liverpool Hospital, Sydney; †The Austin and Repatriation Medical Centre, Melbourne; ‡The Queen Elizabeth Hospital, Adelaide; §Canberra Specialist Centre, Canberra; 㛳Royal Perth Hospital, ¶Sir Charles Gairdner Hospital, Perth; **Royal Melbourne Hospital, Melbourne; ††Prince of Wales Hospital, Sydney; ‡‡Royal Brisbane Hospital, Brisbane; §§North Coast Neurology Centre, Maroochydore; 㛳㛳Westmead Hospital, Sydney, Australia
Summary: Purpose:To evaluate the safety, tolerability, efficacy, and impact on quality of life of gabapentin (Neurontin; GBP) as adjunctive therapy in patients with refractory partial seizures. Methods: AUS-STEPS was an open-label, multicenter, prospective study in patients experiencing partial seizures who were inadequately controlled with one to three concurrent antiepileptic drugs (AEDs). GBP treatment was titrated to a maximum of 4,800 mg/day, over a treatment period of 24 weeks, to achieve an efficacious and tolerable dosage. Efficacy was assessed by seizure-frequency data. Quality of life was evaluated by using the QOLIE-10 questionnaire, and safety was assessed by adverse-event reports and clinical laboratory findings. Results: A total of 176 patients received treatment with GBP,
with 174 evaluable for intention-to-treat (ITT) efficacy analysis. A reduction of >50% in overall seizure frequency was observed in 93 patients (53%). There was a small (4.6%) overall improvement in QOLIE-10 score. The most frequent adverse events were dizziness (31%), fatigue (29%), somnolence (27%), headache (21%), and ataxia (20%), with no major increase seen in adverse events necessitating discontinuation as the dose of GBP was titrated upward. Conclusions. This study indicates that patients with partial epilepsy may be effectively treated with GBP at dosages of 艋4,800 mg/day, without altering the safety profile of the drug. Key Words: Gabapentin—Quality of life—Partial-onset seizures—Refractory epilepsy—Responders.
Gabapentin, (Neurontin; GBP) has been introduced into clinical practice as adjunctive therapy for the management of patients with epilepsy. This study was carried out to investigate the efficacy, safety, tolerability, and impact on quality of life of GBP in an open-label stepwise titration design intended to resemble clinical practice. Controlled phase II and III clinical trials have indicated that GBP is associated with a dose-related increase in efficacy at dosages of 600 to 1,800 mg/day (1,2). Initial studies with higher dosages have indicated good tolerability with long-term use of 艋2,400 mg/day (3), and short-term use of 艋4,800 mg/day (4,5).
More recently, the results of the NEON study in Canada (6) and the US-STEPS study (7,8) have been published, both of which used GBP as adjunctive therapy. The US-STEPS study, examining 2,216 patients treated for 16 weeks, indicated that titration of dosages to 1,800, 2,400, and 3,600 mg/day was well tolerated. Similarly, in the NEON study of 141 patients treated for 20 weeks, the safety profile of GBP appeared to be unaffected by dose titrations from 400 to 2,400 mg/day, although no dosage comparisons were presented. METHODS Study design This study was an open-label, multicenter, prospective study examining step titration of dosage against response. It was conducted in 12 sites in Australia from March 1996 to April 1999. All investigators used the
Revision accepted June 28, 2001. Address correspondence and reprint requests to Dr. D. McLaughlin at Ladhope Chambers, 131 Wickham Terrace, Brisbane, QLD, 4000, Australia.
1335
1336
R. BERAN ET AL.
same protocol, case-report form, and data-collection methods. The ethics committee at each center approved the protocol, and a written statement of informed consent was obtained before study entry. The study consisted of a screening visit, then an 8-week baseline, followed by a 24-week treatment period, with evaluations at the start and end of the baseline phase, and after 2, 4, 8, 12, 16, 20, and 24 weeks of treatment. Patients were aged 12 years or older at study entry and were experiencing more than two partial-onset seizures per month with or without secondary generalization despite therapy with one to three concurrent antiepileptic drugs (AEDs). Seizures were defined according to the classification developed by the International League Against Epilepsy (9). GBP treatment was started at 300 mg/day and titrated to dosage levels of 1,200, 1,800, 2,400, 3,600, or 4,800 mg/day to achieve an effective tolerable dosage. At all dose levels, GBP was administered 3 times daily. Patients could also be titrated beyond the dose at which seizures were reduced by 艌50%, to obtain further improvement or seizure freedom. Not all patients had the opportunity for titration to 4,800 mg/day dosage because use of this dosage was a late amendment, and most had already completed the study. The dosage of concomitant AEDs remained unchanged throughout the study. Assessments Efficacy was assessed by seizure frequency (total number of seizures per 28 days) as calculated from patient diaries. Quality of life was assessed by administering the QOLIE-10 questionnaire at screening, at week 16 of treatment, and at the final visit. Patients were assessed at screening and at weeks 0, 2, 4, 8, 12, 16, 20, and 24 of the treatment phase. Responders were defined as those patients who recorded a 艌50% reduction in seizure frequency compared with the baseline phase. Patients who responded and remained responders at each subsequent study visit, regardless of the dosage received, were classified as responders for the dosage at which they first became a responder. The minimal time for evaluation of efficacy was set to the predetermined clinical assessment visits. Patients who were never responders, or who did not maintain their response, were classified as nonresponders at the highest dosage received. Safety was assessed by adverse-event reports, clinical laboratory findings including full blood count, serum creatinine concentration, and liver function tests, ECG, and overall tolerability assessment by investigators. Adverse events were recorded at each study visit and were coded using the World Health Organization (WHO)-based system of classification. To classify events by the dosage at which they commenced, events were classified as “new” the first time they occurred after start of treatment with GBP and as “worsening” if they increased in severity. The Epilepsia, Vol. 42, No. 10, 2001
number and percentage of adverse events were presented for each WHO body system and preferred term, by dosage at which the event first occurred. Shifts in laboratory parameters and ECG from baseline to study end were checked. Adverse-event dropout rate, defined as the proportion of patients at a given dosage terminating the study prematurely because of adverse events, was also examined. Withdrawals due to adverse events were categorized by the dosage at withdrawal. Statistical methods The primary analysis was designed to calculate the number and percentage of responders with 艌50% seizure-frequency reduction from baseline on respective drug dosages. Descriptive statistics (mean, SD, minimum, median, maximum) were presented for the overall seizures per month and by dosage at baseline and at the last visit, as well as the percentage reduction from baseline. Summary statistics for the changes in types of seizures (simple partial, complex partial, secondarily generalized, other) at baseline and the final visit were also examined. The adverse-event dropout rate at each dosage was presented as the number and percentage of patients. Quality of life, as assessed using the QOLIE- 10 score, was calculated by standardizing each of the 10 question scores (graded from 1, best, to 5, worst) to a scale of 0 to 100 (worst to best). The change in each score during the study was calculated by subtracting the baseline score from the final score (i.e., a positive change represented an improvement). The overall score was calculated for each patient as the mean of the standardized values for all questions answered. Similarly, the overall percentage change was the mean of the percentage changes for each question for each patient. The overall percentage change was tested against no change by using a univariate t test and a comparison of the overall percentage change by dosage categories at which a response was achieved (1,800 mg/day) was performed by using a two-sample t test. RESULTS Patient characteristics A total of 188 patients was enrolled in the study, with 176 receiving treatment with GBP and therefore evaluable for safety assessments, and 174 patients eligible for intent-to-treat (ITT) efficacy analysis. The average age of patients in the ITT population was 41 ± 15 years (range, 13–87 years). The average duration of their epilepsy was 24 ± 14 years (range, 1,800 mg/day to 艋2,400 mg/day dosage, 22 (26.8%) of 82 at the >2,400 mg/day to 艋3,600 mg/day dosage, and three (30%) of 10 at the >3,600 mg/day to 艋4,800 mg/day dosage (Fig. 1). Thirty-six (20.7%) patients were seizure free, with a count of zero seizures per month, at their last visit. At every dosage there was a reduction in the number of seizures per month. The 174 evaluable patients had a mean (SD) of 14.9 (24.5) seizures on entry, and this decreased to 9.7 (19.2) at study end. The number and percentage of patients with seizures of each type and the
1337
frequency per month of these seizures at the final visit is shown in Table 2. Comparing the seizure frequency at study entry with that at the final visit, there was a reduction in simple partial seizures and complex partial seizures of 40.4% and 35.4%, respectively. There was a small but statistically significant improvement in QOLIE score of 4.6 ± 17.8% (p ⳱ 0.0016) in the 157 patients who had completed a baseline and at least one follow-up questionnaire. When responders were analyzed separately, it was found that there was a higher overall improvement (10.9 ± 20.0%) in those responding at higher dosages (>1,800 mg/day) compared with those responding at the lower dosages (艋1,800 mg/day), where the improvement was 3.3 ± 12.3%, p ⳱ 0.03. Safety analysis The rate of discontinuations due to adverse events did not increase as the dosage of GBP was increased. Of the 17 patients who withdrew because of adverse events, five of 176 withdrew at the 1,200-mg/day dosage, three of 159 and four of 138 at each of the next two dosages, four of 105 withdrew at the 3,600-mg/day dosage, and one of 12 patients withdrew at the 4,800-mg/day dosage (Table 3). There was one death attributed to suicide, which was considered not to be related to the study drug. Throughout the study, a total of 165 (94%) patients reported new or worsening adverse events. The most frequent adverse events were dizziness (n ⳱ 54, 31%), fatigue (n ⳱ 51, 29%), somnolence (n ⳱ 47, 27%), headache (n ⳱ 37, 21%), and ataxia (n ⳱ 35, 20%). Less frequent events included amnesia, depression, and nausea. Fifty (28%) patients reported events that were assessed by the investigator as related to the study drug. The most commonly reported events assessed as related to GBP were somnolence (10%), fatigue (10%), dizziness (9%), and ataxia (6%). The number and percentage of new or worsening adverse events commencing at each dosage is presented in Table 3. Increasing the dosage of GBP was not associated with a significant increase in adverse events necessitating withdrawal. Safety was well maintained at each step, with no conclusion being able to be drawn from the 4,800-mg/day dosage because of insufficient
TABLE 1. Seizures at Study Entry–Population: ITT (n = 174) Frequency (seizures/28 days) Seizure typea Seizures per month at final visit Simple partial Complex partial Secondarily generalised Other
n (%) 64 (37.2) 133 (77.3) 54 (31.4) 3 (1.7)
Range
Mean ± SD
Median
0–129 0–191 0–18 0–14
14.9 ± 24.5 (1–191) 5.22 ± 15.33 8.21 ± 19.87 1.02 ± 2.37 0.09 ± 1.07
7.0 0 3.67 0 0
a n ⳱ 172 as two patients did not have seizure details at week 0 and were unable to be included in this analysis.
Epilepsia, Vol. 42, No. 10, 2001
1338
R. BERAN ET AL.
FIG. 1. Analysis of responders by dose at which response was achieved and maintained.
patient numbers. No abnormal clinical laboratory or ECG findings of clinical significance were noted during the study. DISCUSSION This study was designed to assess the safety, efficacy, and tolerability of GBP when used as adjunctive therapy in patients with uncontrolled partial seizures who were already receiving one to three other AEDs. The dosing regimen of the study allowed adjustment of dose in response to inadequate seizure control and/or adverse events 艋3,600 mg/day in all patients and 4,800 mg/day in 12 patients. The use of dosages 艋4,800 mg/day generated data for treatment with GBP at higher doses than previously studied, but this dosage was a late amendment to the study. Many centers had recruited most of their patients by the time this amendment was introduced, hence the low patient numbers treated at this dosage.
This study suggests that the efficacy of GBP improves as the dose increases, without an increase in the rate of adverse events. Patients continued to respond in each dosage group, such that the cumulative percentage of responders increased as the dosage was increased. By the end of the study, 53% of all patients had experienced 艌50% reduction in seizures per month. To be classed as responders, patients had to maintain their response until the end of the treatment phase. As patients spend less time on higher doses, they had less time in which to relapse and become nonresponders. Responder rates at various doses cannot therefore be directly compared. These results are consistent with the findings of the USSTEPS study, where the cumulative percentage of responders increased as the dosage was increased, leading the investigators to conclude that titration of GBP to efficacy is appropriate in patients with partial epilepsy. Some patients chose to escalate dosages even after achieving a 50% reduction in monthly seizures because of their desire to be seizure free.
TABLE 2. Seizures at Final Visit–Population: ITT (n = 174) Frequency (seizures/28 days) Seizure typea Seizures per month at final visit Simple partial Complex partial Secondarily generalised Other
n (%) 40 (23.1) 98 (56.6) 32 (18.5) 1 (0.6)
Range
Mean ± SD
Median
1–85 0–185 0–58 0–2
9.7 ± 19.2 (0–187) 3.11 ± 10.90 5.30 ± 15.72 1.05 ± 4.96 0.01 ± 0.12
3.1 1 0 0 0
a n ⳱ 173, as one patient did not have seizure details at final visit and was unable to be included in this analysis.
Epilepsia, Vol. 42, No. 10, 2001
AUS-STEPS: GABAPENTIN IN PARTIAL SEIZURES
1339
TABLE 3. Total number of new and worsening adverse avents, and adverse event–related and non–adverse event–related withdrawals, by dosage level. Safety population: (n = 176)
Number of patients
Patients with one or more adverse event
ⱕ1,200 mg/day >1,200 to ⱕ1,800 mg/day >1,800 to ⱕ2,400 mg/day >2,400 to ⱕ3,600 mg/day >3,600 mg to ⱕ4,800 mg/day
176 159 138 105 12
117 80 87 65 5
Total, all dosages (% all patients)
176
N/A
Dosage
The tolerability and safety profile of GBP was maintained at the various dosages 艋4,800 mg/day. Increasing the dose of GBP to 3,600 mg did not appear to increase the number of patients withdrawing because ofo adverse events. This result is supported by previous data: in the US-STEPS study, GBP was given to a maximal dosage of 3,600 mg/day, with doses >1,800 mg/day as well tolerated as dosages 艋1800 mg/day in 281 patients. In the Canadian NEON study of 141 patients, no significant drug-related adverse events were reported at doses of 艋2,400 mg/day. The treatment period of 24 weeks of our study is comparable to the 24 weeks of the NEON study and the 16-week period of the US-STEPS study, as was the proportion of withdrawals from each study (AUSTEPS, 17%; US-STEPS, 18%; NEON, 26%). The design of this study was intended to reflect clinical practice. In an open-label study, treatment effect is not as rigorously defined as compared with a randomized controlled trial. However, the estimate of treatment effect provided by this study may not be larger or qualitatively different from those that may have been obtained from a randomized, controlled study (10). The results indicate the likely response, with good tolerability, when GBP is used stepwise, in increasing dosages 艋4,800 mg/day, in adult patients with uncontrolled partial seizures.
Adverse event–related withdrawals 5 3 4 4 1 17 (9.7)
Nonadverse event–related withdrawals 4 2 3 3 1
Total withdrawals 9 5 7 7 2
13 (0.6)
Cumulative number of withdrawals at this and lower dosage groups 9 14 21 28 30
30 (17.0)
N/A
REFERENCES 1. Gabapentin as add-on therapy in refractory partial epilepsy: a double-blind, placebo-controlled, parallel-group study: the US Gabapentin Study Group No. 5. Neurology 1993;43:2292–8. 2. Anhut H, Ashman P, Feuerstein TJ, et al. Gabapentin (Neurontin) as add-on therapy in patients with partial seizures: a double-blind, placebo-controlled study: The International Gabapentin Study Group. Epilepsia 1994;35:795–801. 3. Ojemann LM, Wilensky AJ, Temkin NR, et al. Long-term treatment with gabapentin for partial epilepsy. Epilepsy Res 1992;13: 159–65. 4. Abou-Khalil B, Shellenberger M, Anhut H. Two open-label, multicenter studies of the safety and efficacy of GBP in patients with refractory epilepsy [Abstract]. Epilepsia 1992;33:77. 5. Watemberg N, Pellock J. Efficacy and tolerability of high dose GBP add-on therapy in children and adults [Abstract]. Epilepsia 1995;36:57. 6. Bruni J. Outcome evaluation of gabapentin as add-on therapy for partial seizures: NEON Study Investigators Group, Neurontin Evaluation of Outcomes in Neurological Practice. Can J Neurol Sci 1998;25:134–40. 7. Morrell MJ. Dosing to efficacy with Neurontin: the STEPS trial: study of titration to effect profile of safety. Epilepsia 1999;40 (suppl 6):S23–6; discussion S73–4. 8. McLean MJ, Morrell MJ, Willmore U, et al. Safety and tolerability of gabapentin as adjunctive therapy in a large, multicenter study. Epilepsia 1999;40:965–72. 9. Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for revised classification of epilepsies and epileptic syndromes: Epilepsia 1989;30:389–99. 10. Benson K, Hartz AJ. A comparison of observational studies and randomized, controlled trials. N Engl J Med 2000;342:878–6.
Epilepsia, Vol. 42, No. 10, 2001
