KOMET
Keppra vs Older Monotherapy in Epilepsy Trial: an unblinded, randomised, two parallel-group, stratified trial comparing the effectiveness of levetiracetam with controlled-release carbamazepine and extended-release sodium valproate as monotherapy in patients with newly diagnosed epilepsy
Clinical Question
Is levetiracetam superior to standard antiepileptic drugs (extended-release sodium valproate or controlled-release carbamazepine) as first-line monotherapy in patients with newly diagnosed epilepsy?
Study Overview
Objective
Levetiracetam - To compare the effectiveness of levetiracetam (LEV) with extended-release sodium valproate (VPA-ER) and controlled-release carbamazepine (CBZ-CR) as monotherapy in patients with newly diagnosed epilepsy
Study Summary
- LEV was not superior to standard AEDs for time to treatment withdrawal (HR 0.90, 95% CI 0.74-1.08)
- Standard AEDs showed significantly longer time to first seizure compared to LEV (HR 1.20, p=0.022)
- LEV demonstrated comparable retention rates but modestly lower 12-month seizure freedom rates (53.9% vs 59.9%)
Intervention
Levetiracetam (500-3000 mg/day) vs extended-release sodium valproate (500-2000 mg/day) or controlled-release carbamazepine (200-1600 mg/day)
Patients per Arm
LEV: 841; VPA-ER: 347; CBZ-CR: 500
Bottom Line
Levetiracetam monotherapy was not superior to standard AEDs (valproate or carbamazepine) for time to treatment withdrawal in patients with newly diagnosed focal or generalised seizures. Standard AEDs showed a modest advantage for seizure freedom, while LEV may offer better tolerability, particularly compared to carbamazepine.
Major Points
- LEV was not superior to standard AEDs for the primary outcome of time to treatment withdrawal (HR 0.90, 95% CI 0.74-1.08)
- Standard AEDs showed significantly longer time to first seizure compared to LEV (HR 1.20, 95% CI 1.03-1.39, p=0.022)
- 12-month seizure freedom rates were 53.9% for LEV vs 59.9% for standard AEDs
- In the VPA stratum, time to treatment withdrawal was similar between LEV and VPA-ER (HR 1.02, 95% CI 0.74-1.41)
- In the CBZ stratum, LEV showed a trend toward better retention vs CBZ-CR (HR 0.84, 95% CI 0.66-1.07)
- Discontinuation due to adverse events was lower with LEV (8.3%) than standard AEDs (13.1%)
- LEV may be a viable first-line option for women of childbearing age given lower teratogenicity compared to valproate
- 75% of patients in the ITT population remained on their randomized drug at 12 months
Design
Study Type: Multicentre, unblinded, randomised, controlled, superiority trial with two-parallel-group, stratified design
Randomization: 1
Blinding: Unblinded (open-label). Treatment allocation was concealed using an Interactive Voice Response System via telephone
Enrollment Period: February 2005 to October 2007
Follow-up Duration: 52 weeks
Centers: 269
Countries: Australia, Austria, Belgium, Czech Republic, Finland, France, Germany, Italy, Netherlands, Spain, Sweden, UK, and 12 other European countries (23 European countries + Australia = 24 countries total per paper)
Sample Size: 1688
Analysis: Intention-to-treat analysis (1701 screened, 1698 randomised, 1688 ITT after 10 excluded). Kaplan-Meier survival curves for time to event. Cox proportional hazards regression model for treatment effect (HR with 95% CI). Log-rank test for sample size calculation. Two-sided α=0.05. Power 90%.
Inclusion Criteria
- Age ≥16 years
- Two or more unprovoked seizures in the previous 2 years
- At least one seizure during the previous 6 months
- Seizures classified according to ILAE classification
- Written informed consent
Exclusion Criteria
- Previous treatment with LEV, VPA, or CBZ for any indication
- Treatment for epilepsy with any other AED in the last 6 months
- Acute seizure treatment >2 weeks duration or stopped <1 week before screening
Arms
| Field | Levetiracetam (LEV) | Control | Control |
|---|---|---|---|
| Intervention | LEV starting dose 500 mg/day, up-titrated over 2 weeks to initial target dose of 1000 mg/day, administered twice daily as equal doses. Maximum dose 3000 mg/day if seizures occurred. Median daily dose: 987 mg/day (range 250-2807) | VPA-ER starting dose 500 mg/day, up-titrated over 2 weeks to initial target dose of 1000 mg/day, administered twice daily as equal doses. Maximum dose 2000 mg/day. Median daily dose: 987 mg/day (range 500-2263) | CBZ-CR starting dose 200 mg/day, up-titrated over 2 weeks to initial target dose of 600 mg/day, administered twice daily as equal doses. Maximum dose 1600 mg/day. Median daily dose: 588 mg/day (range 180-1422) |
| Duration | 52 weeks | 52 weeks | 52 weeks |
Outcomes
| Outcome | Type | Control | Intervention | HR / OR / RR | P-value |
|---|---|---|---|---|---|
| Time to treatment withdrawal from study medication calculated from randomisation to the day after the last intake of study medication (LEV vs standard AEDs combined) | Primary | 219 events (25.9%); 12-month withdrawal rate 25.9% | 200 events (23.8%); 12-month withdrawal rate 23.9% | 0.9 | 0.258 |
| Time to treatment withdrawal: LEV vs VPA-ER | Secondary | 12-month withdrawal rate 21.6% (95% CI 17.7-26.4) | 12-month withdrawal rate 22.0% (95% CI 18.0-26.7) | 1.02 | 0.882 |
| Time to treatment withdrawal: LEV vs CBZ-CR | Secondary | 12-month withdrawal rate 28.8% (95% CI 25.1-33.0) | 12-month withdrawal rate 25.2% (95% CI 21.6-29.3) | 0.84 | 0.161 |
| Time to first seizure: LEV vs standard AEDs | Secondary | 305 events (36.0%); 12-month seizure freedom 59.9% | 355 events (42.2%); 12-month seizure freedom 53.9% | 1.2 | 0.022 |
| Time to first seizure: LEV vs VPA-ER | Secondary | 12-month seizure freedom 64.5% (95% CI 58.9-69.5) | 12-month seizure freedom 58.7% (95% CI 53.1-63.9) | 1.19 | 0.167 |
| Time to first seizure: LEV vs CBZ-CR | Secondary | 12-month seizure freedom 56.7% (95% CI 51.8-61.2) | 12-month seizure freedom 50.5% (95% CI 45.8-55.1) | 1.2 | 0.061 |
| Time to first seizure: LEV vs CBZ-CR (focal seizures only) | Secondary | 12-month seizure freedom 56.2% (95% CI 51.0-61.0) | 12-month seizure freedom 48.1% (95% CI 42.9-53.1) | 1.24 | <0.05 (exact p not reported; CI excludes 1.0) |
| Safety Population Denominators | Adverse | Standard AEDs N=841 (VPA-ER N=342; CBZ-CR N=499) | LEV N=835 (VPA stratum N=345; CBZ stratum N=489) | ||
| ≥1 treatment-emergent AE (LEV N=835 vs Standard AEDs N=841) | Adverse | 574 (68.3%) | 587 (70.3%) | ||
| ≥1 treatment-emergent AE (VPA stratum: LEV N=345 vs VPA-ER N=342) | Adverse | 212 (62.0%) | 228 (66.1%) | ||
| ≥1 treatment-emergent AE (CBZ stratum: LEV N=489 vs CBZ-CR N=499) | Adverse | 362 (72.5%) | 359 (73.4%) | ||
| Drug-related AEs (LEV N=835 vs Standard AEDs N=841) | Adverse | 418 (49.7%) | 381 (45.6%) | ||
| Discontinuation due to AEs (LEV N=835 vs Standard AEDs N=841) | Adverse | 110 (13.1%) | 69 (8.3%) | ||
| Discontinuation due to AEs (VPA stratum: LEV N=345 vs VPA-ER N=342) | Adverse | 16 (4.7%) | 21 (6.1%) | ||
| Discontinuation due to AEs (CBZ stratum: LEV N=489 vs CBZ-CR N=499) | Adverse | 94 (18.8%) | 48 (9.8%) | ||
| Serious AEs (LEV N=835 vs Standard AEDs N=841) | Adverse | 61 (7.3%) | 106 (12.7%) | ||
| Severe AEs (LEV N=835 vs Standard AEDs N=841) | Adverse | 92 (10.9%) | 130 (15.6%) | ||
| Headache (LEV N=835 vs Standard AEDs N=841) | Adverse | 170 (20.2%) | 161 (19.3%) | ||
| Fatigue (LEV N=835 vs Standard AEDs N=841) | Adverse | 134 (15.9%) | 120 (14.4%) | ||
| Dizziness (LEV N=835 vs Standard AEDs N=841) | Adverse | 70 (8.3%) | 68 (8.1%) | ||
| Somnolence (LEV N=835 vs Standard AEDs N=841) | Adverse | 48 (5.7%) | 68 (8.1%) | ||
| Weight increased (LEV N=835 vs Standard AEDs N=841) | Adverse | 98 (11.7%) | 47 (5.6%) | ||
| Weight increased (VPA stratum: LEV N=345 vs VPA-ER N=342) | Adverse | 65 (19.0%) | 21 (6.1%) | ||
| Depression (LEV N=835 vs Standard AEDs N=841) | Adverse | 20 (2.4%) | 43 (5.1%) | ||
| Tremor (VPA stratum: LEV N=345 vs VPA-ER N=342) | Adverse | 32 (9.4%) | 4 (1.2%) | ||
| Rash (CBZ stratum: LEV N=489 vs CBZ-CR N=499) | Adverse | 29 (5.8%) | 9 (1.8%) | ||
| Deaths (LEV N=835 vs Standard AEDs N=841) | Adverse | 3 (0.4%) | 5 (0.6%) |
Subgroup Analysis
Post hoc subgroup analysis by seizure type: In VPA stratum, generalised seizures showed non-significant trend favoring VPA-ER for treatment withdrawal (HR 1.16, 95% CI 0.79-1.71) while focal seizures favored LEV (HR 0.73, 95% CI 0.37-1.44). In CBZ stratum, focal seizures showed similar results to overall (HR 0.84, 95% CI 0.65-1.09); generalised seizures favored LEV but with small numbers (HR 0.49, 95% CI 0.16-1.49). For time to first seizure in focal seizures (CBZ stratum), CBZ-CR was significantly better (HR 1.24, 95% CI 1.01-1.52).
Criticisms
- Open-label (unblinded) design introduces potential for bias in outcome assessment and patient/physician behavior
- Selection of patients at the discretion of the physician may have introduced selection bias
- Choice of best recommended treatment (VPA vs CBZ) was not standardized according to expert recommendations
- Neuroimaging and EEG were not mandatory before treatment selection, potentially leading to misclassification
- Initial target dose of CBZ-CR (600 mg/day) may have been unnecessarily high, contributing to higher discontinuation rates
- 12-month follow-up may be insufficient for long-term effectiveness assessment in epilepsy
- Study did not reach target enrollment of 1964 patients (1701 screened; 1698 randomised; 1688 ITT population), though still adequately powered
- Higher incidence of serious AEs with LEV (12.7%) vs standard AEDs (7.3%) requires further investigation
- VPA stratum included approximately 30% patients with focal seizures who may not have received optimal standard treatment
- Trial was designed before SANAD results identified lamotrigine as preferred for focal epilepsy
Funding
UCB Pharma (manufacturer of levetiracetam/Keppra). UCB Pharma was responsible for design and conduct of the study, and collection, management, analysis and interpretation of data. Medical writing assistance funded by UCB Pharma.
Based on: KOMET (Journal of Neurology, Neurosurgery & Psychiatry, 2013)
Authors: Eugen Trinka, Anthony G Marson, Wim Van Paesschen, ..., for the KOMET Study Group
Citation: J Neurol Neurosurg Psychiatry 2013;84:1138–1147
Content summarized and formatted by NeuroTrials.ai.