← Back
NeuroTrials.ai
Neurology Clinical Trial Database

SANAD I

The SANAD study of effectiveness of valproate, lamotrigine, or topiramate for generalised and unclassifiable epilepsy: an unblinded randomised controlled trial

Share Share Copied! Compare

Year of Publication: 2007

Authors: Anthony G Marson, Asya M Al-Kharusi, Muna Alwaidh, ..., on behalf of the SANAD Study group

Journal: The Lancet

Citation: Lancet. 2007 March 24; 369(9566): 1016–1026

Link: https://www.sciencedirect.com/science/ar...140673607604619

PDF: https://www.sciencedirect.com/science/ar...140673607604619

Clinical Question

In patients with generalised onset seizures or seizures difficult to classify, which antiepileptic drug (valproate, lamotrigine, or topiramate) provides the best balance of seizure control, tolerability, quality of life, and cost-effectiveness?

Bottom Line

Valproate is better tolerated than topiramate and more efficacious than lamotrigine, and should remain the drug of first choice for most patients with generalised and unclassified epilepsies. However, because of known potential adverse effects of valproate during pregnancy, the benefits for seizure control in women of childbearing years should be carefully considered.

Major Points

  • Valproate was significantly better than topiramate for time to treatment failure (HR 1.57 [95% CI 1.19-2.08])
  • Valproate was significantly better than lamotrigine for time to 12-month remission (HR 0.76 [95% CI 0.62-0.94])
  • For idiopathic generalised epilepsy subgroup, valproate was significantly better than both lamotrigine (HR 1.55) and topiramate (HR 1.89) for treatment failure
  • Lamotrigine was best tolerated but least efficacious for seizure control
  • Topiramate was least tolerated due to psychiatric and cognitive adverse events
  • More than 80% of patients achieved 12-month remission by 4 years across all groups
  • 63% of patients had idiopathic generalised epilepsy; 27% were unclassified
  • No significant quality of life differences between treatment groups
  • Valproate dominated both lamotrigine and topiramate in cost per seizure avoided analysis
  • Male:female ratio of 60:40 suggests possible reluctance to randomize women of childbearing age to valproate

Design

Study Type: Pragmatic, multicentre, unblinded, randomised controlled trial (Arm B of SANAD)

Randomization: 1

Blinding: Unblinded. Neither patients nor clinicians were masked to drug treatment. Randomization was performed centrally via telephone using a computer-based minimisation procedure stratified by centre, sex, and treatment history.

Enrollment Period: January 12, 1999 to August 31, 2004

Follow-up Duration: Median approximately 2.5 years; follow-up data obtained up to January 13, 2006

Countries: United Kingdom

Sample Size: 716

Analysis: Intention-to-treat and per-protocol analyses. Cox regression for time-to-event outcomes. Log-rank test for treatment comparisons. Hazard ratios with 95% CIs reported. Power calculation: α=0.05, β=0.10 (90% power) required 445 patients per treatment group for non-inferiority margin of 10%.

Inclusion Criteria

  • History of two or more clinically definite unprovoked epileptic seizures in the previous year
  • Recruiting clinician regarded valproate as the better standard treatment option than carbamazepine
  • Patients with newly diagnosed epilepsy
  • Patients who had failed treatment with previous monotherapy (as long as the drug failure did not include one of the study drugs)
  • Patients in remission of epilepsy who had relapsed after withdrawal of treatment
  • Age greater than 4 years

Exclusion Criteria

  • Clinician or patient felt that treatment was contraindicated
  • All seizures were acute symptomatic seizures (including febrile seizures)
  • Age 4 years or younger
  • History of progressive neurological disease

Baseline Characteristics

Lamotrigine (n=239):

  • Sex - Male, n (%): 142 (59.4%)
  • Sex - Female, n (%): 97 (40.6%)
  • Treatment history - Untreated, n (%): 210 (87.9%)
  • Treatment history - Monotherapy (not optimum), n (%): 19 (8.0%)
  • Treatment history - Recent seizures after remission, n (%): 10 (4.2%)
  • Learning disability, n (%): 24 (10.0%)
  • Neurological deficit, n (%): 5 (2.1%)
  • Febrile convulsions history, n (%): 16 (6.7%)
  • Epilepsy in first degree relatives, n (%): 53 (22.2%)
  • Idiopathic partial epilepsy, n (%): 1 (0.4%)
  • Symptomatic or cryptogenic partial, n (%): 18 (7.5%)
  • Idiopathic generalised epilepsy, n (%): 145 (60.7%)
  • Other syndrome, n (%): 9 (3.8%)
  • Unclassified, n (%): 66 (27.6%)
  • Median interval first to most recent seizure, days (IQR): 492 (162-1510)
  • Median interval most recent seizure to randomisation, days (IQR): 11 (1-49)
  • Median number of seizures (IQR): 10 (3-101)
  • Mean age at first seizure, years (SD): 17.5 (12.1)
  • Mean age, years (SD): 22.8 (14.3)

Topiramate (n=239):

  • Sex - Male, n (%): 142 (59.4%)
  • Sex - Female, n (%): 97 (40.6%)
  • Treatment history - Untreated, n (%): 209 (87.5%)
  • Treatment history - Monotherapy (not optimum), n (%): 20 (8.4%)
  • Treatment history - Recent seizures after remission, n (%): 10 (4.2%)
  • Learning disability, n (%): 26 (10.9%)
  • Neurological deficit, n (%): 3 (1.3%)
  • Febrile convulsions history, n (%): 22 (9.2%)
  • Epilepsy in first degree relatives, n (%): 38 (15.9%)
  • Idiopathic partial epilepsy, n (%): 2 (0.8%)
  • Symptomatic or cryptogenic partial, n (%): 11 (4.6%)
  • Idiopathic generalised epilepsy, n (%): 151 (63.5%)
  • Other syndrome, n (%): 8 (3.4%)
  • Unclassified, n (%): 66 (27.7%)
  • Median interval first to most recent seizure, days (IQR): 401 (105-1702)
  • Median interval most recent seizure to randomisation, days (IQR): 13 (2-41)
  • Median number of seizures (IQR): 8 (3-100)
  • Mean age at first seizure, years (SD): 17.6 (11.5)
  • Mean age, years (SD): 22.3 (13.3)

Valproate (n=238):

  • Sex - Male, n (%): 143 (60.1%)
  • Sex - Female, n (%): 95 (39.9%)
  • Treatment history - Untreated, n (%): 209 (87.8%)
  • Treatment history - Monotherapy (not optimum), n (%): 21 (8.8%)
  • Treatment history - Recent seizures after remission, n (%): 8 (3.4%)
  • Learning disability, n (%): 19 (8.0%)
  • Neurological deficit, n (%): 8 (3.4%)
  • Febrile convulsions history, n (%): 21 (8.8%)
  • Epilepsy in first degree relatives, n (%): 38 (16.0%)
  • Idiopathic partial epilepsy, n (%): 0 (0%)
  • Symptomatic or cryptogenic partial, n (%): 20 (8.4%)
  • Idiopathic generalised epilepsy, n (%): 154 (64.7%)
  • Other syndrome, n (%): 5 (2.1%)
  • Unclassified, n (%): 59 (24.8%)
  • Median interval first to most recent seizure, days (IQR): 384 (126-1402)
  • Median interval most recent seizure to randomisation, days (IQR): 13 (1-42)
  • Median number of seizures (IQR): 8.5 (3-100)
  • Mean age at first seizure, years (SD): 18.3 (13.7)
  • Mean age, years (SD): 22.5 (14.5)

Total (N=716):

  • Sex - Male, n (%): 427 (59.6%)
  • Sex - Female, n (%): 289 (40.4%)
  • Idiopathic generalised epilepsy, n (%): 450 (62.9%)
  • Unclassified, n (%): 191 (26.7%)
  • Symptomatic or cryptogenic partial, n (%): 49 (6.9%)
  • Age 5-9 years at randomisation: 91 patients
  • Age 10-15 years at randomisation: 100 patients
  • Mean age, years (SD): 22.5 (14.0)

Arms

FieldControlLamotrigineTopiramate
InterventionSodium valproate monotherapy. Dose and preparation per clinician's usual practice with guidelines provided. Target dose for adults with inadequate seizure control: 600 mg/day initial maintenance, maximum 2000-3000 mg/day. Median dose at last follow-up for patients still on drug: 1081 mg/day (range 300-3000).Lamotrigine monotherapy. Dose and preparation per clinician's usual practice with guidelines provided. Slow titration required. Median dose at last follow-up for patients still on drug: 203 mg/day (range 50-500).Topiramate monotherapy. Dose and preparation per clinician's usual practice with guidelines provided. Median dose at last follow-up for patients still on drug: 171 mg/day (range 25-400).
DurationLong-term follow-up (median ~2.5 years)Long-term follow-up (median ~2.5 years)Long-term follow-up (median ~2.5 years)

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
Co-primary outcomes: (1) Time from randomisation to treatment failure (stopping randomised drug due to inadequate seizure control, intolerable side-effects, or both; or addition of other AEDs); (2) Time from randomisation to 12-month remission of seizuresPrimaryValproate: Treatment failure events in 74/234 (overall analysis); 12-month remission in 180/232Lamotrigine: Treatment failure events in 88/233; 12-month remission in 168/231. Topiramate: Treatment failure events in 102/232; 12-month remission in 178/2306.14%Treatment failure overall test: Log-rank χ²=10.117, df=2, P=0.006; 12-month remission: Log-rank χ²=6.384, df=2, P=0.041
Time to treatment failure - Lamotrigine vs Valproate (ITT)SecondaryValproate referenceLamotrigineHR 1.25 (95% CI 0.94-1.68)Not significant
Time to treatment failure - Topiramate vs Valproate (ITT)SecondaryValproate referenceTopiramateHR 1.57 (95% CI 1.19-2.08)<0.01
Time to treatment failure for unacceptable adverse events - Topiramate vs ValproateSecondaryValproate referenceTopiramateHR 1.55 (95% CI 1.07-2.26)<0.05
Time to treatment failure for unacceptable adverse events - Topiramate vs LamotrigineSecondaryLamotrigine referenceTopiramateHR 2.15 (95% CI 1.41-3.30)<0.01
Time to treatment failure for inadequate seizure control - Lamotrigine vs ValproateSecondaryValproate referenceLamotrigineHR 1.95 (95% CI 1.28-2.98)<0.01
Time to 12-month remission - Lamotrigine vs Valproate (ITT)SecondaryValproate referenceLamotrigineHR 0.76 (95% CI 0.62-0.94)<0.05
Time to 12-month remission - Topiramate vs Valproate (ITT)SecondaryValproate referenceTopiramateHR 0.93 (95% CI 0.76-1.15)Not significant
Time to 12-month remission - Per protocol - Lamotrigine vs ValproateSecondaryValproate referenceLamotrigineHR 0.76 (95% CI 0.60-0.95)<0.05
Time to 12-month remission - Per protocol - Topiramate vs ValproateSecondaryValproate referenceTopiramateHR 0.77 (95% CI 0.61-0.97)<0.05
Time to 24-month remission - Lamotrigine vs Valproate (ITT)SecondaryValproate: 60% at Year 2Lamotrigine: difference -13% (95% CI -23 to -2)<0.05
Time to first seizure - Lamotrigine vs ValproateSecondaryValproate: 57% seizure-free at Year 1Lamotrigine: difference +11% (more seizures) (95% CI 3 to 20)<0.05
Idiopathic generalised epilepsy subgroup: Treatment failure - Lamotrigine vs ValproateSecondaryValproate referenceLamotrigineHR 1.55 (95% CI 1.07-2.24)<0.05
Idiopathic generalised epilepsy subgroup: Treatment failure - Topiramate vs ValproateSecondaryValproate referenceTopiramateHR 1.89 (95% CI 1.32-2.70)<0.001
Idiopathic generalised epilepsy subgroup: 12-month remission - Lamotrigine vs ValproateSecondaryValproate referenceLamotrigineHR 0.68 (95% CI 0.53-0.89)<0.01
Percentage remaining on randomised drug at Year 1SecondaryValproate: 74% (95% CI 68-79)Lamotrigine: 69% (difference -5%); Topiramate: 60% (difference -14%)
Percentage with 12-month remission at Year 4SecondaryValproate: 87% (95% CI 81-92)Lamotrigine: 79% (difference -8%); Topiramate: 87% (difference 0%)
Any clinically important adverse event (ITT)AdverseValproate: 85/238 (36%)Lamotrigine: 88/239 (37%); Topiramate: 107/239 (45%)
Tiredness/drowsiness/fatigue/lethargyAdverseValproate: 18 [12 per protocol]Lamotrigine: 15 [9]; Topiramate: 25 [20]
Other psychiatric symptomsAdverseValproate: 8 [7]Lamotrigine: 7 [4]; Topiramate: 19 [15]
Weight gainAdverseValproate: 17 [16]Lamotrigine: 8 [5]; Topiramate: 7 [2]
Behaviour/personality change/aggressionAdverseValproate: 4 [4]Lamotrigine: 6 [4]; Topiramate: 20 [18]
Worsening of seizuresAdverseValproate: 7 [3]Lamotrigine: 10 [6]; Topiramate: 13 [9]
Memory problemsAdverseValproate: 3 [0]Lamotrigine: 2 [2]; Topiramate: 12 [10]
Weight lossAdverseValproate: 0 [0]Lamotrigine: 3 [0]; Topiramate: 14 [12]
Allergic rashAdverseValproate: 2 [0]Lamotrigine: 13 [12]; Topiramate: 1 [1]
TremorAdverseValproate: 8 [6]Lamotrigine: 4 [2]; Topiramate: 1 [0]
DepressionAdverseValproate: 3 [3]Lamotrigine: 1 [1]; Topiramate: 9 [6]
Confusion/difficulty thinking/disorientedAdverseValproate: 3 [2]Lamotrigine: 3 [2]; Topiramate: 7 [7]
Pins and needles/dysaesthesiaAdverseValproate: 2 [0]Lamotrigine: 0 [0]; Topiramate: 8 [6]

Subgroup Analysis

Interaction testing between treatment and epilepsy syndrome: No significant interaction for treatment failure (p=0.12), but significant interactions for 12-month remission (p=0.04), 24-month remission (p=0.007), and first seizure (p=0.001). Valproate's advantage was more extreme in idiopathic generalised epilepsy subgroup (n=441) compared to unclassified patients (n=186) or partial/other syndromes (n=52). For IGE subgroup: VPA vs LTG HR 0.65 (0.45-0.93), VPA vs TPM HR 0.53 (0.37-0.76) for treatment failure. Common syndromes included: childhood absence epilepsy (15%), juvenile absence epilepsy (10%), juvenile myoclonic epilepsy (26%), generalised epilepsy with tonic clonic seizures on waking (9%), unspecified IGE (37%).

Criticisms

  • Unblinded design - neither patients nor clinicians masked to treatment, which may have affected symptom reporting and tolerability assessments
  • Failed to achieve target recruitment (716 vs target of 1335 based on 445 per arm) though larger than expected treatment differences compensated
  • Male:female ratio of 60:40 suggests possible reluctance to randomize women of childbearing age due to valproate teratogenicity concerns, potentially limiting generalizability
  • Pragmatic design with variable dosing regimens makes it difficult to assess whether drugs were optimally dosed
  • Quality of life measures showed no differences between groups, possibly due to response bias (patients with poorer QoL less likely to return questionnaires)
  • Not powered to examine pregnancy outcomes or safety in women of childbearing potential
  • Limited power to make definitive statements about relative efficacy for individual seizure types and sub-syndromes within idiopathic generalised epilepsies
  • Cost-effectiveness analysis based on incomplete data (165 adults for cost per QALY, 299 for cost per seizure avoided)
  • Median follow-up of approximately 2.5 years may not capture long-term outcomes
  • Study conducted only in the UK, limiting international generalizability

Funding

UK National Health Service Health Technology Assessment Programme (primary funder). Additional 20% contribution from pharmaceutical companies with products assessed (GlaxoSmithKline, Janssen-Cilag, Novartis, Pfizer, Sanofi-Synthelabo, and the Wellcome Trust). Funding sources had no role in study design, data collection, analysis, interpretation, or manuscript writing.

Based on: SANAD I (The Lancet, 2007)

Authors: Anthony G Marson, Asya M Al-Kharusi, Muna Alwaidh, ..., on behalf of the SANAD Study group

Citation: Lancet. 2007 March 24; 369(9566): 1016–1026

Content summarized and formatted by NeuroTrials.ai.