Study Evaluating LAMICTAL Extended-Release Therapy Added To Current Seizure Treatments In Patients With Primary Generalized Tonic-Clonic Seizures (PGTC) Seizures: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study Evaluating LAMICTAL Extended-Release Therapy Added To Current Seizure Treatments In Patients With Primary Generalized Tonic-Clonic Seizures (PGTC) Seizures (NCT00104416), a randomized phase 3 study. The primary outcome is Percent Change From Baseline in Weekly Primary Generalized Tonic-clonic (PGTC) Seizure Frequency During the Entire Double-Blind Treatment Phase.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, double-blind, parallel-group study comparing extended-release lamotrigine with placebo as add-on treatment for primary generalized tonic-clonic seizures. The registry identifies GlaxoSmithKline as the sponsor and records the study as completed. The recorded start was 2004-12, and primary completion was 2008-07. Enrollment was 153.

Participants were assigned to the Double-Blind Phase placebo group or the Double-Blind Phase LTG XR group. The primary outcome was analyzed in the Intent-to-Treat population, defined in the registry as randomized participants who took at least one dose of study drug and had at least one post-baseline efficacy assessment during that phase. The posted primary outcome group sizes are 72.0 for placebo and 69.0 for LTG XR. The registry also notes that one participant in each group had no PGTC seizures during baseline.

The intervention was added to current seizure treatments, rather than described as replacing them. The registry summary explains that placebo tablets resembled the active tablets. Dose schedules, eligibility details beyond the seizure condition, the randomization procedure, and the method used to maintain masking are not reported in the registry record.

The primary observation window extended from baseline through the end of the Double-Blind Treatment Phase, up to Week 19. The study also included a Continuation Phase with transition and open-label periods, and several secondary outcomes referred to these periods. The registry’s primary comparison concerns only the Double-Blind Treatment Phase; results from continuation periods should not be treated as part of that primary comparison.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: The number of par. starting the CP does not equal the number completing the TP, as 1) the CP was optional, 2) not everyone from the TP was eligible to enter the CP, and 3) Baseline Failures were allowed to enter the CP, however they were not included in the "started" count for the TP.

Double-Blind Phase

Reasons for not completing this period:

Continuation Phase

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat (ITT) Population: all randomized participants who took at least one dose of study drug and had at least one post-baseline efficacy assessment in the Double-Blind Treatment Phase. One participant in each treatment group did not have any PGTC seizures during the Baseline Phase.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change From Baseline in Weekly Primary Generalized Tonic-clonic (PGTC) Seizure Frequency During the Entire Double-Blind Treatment Phase, measured over Baseline through end of Double-Blind Treatment Phase (up to Week 19).

The primary endpoint was percent change from baseline in weekly primary generalized tonic-clonic seizure frequency during the entire Double-Blind Treatment Phase. The registry defines the calculation by comparing seizures per week during Treatment Week 1 through up to Week 19 with seizures per week during Baseline Week 1 through Week 8. Under that definition, a positive value means seizure frequency was reduced relative to baseline.

This outcome expresses change relative to each participant’s baseline seizure frequency, rather than simply counting seizures over unequal periods. It is clinically interpretable as a measure of change in seizure burden during treatment, while allowing baseline frequency to serve as the reference. The registry reports the group summaries as medians and gives the full range for each group. A median describes the middle of the observed distribution; it is not the same as an average. The full range describes the lowest and highest recorded values, not the precision of the estimated group contrast.

The primary population was restricted to randomized participants who took at least one dose and had at least one post-baseline efficacy assessment in the double-blind phase. The registry notes one participant in each treatment group with no baseline PGTC seizures, a detail relevant to interpreting a percent-change measure whose reference is baseline frequency.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a change measure: seizure frequency during treatment is compared with seizure frequency during baseline, and the result is expressed as a percent change. The registry states that a positive number means a reduction. This direction convention matters because the sign of a change measure is not self-explanatory unless the subtraction and coding convention are known. The baseline period was Baseline Week 1 through Week 8, and the treatment measurement covered Treatment Week 1 through up to Week 19.

Why report medians?

The registry summarizes the primary outcome using medians and full ranges. General statistical explanation: a median is the central observation after values are ordered, whereas a mean uses all values in an arithmetic average. A median can be useful when data are skewed or include extreme values, because extreme observations have less direct influence on the median than on the mean. The registry does not state why the median was selected for this endpoint or describe the outcome distribution in detail, so the reason for that choice is not reported in the registry record.

A full range gives the minimum and maximum observed values. It indicates how far apart the most extreme observations were, but it is not a confidence interval and does not show how precisely the median or treatment contrast was estimated. The negative lower endpoints in the posted ranges indicate that the recorded percent change could also represent an increase under the stated sign convention. The range alone does not reveal how many participants had increases or how the other observations were distributed.

What does the reported contrast mean?

The sponsor’s analysis is described as a Cochran-Mantel-Haenszel analysis of the median difference in final values. The reported estimate is 31.6, with a 95% confidence interval from 15.8 to 48.1 and p < 0.001. A confidence interval is an interval estimate that conveys uncertainty around an estimated contrast under the method used. It does not show the spread of individual outcomes, and it does not guarantee that the true value lies within the interval. The p-value measures how incompatible the observed result is with a specified null hypothesis under the test assumptions; it is not the probability that the null hypothesis is true.

The registry does not provide additional detail about the exact contrast coding, stratification factors, or other implementation choices for the Cochran-Mantel-Haenszel analysis. Accordingly, the reported estimate should be read as the registry’s stated median difference in final values, not reconstructed as a particular subtraction unless the direction is specified. The group medians help describe each arm, but an inferential comparison should use the sponsor’s reported method and estimate.

What does the analysis population include?

The Intent-to-Treat population is defined as all randomized participants who took at least one dose and had at least one post-baseline efficacy assessment in the Double-Blind Treatment Phase. This is the registry’s operational definition; it is narrower than a definition that includes every randomized participant regardless of treatment exposure or post-baseline assessment. The registry does not state how participants without evaluable post-baseline measurements were handled in other analyses, or whether any missing outcomes were imputed.

The registry notes one participant in each group had no PGTC seizures during baseline. A percent change uses baseline frequency as its reference, so a zero baseline count creates a special measurement situation. The registry does not explain how those participants’ percent changes were calculated or how they were treated in the primary analysis. This detail does not establish that the analysis was invalid, but it is relevant to understanding the endpoint definition and would benefit from further methods detail.

How should secondary planning figures be read?

General statistical explanation: power is the chance that a specified statistical procedure detects an effect of a specified size under stated assumptions. A planning calculation uses assumptions about the effect, variability, allocation, and test to estimate a future study’s sample size. It is not a measure of the probability that the completed trial’s result is correct, and it is not observed power.

The posted sample-size illustrations are based on the observed effect, use a two-sided significance level of 0.05, assume equal allocation, and use a normal approximation. They examine planning for 100%, 75%, and 50% of the observed effect. Such values are illustrative rather than a reconstruction of the trial’s original design. The trial’s own sample-size justification is not reported in the registry record. These calculations also concern secondary continuous outcomes, not the primary median-based analysis.

General statistical explanation: over-dispersion means that observed variability is greater than a model’s assumed variability. When it is present, standard errors can be too small and confidence intervals too narrow if the model does not account for it. The registry does not report an over-dispersion assessment for this primary endpoint, so it should not be assumed to have been present or absent.

10. Analyses Reported by the Sponsor

The registry’s own primary analysis is the Cochran-Mantel-Haenszel result: a median difference in final values of 31.6, with a 95% confidence interval from 15.8 to 48.1 and p < 0.001. This is the principal inferential result to use when describing the trial’s primary endpoint. The group summaries are medians of 32.1 for placebo and 75.4 for LTG XR, with their respective full ranges reported separately.

No independent recalculated primary comparison is reported in the registry record. A simple re-analysis from group summaries would not necessarily reproduce a method-based comparison of medians, particularly when the registry does not provide participant-level data or all implementation details. The reported values therefore should not be treated as though they were checked by a separate calculation.

The registry identifies the comparison method and reports the estimate, interval, and p-value, but it does not state all details of the contrast coding or analytic implementation. The estimate should consequently be quoted as the registry describes it, rather than assigning a direction beyond the information explicitly given. There is no reported discrepancy between a sponsor analysis and an independent recalculation for this primary endpoint.

11. Multiplicity, Missing Data and Analysis Populations

The primary analysis population is defined as randomized participants who took at least one dose of study drug and had at least one post-baseline efficacy assessment during the Double-Blind Treatment Phase. The registry does not give a detailed account of participants excluded from this population, reasons for missing assessments, or how missing primary outcome data were handled. Imputation methods, sensitivity analyses for missing outcomes, and assumptions about the relationship between missingness and seizure frequency are not reported in the registry record.

The registry lists 16 posted secondary outcomes and reports 17 outcome measures overall. When many outcomes are tested, the chance of at least one apparently unusual result can increase. General statistical explanation: multiplicity procedures can control error rates across a family of tests, or an ordered testing plan can designate which tests are confirmatory. Neither a multiplicity adjustment nor a prespecified testing hierarchy is reported in the registry record. The status of secondary p-values as adjusted or unadjusted is also not reported there.

The record does not report whether interim analyses were planned or conducted, whether stopping rules were used, or the complete statistical analysis plan. The sample-size justification for the trial is not reported in the registry record. Planning illustrations based on observed secondary effects are not a substitute for that original justification.

The primary definition includes an efficacy-assessment requirement after randomization and treatment exposure. Thus, the label Intent-to-Treat should be interpreted using the registry’s stated definition rather than assumed to mean every randomized participant was included regardless of data availability. The handling of the two participants noted to have no baseline PGTC seizures is not described beyond the registry’s observation that one was in each treatment group. These omissions limit independent assessment of missing-data and analysis-population choices; they do not establish how those choices affected the estimate.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: ataxia (3.1%), nystagmus (3.1%), suicide attempt (3.1%), abdominal pain (3.1%), nausea (3.1%).

During the Double-Blind Phase, serious adverse events were reported for 0 of 74 participants at risk in the placebo group and 1 of 72 at risk in the LTG XR group. The registry also lists 21 participants with other adverse events among 74 at risk in the placebo group and 27 among 72 at risk in the LTG XR group. These are participant counts, and the registry separates serious adverse events from other adverse events.

The continuation-phase groups and baseline failures are separate entries, not part of the double-blind comparison. The registry reports serious adverse events for 3 of 69 at risk in the Continuation Phase placebo/LTG group, 3 of 67 in the Continuation Phase LTG/LTG group, and 1 of 32 baseline failures. Group labels and phases should be kept distinct when interpreting these counts. Death results are not reported in the registry record.

Several serious-event terms are listed at 3.1% each: ataxia, nystagmus, suicide attempt, abdominal pain, nausea, and vomiting. The registry entries do not identify the group attribution for these terms, so they should not be assigned to a particular treatment group. Adverse-event counts alone do not establish causality or provide a complete comparison of safety; exposure duration, event definitions, severity, and participant-level context also matter.

13. What the Result Does, and Does Not, Mean

The primary result describes a difference between randomized groups in the registry’s measure of change in weekly PGTC seizure frequency during the Double-Blind Treatment Phase. The reported median difference was 31.6, with a 95% confidence interval from 15.8 to 48.1 and p < 0.001. The group medians were 32.1 for placebo and 75.4 for LTG XR; positive values correspond to reductions under the registry’s endpoint definition.

This result does not mean that every participant experienced a reduction, or that participants within a group had similar outcomes. The full ranges include negative values, and a median does not describe the distribution in detail. The reported confidence interval concerns uncertainty in the group contrast, not the range of individual responses.

The result also does not, by itself, establish the clinical importance of the estimated difference. The registry does not report a threshold for a clinically meaningful change, and statistical evidence is not equivalent to a judgment about benefit, burden, or individual treatment choice. No medical advice follows from the statistical summary.

The finding applies to the registry-defined analysis population and the double-blind period. It should not automatically be extended to all randomized participants, all people with epilepsy, or the continuation-phase groups. The study tested an add-on intervention alongside current seizure treatments, and details of those background treatments are not reported in the registry record. The record also does not provide all information needed to assess generalizability, missing-data handling, or the complete analysis plan.

14. Why This Trial Matters Statistically

This trial illustrates why the choice of summary statistic and analysis method matters. The primary endpoint is summarized by medians and full ranges, while the group comparison is reported using a Cochran-Mantel-Haenszel method. A simple contrast between displayed group summaries is not necessarily equivalent to the reported inferential analysis. Keeping descriptive summaries separate from the adjusted or method-based comparison helps prevent overinterpretation.

The endpoint also highlights the importance of direction conventions. The registry explicitly states that a positive percent change represents a reduction in seizure frequency. Without that definition, a positive or negative value could easily be misread. The broad full ranges further show why central estimates should not be mistaken for individual-level outcomes.

The trial contains multiple secondary outcomes spanning seizure activity, participant-reported measures, and other domains. That breadth can be informative, but it makes multiplicity and prespecification important. The registry record does not describe a multiplicity strategy or full analysis plan, so secondary findings should be interpreted with appropriate caution rather than treated as a collection of independent confirmatory tests.

Finally, the record distinguishes a reported primary analysis from planning illustrations based on observed secondary effects. The latter are not evidence of the completed trial’s power and do not explain the original design rationale. This distinction is broadly useful when reading registry results: observed outcomes, inferential analyses, and prospective planning calculations answer different statistical questions.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record provides the primary estimate, confidence interval, p-value, group medians, and full ranges, but does not report all details needed to reproduce the primary analysis. In particular, the contrast coding and further implementation details of the Cochran-Mantel-Haenszel procedure are not described. The participant-level data needed for an independent reconstruction are not reported in the registry record.

The percent-change outcome has a special issue noted by the registry: one participant in each group had no PGTC seizures during baseline. The record does not state how percent change was defined for those participants or how they were handled in the primary analysis. It also does not describe other missing-data patterns, imputation, sensitivity analyses, or the number of randomized participants without a qualifying post-baseline assessment.

The record does not report the trial’s original sample-size justification, the full statistical analysis plan, interim analyses, stopping rules, or a multiplicity strategy for the many secondary outcomes. These omissions constrain assessment of prespecification and robustness. They do not establish that such procedures were absent; they are not reported in the registry record.

The registry identifies treatment as add-on to current seizure treatments but does not describe the background regimens or dosing details. Generalizability is therefore difficult to assess from the record alone. Safety summaries include several study phases and a baseline-failures group; those populations should not be collapsed into the primary randomized comparison. The registry also does not report death results.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study Evaluating LAMICTAL Extended-Release Therapy Added To Current Seizure Treatments In Patients With Primary Generalized Tonic-Clonic Seizures (PGTC) Seizures — NCT00104416.
• Registry record last updated 2017-01-02; status last verified 2016-11.
• Results first posted 2010-05-18.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT00104416 records a completed Phase 3, randomized, double-blind, parallel-group comparison of extended-release lamotrigine added to current seizure treatments with placebo added to current treatments in participants with primary generalized tonic-clonic seizures. Enrollment was 153. The primary endpoint was percent change from baseline in weekly PGTC seizure frequency during the Double-Blind Treatment Phase, up to Week 19.

The primary group medians were 32.1 for placebo and 75.4 for LTG XR, with positive values representing reductions. The registry reports a Cochran-Mantel-Haenszel median difference in final values of 31.6, a 95% confidence interval from 15.8 to 48.1, and p < 0.001. The record includes secondary outcomes and safety counts but does not report the full statistical plan, multiplicity handling, or detailed missing-data methods.