Long Term Follow-up Study With Keppra XR (Levetiracetam XR) for Partial Seizures: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Long Term Follow-up Study With Keppra XR (Levetiracetam XR) for Partial Seizures (NCT00419393), a na phase 3 study. The primary outcome is Number of Subjects Who Experienced at Least 1 Treatment Emergent Adverse Event During the Actual Treatment Period (6 Months-2 Years).

2. Clinical Question

The broad clinical purpose stated in the registry was to provide continued treatment and assess long-term safety in subjects with partial-onset seizures. The record also identifies remaining on Keppra XR monotherapy through 6 months and through 12 months as secondary outcomes. Those questions concern persistence on monotherapy, whereas the primary outcome concerns the occurrence of treatment-emergent adverse events. They should not be conflated: an adverse event is not automatically an event caused by the study drug, and continued treatment is not itself a measure of safety or comparative efficacy.

The registry record does not report a comparator, a blinded assessment, or a treatment-effect estimate against another group. It also does not report the numerical result for the primary endpoint in the outcome entry. The clinical question is consequently descriptive: what adverse events were recorded among people receiving continued treatment in this study, and what can be learned from those observations without a control group?

3. Trial Design

This was a Phase 3, open-label, single-group study sponsored by UCB Pharma. The registry lists allocation as NA and masking as none. It reports enrollment of 190 and a completed status. The recorded start is 2007-12, and primary completion is 2010-03. These dates describe the registry record; they do not by themselves establish the amount of follow-up completed by each participant.

The intervention was Keppra XR (levetiracetam XR), recorded as a drug. The arm description specifies flexible dosing of 1000 - 3000 mg/day throughout the study, with a planned duration of approximately 6 months-3 years. The primary outcome instead describes the actual treatment period as 6 months-2 years. These are distinct descriptions: one is the planned duration in the arm description, and the other is the timeframe attached to the primary outcome.

The design has no parallel comparator arm. Participants and investigators were not masked, and no allocation procedure is reported. In a single-group follow-up, outcome counts describe the experience of the people enrolled and treated in that study. They do not, on their own, show how outcomes would have differed under another treatment or in the absence of treatment. The registry explicitly notes that the arms are not a like-for-like control comparison; the single-group design makes that limitation especially relevant.

Of the 190 subjects who began the study, 189 received at least one dose and entered the Safety Set. That set is the population identified for the primary safety analysis. A person who began the study but did not receive any study medication is therefore not included in that stated set. The registry record does not describe further analysis populations, participant-level follow-up completeness, or rules for handling discontinuation.

The stated purpose was continued treatment and assessment of long-term safety. The two named secondary outcomes concern the percentage of subjects remaining on Keppra XR monotherapy from study entry through 6 months and through 12 months. The record lists five outcome measures overall, but detailed results for every measure and a statistical analysis plan are not reported in the registry record.

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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Of the 190 subjects beginning the study, 189 received at least one dose of study medication, placing them in the Safety Set (SS) analyzed here.

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 Number of Subjects Who Experienced at Least 1 Treatment Emergent Adverse Event During the Actual Treatment Period (6 Months-2 Years), measured over Duration of the Treatment Period (6 months-2 years).

The primary endpoint is the number of subjects who experienced at least one treatment-emergent adverse event during the actual treatment period, described as 6 months-2 years. It counts subjects with one or more events rather than counting every event as a separate subject-level occurrence. A subject with multiple events would still meet the endpoint’s “at least one” criterion once.

The registry’s definition is broad: an adverse event is any untoward medical occurrence in a patient or clinical investigation subject administered a pharmaceutical product, and it does not necessarily have a causal relationship with treatment. “Treatment-emergent” locates the event in the treatment period; it does not establish that the study drug caused it. The endpoint is therefore a measure of events observed during exposure, not a causal attribution measure.

Counting subjects with at least one event is useful for describing how many participants experienced an event during the stated period. It can give a straightforward overview of the frequency of affected participants. It does not convey the event’s severity, duration, recurrence, or clinical importance by itself. Those aspects require additional definitions and results.

The primary outcome entry identifies the measure and the Safety Set of 189 subjects, but it does not give the numerical count for that endpoint. The adverse-event listing separately records 22 subjects in the serious category and 57 in the other category, each among 189 at risk. Those entries are useful safety summaries, but the registry record does not explicitly identify either count as the numerical result of the primary endpoint. They should not be substituted for that missing result.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a single-group design permit?

As a general statistical point, a single-group study can describe outcomes observed among the participants who received the intervention. It has no concurrent randomized group to represent what would have happened under an alternative intervention. In this study, the registry records one treatment group, allocation as NA, and no masking. Consequently, the observed safety figures are descriptive. They do not estimate a contrast between treatments, and they cannot isolate the treatment’s contribution from other influences on event occurrence.

The open-label design means that treatment assignment was not concealed from participants and investigators. As a general consideration, knowledge of treatment can affect reporting, observation, and assessment of events. The registry does not report whether such effects occurred in this study, so they should be treated as possible design considerations, not as documented findings.

What does the primary endpoint count?

The primary measure is the number of subjects who had at least one treatment-emergent adverse event during the actual treatment period. Its unit is subjects, not individual event episodes. This distinction matters because multiple events in one person do not necessarily correspond to multiple people meeting the endpoint. The adverse-event definition includes untoward medical occurrences after administration of a pharmaceutical product and explicitly does not require a causal relationship with treatment.

As a general statistical point, a binary endpoint records whether a defined condition occurred for each participant. The count of affected participants is a summary of those participant-level outcomes. The meaning of that count depends on a clear event definition, a defined observation period, and a specified analysis population. Here, the record identifies a Safety Set of 189 subjects and an actual treatment-period timeframe of 6 months-2 years. It does not state the numerical primary-endpoint count in the primary outcome result.

How should the safety categories be distinguished?

The adverse-event listing separates a serious category from an other category. It records 22 affected subjects among 189 at risk in the serious category and 57 among 189 at risk in the other category. These counts are not interchangeable with the missing primary endpoint value unless the registry explicitly connects them to that endpoint; it does not do so in the outcome entry. The record also lists selected serious-event terms with percentages, but does not report the denominator or selection rules for those term-level percentages.

As a general point, seriousness and severity are different concepts in adverse-event reporting. Seriousness typically concerns regulatory or clinical criteria, while severity describes the intensity of an event. The registry material summarized in the registry record does not give a severity scale or event-level descriptions, so those features cannot be inferred from the serious-category count or the listed terms.

What can confidence intervals and tests contribute?

As a general statistical explanation, a confidence interval describes uncertainty around an estimate under the assumptions of the analysis. A narrower interval indicates greater precision for that estimate, while a wider interval indicates less precision. A hypothesis test assesses how compatible observed data are with a specified null hypothesis under its assumptions. Neither measure converts an observational count into a causal result; design and analysis both matter.

No confidence interval, p-value, or treatment comparison is reported in the registry record. The record lists zero analysis entries. It does not report an inferential model or a statistical analysis plan. Therefore, no inferential finding should be attributed to this study from the registry entries.

How should planning calculations be understood?

As a general statistical explanation, a sample-size calculation is conditional on its assumptions, including the target effect, significance level, power target, allocation, and analysis method. The planning description associated with this study uses a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect expressed as a stated fraction of an observed effect. It describes a planning illustration, not the trial’s own design justification. The study itself is single-group, so the hypothetical equal-allocation comparison does not describe its enrollment design.

Such a calculation is not observed power. Observed power would not resolve the design limitations or recover a missing endpoint result. Any planning illustration depends on the effect used as its reference and on assumptions about the proposed comparison. The registry record does not report a sample-size justification for the actual study, and the primary outcome’s numerical count is not given in its outcome entry. The planning assumptions should therefore not be presented as evidence that the study was adequately powered to answer a comparative question.

For time-to-event outcomes, general methods may account for the number of events rather than only the number of participants, while other endpoint types commonly use participants per arm in planning. The planning description distinguishes these approaches, but the listed primary outcome is a count of participants with an event, not a reported time-to-event result. No sample-size calculation should be mistaken for an observed study result.

What is not documented?

The registry record does not report the primary endpoint count, confidence intervals, p-values, outcome-specific missing-data rules, interim analyses, or a full statistical analysis plan. It does not report how adverse events were adjudicated, how long each person was observed, or the denominator and selection process for the serious-event term percentages. These omissions limit reconstruction of the analysis and the precision of conclusions. They are reporting gaps, not evidence that a particular procedure was or was not used.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Keppra XR")
n      <- c(189)
events <- c(126)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Keppra XR  189  126  66.7  59.5  73.3  59.7  73.0  59.7  73.0

11. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures and names two secondary outcomes: the percentage of subjects remaining on Keppra XR monotherapy from study entry through 6 months and through 12 months. The number of analysis entries is zero. Numerical results for those secondary outcomes and for the primary endpoint are not reported in the registry record.

When a study assesses multiple outcomes, time points, or statistical comparisons, the chance of at least one apparently notable result can increase. As a general statistical point, a prespecified testing hierarchy or adjustment can help control this issue, but the appropriate approach depends on the study’s objectives and analysis plan. The registry record does not report a multiplicity strategy, a hierarchy among outcomes, or whether any adjustment was planned. It is therefore not possible to determine how multiplicity was handled.

The Safety Set includes 189 subjects who received at least one dose, out of 190 who began the study. This identifies the population for the stated safety analysis, but it does not describe whether all participants completed follow-up or had complete outcome information. Missing-data rules, censoring procedures, treatment discontinuation handling, and any imputation method are not reported in the registry record. The amount and pattern of missing outcome data are also not reported.

As a general statistical point, incomplete follow-up can affect estimates when the chance of missing data is related to participant characteristics or outcomes. Without details on completeness and handling, the potential impact cannot be evaluated from the stated counts alone. The record’s Safety Set definition should therefore be distinguished from evidence that every participant contributed complete observations for the entire treatment period.

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: convulsion (2.1%), chest pain (1.1%), angina pectoris (0.5%), abdominal pain (0.5%), drowning (0.5%).

The registry’s adverse-event listing reports 22 subjects affected in the serious category among 189 at risk, and 57 affected in the other category among 189 at risk. These are category-level entries for the Keppra XR group. The primary outcome entry separately defines a broader endpoint—at least one treatment-emergent adverse event—but does not report its numerical result. The category counts should not be treated as that missing endpoint result.

Selected serious-event terms are listed with percentages: Convulsion, 2.1%; Chest pain, 1.1%; and Angina pectoris, Abdominal pain, Drowning, and Non-cardiac chest pain, 0.5% each. These percentages are registry figures, not estimates recalculated from the category counts. The denominator and term-selection rules are not reported in the registry record.

The listing does not establish that the events were caused by Keppra XR. The registry’s adverse-event definition expressly allows events without a necessary causal relationship to treatment. A death count is not reported in the registry record. Seriousness should also not be read as a severity rating: event intensity and details are not reported in these entries.

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

The results describe safety observations among subjects who received Keppra XR in an open-label, single-group follow-up study. The registry identifies 189 treated subjects in the Safety Set and lists serious and other adverse-event category counts, along with selected serious-event terms and their percentages. These are useful as descriptive records of events reported in the study population.

The findings do not establish that Keppra XR caused any listed event. The registry’s definition of an adverse event does not require a causal relationship, and there is no concurrent comparison group to show how event occurrence differed under another treatment or without the study treatment. The design also does not provide a randomized treatment contrast. The record therefore supports description of reported events, not a causal or comparative conclusion.

The primary outcome is defined as the number of subjects with at least one treatment-emergent adverse event during the actual treatment period, but its numerical result is not reported in the registry record. The separate counts of 22 and 57 belong to the serious and other categories in the safety listing and should not be used as substitutes for the primary endpoint. The term-level percentages likewise do not convey event duration, recurrence, severity, or attribution.

The record identifies monotherapy persistence outcomes at 6 months and 12 months, but their results are not reported. It also does not report confidence intervals, p-values, an analysis plan, missing-data handling, or participant-level follow-up completeness. Without those details, the precision and robustness of estimates cannot be assessed. The stated enrollment and Safety Set counts alone do not demonstrate that follow-up was complete for every participant.

In short, the registry characterizes a treated group and reports selected safety counts. It does not answer how those outcomes compare with another treatment, whether treatment caused them, or what the missing primary outcome count would be. Interpretation should remain limited to the descriptive information actually recorded.

14. Why This Trial Matters Statistically

This study is statistically informative because it illustrates the distinction between a safety description and a comparative treatment estimate. Long-term follow-up can collect observations over an extended treatment period, and the registry identifies both an actual treatment timeframe and a planned duration. Yet duration and enrollment do not substitute for a comparator. A single-group design can describe what was recorded among treated participants, but it cannot directly measure how outcomes differ from an alternative group.

The endpoint definition also matters. The primary measure counts subjects with at least one treatment-emergent adverse event, while the safety listing separates serious and other categories and reports selected serious-event terms. Those summaries answer related but not identical questions. Careful interpretation requires keeping the endpoint definition, event category, analysis population, and observation period distinct. The primary endpoint’s numerical result is not reported in the registry record, so its value cannot be reconstructed from the other safety entries.

The record makes the analysis population explicit: 189 of the 190 subjects who began the study received at least one dose and entered the Safety Set. That is useful context for reading the safety counts. At the same time, it does not reveal how complete follow-up was or how missing outcomes were handled. The absence of analysis entries and reported inferential results also means that the precision of any treatment-effect estimate is not documented.

For clinical researchers, this record reinforces a general statistical lesson: a clearly defined endpoint and a defined population are important, but interpretation also depends on design, comparative information, and transparent reporting. The event percentages and counts can be described as registry observations, while causal attribution and comparative claims require evidence the record does not report.

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 study is open-label and single-group, with no comparator. This prevents a like-for-like estimate of differences in adverse-event occurrence between treatments. Because treatment was not masked, the design also leaves open the possibility that knowledge of treatment influenced reporting or assessment; whether that occurred is not reported in the registry record.

The primary outcome is defined, but its numerical result is not reported in the registry record. The safety listing provides category counts and selected serious-event terms, but it does not state how those entries relate numerically to the primary endpoint. The term-level percentages have no reported denominator or selection method in the record described. A death count is not reported in the registry record.

Planned treatment duration is approximately 6 months-3 years, while the primary outcome timeframe describes the actual treatment period as 6 months-2 years. The record does not report participant-level exposure durations or follow-up completeness, so the amount of observation time represented by the group counts cannot be assessed in detail. The registry also does not report missing-data methods, discontinuation handling, event adjudication procedures, or an analysis plan.

The record lists five outcome measures and two named monotherapy persistence outcomes, but does not report their numerical results or a multiplicity strategy. It lists zero analysis entries and no confidence intervals or p-values. Consequently, statistical precision and the handling of multiple outcomes cannot be evaluated from the registry record.

Finally, the sample-size assumptions are described as a planning illustration, not as the study’s own design justification. The study itself has no two-arm allocation, and the registry does not report its sample-size rationale. The planning assumptions therefore do not establish that the actual single-group study was designed to make a comparative inference.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Long Term Follow-up Study With Keppra XR (Levetiracetam XR) for Partial Seizures — NCT00419393.
• Registry record last updated 2014-09-05; status last verified 2011-06.
• Results first posted 2011-07-13.
• 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

NCT00419393 is a completed Phase 3, open-label, single-group long-term follow-up study sponsored by UCB Pharma. It enrolled 190 subjects with partial-onset seizures; 189 received at least one dose and entered the Safety Set. Participants received flexible Keppra XR (levetiracetam XR) dosing of 1000 - 3000 mg/day.

The primary outcome concerns the number of subjects with at least one treatment-emergent adverse event during the actual treatment period, described as 6 months-2 years. Its numerical result is not reported in the registry record. The safety listing reports 22 subjects in the serious category and 57 in the other category, each among 189 at risk, plus selected serious-event terms with percentages. The study has no comparator, and the record does not report an analysis plan, missing-data methods, or inferential results.