Study to Test the Safety and Efficacy of Padsevonil as Adjunctive Treatment of Focal-onset Seizures in Adult Subjects With Drug-resistant Epilepsy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Test the Safety and Efficacy of Padsevonil as Adjunctive Treatment of Focal-onset Seizures in Adult Subjects With Drug-resistant Epilepsy (NCT03370120), a na phase 2 study. The primary outcome is Percentage of Participants With Treatment-Emergent Adverse Events (TEAEs) Reported by the Participant and/or Caregiver or Observed by the Investigator During the Entire Study.

2. Clinical Question

The study is an extension trial, so its central question concerns longer-term safety and tolerability during individualized adjunctive treatment. The primary endpoint definition includes events that began after the first dose in this study, as well as unresolved events already present before that dose if they worsened in intensity afterward. An adverse event is an untoward medical occurrence; the definition does not require that the event be caused by the study treatment.

The record also gives counts for serious adverse events and other adverse events. These are descriptive safety observations from a single treatment group, not evidence of a difference between treatments. They do not answer whether padsevonil changes seizure frequency or improves another measure of epilepsy control. The stated study purpose includes safety and tolerability, but the primary endpoint named in the registry is an adverse-event percentage, and the posted record does not include its outcome value.

3. Trial Design

This is a Phase 2, open-label, multicenter extension study of padsevonil as adjunctive treatment for adults with drug-resistant epilepsy and focal-onset seizures. It is registered as NCT03370120. The study title identifies the extension design; its official title describes evaluation of safety and efficacy. The registry lists one intervention, padsevonil, and a single-group model. Allocation is listed as NA, and masking as none.

The planned treatment approach was individualized rather than a fixed dose shared by all participants. The registry description states that each participant's starting dose was the dose at the end of the parent study, and that further individual dose adjustments were allowed after 1 week. Treatment was open-label, meaning that the study was not masked. The record does not describe a concurrent control group or random assignment.

Enrollment is listed as 406. The study start date is 2018-08-27, and the primary completion date is 2020-12-11. Its status is terminated. The reason for termination and whether termination changed the planned follow-up or analyses are not reported in the registry record. The primary endpoint timeframe runs from the entry visit at Week 0 through the safety follow-up visit, up to approximately 2 years; this is the stated timeframe, not a claim that every participant was followed for that duration.

The registry describes the Safety Set as all enrolled participants who received at least 1 dose of padsevonil, based on the first-dose date recorded in the study medication case report form. The serious and other adverse-event counts each list 406 participants at risk. No randomized comparison can be drawn from this design. Observations in an open-label, single-group extension can describe events reported during treatment, but cannot separate treatment effects from the course of the underlying condition, other treatments, or factors related to selection into and continued participation in the extension.

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: Participant Flow refers to the Safety Set. Participants who had completed a padsevonil (PSL) parent study (EP0091 \[NCT03373383\] or EP0092 \[NCT03739840\]) were enrolled in this study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Safety Set (SS) consisted of all enrolled participants who were administered at least 1 dose of PSL, based on the first dose date from the first administration of study medication CRF.

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.

Population: The Safety Set (SS) consisted of all enrolled participants who were administered at least 1 dose of PSL, based on the first dose date from the first administration of study medication Case Report form (CRF).

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With Treatment-Emergent Adverse Events (TEAEs) Reported by the Participant and/or Caregiver or Observed by the Investigator During the Entire Study, measured over From Entry Visit (Week 0) until the Safety Follow-up Visit (up to approximately 2 years).

The primary endpoint is the percentage of participants with treatment-emergent adverse events reported by the participant or caregiver, or observed by the investigator, during the entire study. Its timeframe begins at the entry visit, Week 0, and continues until the safety follow-up visit, up to approximately 2 years. The registry defines an adverse event as an untoward medical occurrence in a participant administered a pharmaceutical product; it does not require a causal relationship with treatment.

An event qualified as treatment-emergent if it was not present before the first dose in this study, or if an unresolved event that was already present worsened in intensity after exposure. This definition is intended to capture new or worsening events during the treatment period. It does not, on its own, establish that padsevonil caused an event. Clinical assessment of causality is a distinct question.

A percentage endpoint expresses how many people in the defined analysis population experienced at least one qualifying event, rather than counting every event episode. In general, this kind of measure is useful for describing how commonly participants experienced an event over a specified observation period. Interpretation depends on the event definition, the population included, and how long participants were observed. The registry record does not report the primary endpoint's percentage or its uncertainty, so the primary safety result cannot be quantified from the posted record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a percentage endpoint describe?

For a binary outcome such as whether a participant experienced at least one treatment-emergent adverse event, each participant is classified according to whether the event criterion was met during the specified timeframe. The resulting percentage describes the share of the defined population with an event. This is different from counting event episodes: one participant may experience more than one event, while a participant-level percentage counts that person once for the endpoint.

That distinction matters when reading safety data. A participant-level endpoint addresses how many people experienced an event, while an event count addresses how many occurrences were recorded. The registry's primary endpoint is explicitly a percentage of participants. The serious adverse-event entries, by contrast, are presented as counts of participants at risk and affected. The primary endpoint result itself is not reported in the registry record, so its participant-level percentage cannot be interpreted.

Why does the event definition matter?

The endpoint uses a treatment-emergent definition rather than treating every event during the study as new. An event qualifies if it was absent before the first dose in this study, or if an unresolved pre-existing event worsened in intensity after exposure. This definition establishes a time relationship to treatment, but it does not establish causation. The registry's adverse-event definition explicitly says a causal relationship with treatment is not necessary.

The timeframe also shapes what the endpoint captures. It begins at the Week 0 entry visit and continues to the safety follow-up visit, up to approximately 2 years. In general, a longer observation period can allow more opportunity to observe events, so event percentages should be interpreted in relation to follow-up. The record does not state that all participants completed the same duration of observation, nor does it provide participant-level follow-up details.

What can an open-label, single-group design establish?

In general, open-label means treatment assignment is not masked from participants and study personnel. A single-group design observes outcomes in participants receiving the study intervention without a concurrent randomized comparison group. Such a design can describe reported events during treatment, but cannot estimate a treatment difference against a control group. It is also not possible to rule out explanations such as the underlying condition, other treatments, or differences in who entered and remained in the extension.

The study's individualized dosing is part of the treatment context. The starting dose was based on the dose at the end of the parent study, and additional individual dose adjustment was allowed after 1 week. Because doses were individualized, the record describes a treatment strategy rather than a single uniform dose. Dose-specific outcome results are not reported in the registry record.

How should uncertainty be considered?

In general, a confidence interval gives a range of values compatible with the data under a statistical method and communicates uncertainty around an estimate. A narrow interval indicates greater statistical precision than a wide interval, but neither precision nor an interval establishes causality. The registry record does not report a confidence interval for the primary endpoint, and it does not report a statistical test or a model for that endpoint.

The record lists 0 analyses. That means the registry information summarized in the record does not include a reported analysis entry; it does not establish what analyses may have been conducted outside the registry record. No statistical analysis plan, handling of repeated events, or adjustment method is reported in the registry record. Those details should not be inferred from the endpoint title alone.

What do planning illustrations mean?

In general, sample-size planning illustrations use assumptions about the effect to be detected, the significance level, allocation, and statistical approximation to show possible participant requirements. The stated illustration assumptions use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. The illustrations are not this trial's design: this study is single-group and does not have two randomized arms.

Such illustrations are hypothetical planning exercises, not evidence about the study's achieved power or about an observed treatment effect. Sample-size planning numbers are not observed outcomes. The registry record does not report the study's own sample-size justification or statistical plan. It also does not report a primary endpoint result from which the stated observed-effect basis could be assessed. The planning assumptions therefore cannot fill in the missing endpoint result or establish what the study was designed to detect.

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("Padsevonil (SS)")
n      <- c(406)
events <- c(293)
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  Padsevonil (SS)  406  293  72.2  67.5  76.5  67.6  76.3  67.6  76.3

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 3 outcome measures and 0 analyses, but does not report the primary endpoint result. It also does not report a statistical analysis plan, a hierarchy for testing outcomes, or any method for controlling multiplicity. Multiplicity arises when several outcomes or hypotheses are assessed: the chance of at least one false-positive finding can increase if multiple tests are conducted without an appropriate plan. That general concern cannot be quantified for this study from the registry record, because the analyses and testing strategy are not reported.

Missing-data procedures are not reported in the registry record. There is no description of how participants with incomplete follow-up were handled, whether missing outcome information was imputed, or whether a sensitivity analysis was conducted. These details matter for a participant-level percentage, particularly when follow-up may differ among participants. The primary timeframe extends from Week 0 to safety follow-up, up to approximately 2 years, but the record does not give individual follow-up completeness.

The Safety Set is defined as enrolled participants who received at least 1 dose of padsevonil, and the adverse-event entries list 406 participants at risk. This defines the population for those safety counts. The registry record does not report other analysis populations or explain whether the primary endpoint used a different population. No randomized intention-to-treat comparison is described; the study is single-group and allocation is listed as NA.

These reporting gaps do not show that a particular method was or was not used outside the registry record. They mean that readers cannot assess the unreported methods or their influence on the primary endpoint from the registry record alone.

12. Safety Results

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

Deaths

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: epilepsy (1.7%), generalised tonic-clonic seizure (1.2%), status epilepticus (1.2%), seizure (1.0%), focal dyscognitive seizures (0.5%).

The adverse-event record lists the Padsevonil Safety Set with 406 participants at risk for serious adverse events and 406 at risk for other adverse events. It reports 48 participants affected by serious adverse events and 184 affected by other adverse events. The death count is 0. These figures are descriptive safety information from the study group, not comparisons with another treatment.

The listed serious adverse-event terms and percentages are Epilepsy, 1.7%; Generalised tonic-clonic seizure, 1.2%; Status epilepticus, 1.2%; Seizure, 1.0%; Focal dyscognitive seizures, 0.5%; and Partial seizures with secondary generalisation, 0.5%. These percentages are reported for the listed terms. The record does not indicate that the events were caused by padsevonil, and the general adverse-event definition does not require a causal relationship to treatment.

Serious adverse events are not the same as the primary endpoint, which concerns the percentage of participants with treatment-emergent adverse events of any qualifying type. Nor should the serious-event counts be added to the other-event counts: the record does not state that these categories are mutually exclusive. The primary endpoint value, its confidence interval, and a comparative safety estimate are not reported in the registry record.

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

The registry record supports a descriptive account of safety events among participants in an open-label, single-group extension study. It reports the size of the Safety Set, counts for participants with serious and other adverse events, a death count, and selected serious adverse-event terms with percentages. Those observations describe what was recorded in this study population under the stated safety reporting framework.

The record does not report the primary endpoint result, despite defining the endpoint and its timeframe. As a result, the percentage of participants with treatment-emergent adverse events cannot be stated from the registry record. The other safety counts cannot stand in for that result, because they describe separate categories. Nor can the counts establish how event rates would compare with a control group, since there was no randomized comparator.

These data also do not establish that padsevonil caused any particular event. An adverse event is an untoward medical occurrence, and the registry's definition does not require a causal relationship to treatment. In general, an observed event during treatment may be related to treatment, the underlying condition, other treatments, or other circumstances; this study design does not distinguish among those explanations.

The record does not provide seizure-control results or a comparative efficacy estimate. Although the official title includes evaluation of safety and efficacy, the available information does not establish whether seizure outcomes changed. It would therefore be inappropriate to use the listed safety counts to conclude that treatment did or did not improve seizure control, or to make a treatment recommendation.

The terminated status is another important feature of the record, but the reason for termination is not reported. Without that information and without a reported primary result or analysis, readers cannot determine from the registry record whether termination affected data completeness or the planned interpretation. The findings should be read as descriptive, limited registry information rather than a complete estimate of treatment effect.

14. Why This Trial Matters Statistically

This study illustrates an important distinction in clinical-trial interpretation: a clearly defined endpoint is not the same as a reported endpoint result. The registry specifies a participant-level percentage of treatment-emergent adverse events and defines the event and observation period, yet the primary result is not reported in the registry record. Without that result and an account of uncertainty, the central safety question cannot be answered quantitatively from the record.

The extension design also highlights the difference between safety description and comparative inference. A single-group study can document events observed while participants receive treatment, but it cannot provide the same treatment contrast as a randomized comparison. Open-label reporting, individualized dose adjustment, and the absence of a control arm all form part of the context in which the event counts should be interpreted. They do not invalidate the observations, but they limit what those observations can establish.

The record further shows why analysis-population definitions matter. The Safety Set includes enrolled participants who received at least 1 dose, and the serious and other event counts list 406 participants at risk. A clear population definition helps readers understand whose safety experience is being described. At the same time, unreported follow-up and missing-data procedures leave uncertainty about completeness across the stated timeframe.

For clinical researchers and statisticians, the study is a useful example of the boundary between a registry's descriptive safety entries and a full statistical report. Counts and selected term percentages can inform an account of observed events, but they do not substitute for the primary endpoint estimate, uncertainty interval, analysis plan, or a comparative design. Those distinctions are central to communicating what the available record can and cannot support.

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 main limitation for interpretation is that the primary endpoint result is not reported in the registry record. Although the endpoint definition and timeframe are specified, the record does not give the percentage of participants with treatment-emergent adverse events or its statistical uncertainty. The registry also lists 0 analyses, so no primary endpoint analysis is available for evaluation.

The study is single-group, open-label, and has no randomized comparator. Its adverse-event findings can describe events reported during treatment, but they do not estimate a difference between padsevonil and another treatment or no treatment. The design also cannot establish that an event was caused by padsevonil. Dosing was individualized, and the record does not report dose-specific safety outcomes.

The study status is terminated, but the reason is not reported in the registry record. The record does not explain whether termination affected participant follow-up, data collection, or analyses. It also does not report missing-data methods, individual follow-up completeness, a multiplicity strategy, or the statistical analysis plan. These omissions limit assessment of the completeness and precision of outcome reporting.

The primary timeframe extends from entry at Week 0 to safety follow-up, up to approximately 2 years. That stated timeframe does not establish that every participant was observed for the same duration. The record does not give further detail about follow-up by participant. Finally, the posted safety counts and selected adverse-event terms do not provide comparative seizure-control results; efficacy conclusions are not supported by the information reported.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study to Test the Safety and Efficacy of Padsevonil as Adjunctive Treatment of Focal-onset Seizures in Adult Subjects With Drug-resistant Epilepsy — NCT03370120.
• Registry record last updated 2021-12-29; status last verified 2021-11.
• Results first posted 2021-12-29.
• 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

NCT03370120 is a Phase 2, open-label, multicenter extension study of padsevonil as adjunctive treatment for adults with drug-resistant epilepsy and focal-onset seizures. It uses a single-group design, with allocation listed as NA and masking listed as none. Enrollment is 406; the study is terminated. It began on 2018-08-27, and its primary completion date is 2020-12-11.

The primary endpoint is the percentage of participants with treatment-emergent adverse events from entry at Week 0 to safety follow-up, up to approximately 2 years. The Safety Set includes enrolled participants who received at least 1 dose. The registry reports 48 participants with serious adverse events, 184 with other adverse events, and 0 deaths; the primary endpoint result and its analysis are not reported in the registry record. No randomized comparator is described, so the safety entries are descriptive rather than comparative.