A Study of Soticlestat as an Add-on Therapy in Children and Adults With Dravet Syndrome or Lennox-Gastaut Syndrome: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Soticlestat as an Add-on Therapy in Children and Adults With Dravet Syndrome or Lennox-Gastaut Syndrome (NCT05163314), a na phase 3 study. The primary outcome is Number of Participants With At Least One Treatment Emergent Adverse Event (TEAE).

2. Clinical Question

The study’s brief summary says its main aim was to learn whether add-on soticlestat reduces seizure numbers in children and adults with Dravet syndrome or Lennox-Gastaut syndrome. The listed primary endpoint, however, is a safety measure: the number of participants with at least one treatment-emergent adverse event. These are related parts of the study description, but they are not the same statistical question. A seizure-frequency outcome would describe changes in seizures; the primary endpoint counts participants with a defined adverse event.

The registry lists secondary measures of change in total seizure frequency, convulsive seizure frequency for the Dravet syndrome cohort, and major motor drop seizure frequency for the Lennox-Gastaut syndrome cohort. It also lists global-impression, seizure intensity and duration, nonseizure symptoms, and quality-of-life measures. Their inclusion indicates the kinds of outcomes the study record names, not what the results were. No efficacy estimate or comparison with a control group is reported.

3. Trial Design

This is a phase 3, prospective, open-label, multisite extension study of soticlestat as adjunctive therapy for people with Dravet syndrome or Lennox-Gastaut syndrome. The official title identifies it as an extension of phase 3 studies and names the study ENDYMION 2. The registry lists Takeda as the sponsor, an enrollment of 352, a start date of 2022-03-04, a primary completion date of 2025-09-24, and a terminated status.

The design is single-group, with masking listed as none and allocation as not applicable. Participants received soticlestat alongside their standard anti-seizure therapy. Because there is no separate control arm, the study does not provide a randomized, concurrent comparison of soticlestat with placebo or another treatment. The record does not report a reason for termination, so the status alone cannot explain why the study ended.

The primary endpoint timeframe is up to the end of the study, approximately 3.6 years. This describes the stated follow-up window for the endpoint; it does not mean every participant was followed for that entire period. Participant-level follow-up duration and the amount of missing follow-up are not reported in the registry record.

The arm description begins to describe weight-based titration and mini-tablets, but the posted text is truncated. The full dose schedule and any additional administration details are therefore not reported in the registry record. The study is explicitly an extension, but details of the preceding studies, eligibility for entry into this extension, and any requirements for prior treatment response are not reported in the registry record.

The registry enrollment and the safety summary’s at-risk count are both 352. These figures describe the record’s enrollment and safety summary, respectively; they do not establish that every participant completed follow-up or contributed the same observation time. The safety population is defined as participants who received at least one dose of study drug. The record does not give further details about participant flow, withdrawals, or how many participants contributed data to each secondary outcome.

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: A total of 352 participants with Dravet Syndrome (DS) or Lennox-Gastaut Syndrome (LGS) who had participated in phase 3 clinical studies, TAK-935-3001 (NCT04940624), or TAK-935-3002 (NCT04938427), respectively, and who continued to receive standard-of-care (SOC) antiseizure therapy were enrolled in the study for added treatment with soticlestat.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety analysis set included participants who received at least 1 dose of study drug.

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: Safety analysis set included participants who received at least 1 dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With At Least One Treatment Emergent Adverse Event (TEAE), measured over Up to end of study (approximately 3.6 years).

The primary endpoint counts participants with at least one treatment-emergent adverse event, or TEAE. The registry defines an adverse event as an unfavorable or unintended medical occurrence temporally associated with use of a drug, whether or not it is considered related to the drug. A TEAE is an adverse event that started or increased in severity after the first dose in this open-label extension study. The endpoint is therefore based on timing and occurrence, not on a determination that soticlestat caused the event.

This kind of endpoint is used to summarize how many people in the safety population experienced an event during the specified observation period. It is a participant count, rather than a measure of seizure change or a count of all event episodes. A person with more than one event can still contribute once to an endpoint defined as having at least one event. The posted endpoint description does not say that all events were considered equally serious or related to study treatment.

For an extension study, collecting adverse-event information over continued treatment can describe safety experience beyond the initial phase 3 studies. But a single-group count cannot by itself distinguish events associated with treatment from events reflecting participants’ underlying conditions, background therapy, or other circumstances. The endpoint’s interpretation also depends on how long participants were observed and how completely events were captured. The timeframe is listed as up to the end of study, approximately 3.6 years, while individual exposure and follow-up details are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies the design as prospective, open-label, multisite, and single-group. These descriptions answer different methodological questions. Prospective means participants are followed forward for information collected during the study. Open-label means masking is listed as none. Multisite describes the study setting. Single-group means the record does not present separate treatment arms for a randomized comparison. Together, these features frame what can be learned from the posted results: they can describe observations in the treated group, but do not isolate a treatment effect against a concurrent control.

What does a participant-count endpoint summarize?

The primary outcome is a count of participants with at least one TEAE. General statistical explanation: an endpoint based on whether a participant experienced an event reduces each person’s experience to an event indicator for the defined period. It is different from counting every episode, measuring how long an event lasted, or rating its severity. A participant can experience multiple events and still meet the endpoint as a participant with at least one event. That makes the endpoint straightforward to describe, but it does not show event frequency or the detailed course of each event.

The TEAE definition also matters. An event qualifies if it started or increased in severity after the first dose in this extension. The definition is temporal and does not require that the event be judged caused by soticlestat. The registry describes an adverse event as an unfavorable or unintended occurrence temporally associated with drug use, whether or not considered related. Thus, an event count should not be translated into a count of treatment-caused harms.

Why does the lack of a control group matter?

General statistical explanation: a treatment comparison needs a basis for estimating what might have happened under an alternative treatment or without the treatment under study. A concurrent randomized control group can help make that comparison by balancing, on average, factors that influence outcomes. This extension has one group, and allocation is listed as not applicable. It therefore supports description of the experience recorded in this group but cannot provide a randomized estimate of a difference between treatment strategies.

Participants also received standard anti-seizure therapy, so observed events and clinical outcomes occurred in the context of add-on treatment and background therapy. The registry does not provide enough information to separate the contribution of each component. Changes over time might also reflect factors not represented in a one-group comparison. These are general limitations of interpretation for uncontrolled observations, not claims about the cause of any event in this study.

How should follow-up and exposure be considered?

The endpoint timeframe is up to the end of study, approximately 3.6 years. This is a stated study-level window, not an assurance that every participant had that duration of observation. General statistical explanation: when participants have different follow-up durations, a simple participant count does not express how much time each person was observed. Time-to-event methods can account for varying follow-up under specified assumptions, but no such primary analysis is described in the registry record. Individual follow-up, exposure duration, and censoring information are not reported in the registry record.

Similarly, the study’s enrollment of 352 does not establish that 352 participants had complete endpoint information for every analysis. The safety analysis set is defined as participants who received at least one dose. The registry’s safety summary lists 352 at risk. Further participant flow and outcome-specific analysis populations are not reported in the registry record. This distinction matters because enrollment, treatment exposure, and availability of outcome data are not automatically identical concepts.

What do the listed percentages and counts tell us?

The serious-event summary gives participant figures, and the listed serious terms have percentages. General statistical explanation: a percentage for a specific event term describes how common that listed term was in the stated summary context; it does not show whether the event was caused by treatment, how severe it was for a particular participant, or whether it happened once or repeatedly. The record does not provide enough detail to reconstruct event-level patterns. Counts for different event categories should not be added unless the categories are known to be mutually exclusive, which is not reported.

Confidence intervals would communicate sampling uncertainty around a statistical estimate. General statistical explanation: an interval is not a guarantee that the true value lies within it, but a way to express precision under an analysis model and its assumptions. No confidence interval is reported for the primary endpoint in this record. Without an interval and a defined analysis, the posted counts remain descriptive rather than a basis for judging statistical precision.

What is known about the analysis plan?

The registry lists 14 outcome measures and 0 analyses. It does not report a statistical analysis plan, a method for handling repeated measures, a model for seizure counts, or a method for dealing with missing outcomes. Such methods should not be inferred from the outcome names. General statistical explanation: count outcomes can show greater variability than a simple model assumes, a feature called over-dispersion. If present, it can affect standard errors and confidence intervals; the record does not say whether over-dispersion was assessed or how seizure-frequency outcomes were analyzed.

The sample-size section is a planning illustration based on the observed effect, using a two-sided significance level of 0.05, equal allocation, and a normal approximation. The stated effect is a fraction of the observed effect. These calculations are not the trial’s own design, and they are not observed power. Since this study has one group and no like-for-like control comparison, the illustration’s two-arm assumptions do not describe the actual trial design. The registry does not report its own sample-size justification.

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("Soticlestat")
n      <- c(352)
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  Soticlestat  352  293  83.2  78.9  87.0  79.0  86.8  79.0  86.8

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 14 outcome measures, including the primary safety endpoint and secondary seizure-frequency, global-impression, and quality-of-life outcomes. It lists 0 analyses. The record does not report how the study handled multiplicity, such as whether testing across several outcomes was controlled or whether some outcomes were treated as descriptive. Without a stated plan, the listed outcomes should not be read as a set of confirmatory tests with a known error rate.

General statistical explanation: testing many outcomes can increase the chance that at least one result appears notable by chance. A multiplicity strategy can address that concern, but the appropriate approach depends on the analysis plan and the questions being tested. No such strategy is reported in the registry record. This is especially relevant when interpreting multiple seizure measures, separate syndrome cohorts, and several impression or quality-of-life measures. Their presence on the outcome list does not establish which findings, if any, were statistically supported.

Missing-data methods are also not reported in the registry record. The endpoint timeframe extends up to the end of study, approximately 3.6 years, but the record does not state how incomplete follow-up, missed visits, withdrawal, or unavailable outcome assessments were handled. General statistical explanation: missingness can affect estimates when the likelihood or pattern of missing data is related to participants’ outcomes or characteristics. The direction and extent of any such effect cannot be determined without information about missingness and the analysis approach.

The safety population is defined as participants who received at least one dose. The record does not provide outcome-specific analysis populations or details about how many participants contributed to each secondary outcome. These gaps limit assessment of whether the reported summaries represent all enrolled participants, all treated participants, or a subset with available data for a particular measure.

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: pneumonia (4.8%), seizure (4.5%), status epilepticus (3.1%), covid-19 (1.4%), pneumonia aspiration (1.1%).

The posted summary gives 352 at risk, 86 participants with serious adverse events, and 214 participants with other adverse events in the Soticlestat group. It also lists 4 deaths. These are counts, not a comparison against another group. The record does not report a causal assessment for the deaths, so their presence in the summary should not be treated as proof that they were caused by study treatment.

The listed serious adverse-event terms and percentages are pneumonia at 4.8%, seizure at 4.5%, status epilepticus at 3.1%, COVID-19 at 1.4%, pneumonia aspiration at 1.1%, and seizure cluster at 1.1%. These figures describe the listed terms in the posted summary. They are not a complete narrative of all events, and the registry record does not provide corresponding counts or detailed event histories.

The serious and other adverse-event figures should not be added together to obtain a total. The record does not say that these categories are mutually exclusive, and one participant may have experienced more than one event or category. Nor do the listed percentages establish treatment causation. General interpretation: adverse-event summaries describe what was recorded during observation; comparing risk between treatments requires a suitable comparator and analysis. Neither is available in this single-group record.

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

The record documents a prospective, open-label, single-group extension study of soticlestat added to standard anti-seizure therapy in people with Dravet syndrome or Lennox-Gastaut syndrome. It reports safety-summary counts, including 86 participants with serious adverse events, 214 with other adverse events, and 4 deaths, in a group with 352 at risk. It also lists several serious-event terms with percentages. These are the safety observations that can be described from the posted information.

They do not establish whether soticlestat caused a particular event or whether event occurrence differed from what would have happened with another treatment. The absence of a control group prevents a randomized comparison. The primary endpoint is a TEAE participant count, but a full primary endpoint estimate and its precision are not reported in the registry record. The serious-event summary is not, by itself, enough to infer the complete primary endpoint result.

The study’s stated aim includes learning whether seizure numbers are reduced, and the record lists seizure-frequency outcomes as secondary measures. However, outcome results for seizure frequency and the other listed secondary measures are not reported in the registry record. The stated aim and endpoint list do not establish a result. No conclusion about the direction, magnitude, or precision of seizure changes can be drawn from the information posted.

The registry status is terminated. It does not report the reason for termination, whether the planned study follow-up was completed, or the implications for interpreting the analyses. Those omissions should remain uncertainties rather than invitations to infer a cause. The record also does not report the study’s own sample-size justification or a statistical analysis plan.

Overall, the posted material supports a description of a long-term safety-focused extension and its reported adverse-event summaries. It does not establish comparative safety, treatment causation, or efficacy. Any displayed planning illustrations are not observed power and do not transform a single-group extension into a controlled trial.

14. Why This Trial Matters Statistically

This study illustrates why a trial’s aim, primary endpoint, and design need to be read together. The brief summary describes an interest in seizure reduction, while the primary endpoint is the number of participants with at least one TEAE. The registry also names seizure-frequency and other clinical measures as secondary outcomes. Keeping these questions distinct prevents a safety endpoint from being mistaken for an efficacy result, or an outcome’s presence in the registry from being mistaken for evidence of a result.

The extension design addresses a different statistical setting from a randomized comparison. Its one-group safety summaries can describe events recorded among participants receiving add-on soticlestat in the context of standard anti-seizure therapy. They cannot, on their own, show how those observations compare with a concurrent alternative. This distinction is central when interpreting adverse-event counts and when considering how background illness and treatment may contribute to observed events.

The record also makes the completeness of reporting visible. It lists 14 outcomes but 0 analyses, and does not report confidence intervals, missing-data methods, multiplicity handling, or a statistical analysis plan. These omissions limit judgments about precision and the strength of conclusions across multiple outcomes. A careful account should say what is present, identify what remains unavailable, and avoid filling gaps with assumed methods.

Finally, the terminated status and the long stated endpoint window make follow-up context important. The reason for termination and participant-level follow-up are not reported in the registry record. Without these details, readers cannot tell from the record alone how termination relates to follow-up or the posted summaries. Treating that uncertainty explicitly is part of sound statistical interpretation.

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 key limitation is the single-group design. There is no randomized or concurrent comparator, so the posted counts describe the Soticlestat group but do not estimate a difference between treatment strategies. The study is open-label, with masking listed as none. This means the record does not describe outcome assessment under masking, although the registry does not provide enough detail to characterize how awareness of treatment may have affected any particular outcome.

The registry status is terminated, but the reason is not reported in the registry record. Whether termination affected follow-up, outcome collection, or analysis is also not reported. The stated primary endpoint timeframe is up to the end of study, approximately 3.6 years, but individual follow-up and treatment exposure are not reported. The enrollment of 352 and the at-risk count of 352 do not establish complete follow-up for each participant or for every outcome.

There is no primary endpoint confidence interval or full endpoint analysis reported in the registry record. The record lists secondary outcomes but gives no numerical results for them. It also does not report a statistical analysis plan, multiplicity approach, missing-data method, or sample-size justification. Consequently, readers cannot assess the precision of outcome estimates, the handling of incomplete data, or how conclusions across multiple measures were intended to be controlled.

The arm description is truncated, so full dosing and administration details are not available. The registry does not provide detailed event narratives or causal attribution for the adverse-event summaries, including the 4 deaths. The serious-event terms and percentages are descriptive; they do not identify causation or provide a controlled risk comparison. These limitations restrict interpretation to the reported study design and summaries rather than broader conclusions about treatment effects.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of Soticlestat as an Add-on Therapy in Children and Adults With Dravet Syndrome or Lennox-Gastaut Syndrome — NCT05163314.
• Registry record last updated 2026-09-04; status last verified 2026-08.
• Results first posted 2026-06-09.
• 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

This phase 3 study is a prospective, multisite, open-label, single-group extension of soticlestat as add-on therapy for people with Dravet syndrome or Lennox-Gastaut syndrome. The registry lists an enrollment of 352, a start date of 2022-03-04, a primary completion date of 2025-09-24, and a terminated status. The reason for termination is not reported in the registry record.

The primary endpoint is the number of participants with at least one TEAE, with follow-up up to the end of study, approximately 3.6 years. The safety summary lists 86 participants with serious adverse events, 214 with other adverse events, and 4 deaths among 352 at risk. The record lists 14 outcome measures and 0 analyses, but does not report results for the named secondary seizure and clinical outcomes, a statistical analysis plan, or methods for multiplicity and missing data.

The posted information is descriptive safety evidence from one group, not a randomized treatment comparison. It does not establish treatment causation or provide a reported estimate of seizure reduction.