A Study of Soticlestat Tablets in Healthy Adults: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Soticlestat Tablets in Healthy Adults (NCT05284760), a randomized phase 1 study. The primary outcome is Cmax: Maximum Observed Plasma Concentration for Soticlestat.

2. Clinical Question

3. Trial Design

This was a Phase 1, randomized, open-label crossover study in healthy adults. It enrolled 96 participants and is listed as completed. The listed study dates are 2022-03-11 and 2022-05-22 for study start and primary completion. Participants were assigned to sequences of treatment conditions, so comparisons were made within the crossover design rather than between a treatment group and a separate parallel control group.

The study had two parts. Part A examined three fasted formulations, each delivering 300 mg of soticlestat: T4 tablets, T3 mini-tablets, and T3 commercial tablets. Part B examined the T4 tablet in fasted and fed conditions, as well as a fasted T4 tablet crushed and mixed with applesauce. The registry describes six treatment sequences in Part A and six in Part B, with treatments given in different orders. The descriptions indicate dosing on Day 1 of each period, but the full sequence text is truncated in the registry record.

The crossover structure allows each participant’s measurements under one condition to be considered alongside that participant’s measurements under another condition. In general, this can reduce the influence of stable between-person differences when estimating treatment-condition contrasts. The record does not report details such as washout duration, handling of period or sequence effects, or a full protocol for the crossover analysis. Those details are not reported in the registry record.

The primary endpoint was maximum observed plasma concentration (Cmax), assessed from pre-dose through multiple time points up to 96 hours after dosing. The primary pharmacokinetic analysis population was restricted to participants with sufficient protocol compliance and an evaluable pharmacokinetic profile. The record lists 72 participants for Treatment A, 72 for Treatment B, and 71 for Treatment C in Part A; it lists 24 for each of Treatments D, E, and F in Part B. These are the numbers shown for the endpoint analyses, not a statement that every enrolled participant completed every treatment condition.

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.

Only the first 6 of 12 groups are shown.

Overall Study

Analysis population for the primary outcome, as defined in the registry: All participants who complied sufficiently with the protocol and displayed an evaluable Pharmacokinetic (PK) profile (eg, exposure to treatment, availability of measurements, and absence of major protocol violations) were included in the PK set.

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: All participants who received at least one dose of the study drug(s) were included in the safety set.

Only the first 6 of 13 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Cmax: Maximum Observed Plasma Concentration for Soticlestat, measured over Day 1 pre-dose and at multiple time points (up to 96 hours) post-dose.

Maximum observed plasma concentration, or Cmax, is the highest measured concentration of a drug in plasma after a dose. It is a pharmacokinetic measure of peak exposure. In this study, the endpoint was measured in nanograms per millilitre (ng/ml) from Day 1 pre-dose through multiple post-dose time points up to 96 hours.

Cmax is useful when comparing how different formulations or administration conditions affect the concentration profile after dosing. For example, a tablet formulation, food, or crushing and mixing a tablet with applesauce could change the rate or extent of drug absorption. Cmax captures the observed peak, but it does not describe the entire concentration-time profile on its own. Other pharmacokinetic measures, such as total exposure over time, address different aspects of that profile; their results should not be inferred from Cmax.

The posted arm results are geometric means accompanied by geometric coefficients of variation. A geometric mean is commonly used for pharmacokinetic measurements because concentrations can be right-skewed and treatment comparisons are often expressed as ratios. The coefficient of variation describes relative dispersion on a percentage scale, rather than giving a range of individual participant values.

The endpoint is a measure of drug exposure, not a measure of clinical benefit. It also does not establish that two conditions are equivalent unless the analysis and decision criteria support that conclusion. Detailed definitions of the sampling schedule and the sponsor’s statistical model are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a geometric mean represent?

Pharmacokinetic concentrations are often analyzed on a logarithmic scale because their distributions may be skewed and because many treatment effects are naturally expressed as ratios. A geometric mean summarizes values after this log transformation and return to the original scale. It is not the same as an arithmetic average. Here, the posted Cmax values are geometric means in ng/ml.

A geometric coefficient of variation expresses dispersion relative to the geometric mean. Larger values indicate more relative variability in the measurements. It does not provide the minimum and maximum observed concentrations, and it should not be read as a confidence interval for the mean. In this study, the reported geometric coefficients of variation vary across treatment conditions, indicating that the observed spread differs by condition.

How should a geometric mean ratio be read?

A GMR expresses the relationship between the geometric means for two conditions. Under general statistical interpretation, a ratio near 100 on a percentage scale represents similar geometric means, while values farther from 100 indicate a larger relative separation. The registry reports GMRs and confidence intervals, but does not explicitly state the ratio orientation in the posted results. It is therefore appropriate to identify the treatment pair associated with each GMR without assigning a direction beyond what the record states.

The registry labels the two Part A comparisons as equivalence analyses. In general, an equivalence analysis asks whether the estimated difference is sufficiently small to lie within prespecified acceptable bounds. The confidence interval is assessed against those bounds. The bounds and the detailed analysis plan are not reported in the registry record. Consequently, the label alone does not show which margins were used or whether an independent reader can verify the sponsor’s equivalence assessment from the posted results.

What does the confidence interval contribute?

A confidence interval gives a range of values compatible with the data and the statistical model at its stated confidence level. It communicates precision: a wider interval reflects greater uncertainty in the estimated contrast than a narrower interval. The Part A and Part B comparisons are reported with 90% confidence intervals. These intervals are not a guarantee that the true value lies within their endpoints, and they do not account for unreported design choices, missing measurements, or possible protocol deviations.

The registry gives no p-values for the four reported comparisons. A confidence interval can provide information about the estimated contrast and its uncertainty without a p-value being displayed. For an equivalence question, the important interpretation depends on the interval and the prespecified margins, not on treating a conventional test of no difference as equivalent to an equivalence test.

Why does the crossover design matter?

In a crossover study, participants receive multiple conditions in a sequence. General advantages include comparing each participant with themselves, which can reduce the impact of stable differences between participants. The order of conditions can also matter. Period effects, carryover, and sequence effects are general concerns for crossover analyses, and study protocols commonly address them through design and analysis choices. The registry record does not report the washout details or how these effects were handled statistically, so no specific method should be assumed.

The reported arm-level geometric means are descriptive summaries. Pairwise GMR analyses are the more direct reported comparisons for the selected treatment pairs, but the registry does not name the statistical model or its covariates. The number of endpoint observations also varies across Part A conditions: 72 for Treatment A, 72 for Treatment B, and 71 for Treatment C. A reader should not treat those arm summaries as if they were independent parallel groups; the study is described as a crossover.

How should variability and precision be distinguished?

The geometric coefficient of variation describes dispersion among measurements, while the confidence interval describes uncertainty in an estimated comparison. These quantities answer different questions. A relatively large coefficient of variation can coexist with a ratio estimate whose confidence interval is informative, depending on the design and amount of data. The registry does not report individual-level data, so the posted summaries cannot show the full distribution or how within-participant correlation affected precision.

These statistical concepts are general explanations. For the study-specific analyses, the registry record does not report the full statistical analysis plan, model specification, handling of missing measurements, or equivalence margins. The reported numerical estimates should therefore be read as the posted sponsor analyses, with the limits of those details kept in view.

10. Analyses Reported by the Sponsor

The registry reports sponsor analyses for four Cmax comparisons: Treatment A with Treatment C and Treatment B in Part A, and Treatment D with Treatment E and Treatment F in Part B. The first two are labeled equivalence analyses and have reported GMRs of 97.28 and 81.04, each with a 90% confidence interval. The Part B GMRs are 37.48 and 89.45, also with 90% confidence intervals, and are labeled other analyses.

The registry record does not report a separate recalculated estimate alongside these sponsor analyses. It also does not report the detailed model, ratio orientation, or equivalence margins needed to recreate or fully audit the comparisons from the arm-level summaries. The posted geometric means and coefficients of variation are descriptive summaries; they are not a substitute for the sponsor’s crossover analysis.

No discrepancy between a sponsor result and a recalculation is identified in the registry record. Where a recalculation is available, it is an unadjusted calculation from posted summary statistics and does not reproduce a potentially model-based crossover analysis. In general, if such estimates disagree, the sponsor’s own analysis takes precedence for describing the trial’s reported result. Here, interpretation should remain anchored to the registry’s reported GMRs and their stated confidence intervals, without claiming an independent replication of the sponsor’s model.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary endpoint and one secondary outcome: the number of participants reporting one or more treatment-emergent adverse events. It also reports four pairwise Cmax analyses across the two study parts. Multiple comparisons raise a general statistical issue: examining several contrasts can increase the chance of findings that appear notable by chance unless the analysis plan specifies how multiplicity is handled. The registry record does not report whether any multiplicity adjustment was planned or applied, how the four comparisons were prioritized, or whether additional exploratory comparisons were conducted.

Missing-data handling is also important in pharmacokinetic studies because a participant may lack measurements needed to estimate an endpoint. The primary population was described as participants with sufficient protocol compliance and an evaluable pharmacokinetic profile, including treatment exposure, available measurements, and absence of major protocol violations. The record does not report how many participants were excluded from each analysis, why any measurements were unavailable, or how missing observations were handled statistically. The different analysis counts across Part A conditions are visible, but the reason for the difference is not reported in the registry record.

The registry record does not report a full statistical analysis plan, interim analyses, a sample-size justification, or the details of any adjustment for period, sequence, or carryover effects. These omissions do not prove that such procedures were absent from the study; they mean the methods are not reported in the registry record. Without those details, readers cannot fully assess prespecification, multiplicity control, or the sensitivity of the results to analytic decisions.

The listed primary pharmacokinetic population is a selected analysis population rather than necessarily every enrolled participant. The distinction matters because the endpoint estimates describe those with evaluable profiles under the stated criteria. The registry does not provide enough detail to assess how exclusions may have affected the comparisons.

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 4% reporting threshold.

Most frequent non-serious adverse event terms

The adverse-event results list no participants with serious adverse events in any of the six treatment conditions: Treatments A, B, and C in Part A and Treatments D, E, and F in Part B. The listed numbers at risk were 72 for each Part A condition and 24 for each Part B condition. No deaths are listed in any condition.

The counts for participants with other adverse events were 4 for Treatment A, 5 for Treatment B, and 0 for Treatment C in Part A; in Part B, the counts were 0 for Treatment D, 2 for Treatment E, and 1 for Treatment F. These are counts attached to treatment conditions in a crossover study, and they should not be added across conditions to infer a unique number of affected participants. A participant may have received multiple conditions.

Zero reported serious events in these groups means none were recorded in the posted results for the listed conditions. It does not establish that the probability of a serious event is zero, particularly in a study of this size and duration. The registry’s listed secondary outcome concerns participants reporting one or more treatment-emergent adverse events. Further detail about event types, timing, severity, relatedness, and how events were attributed across crossover periods is not reported in the registry record.

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

The posted results address how Cmax varied across selected soticlestat formulation and administration conditions in healthy adults. They provide geometric mean concentrations, measures of relative dispersion, and sponsor-reported GMRs with 90% confidence intervals. Part A compares tablet formulations under fasted conditions; Part B compares fasted, fed, and crushed-tablet conditions. The reported estimates therefore concern pharmacokinetic exposure under those study conditions.

These results do not answer whether soticlestat improves a health outcome. The trial was conducted in healthy adults and reports a pharmacokinetic endpoint, not a clinical efficacy outcome. The results should not be generalized automatically to people with a medical condition or to dosing situations not examined in the study.

The Part A analyses are labeled equivalence comparisons, but the registry does not report the margins or full statistical methods. The GMRs and intervals alone should not be treated as a complete independent demonstration of equivalence. Part B comparisons are labeled other analyses, and their ratio orientation is not stated in the registry record. The data can describe the reported estimates, but do not support assigning unreported methods or conclusions.

Arm-level means and variability are summaries, not descriptions of every participant’s concentration profile. The confidence intervals express statistical uncertainty under the underlying analysis, but do not resolve uncertainty from missing methods, missing measurements, or design details not reported in the registry record. Similarly, the absence of recorded serious events does not establish that no risk exists.

The appropriate conclusion is limited: the registry posts Cmax results for the tested treatment conditions and adverse-event counts for the listed groups. It does not provide the full protocol or analysis plan needed to answer every question about study conduct, precision, equivalence criteria, or the robustness of the analyses.

14. Why This Trial Matters Statistically

This trial is statistically useful because it illustrates a pharmacokinetic crossover design in which formulation and administration conditions are compared within a study rather than against an unrelated parallel control group. The design can make within-participant comparisons informative when stable individual differences would otherwise add variability. At the same time, crossover studies require attention to treatment order, period effects, and possible carryover; those details are not reported in the registry record.

The results also show why descriptive summaries and comparative analyses should be kept distinct. Geometric means and geometric coefficients of variation describe each listed treatment condition. GMRs with confidence intervals address selected pairwise contrasts. A reader should not infer the sponsor’s model or rebuild its analysis from arm summaries alone, especially when the study uses a crossover structure.

The registry labels two Part A contrasts as equivalence analyses. That label highlights the importance of prespecified margins: equivalence is not established merely because estimates appear similar or because a confidence interval includes a particular value. Since the margins and detailed methods are not reported in the registry record, the posted results cannot independently resolve that methodological question.

Finally, the study underscores the limits of safety interpretation from small treatment-condition groups and event counts. The absence of serious adverse events in the posted results is a descriptive finding, not proof of zero risk. In combination, these features make the record a useful example of how to separate exposure estimation, equivalence claims, design structure, and safety reporting.

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.

Several limitations are specific to interpretation of this trial’s registry results. First, the study involved healthy adults and focused on Cmax. That limits what can be concluded about pharmacokinetics in other populations and does not address clinical outcomes. Second, the crossover sequences are described only in part: the listed text is truncated, and details of washout duration and the handling of period, sequence, or carryover effects are not reported in the registry record.

Third, the sponsor’s reported GMRs are not accompanied by the full statistical model or a stated ratio orientation. The equivalence margins for the two Part A analyses are also not reported in the registry record. These omissions constrain independent assessment of the comparisons, even though the registry provides point estimates and 90% confidence intervals.

Fourth, the primary analysis population excludes participants without sufficient compliance or an evaluable pharmacokinetic profile, but the record does not give the number or reasons for exclusions by condition. The Part A endpoint counts are not identical across all conditions. Missing-measurement handling and sensitivity analyses are not reported in the registry record.

Fifth, adverse-event counts are associated with multiple treatment conditions in a crossover study. The record does not give enough detail to determine whether event attribution or participant overlap can be reconstructed from the posted counts. Event descriptions, timing, and detailed severity or relatedness information are not reported in the registry record.

Finally, the record does not report the sample-size justification, interim analyses, full statistical analysis plan, or multiplicity approach. These details may affect assessment of precision and prespecification, but they should not be inferred from the results alone.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of Soticlestat Tablets in Healthy Adults — NCT05284760.
• Registry record last updated 2026-09-10; status last verified 2026-08.
• Results first posted 2024-02-28.
• 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

NCT05284760 is a completed Phase 1, randomized, open-label crossover study sponsored by Takeda. It enrolled 96 healthy adults and examined soticlestat tablet formulations, food, and a crushed tablet mixed with applesauce. The primary endpoint was maximum observed plasma concentration, assessed from pre-dose through multiple time points up to 96 hours after dosing.

The registry reports geometric mean Cmax values and geometric coefficients of variation for three treatment conditions in each study part. It also reports four GMR comparisons with 90% confidence intervals. The two Part A comparisons are labeled equivalence analyses; the Part B comparisons are labeled other analyses. Serious adverse events and deaths are listed as zero across the six treatment conditions, while other adverse-event counts vary by condition.

Interpretation is limited by missing reporting on the statistical model, equivalence margins, analysis details, and several aspects of missing-data and crossover handling. The results describe pharmacokinetic measurements in the posted analysis population and do not establish clinical benefit.