A Study to Explore the Safety And Tolerability of Doses of E2007 Up to a Maximum of 8 mg In Patients With Parkinson's Disease Who Experience End-of-Dose Wearing Off Motor Fluctuations: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Explore the Safety And Tolerability of Doses of E2007 Up to a Maximum of 8 mg In Patients With Parkinson's Disease Who Experience End-of-Dose Wearing Off Motor Fluctuations (NCT00165789), a randomized phase 2 study. The primary outcome is Number of Participants With Any TEAE.

2. Clinical Question

The clinical question is therefore about safety and tolerability during treatment, rather than whether the intervention changes a motor-function outcome. The record lists secondary outcomes involving UPDRS scores and measures of “on” and “off” time, among other measures, but the primary endpoint is the occurrence of any treatment-emergent adverse event. The reported counts describe how many participants in each cohort had at least one such event. They do not, by themselves, establish whether the groups differ beyond what could arise from sampling variability. A direct comparative estimate and its uncertainty are not reported in the registry record.

3. Trial Design

This was a Phase 2 randomized, double-blind, placebo-controlled, parallel-group study. The registry describes two treatment groups and a planned allocation ratio of 3 to 1 between E2007 and placebo. Participants were to receive study treatment for ten weeks, from Days 1 to 70. The study is marked completed, with enrollment of 75. The registry lists the condition as Parkinson’s disease and characterizes the target population as patients experiencing end-of-dose wearing-off motor fluctuations.

The posted outcome counts retain two cohort labels. Cohort 1 includes 8 participants assigned to placebo and 20 assigned to the Perampanel-labeled arm. Cohort 2 includes 12 assigned to placebo and 35 assigned to the Perampanel-labeled arm. For the primary outcome, the respective event counts are 7, 18, 8, and 32. The record does not explain why results are presented in separate cohorts or state whether a pooled comparison was planned. The arm labels identify the active treatment as Perampanel, while the intervention listing names E2007; the registry record does not clarify this naming difference.

The primary analysis population is defined as the Safety Population: subjects who completed the Baseline Phase and received at least one dose of double-blind study medication. This definition is relevant because the primary outcome concerns events occurring after treatment begins. The record does not provide a separate definition for an efficacy population. It also does not report the detailed randomization procedure, concealment methods, dose escalation schedule, or protocol rationale for the cohort structure. Those design details should not be inferred from the broad randomized, double-blind, parallel-group description.

The registry lists secondary outcomes that include changes in UPDRS scores, “off” time, “on” time, dyskinesia measures, and Goetz/Rush score. The primary result remains the count of participants with any treatment-emergent adverse event through the end of study. No inferential analysis entries are 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: Safety Population was the primary population for analysis defined as all subjects who completed the Baseline Phase and who received at least 1 dose of double-blind study medication.

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 Participants With Any TEAE, measured over Through end of study.

The primary endpoint is the number of participants with any treatment-emergent adverse event through the end of the study. The registry defines a treatment-emergent adverse event as an adverse event that started on or after the first dose of study medication and continued to be recorded through the end of the study. The description says adverse-event information was recorded throughout the study after informed consent, including information from clinical laboratory tests, vital-sign measurements, and physical examinations. It also directs readers to the Adverse Event section for safety and tolerability information.

This endpoint is a count of participants who experienced at least one event, rather than a count of all event episodes. As a general statistical point, this kind of participant-level summary is useful for describing how many people in each group had an event during the observation period. It does not convey the number of events each participant experienced, the duration or severity of each event, or the timing of events. Those details require other outcome information.

The posted counts are 7 of 8 in Cohort 1 placebo, 18 of 20 in Cohort 1 Perampanel, 8 of 12 in Cohort 2 placebo, and 32 of 35 in Cohort 2 Perampanel. These are descriptive observations in the Safety Population. The registry record does not report a comparative estimate, confidence interval, or hypothesis test for this endpoint. The counts should therefore not be read as a statistically established treatment difference.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is binary at the participant level: each participant either had at least one treatment-emergent adverse event or did not, according to the registry’s definition. The results are expressed as counts of participants with an event and the group denominators. This format makes the observed group data easy to describe, but an observed count alone is not an estimate of how much treatment changes event risk relative to placebo.

What would a comparative analysis estimate?

In general, a binary-outcome comparison can be expressed using an absolute difference in event proportions or a relative measure comparing event proportions. These answer different questions: an absolute contrast describes separation on the probability scale, while a relative contrast describes the event frequency in one group in relation to another. Both require a defined comparison and a method for quantifying uncertainty. The registry record does not report either type of estimate for this endpoint. The posted counts therefore should not be converted into a claimed comparative effect.

A confidence interval is a statistical range intended to show the precision of an estimate under the assumptions of the analysis. A narrow interval indicates greater precision than a wide interval, though interpretation also depends on the estimand and the study design. No confidence interval for the primary comparison is reported. Likewise, no p-value or other inferential result is posted for this endpoint. It is not possible to judge from the registry entry whether a difference is compatible with a wide range of effects or is estimated precisely.

Why do the cohort labels matter?

The outcome is reported separately for Cohort 1 and Cohort 2. A pooled analysis might account for cohort explicitly, or it might combine data under other assumptions. The appropriate method depends on the design and analysis plan. The registry record does not describe how cohorts were intended to be handled in a comparative analysis, and it reports no analysis entries. Therefore, the cohort-specific counts should remain separate in interpretation rather than being combined through an unreported method.

The arm labels also warrant care. The intervention listing names E2007, while the active arms in the results are labeled Perampanel. The record does not explain this difference. The reported counts can be attributed to the Perampanel-labeled cohorts as recorded, but the naming discrepancy should not be resolved by assumption.

What does randomization contribute?

As a general statistical principle, random allocation aims to make assigned treatment groups comparable with respect to both measured and unmeasured baseline characteristics, on average across repeated trials. Double blinding is intended to limit the influence of treatment knowledge on participant experience, care, and assessment. These design features support the rationale for a treatment comparison, but they do not remove sampling variability, guarantee balance in a particular trial, or replace a prespecified analysis. The registry record does not report baseline balance statistics or the details of the randomization and masking procedures.

The study description states that allocation was planned in a 3 to 1 ratio. The posted cohort denominators are not presented as one pooled comparison, and the record does not state how the planned allocation was applied across cohorts. No explanation of the observed group sizes or the cohort structure is reported. Any interpretation beyond the stated design and posted counts would require information not available in the registry record.

How should the planning illustrations be read?

Planning illustrations are not observed treatment effects and are not “observed power.” The record describes such illustrations as using the observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect set as a stated fraction of the observed effect. These are assumptions for hypothetical planning calculations, not a description of the trial’s own design or its sample-size justification. The registry record does not report the trial’s own sample-size justification.

Equal allocation is a planning assumption for those illustrations, whereas the study description gives a planned 3 to 1 allocation. The methods and purpose differ: a planning calculation asks what sample size might be needed under specified assumptions, while the trial record describes the study that was conducted. Such illustrations do not show that the study had a particular probability of detecting an effect, and they do not alter the interpretation of the posted outcome counts. No numerical planning estimates are stated in this interpretation.

Why does the missing comparison limit inference?

The registry record explicitly indicates that values are missing or not estimable for a two-arm comparison of this binary endpoint. With no reported effect estimate, confidence interval, or test, readers cannot use the primary outcome entry to quantify the magnitude or precision of a group difference. The counts remain informative as descriptions of what was recorded in the four groups, but they are not a substitute for the missing analysis. The record does not report whether the absence reflects an analysis decision, data limitations, or another reason.

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("Cohort 1 - Placebo", "Cohort 1 - Perampanel", "Cohort 2 - Placebo", "Cohort 2 - Perampanel")
n      <- c(8, 20, 12, 35)
events <- c(7, 18, 8, 32)
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  Cohort 1 - Placebo  8  7  87.5  47.3  99.7  52.9  97.8  51.8  97.2
ARM  Cohort 1 - Perampanel  20  18  90.0  68.3  98.8  69.9  97.2  69.6  97.0
ARM  Cohort 2 - Placebo  12  8  66.7  34.9  90.1  39.1  86.2  38.6  86.1
ARM  Cohort 2 - Perampanel  35  32  91.4  76.9  98.2  77.6  97.0  77.5  96.9

11. Multiplicity, Missing Data and Analysis Populations

The registry lists secondary outcomes involving UPDRS scores, “off” time, “on” time, dyskinesia measures, and Goetz/Rush score. Ten secondary outcomes are posted. When a trial evaluates multiple outcomes, general statistical practice recognizes that repeated testing can increase the chance of finding at least one apparently notable result by chance. A prespecified multiplicity strategy can address this issue, for example by defining which outcomes are primary or by adjusting inference across tests. The multiplicity plan for this study is not reported in the registry record. No analysis entries are reported, so there are no posted secondary-outcome comparisons to interpret.

The registry defines the Safety Population for the primary outcome as participants who completed the Baseline Phase and received at least one dose of double-blind study medication. It does not report a separate efficacy analysis population. The record also does not describe how missing primary-outcome information was handled, whether follow-up was complete for all participants, or what assumptions were used if event status was unavailable. The listed cohort denominators provide the group sizes for the posted summaries, but do not by themselves establish how missing data were managed.

The record does not report an interim-analysis plan, stopping rules, or a statistical analysis plan for the primary endpoint. These details should not be inferred from the Phase 2 designation or from the study’s safety focus. Without a reported multiplicity approach, missing-data rules, and comparative analysis, the primary counts remain descriptive, and the statistical uncertainty around any between-group effect cannot be assessed from the registry entry.

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: fall (5.0%), hip fracture (5.0%), lumbar vertebral fracture (5.0%), pelvic fracture (5.0%), tibia fracture (5.0%).

Serious adverse events are reported separately from the primary endpoint of any treatment-emergent adverse event. The posted serious-event counts are 0 of 8 in Cohort 1 placebo, 4 of 20 in Cohort 1 Perampanel, 0 of 12 in Cohort 2 placebo, and 4 of 35 in Cohort 2 Perampanel. These figures count participants with a serious adverse event in each group. They should not be added to the any-event counts or treated as a measure of all adverse-event episodes.

The adverse-event record also lists serious-event terms and percentages, including FALL, HIP FRACTURE, LUMBAR VERTEBRAL FRACTURE, PELVIC FRACTURE, TIBIA FRACTURE, and ELECTROCARDIOGRAM ABNORMAL. The record does not provide enough context in this entry to interpret those term percentages as comparative risks across treatment groups. They should be read as listed event-term summaries, not as evidence of a causal relationship or a between-group effect.

Death counts are not reported in the registry record. Missing death information must not be interpreted as evidence that no deaths occurred. Serious-event counts describe the posted observations; without a comparative analysis and uncertainty estimates, they do not establish whether serious-event occurrence differed between groups.

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

The result establishes that the registry recorded treatment-emergent adverse events among participants in each of the four cohorts. The primary endpoint counts were 7 of 8 in Cohort 1 placebo, 18 of 20 in Cohort 1 Perampanel, 8 of 12 in Cohort 2 placebo, and 32 of 35 in Cohort 2 Perampanel. The endpoint concerns participants with at least one event from the first dose through the end of the study. It is a descriptive safety summary for the defined Safety Population.

The result does not establish a treatment effect. The registry record does not report a comparative estimate, confidence interval, or p-value for the binary endpoint and notes that values are missing or not estimable for a two-arm comparison. Without those results, the magnitude and precision of any difference cannot be evaluated. The counts alone cannot determine whether observed variation between groups reflects a treatment effect, chance, cohort characteristics, or other factors.

The result also does not describe the full burden of adverse events. A participant-level any-event count does not show how many events each person experienced, how long events lasted, their severity, or their timing. Serious adverse events are reported separately; neither the primary counts nor the serious-event counts establish causation. Deaths are not reported in the registry record.

The randomized, double-blind design is an important feature of the study, but it does not compensate for the lack of a posted comparative analysis. The planned allocation ratio, cohort-specific presentation, and difference between the E2007 intervention name and Perampanel arm labels are additional context that the record does not fully explain. No conclusions about benefit or harm beyond the recorded descriptive data are warranted from these results alone.

14. Why This Trial Matters Statistically

This trial is statistically useful as an example of the distinction between a registered study design and an interpretable treatment-effect estimate. The record describes randomization, double blinding, a placebo comparator, and parallel groups, but the primary binary outcome is posted only as cohort-specific counts. The presence of a trial design intended to support comparison does not ensure that a comparative result is available in the registry.

The outcome itself illustrates why endpoint definition matters. Counting participants with at least one treatment-emergent adverse event gives a clear participant-level safety summary. It does not capture event frequency per participant, severity, duration, or timing. A different endpoint would be needed to address those aspects of the safety profile. The separate serious-event counts serve a distinct descriptive purpose and should not be merged conceptually with the any-event endpoint.

The two-cohort reporting also draws attention to analysis structure. An overall contrast could require assumptions or a method that accounts for cohort. Since the registry record does not state how cohorts were to be handled and reports no inferential analysis entries, readers should not create an unreported pooled result. This restraint is a core part of statistical interpretation: numerical group counts are observations, while comparative estimates require explicit analytical methods.

Finally, the record distinguishes trial-specific design from hypothetical sample-size planning. Planning illustrations based on an observed effect and specified assumptions do not explain the original study’s sample-size justification and are not observed power. The registry record does not report that justification. Keeping these concepts separate prevents a planning exercise from being mistaken for evidence about the strength or certainty of the observed findings.

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 primary limitation for interpretation is the absence of an estimable two-arm comparison for the binary endpoint. The registry reports counts by cohort and treatment label but no effect estimate, confidence interval, or p-value. The reason that values are missing or not estimable is not reported in the registry record. As a result, the available outcome entry supports description but not a quantified inference about between-group differences.

The cohort structure is not explained in detail. The record does not say why the outcome is separated into Cohort 1 and Cohort 2, whether cohorts differed in treatment procedures or participant characteristics, or how they were intended to be handled analytically. The intervention is named E2007 in the intervention listing, while active arms are labeled Perampanel; the record does not reconcile those labels. These omissions limit interpretation of the group structure and terminology.

The outcome description defines treatment-emergent events and the Safety Population, but the registry record does not report missing-data rules, completeness of follow-up, or any sensitivity analyses. It does not report an analysis plan, multiplicity strategy, interim-analysis plan, or sample-size justification. The record also does not provide the details needed to interpret serious-event term percentages as comparative group risks. Deaths are not reported in the registry record.

Finally, the any-event endpoint is a broad summary. It does not give event-level information on severity, duration, timing, recurrence, or causal assessment. These limitations do not invalidate the posted counts, but they define what can be concluded from them. In particular, the registry findings should not be treated as a complete account of safety or as a statistical demonstration of a treatment difference.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study to Explore the Safety And Tolerability of Doses of E2007 Up to a Maximum of 8 mg In Patients With Parkinson's Disease Who Experience End-of-Dose Wearing Off Motor Fluctuations — NCT00165789.
• Registry record last updated 2015-06-02; status last verified 2015-04.
• Results first posted 2015-06-02.
• 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

The registry record describes a completed Phase 2 randomized, double-blind, placebo-controlled, parallel-group study in Parkinson’s disease among patients experiencing end-of-dose wearing-off motor fluctuations. Enrollment was 75. The study description states a planned 3 to 1 allocation and ten weeks of treatment, from Days 1 to 70.

The primary endpoint was the number of participants with any treatment-emergent adverse event through the end of study. The Safety Population included subjects who completed the Baseline Phase and received at least one dose of double-blind study medication. Results are reported separately for Cohort 1 and Cohort 2, with placebo and Perampanel-labeled arms.

The registry reports descriptive event counts but no inferential analysis entries. Values for a two-arm comparison of the binary endpoint are missing or not estimable. The record does not report a comparative estimate, confidence interval, p-value, missing-data approach, multiplicity plan, or sample-size justification. The intervention name and active-arm labels are not reconciled in the record.