Efficacy, Safety and Tolerability of E2007 in Levodopa Treated Parkinson's Disease Patients With Motor Fluctuations: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy, Safety and Tolerability of E2007 in Levodopa Treated Parkinson's Disease Patients With Motor Fluctuations (NCT00360308), a randomized phase 3 study. The primary outcome is Mean Change From Baseline in Total Daily OFF Time (Hours) to Week 18 (Including LOCF Data).

2. Clinical Question

The endpoint captures time during the day when participants recorded that medication had worn off and was no longer providing benefit with regard to stiffness, slowness and tremor. It was assessed from home diary entries. Participants recorded their state every 30 minutes during waking hours for three consecutive days immediately before selected visits. The primary outcome timeframe was baseline and Week 18, and its title specifies that the result includes LOCF data.

This is a three-arm, parallel-group, double-blind randomized trial. The registry lists an enrollment of 723 and a terminated status. The central statistical question is not whether the arms changed from baseline individually, but whether the changes differed between randomized groups. Arm-specific estimates describe each group’s change; a treatment comparison requires a contrast between groups and its uncertainty. Such a contrast is not reported in the primary results shown in the registry record.

3. Trial Design

The registry describes a phase 3, randomized, double-blind, parallel-group study of E2007, entacapone and placebo in levodopa-treated patients with Parkinson’s disease and motor fluctuations. The official title also describes the study as multicentre and double-dummy. Participants were allocated in a 1:1:1 manner after screening, according to the brief study summary. Equal allocation was intended across the three arms; the posted primary outcome counts differ between arms.

The E2007 arm description specifies 2 mg once daily in the evening during Weeks 0 to 2 and 4 mg once daily in the evening during Weeks 2 to 18. The entacapone arm received 200 mg with each dose of levodopa. The placebo arm received matching E2007 and matching entacapone. The summary says patients in the E2007 arm were titrated from 2 mg to 4 mg during the first 4 weeks, while the arm description places the change at Week 2. The registry record does not explain this difference, so the exact titration schedule cannot be resolved from these descriptions alone.

During the double-blind phase, the planned treatment period ended at Week 18. The summary lists visits at Weeks 2, 4, 6, 10, 14 and 18, with a follow-up visit at Week 22. Diary entries were collected at baseline and before visits at Weeks 6, 10, 18 and 22. UPDRS Parts I, II, III and IV were assessed at baseline and Weeks 10 and 18. At Week 18, participants were to complete efficacy and safety assessments and stop study medication, followed by the Week 22 visit.

The registry lists an enrollment of 723, a start in 2006-11 and primary completion in 2008-04. The status is terminated. The reason for termination is not reported in the registry record. Although the record lists the trial as randomized and double-blind, further details about the randomization sequence, allocation concealment and masking checks are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Intent-to-Treat (ITT) Population comprised all randomised patients who took at least one dose of study medication or placebo and who had a valid baseline efficacy measure and at least one post-baseline efficacy measure.

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 Mean Change From Baseline in Total Daily OFF Time (Hours) to Week 18 (Including LOCF Data), measured over Baseline and Week 18.

The primary endpoint is mean change from baseline in total daily OFF time, measured in hours, to Week 18, including LOCF data. OFF time is the diary-recorded period when medication had worn off and was no longer providing benefit with respect to stiffness, slowness and tremor. The complementary ON state is when medication is providing benefits to mobility, slowness and stiffness. Diary entries were made every 30 minutes during waking hours on three consecutive days before the specified assessment visits.

Daily OFF time is relevant to the trial’s stated population because participants were selected for levodopa-treated Parkinson’s disease with motor fluctuations. A continuous measure expressed in hours can describe the magnitude and direction of change, rather than only whether a participant crossed a threshold. A reduction in the measured OFF-time value corresponds to a negative change from baseline, as in the posted arm estimates. The endpoint therefore summarizes diary-recorded change over the assessment period, while remaining dependent on participants’ reports and the diary assessment schedule.

The registry reports least-squares mean changes and 95% confidence intervals for the arms. These are arm-level estimates, not direct estimates of the difference between treatments. A between-arm contrast would be needed to quantify how much the change differed between a treatment and its comparator, along with uncertainty for that contrast. That comparison is not reported in the primary results represented in the registry record. The endpoint title specifies LOCF data, but additional details of the missing-data procedure are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a least-squares mean represent?

A least-squares mean is a model-based estimate of a group mean. In general, such estimates can account for variables included in the statistical model and may differ from a raw average calculated directly from individual observations. The registry identifies the primary parameter as a least-squares mean, but the specific model, included covariates and handling of repeated measurements are not reported in the registry record. The posted estimates should therefore be described as model-based arm means without assuming how the model was specified.

How should the confidence intervals be read?

A confidence interval expresses uncertainty around an estimated quantity. In general, a narrower interval indicates greater precision than a wider interval, while the interval’s location reflects the range of values compatible with the estimate under the method used. The registry reports 95% confidence intervals around the arm-specific mean changes. These intervals concern each arm’s estimated change from baseline, not the difference between arms. A direct comparative interval would be needed to describe uncertainty around a treatment contrast.

The endpoints of the posted intervals are -1.16 and -0.48 hours for placebo, -1.63 and -0.96 hours for entacapone, and -1.27 and -0.57 hours for the arm labelled Perampanel. These limits are reported as registry values and are already rounded. They should not be read as participant-level ranges, nor as limits for a treatment difference. The registry does not report a p-value for the primary contrast in the results represented.

Why does the randomized comparison matter?

Randomization is intended to make treatment groups comparable on average at baseline, including with respect to factors that may affect outcomes. In general, a treatment effect is assessed by comparing outcomes across randomized groups, rather than by testing whether each group changed from its own baseline. A group can improve from baseline for reasons that are not specific to its assigned treatment. The arm estimates therefore provide descriptive context, while an estimated between-group contrast would address the comparative question more directly.

The record states that allocation was randomized in equal proportions across three arms and that the study was double-blind. The official title describes a double-dummy design, and the summary says matching dummy titrations were used to maintain masking for the entacapone and placebo arms. The randomization method, concealment process and details of how masking was maintained beyond those descriptions are not reported in the registry record.

What does the endpoint capture?

The endpoint is a continuous measure: total daily OFF time in hours, summarized as mean change from baseline to Week 18. Continuous outcomes retain information about the amount of change. In general, this can be more informative than reducing a measure to a simple category, although interpretation depends on measurement quality and the analysis method. Here, participants recorded their state every 30 minutes during waking hours on three consecutive days before selected visits. Diary-based measurement captures participants’ reported experience over sampled days; it is not a continuous device measurement of every day throughout follow-up.

The registry defines OFF as the time when medication had worn off and was no longer providing benefits with regard to stiffness, slowness and tremor. It defines ON as the time when medication was providing benefits to mobility, slowness and stiffness. Other diary categories included ON without dyskinesias, ON with non-troublesome dyskinesias, ON with troublesome dyskinesias and asleep. The primary endpoint focuses on total OFF time rather than combining all diary categories into a single score.

What does LOCF imply?

The primary endpoint title says that the result includes LOCF data. In general, last observation carried forward replaces a missing later measurement with the participant’s last observed measurement. This approach imposes an assumption that the carried-forward value is a suitable stand-in for the unobserved value. If participants’ outcomes would have changed after their last observation, that assumption may not hold; the direction and size of any effect on estimates depend on the pattern of missingness and the actual course of outcomes.

The registry record does not report the amount or pattern of missing outcome data, the exact occasions to which LOCF was applied, or sensitivity analyses using alternative assumptions. It also does not describe whether other missing-data methods were used. Consequently, the potential influence of missingness on the posted estimates cannot be assessed fully from this record.

Why can over-dispersion matter for count outcomes?

Over-dispersion is a general issue in count data: observed variability can exceed what a simple count model assumes. When that occurs, a model that fails to account for the extra variability may give confidence intervals or tests that are too optimistic. This concept is not the main analysis issue for the primary endpoint, which is continuous, and the registry record does not report an over-dispersion assessment for this outcome. It is included as a general statistical consideration rather than a claim about the trial’s analysis.

The record identifies a continuous primary outcome, arm-specific least-squares means and 95% confidence intervals. It does not report the model formula, assumptions checks, exact contrast specification, test statistic, p-value or treatment-difference estimate. Those details would be needed to reproduce or fully audit the primary inferential analysis. The posted numbers support interpretation of the estimated changes within arms and their reported precision, but not reconstruction of a complete comparative test.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures, including the primary outcome and three secondary outcomes. The secondary measures concern change in UPDRS Part II score, UPDRS Part III score in the ON state, and total daily ON time without dyskinesias or with non-troublesome dyskinesias. The record reports three secondary outcomes as posted. It does not provide the results of those secondary outcomes in the registry record described for this analysis, so their estimates should not be inferred from the primary outcome.

When several outcomes or treatment comparisons are tested, the chance of at least one false-positive finding can increase. General approaches include prespecifying a hierarchy, adjusting significance thresholds or interpreting findings as exploratory. The registry record does not report a multiplicity adjustment, testing hierarchy or other strategy for handling multiple outcomes or comparisons. It also does not report the statistical analysis plan, so the planned order and formal status of secondary analyses cannot be determined from the record.

The primary outcome title states that Week 18 results include LOCF data. The exact implementation, number of participants with missing values, reasons for missingness and sensitivity analyses are not reported in the registry record. The record defines the ITT population for efficacy as randomized patients who took at least one dose and had a valid baseline measure and at least one post-baseline efficacy measure. This is the registry’s stated efficacy population definition; it should not be assumed to be identical to every other convention called intention-to-treat.

Outcome counts are 228 for placebo, 221 for entacapone and 213 for the arm labelled Perampanel. These counts describe the posted primary outcome and are not evidence that every randomized participant had complete follow-up. The record does not report the number excluded from the efficacy population, the reasons for exclusion or the extent to which participants contributed observations at each assessment.

11. 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: cellulitis (0.4%), femoral neck fracture (0.4%), back injury (0.4%), head injury (0.4%), transient ischaemic attack (0.4%).

The safety record reports serious adverse events for 6 of 247 participants in the placebo group, 8 of 234 in the entacapone group and 11 of 242 in the group labelled Perampanel. These are counts of affected participants alongside the respective numbers at risk in the safety record. They should be read as group-specific counts, not as the number of distinct serious-event episodes. The registry record does not give a comparative statistical analysis of these figures.

Other adverse events are reported separately: 39 of 247 participants in the placebo group, 63 of 234 in the entacapone group and 66 of 242 in the group labelled Perampanel were affected. The record distinguishes these from serious adverse events; the two categories should not be treated as interchangeable. It does not report deaths in the safety data represented here, so death counts are not reported in the registry record.

The listed serious-event terms are Cellulitis, Femoral Neck Fracture, Back Injury, Head Injury, Transient Ischaemic Attack and Angina Unstable, each reported as 0.4%. The record does not specify in the presented information which arm or arms correspond to each listed term. These terms and percentages are not a complete description of all safety outcomes, and they do not by themselves establish a comparative safety profile. Differences in counts should be interpreted in light of the group denominators, event definitions, follow-up and the absence of a reported comparative analysis.

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

The primary outcome estimates describe mean change in diary-recorded daily OFF time from baseline to Week 18. The posted changes are -0.82 hours for placebo, -1.29 hours for entacapone and -0.92 hours for the E2007 arm labelled Perampanel. Their corresponding 95% confidence intervals describe uncertainty around each arm-specific estimate. These values show that the estimated mean change was negative in each group on the stated measure.

That pattern does not answer whether a treatment was better than another. A within-group change is not a randomized treatment contrast. For a comparative conclusion, readers would need the estimated difference between groups, its confidence interval and the analysis used to test or estimate that difference. The registry record does not report those comparative results in the primary outcome information. Nor does it report a primary-comparison p-value. It is therefore not possible to determine from these arm-specific results alone the size, precision or statistical significance of the differences between treatments.

The confidence intervals also have a specific scope: they quantify uncertainty around the arm means, not the range of changes among individual participants and not the interval for a treatment difference. They cannot establish how much participants varied, how many experienced a particular amount of change, or whether a difference would be important to patients. No threshold for a clinically meaningful change is reported in the registry record.

The result is based on a diary outcome with LOCF data. The registry record does not report the extent of missingness or sensitivity analyses, which limits assessment of how missing observations may have affected the estimates. The least-squares means are model-based, but model details are not reported. These omissions make it difficult to assess the analysis assumptions independently or reproduce the estimates.

The safety counts are descriptive and use their own group denominators. They do not provide a statistical comparison or account by themselves for event duration, severity beyond the serious-event classification, or individual circumstances. Taken together, the posted efficacy and safety figures describe selected trial outcomes. They do not, without comparative estimates and fuller analysis details, support a definitive statement about treatment effects or comparative safety.

13. Why This Trial Matters Statistically

This trial raises several useful statistical issues for clinical researchers. First, it distinguishes change within a group from the effect of assignment to a treatment. The posted primary results are arm-specific changes, while the clinical question is comparative. Keeping those quantities separate helps prevent a common interpretive error: treating improvement from baseline as proof that one randomized treatment outperformed another.

Second, the endpoint depends on repeated participant diary entries. The measurement represents reported daily OFF time on sampled days before specified visits, rather than an uninterrupted record of every day in the treatment period. That design makes measurement procedures, adherence and missing entries relevant to interpretation. The registry specifies the diary schedule and includes LOCF in the endpoint title, but does not report the degree of missingness or sensitivity analyses.

Third, least-squares means and confidence intervals are useful only when their estimand and model are understood. The registry identifies the summary measure but does not report the model specification or direct treatment contrasts. This illustrates why numerical estimates alone may not convey the full inferential result: the contrast, its uncertainty and the assumptions behind the analysis are needed to answer the comparative question.

Finally, the record includes multiple efficacy outcomes and descriptive safety counts. Without a reported multiplicity strategy, analysis plan and comparative safety analysis, readers should distinguish the posted measures from a complete account of prespecified inference. The trial’s randomized, double-blind, parallel design provides an important basis for comparing assigned groups, but the interpretive value of that design depends on transparent reporting of the relevant contrasts and methods.

14. 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 most important limitation of the posted primary result is that it reports arm-specific mean changes rather than a treatment contrast. As a result, the estimates and confidence intervals do not establish the comparative effect of E2007 or entacapone versus placebo, or of either active treatment versus the other. The record does not report a primary comparison estimate or p-value.

Several analysis details are also absent. The model used to estimate least-squares means, its covariates and assumptions, and the handling of repeated observations are not reported in the registry record. Although the endpoint title says the result includes LOCF data, the implementation, missing-data extent and sensitivity analyses are not described. The registry also does not report the multiplicity strategy for the primary and secondary outcomes.

The trial descriptions are not fully consistent about E2007 titration. The arm description gives 2 mg during Weeks 0 to 2 and 4 mg during Weeks 2 to 18, whereas the brief summary describes titration during the first 4 weeks. The registry record does not resolve this difference. The status is terminated, but the reason for termination is not reported in the registry record.

The diary endpoint relies on participant reports collected on specified days, and the record does not describe completeness or adherence to diary completion. The displayed adverse-event counts are descriptive; deaths are not reported in the registry record, and the listed serious-event terms are not assigned to specific arms in the information available. These limitations constrain interpretation of both efficacy and safety summaries.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Efficacy, Safety and Tolerability of E2007 in Levodopa Treated Parkinson's Disease Patients With Motor Fluctuations — NCT00360308.
• Registry record last updated 2014-07-11; status last verified 2013-08.
• Results first posted 2012-11-22.
• 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.

18. Record Summary

NCT00360308 is a phase 3 randomized, double-blind, parallel-group study in levodopa-treated patients with Parkinson’s disease and motor fluctuations. The registry lists an enrollment of 723 and a terminated status. Participants were assigned to E2007, entacapone or placebo; the listed primary endpoint is mean change in total daily OFF time from baseline to Week 18, including LOCF data.

The posted arm-specific least-squares mean changes were -0.82 hours for placebo, -1.29 hours for entacapone and -0.92 hours for the arm labelled Perampanel, with 95% confidence intervals reported for each. The registry record represented does not report a direct treatment contrast or its p-value. It lists three secondary outcomes and describes safety counts by group. Missing-data details, the statistical model, multiplicity handling and the reason for termination are not reported in the registry record.