This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Perampanel in Seizure Patients With Primary Glial Brain Tumors (NCT02363933), a na phase 4 study. The primary outcome is Percentage of Patients With ≥50% Seizure Reduction During the Maintenance Period Compared With Seizure Frequency Before Initiation of Perampanel.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02363933 |
| Phase | Phase 4 |
| Condition | Brain Tumor, Primary |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 9 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Duke University |
| Results posted | Yes |
2. Clinical Question
Population
Patients with primary glioma presenting refractory partial onset seizure activity. The study description defines refractory activity as 3 or more seizures in a 28-day period. The primary outcome description notes that seizure frequency was not systematically gathered at screening, which prevented calculation of the planned responder rate.
Intervention
Perampanel added to the participants’ current anti-epileptic drug (AED) treatment. The study description gives a starting dose of 2 mg daily, with planned titration toward 8 mg daily if tolerated, followed by a maintenance period. Dose modification was permitted according to reported tolerability.
Comparator
There was no separate comparison group. The study is recorded as single-group, with allocation listed as not applicable and masking listed as none. The primary outcome was framed as a comparison with seizure frequency before perampanel initiation, rather than a comparison between randomized groups.
Primary question
What percentage of patients had at least a 50% reduction in seizure frequency during the maintenance period compared with seizure frequency before starting perampanel? The registry record states that this percentage could not be computed because baseline seizure frequency was not systematically gathered at screening.
The clinical question is therefore stated, but the record does not provide a calculable primary responder result. The planned comparison depended on having an adequate baseline measure for each participant. Without systematically collected baseline seizure frequency, the registry record cannot establish which participants met the responder definition or what proportion met it.
3. Trial Design
The registry record describes a completed, single-group study sponsored by Duke University. Enrollment is listed as 9. The study is open label, with masking listed as none, and allocation listed as not applicable. The intervention arm combined perampanel with participants’ current anti-epileptic drug treatment; there is no separate control arm for a like-for-like comparison.
The title and study description identify the study as Phase II, while the registry phase entry lists Phase 4. The record does not explain this difference. The study dates are listed as 2015-04-07 for start and 2017-03-14 for primary completion, and the status is completed.
The described regimen began with perampanel at 2 mg daily for 2 weeks, followed by titration in 2 mg increments per week toward a planned dose of 8 mg daily if tolerated. The arm description says participants received perampanel with their current AED for a total of 20 weeks and describes a maintenance dose through 16 weeks. The brief description also says treatment assessments were planned at enrollment, 8 weeks, 16 weeks, and 24 weeks. Some of the regimen description is incomplete in the registry record, so the remaining taper details are not reported in the registry record.
Study assessments were described as including seizure history, physical and neurological examinations, Karnofsky Performance Status, patient-reported quality of life, and computer-based neurocognitive testing. The record does not report a randomized design, masking, or an external control group. Consequently, changes over time would not by themselves distinguish the effect of perampanel from other influences, including the background treatment and the course of illness.
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: 1 subject decided not to participate-had no more seizures after enrolling/while waiting to start
Overall Study
| Stage | Perampanel + Current Anti-Epileptic Drug |
|---|---|
| STARTED | 8 |
| COMPLETED | 2 |
| NOT COMPLETED | 6 |
Analysis population for the primary outcome, as defined in the registry: Seizure frequency was not systematically gathered at screening. As the 50% responder rate requires an adequate baseline comparison, it is not possible to compute the 50% responder rate.
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.
| Characteristic | Unit | Perampanel + Current Anti-Epileptic Drug |
|---|---|---|
| Participants analysed | 8 | |
| Age, Categorical: <=18 years | Participants | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 8 |
| Age, Categorical: >=65 years | Participants | 0 |
| Age, Continuous | years | 45.0 (SD 9.1) |
| Sex: Female, Male: Female | Participants | 2 |
| Sex: Female, Male: Male | Participants | 6 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 0 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 8 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 |
| Race (NIH/OMB): Asian | Participants | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 |
| Race (NIH/OMB): Black or African American | Participants | 0 |
| Race (NIH/OMB): White | Participants | 8 |
| Race (NIH/OMB): More than one race | Participants | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 |
| Region of Enrollment: United States | Participants | 8 |
6. Primary Endpoint
The registry lists the primary outcome as Percentage of Patients With ≥50% Seizure Reduction During the Maintenance Period Compared With Seizure Frequency Before Initiation of Perampanel, measured over 20 Weeks.
| Element | Registry entry |
|---|---|
| Result type | N/A |
| Measure of spread | not applicable |
| Unit | not stated |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was the percentage of patients with at least a 50% reduction in seizure frequency during the maintenance period compared with seizure frequency before perampanel was started. This is a responder-rate endpoint: it classifies each participant according to whether the specified reduction threshold was reached, then expresses the result as a percentage of the relevant group. The registry record lists a timeframe of 20 Weeks.
A responder definition is one way to translate changes in a count outcome into a result that is easier to interpret at the participant level. It asks whether an individual crossed a prespecified threshold, rather than only describing an average change across all participants. That threshold does not capture every possible pattern of change, and the percentage depends on having comparable seizure-frequency information for the baseline and maintenance periods.
In this study, the registry record says seizure frequency was not systematically gathered at screening. It further states that an adequate baseline comparison was required and that the 50% responder rate could not be computed. Thus the endpoint is described, but a primary responder percentage is not reported in the registry record. The record does not report a substitute primary analysis that resolves the missing baseline comparison.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The central interpretive point is that the primary outcome was not calculable from the information recorded. The study intended to measure the percentage of participants whose seizure frequency fell by at least 50% during maintenance compared with frequency before perampanel initiation. However, the registry record explicitly says screening seizure frequency was not systematically gathered and that this made the responder rate impossible to compute. No primary responder percentage is reported.
Enrollment is listed as 9, while the safety group lists 8 participants at risk. These are distinct record entries and should not be treated as interchangeable denominators. The registry record does not explain the difference. It also does not report a primary analysis population or explain how participants with missing assessments, incomplete treatment, or other deviations were handled.
The single-group design means that there is no concurrent control group to account for changes that might occur over time or alongside current AED treatment. Even if a within-participant change had been available, this design alone would not establish that perampanel caused that change. The absent baseline measure is an additional obstacle: the planned responder definition specifically requires comparison with pretreatment seizure frequency.
Safety information is recorded separately. The adverse-event summary lists 8 participants at risk for serious adverse events and 1 affected, and lists 8 at risk for other adverse events and 8 affected. It also lists 1 death. The record does not provide enough detail in these figures to determine the cause of the death, establish whether an event was attributable to perampanel, or compare safety with an untreated or alternative-treatment group. A secondary outcome concerns events possibly, probably, or definitely attributable to perampanel, but its result is not reported in the registry record.
The record lists two outcome measures and no analyses. It therefore does not provide inferential estimates, confidence intervals, or a reported statistical test for the primary endpoint. This is not evidence that the planned responder criterion was or was not met; it means the registry record does not provide a computable result. Any planning illustrations should also be distinguished from observed evidence: sample-size calculations are planning exercises, not observed power, and cannot replace an unavailable endpoint result.
Overall, the record supports a description of the intended question, study structure, regimen, and posted safety counts. It does not support a quantitative conclusion about the primary seizure-reduction endpoint. That distinction matters: a well-defined endpoint and a completed study do not ensure that the data needed to estimate the endpoint were collected or reported.
9. Statistical Methods Explained
The statistical interpretation starts with the endpoint’s data requirements. The primary outcome is a responder rate based on each participant’s seizure frequency during maintenance relative to seizure frequency before treatment. To classify a participant, both periods must have usable and sufficiently comparable frequency measurements. The registry record says screening seizure frequency was not systematically gathered and explicitly states that the responder rate cannot be computed. This is a limitation in the information needed for the endpoint, not a finding that participants did or did not respond.
What does a responder rate represent?
In general, a responder rate is the proportion of an analysis group meeting a defined criterion. Here, the criterion was at least a 50% reduction in seizure frequency during maintenance compared with the pretreatment frequency. The cutoff turns a change in a count into a yes-or-no classification for each participant. The resulting percentage would describe how many met that definition, but it would not show the full range of individual seizure-frequency changes. It would also depend on the baseline measure being collected consistently enough to make the comparison.
Why does the single-group design matter?
In general, a concurrent control group helps separate changes associated with an intervention from changes that could occur for other reasons. This study is recorded as single-group, with allocation not applicable and masking none. The planned primary comparison was with participants’ own pretreatment seizure frequency, not with a separate control group. A before-and-after comparison can describe change when measurements are available, but without a control group it cannot by itself distinguish an intervention effect from time trends, other treatment, or other influences. In this record, the baseline measurement problem prevents the stated responder comparison as well.
What can the sample-size illustrations establish?
In general, sample-size calculations describe how many participants might be needed under chosen assumptions to detect a specified effect with a selected statistical approach. The registry record characterizes the planning illustrations as based on the observed effect and specifies a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It also distinguishes participant counts per arm from total event counts for time-to-event calculations. These are planning assumptions, not evidence that the trial had a particular probability of detecting an effect. They are not observed power.
The trial itself is not recorded as a two-arm study, so an equal-allocation, two-arm planning setup should not be mistaken for its design. The primary responder rate is stated as not computable, and no numeric primary effect estimate is reported in the registry record. Accordingly, the planning illustrations do not supply a substitute for an observed primary result. The registry record does not report the original sample-size justification or explain how the planning assumptions relate to the study’s feasibility objective.
How should uncertainty be understood?
In general, a confidence interval gives a range of values compatible with the data and a stated statistical procedure, conveying the precision of an estimate. A narrow interval generally indicates greater precision than a wide interval, while neither an interval nor a point estimate alone resolves design limitations. No primary responder estimate or confidence interval is reported in the registry record. Without a computable estimate, the precision of the intended primary result cannot be evaluated.
In general, statistical tests assess how compatible observed data are with a specified null model under the test’s assumptions. A p-value is not the probability that a treatment works, and it does not measure clinical importance. The registry record lists no analyses and reports no primary p-value. It is therefore not possible to describe a reported statistical test for the responder endpoint.
What do the adverse-event counts describe?
The posted safety figures describe counts of participants at risk and affected. They are descriptive safety summaries, not randomized comparisons. In general, the number affected provides context for an event summary, but interpretation also depends on definitions, observation periods, event-level detail, and attribution. The record lists serious adverse-event terms and percentages, but those listings do not establish that perampanel caused the events. The secondary outcome about events considered possibly, probably, or definitely attributable to treatment has no result reported in the registry record.
The record lists 9 enrolled and 8 at risk in the adverse-event group. Statistical interpretation should preserve those separate figures rather than assume that every enrolled participant contributed to every safety summary. The reason for the difference, the safety analysis population, and handling of missing safety information are not reported in the registry record.
10. Multiplicity, Missing Data and Analysis Populations
The registry record lists two outcome measures: the primary seizure-reduction responder rate and a secondary outcome concerning adverse events possibly, probably, or definitely attributable to perampanel. It lists no analyses. A result for the primary endpoint and a result for the stated secondary outcome are not reported in the registry record.
Multiplicity refers, in general, to the increased chance of finding an apparently notable result when several statistical questions or tests are considered. A statistical plan may specify which endpoint is primary and how multiple comparisons are handled. The registry record does not report a multiplicity adjustment, a testing hierarchy, or whether any adjustment was planned. With no analyses reported, it is not possible to determine how multiplicity was handled in practice.
The primary endpoint requires a baseline seizure-frequency measure. The record states that seizure frequency was not systematically gathered at screening and that the responder rate could not be computed. It does not report an alternative method for addressing missing baseline seizure-frequency data, an imputation approach, or a sensitivity analysis. It also does not describe missingness in other assessments.
The record does not report the analysis populations used for efficacy or safety, or how participants who did not complete treatment or assessments were handled. Enrollment is listed as 9 and the adverse-event group lists 8 at risk; the reason for that difference is not reported in the registry record. These omissions limit assessment of who contributed to each analysis and how robust any result would be to missing information. The planned analysis details, including any interim analyses, are not reported in the registry record.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Perampanel + Current Anti-Epileptic Drug | 1 | 8 | 12.5% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Perampanel + Current Anti-Epileptic Drug | 1 of 8 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Perampanel + Current Anti-Epileptic Drug | 8 of 8 |
Most frequent serious adverse event terms
| Event term | Organ system | Perampanel + Current Anti-Epileptic Drug |
|---|---|---|
| Multi-organ failure | General disorders | 1 of 8 |
| Lung infection | Infections and infestations | 1 of 8 |
| Fall | Injury, poisoning and procedural complications | 1 of 8 |
| Hypoxia | Respiratory, thoracic and mediastinal disorders | 1 of 8 |
| Respiratory failure | Respiratory, thoracic and mediastinal disorders | 1 of 8 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Perampanel + Current Anti-Epileptic Drug |
|---|---|---|
| Fatigue | General disorders | 6 of 8 |
| Platelet count decreased | Investigations | 3 of 8 |
| Hyperglycemia | Metabolism and nutrition disorders | 3 of 8 |
| Nausea | Gastrointestinal disorders | 2 of 8 |
| Mucosal infection | Infections and infestations | 2 of 8 |
| Creatinine increased | Investigations | 2 of 8 |
| Hypokalemia | Metabolism and nutrition disorders | 2 of 8 |
| Dizziness | Nervous system disorders | 2 of 8 |
| Nervous system disorders - Other, specify | Nervous system disorders | 2 of 8 |
| Pyramidal tract syndrome | Nervous system disorders | 2 of 8 |
Serious events with the highest arm-level rate: multi-organ failure (12.5%), lung infection (12.5%), fall (12.5%), hypoxia (12.5%), respiratory failure (12.5%).
The serious adverse-event summary lists 8 participants at risk and 1 affected. The record also lists 1 death. These figures describe the recorded safety group; they do not establish the cause of an event or whether it was related to perampanel. The registry record does not report the cause of death in the information, nor does it provide a treatment-versus-control comparison.
The listed serious-event terms are Multi-organ failure, Lung infection, Fall, Hypoxia, and Respiratory failure, each accompanied by 12.5%. These are registry-listed terms and percentages; they should not be read as evidence that perampanel caused those events. The record does not provide a detailed event-level account or explain the basis for attribution in this summary.
For other adverse events, the group lists 8 at risk and 8 affected. The secondary outcome was the percentage of patients with an adverse event possibly, probably, or definitely attributable to perampanel, but a result for that outcome is not reported in the registry record. The distinction between an event being recorded and an event being attributed to treatment is important: the first is a safety observation, while the second requires an attribution assessment. The posted counts alone do not answer the attribution question.
12. What the Result Does, and Does Not, Mean
The record establishes what the study set out to measure: the percentage of patients with at least a 50% reduction in seizure frequency during the maintenance period compared with frequency before perampanel initiation. It also describes a single-group regimen adding perampanel to current AED treatment and reports safety-group counts. These are useful descriptions of the study question and what was recorded.
The primary efficacy result, however, is not available. The registry record says screening seizure frequency was not systematically gathered, making it impossible to compute the stated responder rate. No substitute estimate, confidence interval, or primary statistical test is reported. The absence of a computable endpoint is not a zero responder rate, and it does not show that no participant had a reduction. It means that the record does not support calculation of the planned percentage.
The design also limits what any before-and-after change could establish. There was no separate control group, and masking was listed as none. Even with complete baseline and maintenance measurements, a single-group comparison would describe change within the group but would not isolate the contribution of perampanel from other influences. The registry record does not report an adjusted analysis that addresses this limitation.
The safety counts should likewise be read descriptively. The record lists serious adverse events, other adverse events, and a death, but the counts do not establish causation. A secondary outcome intended to capture events attributed to perampanel has no reported result. The available safety information therefore does not provide a comparative estimate of treatment-related risk.
In general, statistical evidence depends on the connection between a clearly defined question, appropriate measurements, and an analysis that is reported with enough detail to assess uncertainty. This record defines a responder threshold but reports that a required baseline measure was not systematically collected. Consequently, the main efficacy question remains unanswered by the posted primary outcome data. The record should not be interpreted as demonstrating either benefit or lack of benefit.
13. Why This Trial Matters Statistically
This study is statistically informative because it illustrates how a clearly stated endpoint can still become unestimable when a required measurement is not collected systematically. The planned primary outcome was specific: a responder had to meet the at-least-50% seizure-reduction criterion during maintenance relative to pretreatment frequency. The registry record says the baseline frequency data were not systematically gathered and therefore the rate could not be computed. Endpoint definitions need to be matched by data collection that supports the intended comparison.
The case also highlights the difference between a within-participant comparison and a controlled comparison. A single-group design can describe what happened to measured participants over time, but it does not provide a concurrent estimate of what would have happened under a different treatment strategy. The lack of masking is also part of the design context. These features do not automatically invalidate descriptive observations, but they limit causal interpretation.
Denominators matter as well. Enrollment is listed as 9, whereas the adverse-event group lists 8 participants at risk. The registry record does not explain the difference. Keeping the denominators distinct avoids treating the enrollment total as though it necessarily applies to every outcome or safety summary.
Finally, the record distinguishes planned outcomes from reported analyses. It lists two outcome measures but no analyses, and it does not report a primary estimate or uncertainty measure. Statistical interpretation must therefore avoid filling gaps with assumptions. This is a useful example of why registry descriptions of a question, a design, and a result are separate pieces of evidence: a completed study and a prespecified outcome do not, on their own, provide an interpretable estimate.
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 direct limitation is the missing systematic baseline seizure-frequency measurement. The primary endpoint required comparison with pretreatment frequency, and the registry record states that this prevented computation of the responder rate. No alternative primary endpoint result is reported.
The study had a single group and no masking. There was no separate randomized comparator to provide a concurrent reference for changes over time. The record therefore cannot support a direct between-treatment comparison, and the design would constrain causal interpretation even if the planned within-participant endpoint were available.
The registry lists 9 enrolled participants but 8 at risk in the adverse-event group. The reason for this difference is not reported in the registry record. The analysis populations, missing-data methods, treatment completion details, and handling of incomplete assessments are also not reported. These gaps make it difficult to determine exactly who contributed to each outcome.
The record lists a secondary adverse-event outcome focused on attribution, but does not report its result. It also lists serious-event terms and a death without enough information in the summary to determine their causes or relationship to treatment. The primary statistical analysis, confidence intervals, p-values, and multiplicity procedures are not reported in the registry record.
There is a further inconsistency in the study phase description: the title and narrative describe Phase II, while the registry phase entry lists Phase 4. The record does not explain the difference. Some treatment-description text is incomplete, including the taper details, which are not reported in the registry record. These limitations affect interpretation of the registry record and do not establish what may have been documented elsewhere.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Perampanel in Seizure Patients With Primary Glial Brain Tumors — NCT02363933.
- Registry record last updated 2018-04-27; status last verified 2018-03.
- Results first posted 2018-04-02.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
The registry record describes a completed, open-label, single-group study of perampanel added to current AED treatment in patients with primary glioma and refractory partial onset seizures. Enrollment is listed as 9. The planned primary endpoint was the percentage with at least a 50% seizure-frequency reduction during maintenance compared with pretreatment frequency.
The primary responder rate could not be computed because seizure frequency was not systematically gathered at screening. The record lists two outcome measures and no analyses. The safety group lists 8 participants at risk for serious adverse events, 1 affected, 8 at risk for other adverse events, 8 affected, and 1 death. These figures do not establish treatment attribution. No primary efficacy estimate or result for the stated attribution outcome is reported in the registry record.