Efficacy, Safety, Tolerability, and Pharmacokinetics of NBI-827104 in Pediatric Subjects With Epileptic Encephalopathy With Continuous Spike-and-Wave During Sleep: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy, Safety, Tolerability, and Pharmacokinetics of NBI-827104 in Pediatric Subjects With Epileptic Encephalopathy With Continuous Spike-and-Wave During Sleep (NCT04625101), a randomized phase 2 study. The primary outcome is Ratio of Spike-Wave Index (SWI) During First Hour of Nonrapid Eye Movement (NREM) Sleep at Week 6.

2. Clinical Question

The endpoint compares the Week 6 SWI value with baseline. SWI was defined as the percentage of seconds with at least 1 spike-wave complex during defined periods of overnight NREM sleep, using centralized video-EEG reading. The registry record identifies this as a phase 2 efficacy assessment, alongside safety, tolerability, and pharmacokinetic aims. It also lists caregiver and clinician global-impression responder outcomes and a clinical global-impression severity outcome as secondary measures.

3. Trial Design

This was a randomized, triple-masked, parallel-group phase 2 study of NBI-827104 and placebo. Its official title describes a double-blind, placebo-controlled design. The record lists an enrollment of 24 and identifies Neurocrine Biosciences as the sponsor. The study started on 2021-04-26, reached primary completion on 2022-08-08, and is recorded as completed.

Both treatments were administered orally for 13 weeks. The brief summary specifies once-daily administration of NBI-827104, while the arm descriptions state that both NBI-827104 and placebo were administered orally for 13 weeks. The primary endpoint was assessed from baseline to Week 6, so the primary measurement point falls within the treatment period.

The primary analysis population is described as the Full Analysis Set: randomized participants who had at least 1 dose of study treatment and at least 1 efficacy video EEG assessment. The record says the number analyzed for this outcome equals the number evaluable for it. It reports 14 analyzed for the NBI-827104 estimate and 6 for the placebo estimate. These endpoint counts should not be assumed to describe every randomized participant or every safety analysis; the record gives separate safety denominators.

The record reports a single ANCOVA analysis for the primary comparison. It does not report the model covariates, the allocation ratio, the randomization procedure, or a detailed statistical analysis plan. The record also does not report the protocol’s sample-size justification or whether interim analyses were planned or conducted.

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: Full Analysis Set included all randomized participants who had at least 1 dose of study treatment and at least 1 efficacy video EEG assessment. The number of participants analyzed = the number of participants who were evaluable for this outcome 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.

Population: Safety Analysis set included all participants who received at least one dose of study treatment.

6. Primary Endpoint

The registry lists the primary outcome as Ratio of Spike-Wave Index (SWI) During First Hour of Nonrapid Eye Movement (NREM) Sleep at Week 6, measured over Baseline to Week 6.

The primary endpoint is the ratio of SWI during the first hour of NREM sleep at Week 6 relative to baseline, expressed on a log base 10 scale. The registry record describes the timeframe as baseline to Week 6. Baseline was the last value measured before study treatment began on Day 1. SWI was defined as the percentage of seconds containing at least 1 spike-wave complex during defined periods of overnight NREM sleep, based on centralized video-EEG reading.

Using a ratio places the follow-up measurement in relation to the participant’s baseline value, while a logarithmic transformation represents ratios on a scale suitable for statistical analysis. In general, log transformation can make multiplicative changes easier to model and may help when measurements are skewed, although the reason for choosing this transformation is not reported in the registry record.

The primary outcome is a continuous measure, and the reported group summaries are least-squares means with standard errors. The record lists the NBI-827104 estimate as -0.02 with standard error 0.02, and the placebo estimate as 0.01 with standard error 0.04. These are group estimates on the stated log-transformed scale, not raw SWI percentages. The endpoint captures an EEG measure of sleep spike-wave activity; it is distinct from the caregiver and clinician global-impression outcomes listed as secondary measures.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a least-squares mean represent?

The reported arm values are least-squares means, a type of model-based mean. In general, a least-squares mean is estimated from a statistical model and can reflect adjustment for model terms, rather than being a simple arithmetic average of individual observations. The record identifies ANCOVA as the method for the primary comparison, but it does not report the covariates or the exact model specification. The values -0.02 and 0.01 should therefore be read as the posted model-based group summaries, not as enough information to reconstruct the adjusted contrast.

The standard errors, 0.02 for NBI-827104 and 0.04 for placebo, describe the estimated sampling uncertainty around each arm’s least-squares mean under the model. In general, a standard error is not itself a confidence interval and does not communicate the full uncertainty around the difference between groups. The record does not post confidence limits for either arm or the adjusted comparison.

What does ANCOVA do?

ANCOVA, or analysis of covariance, models an outcome while accounting for one or more covariates. In general, adjustment can improve precision when covariates are relevant and appropriately specified. The registry record names ANCOVA for the primary comparison, but does not report which covariates were included, how the baseline measurement entered the model, or whether any assumptions were checked. Those details are not reported in the registry record, so the exact adjusted comparison cannot be independently reconstructed from the arm summaries.

The sponsor’s reported p-value is 0.433. In general, a p-value is calculated under a statistical model and a null hypothesis, and it reflects the extremeness of data such as those observed if that null model were correct. It is not an estimate of effect size and cannot substitute for a treatment contrast with a confidence interval. The record does not report the confidence interval or the adjusted estimate, limiting assessment of the effect’s precision and range.

Why is a recalculation unavailable?

A simple unadjusted recalculation requires sufficient posted information to estimate a treatment contrast and its uncertainty. The registry record marks the continuous two-arm comparison as not estimable from the available values. Accordingly, no recalculated contrast, p-value, or confidence interval is reported. The arm-specific estimates and standard errors do not supply all the information needed to recreate the sponsor’s ANCOVA, particularly because the model specification and participant-level data are not reported.

How should the transformed scale be read?

The endpoint is a log base 10 transformed ratio of Week 6 SWI to baseline. In general, a logarithm transforms a ratio so that multiplicative changes are represented on an additive scale. A value on this transformed scale is not a percentage-point change in SWI. Its interpretation depends on the ratio definition and transformation, and the posted arm means should be understood on that transformed scale. No conversion to raw-scale ratios or percentage changes is made.

The endpoint also uses a centralized video-EEG assessment and a defined sleep period: the first hour of NREM sleep. Standardized measurement can reduce some sources of variation in outcome assessment. The record does not provide further details about EEG reading procedures, adjudication, reliability, or handling of recordings that were not evaluable.

What does a confidence interval add?

In general, a confidence interval gives a range of values compatible with an estimate under the statistical method used. It helps readers assess both the direction and precision of a treatment contrast. A narrow interval suggests greater precision than a wide one, but neither width nor coverage can be judged without the actual interval. Since the registry record does not provide a confidence interval for the ANCOVA comparison, the precision of the treatment contrast cannot be described from the p-value alone.

What can the reported analysis establish?

The reported ANCOVA is the sponsor’s inferential analysis for the primary comparison. The record supplies its p-value but not its estimated contrast, confidence interval, model terms, or full analysis plan. General statistical interpretation therefore has to distinguish the available model-based result from information that is not reported. The arm estimates, standard errors, analyzed counts, and p-value can be described as posted, but they do not permit a full independent replication or a complete assessment of the model assumptions.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("NBI-827104", "Placebo")
n   <- c(14, 6)
m   <- c(-0.02, 0.01)
sp  <- c(0.02, 0.04)
sd  <- sp * sqrt(n)      # standard errors converted to SDs
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  NBI-827104  14  -0.020  0.075  -0.063  0.023
CARM  Placebo  6  0.010  0.098  -0.093  0.113

11. Analyses Reported by the Sponsor

The sponsor’s reported primary comparison used ANCOVA and gave p = 0.433. The registry record does not report the adjusted between-group estimate or its confidence interval, so the reported p-value cannot be paired with a stated magnitude or interval for the contrast. The arm-specific least-squares means are -0.02 for NBI-827104 and 0.01 for placebo, with standard errors of 0.02 and 0.04, respectively.

A recalculated unadjusted comparison is not available from the posted continuous-endpoint summary statistics. The registry record explicitly identifies the two-arm comparison as not estimable from the available values. Consequently, there is no recalculated result to compare numerically with the sponsor’s ANCOVA result, and no discrepancy between two estimates can be assessed. The sponsor’s result remains the only reported inferential comparison for this endpoint.

The registry record does not describe the ANCOVA covariates or provide participant-level data, so it is not possible to reproduce the model or verify how adjustment affected the comparison. The p-value should be read as the reported output of that sponsor analysis, not as a measure of effect magnitude or a substitute for the missing estimate and confidence interval.

12. Multiplicity, Missing Data and Analysis Populations

The registry record lists four secondary outcomes: the ratio of SWI during the first hour of NREM sleep at Week 12; caregiver global-impression responders; clinician global-impression responders; and clinical global-impression severity scores. It also identifies the Week 6 SWI ratio as the primary outcome. The record does not report a multiplicity procedure, such as an adjustment across outcomes or a hierarchy for testing them. Without that information, the control of false-positive findings across multiple outcomes cannot be assessed.

The Full Analysis Set is defined as randomized participants who had at least 1 dose of study treatment and at least 1 efficacy video EEG assessment. For the primary outcome, the registry record states that the number analyzed equals the number evaluable. It reports 14 analyzed for NBI-827104 and 6 for placebo. The detailed reasons for participants not contributing an evaluable endpoint, and the handling of missing Week 6 measurements, are not reported in the registry record.

The record does not describe imputation, sensitivity analyses for missing data, or assumptions about why measurements might be unavailable. In general, missing-data methods can affect estimates when the chance of an observation being missing is related to outcomes or treatment. Without the analysis details, readers cannot determine whether such methods were used or how robust the endpoint result is to missingness.

The record gives an enrollment figure of 24 but does not report a sample-size justification or the planned power for the primary endpoint. It also does not report an interim-analysis plan, a detailed statistical analysis plan, or a primary-comparison significance threshold. The absence of these details does not establish that such procedures were absent; it means they are not reported in the registry record. The available information supports a description of the stated analysis population and reported outcome, but not a complete assessment of prespecification or multiplicity control.

13. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: seizure (6.2%).

The safety record lists serious adverse events among 1 of 16 participants at risk in the NBI-827104 group and 0 of 8 at risk in the placebo group. It reports 0 deaths in each group. These are event counts and denominators as recorded; they are not estimates of comparative risk adjusted for follow-up or other factors.

The record lists a serious adverse event term, Seizure, with 6.2% as the top serious term. It does not provide further detail in the registry record about the event’s timing, outcome, relationship to study treatment, or the group in which it occurred. The percentage should not be interpreted as a treatment-group rate without that context.

For other adverse events, the record reports 11 of 16 at risk for NBI-827104 and 8 of 8 at risk for placebo. These figures are distinct from the serious-event figures. The record does not provide event-specific details or exposure-adjusted analyses in the registry record. Differences in counts or denominators alone do not establish that an event was caused by treatment or that one treatment has a particular safety profile.

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

The posted results describe a randomized, masked comparison of NBI-827104 with placebo in pediatric subjects with epileptic encephalopathy with continuous spike-and-wave during sleep. The primary endpoint is a log-transformed ratio of sleep spike-wave index at Week 6 relative to baseline. The registry record reports arm-specific least-squares means and standard errors, together with an ANCOVA p-value of 0.433.

The result does not provide a reported adjusted treatment contrast or confidence interval. Without those estimates, readers cannot judge the range of values compatible with the sponsor’s comparison or quantify its precision from the posted result. The recalculated comparison is also not estimable from the available summary statistics. The reported p-value is therefore not enough to establish the size or clinical importance of any difference.

These results do not establish that the treatments are equivalent, nor do they demonstrate that no difference exists. In general, a result that does not provide a precise estimate can leave a range of possible effects unresolved. Equivalence or noninferiority conclusions require suitable design features and prespecified margins, which are not reported in the registry record.

The endpoint is an EEG measure at a particular time point, not a direct summary of every clinical outcome. The record lists caregiver and clinician global-impression outcomes and a severity measure as secondary outcomes, but their results are not included in the registry record for this analysis. Findings for the primary measure should not be assumed to represent those secondary outcomes, later time points, safety, tolerability, or pharmacokinetics.

Interpretation is also limited by the analyzed endpoint counts and the lack of detail about missing measurements, model covariates, and analysis assumptions. The record supports a careful description of what was posted, but it does not provide enough detail to reconstruct the sponsor’s ANCOVA or fully evaluate the uncertainty and robustness of its result.

15. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates the difference between arm-level summaries and a treatment comparison. The least-squares means and their standard errors describe each group separately; the ANCOVA p-value describes an inferential comparison under the sponsor’s model. Neither component alone supplies the adjusted difference and its uncertainty. Those quantities are important for evaluating magnitude and precision, not just whether a test produced a particular p-value.

The endpoint also illustrates how a continuous measure can be expressed as a transformed ratio rather than a raw change in the original measurement. The Week 6 SWI ratio uses baseline as its reference and applies a log base 10 scale. Readers need to keep that scale in mind: the reported least-squares means are not percentages or raw changes in SWI. The registry record defines the EEG measure and reading approach, making the outcome’s measurement target explicit.

Another statistical issue is the distinction between enrollment and the population contributing to an endpoint. The registry record specifies a Full Analysis Set definition and reports endpoint-specific analyzed counts. It does not explain all exclusions or missing measurements. That distinction matters when judging how broadly an estimate applies to the randomized study population.

Finally, the trial record lists multiple secondary outcomes but does not report how multiplicity was handled. This makes it important to separate the primary comparison from other planned measures and to avoid treating any single reported result as a complete account of efficacy or safety. The available record is useful for describing the design and posted analysis, while several details needed for full statistical evaluation are not reported.

16. 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 comparison is incompletely reported for independent statistical interpretation. The registry record gives an ANCOVA p-value but does not give the adjusted between-group estimate, its confidence interval, the model covariates, or a detailed model specification. The continuous summary statistics are marked as insufficient to estimate a recalculated two-arm comparison. Thus, the sponsor result cannot be independently reproduced from the posted arm summaries.

The outcome population is defined, and the endpoint-specific analyzed counts are reported, but the record does not give a participant flow accounting for the difference between enrollment and endpoint analysis. Reasons for missing or unevaluable EEG assessments, methods for handling missing data, and sensitivity analyses are not reported in the registry record. These omissions limit assessment of potential selection effects and robustness.

The record does not report the primary comparison’s significance threshold, sample-size justification, or whether interim analyses were planned or conducted. It also does not describe adjustment for testing multiple outcomes, despite listing four secondary outcomes. These absences prevent a full assessment of design assumptions and multiplicity control; they should not be taken to mean that such plans or procedures did not exist.

The primary outcome is a specific EEG measure at Week 6. The registry entries available do not include results for the secondary outcomes, further time points, or details needed to interpret the serious-event listing. The record therefore supports a focused interpretation of the Week 6 endpoint and posted safety counts, but not broader conclusions across all measures or the full clinical course.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Efficacy, Safety, Tolerability, and Pharmacokinetics of NBI-827104 in Pediatric Subjects With Epileptic Encephalopathy With Continuous Spike-and-Wave During Sleep — NCT04625101.
• Registry record last updated 2025-09-05; status last verified 2025-09.
• Results first posted 2025-09-05.
• 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.

20. Record Summary

NCT04625101 is a completed phase 2 randomized, triple-masked, parallel-group study sponsored by Neurocrine Biosciences. It enrolled 24 pediatric subjects with epileptic encephalopathy with continuous spike-and-wave during sleep and compared oral NBI-827104 with oral placebo over 13 weeks.

The primary endpoint was the log-transformed ratio of SWI during the first hour of NREM sleep at Week 6 relative to baseline, measured using centralized video-EEG reading. The registry record reports least-squares means of -0.02 for NBI-827104 and 0.01 for placebo, with standard errors of 0.02 and 0.04, and an ANCOVA p-value of 0.433. It does not report an adjusted treatment estimate or confidence interval, and the available summary statistics do not permit a recalculated comparison.

The record also lists four secondary outcomes and safety results. Details of multiplicity handling, missing-data methods, model specification, and sample-size justification are not reported in the registry record.