Brexpiprazole for the Long-term Treatment of Patients With Agitation Associated With Dementia of the Alzheimer's Type: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Brexpiprazole for the Long-term Treatment of Patients With Agitation Associated With Dementia of the Alzheimer's Type (NCT03724942), a na phase 3 study. The primary outcome is The Frequency of Subjects With Treatment-Emergent Adverse Events (TEAEs).

2. Clinical Question

The study also listed changes in the Cohen-Mansfield Agitation Inventory score and Clinical Global Impression of Severity score, along with the Clinical Global Impression of Improvement score, at 14 weeks after dosing. These outcomes concern agitation and clinical status, but numerical results for them are not reported in the registry record. The available design details therefore frame the question as one about safety during extended treatment, rather than a controlled estimate of brexpiprazole's effect compared with placebo or another treatment.

3. Trial Design

This was a Phase 3, multicenter, uncontrolled, open-label, single-group trial. It evaluated extended brexpiprazole treatment in patients with agitation associated with dementia of the Alzheimer's type. The study enrolled 164 participants, and its status is recorded as completed. The registry lists a start date of 2018-11-09 and a primary completion date of 2021-12-02.

Participants rolled over from Trial 331-102-00088 after completing a double-blind trial. The registry description says this study evaluated brexpiprazole safety for a maximum of 24 weeks, including the treatment period of the earlier trial. The primary outcome timeframe is described separately as baseline to week 14. Brexpiprazole 1 mg or 2 mg is described in the study summary as the treatment being evaluated after a 14 week regimen in the earlier trial.

Open-label means treatment assignment was not masked in this study. Single-group and uncontrolled mean that the extension did not include a concurrent control arm for estimating a treatment contrast. Although adverse-event records are organized under ROLLOVER BREX and ROLLOVER PLACEBO, those labels do not make the extension a randomized comparison. The cohorts reflect rollover categories, and the registry specifically cautions that they are not a like-for-like control comparison.

The registry record names one intervention, brexpiprazole, and lists four outcome measures. It reports no analyses. The registry record does not report the method of assigning the rollover treatment, the rationale for the planned enrollment, or a statistical analysis plan. Without a concurrent randomized control group and posted comparative analyses, the design principally supports a descriptive account of events observed during follow-up, not a causal comparison of treatment effects.

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 safety analysis set comprised subjects who have received at least 1 dose of the IMP.

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: This trial was conducted in 164 participants in Japan.

6. Primary Endpoint

The registry lists the primary outcome as The Frequency of Subjects With Treatment-Emergent Adverse Events (TEAEs), measured over From baseline to week 14.

The primary endpoint is the frequency of subjects with treatment-emergent adverse events, with the stated timeframe from baseline to week 14. A treatment-emergent adverse event is an event recorded after the relevant baseline point during treatment or follow-up; the term alone does not establish that treatment caused the event. The registry describes the endpoint using a percentage, while the listed parameter is a number. Numerical results for this primary endpoint are not reported in the registry record.

Safety studies commonly track treatment-emergent events because systematic recording can describe what happened after treatment began, including events that are serious and those that are not. Such monitoring is useful in an extension, where participants continue treatment beyond an earlier trial period. It can identify event types and the number of participants reported with events over the stated observation period. That information is descriptive: interpretation depends on the population, duration of observation, event definitions, and completeness of follow-up.

Here, the safety analysis set is defined as subjects who received at least one dose of the investigational medicinal product. The endpoint description also notes that the trial enrolled subjects rolling over from Trial 331-102-00088 and that safety was evaluated over a maximum of 24 weeks including the earlier trial's treatment period. The primary outcome's stated timeframe is baseline to week 14, so the endpoint window and the broader safety description should not be treated as interchangeable. The registry record does not report a numerical primary endpoint estimate, its uncertainty interval, or an inferential test.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The design determines the kinds of statistical conclusions that are appropriate. This was an uncontrolled, open-label, single-group extension. A single-group study can describe events observed among participants who receive treatment, but without a concurrent randomized comparator it cannot separate treatment-associated changes from other explanations, such as the natural course of illness, background risks, or differences between participants. The rollover cohorts do not supply a valid randomized comparison within this extension, as the registry record cautions.

What does a frequency endpoint describe?

A frequency endpoint counts or summarizes how many participants experienced an event during a defined observation window. The primary outcome is described as the frequency of subjects with treatment-emergent adverse events from baseline to week 14. That is a participant-level concept: it concerns whether participants had an event during the window, not necessarily the total number of event episodes. The registry record does not report a numerical primary-outcome result, so neither the event frequency nor its uncertainty can be evaluated from the posted primary outcome information.

In general, a proportion is interpreted in relation to its denominator and observation period. Follow-up duration matters because a participant observed for longer has more opportunity to experience an event. A frequency also depends on event definitions, ascertainment, and whether participants with incomplete follow-up are included. The record defines the safety analysis set as participants who received at least one dose, but does not report the missing-data approach or detailed handling of different follow-up durations.

How should uncertainty be understood?

In general, a confidence interval communicates statistical uncertainty around an estimate under the assumptions of the chosen method. For a single-group frequency, an interval can show how imprecisely the observed proportion estimates the underlying event frequency. An interval does not account for all possible biases, and it does not create a control group. The registry record does not report a confidence interval for the primary endpoint or a statistical test for it.

The listed serious adverse-event counts and other event counts are descriptive. Counts alone do not express uncertainty, and the cohort totals do not establish that any difference between the rollover groups reflects a treatment effect. A formal contrast would require an appropriate design and analysis; the registry record does not report such an analysis. The displayed percentages for selected serious-event terms are registry values, not causal risk estimates.

Why does the open-label extension matter?

In an open-label study, participants and study personnel are not masked to treatment. In general, knowledge of treatment can affect reporting, assessment, and decisions about whether an event is recorded. This does not mean that a recorded event is invalid; it means event ascertainment should be interpreted in light of the study's open-label design. The registry does not report further details about event adjudication or masking of outcome assessors.

The extension enrolled participants who had completed an earlier double-blind trial. This rollover approach allows continued observation in a group with prior trial participation, but its results describe that selected group. They need not represent all patients with the condition. The record does not report details of the rollover selection process beyond the description that participants completed the earlier trial.

What can be inferred from the outcome list?

The secondary outcomes include mean change from baseline in the Cohen-Mansfield Agitation Inventory score and Clinical Global Impression of Severity score, plus the Clinical Global Impression of Improvement score, all at 14 weeks after dosing. These measures concern symptoms or clinical ratings, but listing an outcome is not the same as reporting its result. The registry record lists four outcome measures and no analyses. Numerical estimates, uncertainty intervals, analysis models, and missing-data procedures for these outcomes are not reported in the registry record.

In general, a mean change summarizes average within-participant change, while a clinical rating may also be reported as a score at follow-up. A within-group change is not equivalent to a treatment effect relative to a comparator: without a control group, change can reflect treatment, time, natural variation, or other influences. No efficacy conclusion can be drawn from the outcome names alone.

What about planning calculations?

In general, sample-size planning asks how many participants would be needed to detect a specified effect under stated assumptions, such as a significance level, allocation, and statistical model. A planning illustration is not an observed result and does not measure the probability that a completed study found an effect. The record does not report the trial's sample-size justification or an estimated treatment effect for a controlled comparison. Because this study is single-group and lacks a concurrent randomized comparator, comparative planning illustrations should not be read as evidence of treatment efficacy. The registry record does not report an interim-analysis plan, a multiplicity procedure, or a detailed statistical analysis plan.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists several outcomes: the primary treatment-emergent adverse-event frequency and secondary measures involving the Cohen-Mansfield Agitation Inventory, Clinical Global Impression of Severity, and Clinical Global Impression of Improvement. Having multiple outcomes raises a general statistical issue: repeated testing can increase the chance of at least one apparently unusual result when no adjustment is used. The registry record does not report a multiplicity adjustment or a testing hierarchy. It also reports no analyses, so there are no posted outcome tests to interpret as adjusted or unadjusted.

The registry defines the safety analysis set as participants who received at least one dose of the investigational medicinal product. It does not report the analysis population for the secondary outcomes, whether all enrolled participants were included in each outcome analysis, or how departures from the planned observation schedule were handled. The registry record also does not report a missing-data method, imputation approach, or sensitivity analysis. For a general interpretation, missing observations can affect estimates when the participants with observed follow-up differ from those without it; the direction and size of any such effect cannot be determined from the record.

The distinction between the stated 14-week primary outcome timeframe and the safety description covering a maximum of 24 weeks including the earlier trial period also matters. The record does not provide enough detail to reconstruct participant-level exposure or follow-up for each analysis. No statistical analysis plan, interim-analysis plan, or sample-size justification is reported in the registry record. Consequently, the choice of analysis population, treatment of missing data, multiplicity control, and planned inferential procedures cannot be verified from the registry entry.

11. 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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumonia aspiration (3.2%), gait disturbance (2.0%), pneumonia (1.6%), femur fracture (1.6%), pneumothorax (1.6%).

The serious adverse-event figures are separated by rollover cohort. In ROLLOVER BREX, 7 serious events were recorded among 102 participants at risk; in ROLLOVER PLACEBO, 4 were recorded among 62. The record also lists other adverse events for 90 of 102 and 55 of 62 participants, respectively. One death is recorded for ROLLOVER PLACEBO and none for ROLLOVER BREX.

These counts describe what the registry records for the named cohorts. They should not be read as a randomized comparison of brexpiprazole with placebo in this extension: the trial itself was open-label, uncontrolled, and single-group, and the registry warns that the rollover groups are not a like-for-like control comparison. The counts do not show whether an event was caused by treatment, and the death counts alone do not establish a treatment-related cause.

The registry lists selected serious-event terms with percentages: Pneumonia aspiration, 3.2%; Gait disturbance, 2.0%; Pneumonia, 1.6%; Femur fracture, 1.6%; Pneumothorax, 1.6%; and Femoral neck fracture, 1.0%. These entries identify reported event categories and their listed percentages. Without further context, they should not be treated as comparative risks or as estimates of treatment-attributable harm.

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

The study contributes descriptive safety information from an open-label extension involving participants who completed an earlier double-blind trial. Its primary question concerns treatment-emergent adverse events over a stated baseline-to-week-14 window, and the registry also describes safety observation for a maximum of 24 weeks when the earlier trial's treatment period is included. The record defines the safety analysis set as participants who received at least one dose of the investigational medicinal product.

The posted serious-event counts and selected serious-event terms show that events were recorded in both rollover cohorts. They do not establish that treatment caused those events. Nor can the cohort counts be interpreted as a valid treatment comparison: this extension had no concurrent control group, and the registry says the rollover categories are not like-for-like control arms. A descriptive record of events is useful for understanding what was observed, but it does not by itself estimate what would have happened to comparable participants under a different treatment.

The registry record does not report numerical results for the primary adverse-event frequency or the listed secondary clinical outcomes. It therefore does not allow readers to quantify the primary endpoint, assess its precision, or determine the magnitude of any change in agitation or clinical ratings. Missing numerical results should not be interpreted as evidence of no events, no change, or no effect.

These results also apply to a selected rollover population, rather than necessarily to every person with agitation associated with dementia of the Alzheimer's type. The registry record does not report sufficient information to assess how participants who did not enter or complete the extension might differ. Generalization beyond the enrolled group should therefore be cautious. Overall, the design and posted information support a limited descriptive reading of safety events, not a causal efficacy conclusion or a direct comparison between brexpiprazole and placebo.

13. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates the distinction between describing events and estimating comparative effects. An uncontrolled extension can record safety outcomes during continued treatment and identify event types reported among participants. Such information has value, particularly when follow-up continues beyond an earlier trial. Its interpretation, however, depends on who entered the extension, how long participants were observed, how events were assessed, and how missing observations were handled.

The rollover structure makes the distinction especially important. Cohort labels retain information about participants' earlier trial grouping, but they do not turn the extension into a concurrent randomized comparison. The registry explicitly cautions that the ROLLOVER BREX and ROLLOVER PLACEBO cohorts are not a like-for-like control comparison. Treating their counts as if they arose from randomized arms could overstate what the design can establish.

The record also demonstrates that outcome registration and outcome reporting are different. It names a primary safety outcome and several clinical measures, but numerical outcome results and analyses are not reported in the registry record. Without estimates and their uncertainty, readers cannot assess the precision of the primary outcome or the size of any observed clinical changes. Listing a measure signals what was intended to be assessed; it is not itself evidence of a result.

For biostatistical interpretation, the trial underscores why design, analysis population, outcome window, and reporting completeness should be considered together. A clearly defined safety set and observation period help describe the intended analysis, but missing analysis details limit reproducibility and assessment of bias. The registry record allows a careful account of design and some adverse-event counts, while leaving key questions about inferential methods and outcome estimates unanswered.

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 central limitation is the single-group, uncontrolled, open-label design. There is no concurrent randomized comparator, so the study cannot isolate the effect of brexpiprazole from other influences on observed events or clinical changes. The rollover categories do not resolve this limitation; the registry states that they are not a like-for-like control comparison.

Participants came from an earlier double-blind trial, so the extension describes a population selected through prior trial participation and completion. The registry record does not report enough information to determine how participants who did not roll over differed from those who entered this study. The results may not represent all patients with agitation associated with dementia of the Alzheimer's type.

The primary endpoint timeframe is baseline to week 14, while the safety description refers to a maximum of 24 weeks including treatment in Trial 331-102-00088. The record does not provide participant-level follow-up details that would resolve how these windows relate for each analysis. It also does not report numerical primary or secondary outcome results, uncertainty intervals, the statistical analysis plan, missing-data methods, multiplicity handling, or interim analyses.

The registry lists serious-event counts by rollover cohort and selected serious-event terms, but the information is insufficient to attribute events to treatment or establish comparative risk. Details about event adjudication and the circumstances of the recorded death are not reported in the registry record. These limitations constrain both causal interpretation and assessment of precision.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Brexpiprazole for the Long-term Treatment of Patients With Agitation Associated With Dementia of the Alzheimer's Type — NCT03724942.
• Registry record last updated 2024-09-19; status last verified 2024-05.
• Results first posted 2024-09-19.
• 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

NCT03724942 is a completed Phase 3, multicenter, open-label, uncontrolled, single-group trial of extended brexpiprazole treatment for agitation associated with dementia of the Alzheimer's type. The registry reports 164 participants enrolled. Participants rolled over from Trial 331-102-00088, and the safety description covers a maximum of 24 weeks including the earlier trial's treatment period.

The primary outcome was the frequency of subjects with treatment-emergent adverse events from baseline to week 14. The safety analysis set comprised participants who received at least one dose of the investigational medicinal product. The record also lists secondary clinical outcomes at 14 weeks after dosing. It reports no analyses, and numerical results for the primary and secondary outcomes are not reported in the registry record.

Serious adverse-event counts are listed for ROLLOVER BREX and ROLLOVER PLACEBO, but these cohorts are not a like-for-like control comparison. The record does not report a statistical analysis plan, missing-data methods, multiplicity handling, or sample-size justification.