A Study of Elpipodect (MK-8189) in Participants With Bipolar I Disorder (MK-8189-020): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Elpipodect (MK-8189) in Participants With Bipolar I Disorder (MK-8189-020) (NCT06273774), a randomized phase 1 study. The primary outcome is Number of Participants Who Experience One or More Adverse Events (AEs).

2. Clinical Question

3. Trial Design

This was a randomized, triple-masked, sequential phase 1 clinical study sponsored by Merck Sharp & Dohme LLC. It studied elpipodect, also called MK-8189, and matching placebo in participants with stable bipolar I disorder. The registry record lists enrollment of 34 and a completed status. The study dates recorded are 2024-04-08 for start and 2024-08-21 for primary completion.

The intervention was organized into panels with different dose schedules. Panel A used 24 mg once daily for 14 days. Panel B was planned to begin at 16 mg once daily on Days 1 to 3 and increase to 24 mg on Days 4 to 14. Panel C began at 8 mg on Day 1, then 16 mg on Day 2, followed by 24 mg once daily on Days 3 to 14. Placebo was described as matching each panel’s MK-8189 regimen. The record says no participants were enrolled in Panel B and no data were collected for that panel.

Although randomization and triple masking are recorded, details of the randomization procedure, masking roles, allocation ratios, and how the sequential structure governed participant assignment are not reported in the registry record. A detailed protocol, including rationale for dose selection and any rules for moving between panels, is not reported in the registry record. The posted data list arm-level outcome counts, but those counts should not be treated as proof that each listed dose group represents a separate randomized comparison. No direct statistical comparison is posted.

The study’s brief summary says the goal was to evaluate safety and tolerability, and the official title also identifies pharmacokinetics. The listed primary outcome concerns adverse events. Pharmacokinetic results are not reported in the registry record in the registry record for this account. The record states that there was no hypothesis testing.

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: All participants who received at least one dose of treatment were included in the analysis. No participants were enrolled in Panel B and no data was collected for Panel B. One participant in Panel C receiving 24 mg MK-8189 experienced an adverse event and was down-titrated to 12 mg MK-8189 per protocol.

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: No participants were enrolled in Panel B and no data was collected for this panel.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants Who Experience One or More Adverse Events (AEs), measured over Up to 28 days.

The primary endpoint counts participants who experienced one or more adverse events, with assessment up to 28 days. The registry defines an adverse event as an untoward medical occurrence during a clinical study that is temporally associated with use of the study intervention, whether or not the event is considered related to that intervention. The endpoint therefore counts events regardless of whether investigators judged them to be caused by treatment.

A participant is counted for this outcome if at least one adverse event was recorded; the measure is not the total number of separate events. This distinction matters because one participant may experience more than one event, while the endpoint as named focuses on whether a participant experienced one or more. The outcome is expressed in participants, and the registry says the analysis population included all participants who received at least one dose of treatment.

As a general statistical point, an adverse-event count is useful in an early clinical study because it gives a direct descriptive view of observed tolerability during the defined observation period. It does not by itself establish whether an event was caused by the intervention, how severe the overall burden was, or whether treatment groups differ beyond chance variation. Those questions require additional definitions, context, and analyses. The registry record describes the study as having no hypothesis testing, so the primary outcome should be read as a safety description rather than a confirmatory comparison.

The registry also reports serious adverse events separately. Seriousness is distinct from the broader adverse-event definition: a serious event is recorded under the registry’s serious-event reporting category, while the primary outcome includes adverse events more broadly. The record’s detailed criteria for event severity and causality are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary outcome as a binary participant-level measure: whether a participant experienced one or more adverse events during the stated observation period. The posted result is a count of participants with an event within each listed arm. The corresponding arm counts indicate the number included in that outcome entry. No continuous outcome summary or event-rate model is reported for this primary measure.

What does a participant count mean?

As a general statistical explanation, a binary outcome reduces each participant’s experience to one of two states for the analysis: an event was recorded, or it was not. For an endpoint defined as one or more adverse events, the participant is counted once for that endpoint even if more than one event occurred. The count can be understood only alongside the number of participants represented in the arm. The registry posts both quantities for the listed arms, but it does not report a formal estimate comparing them.

This endpoint is different from counting every event. Event totals can be larger than participant counts when one participant experiences multiple events. The posted primary outcome title specifically concerns participants who experienced one or more adverse events, and its unit is participants. The separate serious-event information also reports participants affected, along with serious event terms. The registry record does not give individual event histories or clarify repeated events in enough detail for an event-level analysis.

Why are confidence intervals and tests important?

As a general statistical explanation, a confidence interval describes the precision of an estimated quantity under a statistical model. For a group difference, it can convey a range of values compatible with the data and model; a test addresses a related question about evidence against a specified null hypothesis. Neither is a substitute for study design or a clear estimand. In this study, no confidence intervals, treatment-effect estimates, or hypothesis tests are reported in the registry record. The brief summary explicitly says there was no hypothesis testing. Consequently, the posted counts do not support a registry-reported inferential claim about a difference between treatment and placebo.

The record identifies the study as randomized and triple-masked, but the statistical details of the randomization and the planned analysis are not reported in the registry record. The specific method used to summarize the binary endpoint beyond posting counts is not reported in the registry record. Adjustment for covariates, handling of repeated events, and any model-based analysis are also not reported in the registry record. These details should not be inferred from the design labels.

How should the sequential dose information be read?

As a general principle, a sequential dose design may involve staged evaluation, but the meaning of a sequence depends on its protocol and assignment rules. The registry lists dose regimens for Panels A, B, and C and says no participants were enrolled in Panel B. It does not report the decision criteria for progressing between panels or explain whether every arm entry represents a standalone randomized group. The Panel C schedule itself describes increasing doses over time, while the outcome listing contains dose-labeled entries. Without further analysis details, these entries should not be treated as independent head-to-head comparisons.

The record’s statement that one participant in Panel C was down-titrated from 24 mg to 12 mg after an adverse event adds an interpretive complication. A dose reduction during treatment is not necessarily equivalent to assignment to a separate dose arm. The registry does not explain how this participant’s exposure and outcome were categorized for the arm-level results. That uncertainty argues against interpreting the 12 mg outcome entry as a distinct dose comparison.

What does the missing comparison mean?

The record indicates that values are missing or not estimable for a two-arm comparison of the binary endpoint. This is consistent with the absence of a posted treatment-effect estimate. A recalculated unadjusted comparison from summary counts would still require a defensible choice of groups and assumptions about their relationship. The registry does not provide that basis, and no such comparison should be treated as the sponsor’s analysis. The record states there was no hypothesis testing, and no sponsor analysis is posted for the primary endpoint.

As a general matter, small or uneven group counts can make estimates imprecise, especially when events are uncommon or when a group contains very few participants. That general caution does not create a new estimate for this trial. No confidence interval or sample-size result is reported in the registry record for the primary comparison. Any planning illustration based on an observed effect would be a hypothetical design calculation, not observed power; no such numerical planning result is reported in the registry record.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Panel A: MK-8189 24 mg", "Panel A: Placebo", "Panel C: MK-8189 8 mg", "Panel C: 16 mg MK-8189", "Panel C: MK-8189 24 mg", "Panel C: MK-8189 12 mg")
n      <- c(12, 5, 13, 13, 13, 1)
events <- c(9, 4, 1, 4, 6, 0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Panel A: MK-8189 24 mg  12  9  75.0  42.8  94.5  46.8  91.1  46.2  90.9
ARM  Panel A: Placebo  5  4  80.0  28.4  99.5  37.6  96.4  35.9  95.7
ARM  Panel C: MK-8189 8 mg  13  1  7.7  0.2  36.0  1.4  33.3  1.8  33.9
ARM  Panel C: 16 mg MK-8189  13  4  30.8  9.1  61.4  12.7  57.6  12.8  58.1
ARM  Panel C: MK-8189 24 mg  13  6  46.2  19.2  74.9  23.2  70.9  23.0  71.1
ARM  Panel C: MK-8189 12 mg  1  0  0.0  0.0  97.5  0.0  79.3  1.3  84.2

11. Multiplicity, Missing Data and Analysis Populations

The registry record states that there was no hypothesis testing. It lists one primary adverse-event outcome, two outcome measures overall, no statistical analyses, and no posted secondary outcomes. A multiplicity adjustment is therefore not reported. The registry record does not describe a family of confirmatory hypotheses, a hierarchy among outcomes, or a prespecified procedure for controlling type I error.

As a general statistical point, multiplicity becomes relevant when several outcomes, treatment contrasts, time points, or subgroups are tested and interpreted as evidence of treatment effects. Repeated testing can increase the chance of a false-positive result unless the analysis plan addresses it. That general concern should not be mistaken for evidence that such testing occurred in this study. The available record says there was no hypothesis testing and does not report inferential analyses.

The primary outcome population is described as all participants who received at least one dose of treatment. The registry does not report whether all treated participants had complete primary outcome data, how missing outcome information was handled, or whether any participants were excluded from the analysis. It also does not report a missing-data sensitivity analysis or a rule for handling participants who discontinued treatment.

Panel B had no enrolled participants and no collected data. This is an explicit absence of data for that panel, not a result indicating no adverse events. For the other arm entries, the registry reports counts, but does not provide a narrative account of missing observations or explain the relationship between the overall enrollment figure and the outcome denominators. The record also does not provide enough detail to determine how dose changes were represented in the analysis population. These omissions limit assessment of completeness and analysis-population handling.

12. 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: bipolar i disorder (8.3%), homicidal ideation (8.3%).

The registry reports serious adverse events separately from the broader primary outcome of one or more adverse events. In Panel A, serious adverse events were recorded for 2 participants among 12 at risk in the MK-8189 24 mg arm and 0 among 5 at risk in the placebo arm. In Panel C, the serious-event count was 0 among 13 at risk in each listed MK-8189 8 mg, 16 mg, and 24 mg arm, 0 among 1 at risk in the 12 mg entry, and 0 among 4 at risk in the placebo arm. The registry reports 0 deaths in every listed group.

The registry lists Bipolar I disorder and Homicidal ideation as serious adverse event terms, each with 8.3%. It does not report the term-level participant counts, their relationship to the group-level serious-event entries, or whether the terms were considered related to study treatment. The serious-event entries therefore describe what was recorded, not whether treatment caused those events.

As a general statistical point, serious adverse events are uncommon outcomes in many studies, and small group counts can leave substantial uncertainty about the underlying frequency. A zero recorded count does not establish that an event cannot occur, and a nonzero count does not establish causality. No inferential comparison or confidence interval for serious adverse events is reported in the registry record.

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

The record shows that adverse events were reported among participants in several listed study arms during the defined follow-up. It gives the numbers of participants with one or more adverse events for Panel A and Panel C, and it reports no participant or outcome data for Panel B. It also records serious adverse events and zero deaths in the listed groups. These are descriptive observations from a phase 1 study focused on safety and tolerability, with pharmacokinetics named in its official title.

The results do not establish that elpipodect caused or prevented any event. The primary outcome definition includes events whether or not they were considered related to the intervention, and the registry does not provide a formal treatment-effect estimate. The record expressly states there was no hypothesis testing. Accordingly, differences between the displayed counts should not be interpreted as statistically demonstrated treatment effects.

The results also do not establish a definitive comparison between dose levels. The study used panel-specific regimens, the registry does not report the assignment and analysis details needed to interpret every listed arm as an independent comparison, and the Panel C dose-change description complicates interpretation of the 12 mg entry. The absence of Panel B data cannot be read as evidence about safety in that panel.

More broadly, this record does not establish efficacy or clinical benefit. The primary outcome is an adverse-event count, not a measure of symptom improvement or disease control. Pharmacokinetic results and detailed participant characteristics are not reported in the registry record. The brief summary states that the study had no hypothesis testing, so the posted counts are best understood as a limited descriptive account of observed adverse events in the study participants.

As a general statistical point, a small early clinical study can identify observations that warrant further evaluation, but its descriptive counts do not settle how often events would occur in a broader population. The record does not report precision estimates or a planned inferential analysis. Conclusions should remain within the scope of the recorded groups, outcome definition, and follow-up period.

14. Why This Trial Matters Statistically

This study illustrates why the design and reporting context matter when reading safety counts. The registry labels the trial randomized and triple-masked, and the study was organized into sequential panels with different dosing schedules. Those design features are relevant, but they do not by themselves supply the details needed to interpret each arm-level count as a causal comparison. The record does not describe the allocation procedure, panel progression rules, or analysis plan.

The outcome is also a useful example of the distinction between a participant-level binary endpoint and an event count. The primary measure counts participants who had one or more adverse events, while separate serious-event information describes serious events and listed terms. These are related safety summaries, but they answer different descriptive questions. A reader should not equate the number of affected participants with the total number of events.

The registry’s explicit statement that there was no hypothesis testing is central to interpretation. Numerical differences among arms can attract attention, but without a planned statistical comparison, inferential estimates, or uncertainty intervals, those differences remain descriptive. This is particularly important when group sizes differ and when some entries represent a very small number of participants.

The Panel C dose-reduction narrative further demonstrates how treatment exposure and outcome categorization can affect interpretation. The record notes a participant who was down-titrated after an adverse event, while the outcome listing separately includes a 12 mg entry. The relationship between these details is not explained. A statistical summary is only as interpretable as its definitions and mapping of participants to groups.

Finally, the record gives a completed study status and posted safety counts but does not report a sample-size justification or precision analysis. The design should therefore be read as a clinical study that describes observations, not as a confirmatory test. This distinction helps keep conclusions aligned with what the record actually documents.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The record does not report a sample-size justification, so the reason for the planned or achieved enrollment cannot be assessed. It also does not provide confidence intervals or an inferential analysis for the primary adverse-event outcome. Because the registry states that there was no hypothesis testing, the arm-level counts do not establish statistically supported differences.

Outcome denominators vary across the listed arms, and no participants or outcome data are listed for Panel B. The registry does not explain the relationship between the enrollment figure and every outcome denominator. The analysis population is described as participants who received at least one dose, but detailed participant flow, exclusions, treatment discontinuations, and missing outcome handling are not reported in the registry record.

Interpretation of Panel C is limited by the sequential dosing description and the account of a participant who experienced an adverse event at 24 mg and was down-titrated to 12 mg. The registry does not explain how that participant was categorized in the posted arm-level outcome entries. It is therefore not clear whether the dose-labeled entries correspond to separate assignment groups or describe exposure categories. The record does not provide enough information to resolve this issue.

The adverse-event definition includes occurrences regardless of whether they were considered related to study intervention. Detailed judgments about causality, event duration, severity, recurrence, and timing are not reported in the registry record. For the serious-event terms, the record lists percentages but does not give further term-level attribution or context. These limitations prevent a more detailed assessment of safety patterns.

Finally, the record does not provide pharmacokinetic results, detailed baseline characteristics, or information needed to assess how representative the participants were of people with bipolar I disorder more broadly. The reported observation period for the primary outcome is up to 28 days. The findings should therefore be interpreted within the recorded study population, treatment schedules, and follow-up rather than generalized beyond them.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of Elpipodect (MK-8189) in Participants With Bipolar I Disorder (MK-8189-020) — NCT06273774.
• Registry record last updated 2026-04-29; status last verified 2026-04.
• Results first posted 2025-08-17.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT06273774 is a completed, randomized, triple-masked, sequential phase 1 study of elpipodect, also known as MK-8189, in participants with stable bipolar I disorder. The registry lists enrollment of 34 and matching placebo groups across Panels A, B, and C. No participants were enrolled in Panel B, and no data were collected for that panel.

The primary outcome was the number of participants who experienced one or more adverse events up to 28 days. The registry states that there was no hypothesis testing. Posted results include adverse-event counts for the listed Panel A and Panel C arms, plus separate serious-event counts and terms. The record does not report a formal treatment-effect estimate, confidence interval, or inferential comparison. Interpretation of the dose entries is limited by the sequential regimens and the reported down-titration in Panel C.