A Study of Selexipag in Participants Who Participated in a Previous Selexipag Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Selexipag in Participants Who Participated in a Previous Selexipag Study (NCT04565990), a na phase 3 study. The primary outcome is Number of Participants With Treatment-emergent Adverse Events (TEAEs).

2. Clinical Question

The study asked about the safety of continuing selexipag in participants who had previously taken part in an Actelion-sponsored selexipag study. The registry describes the intended participants as people who derived benefit from selexipag in indications for which a positive benefit-risk had been established. It does not report further eligibility details in the record.

This is a safety follow-up question, not a randomized comparison of selexipag with another treatment or with no treatment. Consequently, the results describe adverse events recorded among participants receiving selexipag in this study. They do not, by themselves, establish whether an event was caused by selexipag. The registry definition explicitly says that an adverse event did not necessarily have a causal relationship with the pharmaceutical or biological agent under study.

The registry reports enrollment of 43 participants. The primary endpoint timeframe was from Day 1 up to 3 days after the last dose, for up to 28 months 3 days. This gives the context for interpreting the reported safety counts: they concern the study’s stated observation period, not an estimate of risk outside that period.

3. Trial Design

This was a Phase 3, multicenter, single-arm, open-label, long-term follow-up safety study. The registry lists the study as completed, with a start date of 2021-05-03 and a primary completion date of 2023-11-10. Enrollment was 43. The condition listed was pulmonary hypertension.

All participants in the study received selexipag. The planned regimen was tablets twice daily, at a dose strength corresponding to the participant’s individual maximum tolerated dose from the parent study. The record does not describe a randomized allocation procedure, and allocation is listed as not applicable. Masking is listed as none. There was no concurrent control arm, so the study was not designed as a like-for-like comparison between treatment groups.

The study’s stated purpose was to assess long-term safety while allowing continued selexipag treatment for participants who had previously participated in an Actelion-sponsored study and who derived benefit from selexipag in the relevant indications. This design focuses on participants continuing treatment after an earlier study. It therefore addresses safety observations in that follow-up setting, rather than a broad comparison of treatment strategies in a newly recruited population.

With an open-label design, participants and study personnel were not masked to treatment assignment. Because the study had a single group, there was no separate randomized group to provide a concurrent benchmark for event frequency. These design features matter when interpreting adverse-event counts: they describe events observed during this study, but they cannot isolate a treatment effect from other possible influences.

The record does not report the recruitment process, detailed eligibility criteria, the distribution of follow-up duration among participants, or the reasons participants stopped treatment. It also does not report a protocol-level statistical analysis plan in the information summarized in the registry record. The study design and available endpoint information support a descriptive reading of safety observations, rather than an estimate of comparative efficacy or a causal treatment effect.

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: Safety analysis set included all participants who received at least 1 dose of study intervention in this study..

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Treatment-emergent Adverse Events (TEAEs), measured over From Day 1 up to 3 days after last dose of drug (up to 28 months 3 days).

The primary endpoint was the number of participants with treatment-emergent adverse events, expressed as a count of participants. An adverse event was defined as any untoward medical occurrence in a clinical study participant who received a pharmaceutical product, whether investigational or non-investigational. The definition did not require a causal relationship with the study agent.

Treatment-emergent adverse events were defined as events occurring at or after the initial administration of study intervention through the day of the last dose plus 3 days. The timeframe ran from Day 1 up to 3 days after the last dose, for up to 28 months 3 days. The endpoint included serious and non-serious events.

Counting participants with an event is a descriptive safety measure: it indicates how many people in the safety analysis set had at least one event meeting the definition. It is distinct from counting every event, since one participant may experience more than one event. A participant count also does not, on its own, convey event severity, duration, recurrence, or the investigator’s assessment of causality.

Such an endpoint is used in safety follow-up because it records the occurrence of medical events during exposure and the defined period after treatment. In this single-group study, however, the count is not a comparison with an untreated or alternative-treatment group. The registry record does not report an endpoint estimate for the primary measure in the information; the reported adverse-event counts should not be substituted for that primary endpoint unless their definitions and relationship to it are specified.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods should be interpreted in light of the study design and the form of the endpoint. This study had one treatment group, was open-label, and did not have a concurrent control arm. Its primary endpoint was a participant count, not a mean, rate ratio, or time-to-event comparison. The registry record lists 0 analyses, and it does not report an inferential analysis for the primary endpoint in the information.

What does a participant count describe?

A participant count records how many people met an event definition during the stated observation window. It is not the same as a count of all event episodes: one participant can have multiple events, and a participant-level endpoint generally counts that person once if the endpoint is whether any qualifying event occurred. The posted record identifies the primary endpoint as a participant count, but the registry information does not give the primary endpoint result itself.

For the adverse-event categories, the record lists 7 participants with serious adverse events and 20 with other adverse events. These figures should be retained as the categories reported; the record does not establish whether a participant could appear in both categories. Accordingly, combining them would risk double-counting or misrepresenting the registry’s classification. The number of deaths is reported separately as 5, without further detail about overlap with the adverse-event categories.

Why does the single-group design matter?

In a randomized comparison, outcomes across groups can be contrasted under the study’s allocation scheme. Here there was no second randomized group. A count within the selexipag group describes what was recorded, but it does not show what would have occurred in the same participants under another treatment. Differences from an external population, if any, could reflect differences in participant characteristics, eligibility, follow-up, event definitions, or other factors. The registry record does not report an external comparison.

Open-label treatment means masking was absent. For safety outcomes, awareness of treatment can affect reporting, evaluation, or classification of events. That general possibility does not show that any particular event in this study was misclassified; it is a design consideration when interpreting unblinded safety data.

What do confidence intervals add?

In general, a confidence interval expresses uncertainty around an estimated quantity under a statistical model and analysis method. It helps readers assess how compatible the data are with a range of values, rather than relying on a single estimate. The registry information does not report a confidence interval for the primary endpoint or the adverse-event counts. No interval should therefore be inferred from the counts alone.

In general, uncertainty around event proportions can also depend on the number of participants observed and on how completely follow-up is captured. A count without a clearly reported denominator, follow-up distribution, or interval estimate offers less information about precision than a fully specified analysis. The record gives enrollment of 43 and adverse-event category counts, but it does not report enough detail to reconstruct an adjusted or exposure-time analysis.

How should the planning illustration be read?

The sample-size assumptions describe a planning illustration based on an observed effect, not the design justification for this study. The stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. For non-time-to-event outcomes, the illustration is expressed as participants per arm for 80% power or other stated planning targets, while time-to-event illustrations concern total events. These are prospective planning quantities, not observed power, and they do not change the interpretation of the single-group safety results.

In general, power is a design property tied to assumptions about the effect, variability, significance level, and analysis. A planning calculation based on an observed effect can be useful as an illustration, but it is not a measure of how persuasive the completed study’s results are. The registry record does not report that this study used the planning assumptions as its protocol design or sample-size justification.

Why might over-dispersion matter?

In general, over-dispersion occurs when variation in count data is greater than a simple model assumes. If event counts are analyzed with a model that does not account for that extra variation, uncertainty estimates can be too small. This issue is relevant to repeated-event counts, but the primary endpoint is defined as the number of participants with adverse events, rather than the total number of event episodes. The registry record does not report a model for repeated events, an over-dispersion assessment, or an inferential analysis for this endpoint.

Because the endpoint is a safety count and no comparative analysis is reported, the clearest statistical reading is descriptive. The data identify reported event counts and the definitions and timeframe attached to them. They do not provide a treatment-effect estimate, a formal test against a comparator, or a model-based estimate of risk. Any such analysis would require details not 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("Selexipag")
n      <- c(43)
events <- c(22)
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  Selexipag  43  22  51.2  35.5  66.7  36.8  65.4  36.7  65.4

11. Multiplicity, Missing Data and Analysis Populations

The registry record does not report a multiplicity strategy. Multiplicity concerns the increased chance of at least one false-positive finding when many outcomes, event categories, time points, or comparisons are tested. In general, a prespecified approach can control or otherwise address that issue. The record does not state whether any adjustment was planned or applied.

The registry lists 5 outcome measures and 0 analyses. It identifies the primary endpoint as the number of participants with treatment-emergent adverse events, but the statistical analysis plan, outcome hierarchy beyond that designation, and any confirmatory testing procedure are not reported in the registry record. The presence of multiple recorded safety terms should not be treated as evidence that formal hypothesis tests were performed.

Missing-data handling is not reported in the registry record. The record does not describe how incomplete adverse-event information, missed visits, withdrawal, loss to follow-up, or treatment discontinuation were handled. It also does not report the amount or pattern of missing data, so the impact of missingness on the observed safety counts cannot be assessed from the available information.

The defined safety analysis set included all participants who received at least 1 dose of study intervention in this study. The record does not provide further detail on exclusions, the handling of participants without complete follow-up, or whether every endpoint used that same analysis population. A clearly defined analysis population is important because estimates can depend on who is included and how exposure and observation time are handled.

Consequently, the available information supports a descriptive account of the recorded safety observations, but not an assessment of multiplicity control, missing-data bias, or the robustness of an inferential analysis. Those aspects are not reported in the registry record.

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: pulmonary arterial hypertension (7.0%), cardiac failure acute (2.3%), covid-19 pneumonia (2.3%), pneumonia (2.3%), subacute endocarditis (2.3%).

The registry reports 7 participants with serious adverse events, 20 with other adverse events, and 5 deaths in the Selexipag group. These are distinct reported entries; the record does not explain whether the serious and other-event categories overlap, so they should not be summed. The primary endpoint includes serious and non-serious treatment-emergent adverse events, but the registry information does not state the primary endpoint count.

The listed serious adverse event terms are Pulmonary Arterial Hypertension at 7.0%; Cardiac Failure Acute at 2.3%; Covid-19 Pneumonia at 2.3%; Pneumonia at 2.3%; Subacute Endocarditis at 2.3%; and Syphilis at 2.3%. The percentages are reported as registry values. The record does not provide the event-level detail needed to determine whether terms recur in the same participant or whether participants experienced more than one term.

Serious adverse events are not synonymous with events caused by treatment. The study’s adverse-event definition does not require a causal relationship with the study drug. The record does not report the investigator’s causality assessments for these events, nor does it describe the seriousness criteria applied to each term. The circumstances and causes of the 5 deaths are also not reported in the registry record.

These data are best read as a description of events recorded during follow-up in a single treatment group. They do not, without a comparator and further event detail, establish that selexipag caused an event or quantify a comparative safety difference.

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

The results describe safety observations among participants receiving selexipag in a long-term follow-up study. The primary endpoint was defined as the number of participants with treatment-emergent adverse events, from Day 1 through 3 days after the last dose, for up to 28 months 3 days. The registry’s adverse-event information reports 7 participants with serious events, 20 with other events, and 5 deaths. These categories are not established as mutually exclusive, and the primary endpoint count is not reported in the information.

The findings do mean that the registry recorded serious and other adverse events and deaths in the study group. They offer descriptive information about the safety experience captured during this study’s observation period. The listed serious event terms provide additional descriptions of reported events, with the registry percentages attached to those terms.

The findings do not establish that selexipag caused those events. The registry definition of an adverse event explicitly does not require a causal relationship to the study intervention. The findings also do not show what event frequency would have occurred with another treatment or without treatment, because there was no concurrent comparator group.

Nor do the counts alone establish the frequency of events in a broader population or over a different duration of use. The participants came from a previous selexipag study, and the follow-up design was focused on continued treatment in that context. The registry does not report detailed follow-up distribution or enough event-level information to characterize recurrence, duration, or individual exposure.

The number of deaths should not be interpreted as a treatment-attributable death count: causes, timing, and causality assessments are not reported in the registry record. Similarly, the serious-event terms do not establish a causal link. These are limits on interpretation, not evidence for or against causation.

In short, this is a descriptive safety record from a single-group follow-up study. It offers event counts and terms, but not a comparative effect estimate, a formal causal analysis, or a complete account of uncertainty. The posted data should be interpreted within those boundaries.

14. Why This Trial Matters Statistically

This study matters statistically because it illustrates the difference between documenting safety events and estimating a comparative treatment effect. Its single-group design is suited to recording what happened during follow-up among participants continuing selexipag. Without a concurrent comparison group, the same data cannot isolate the contribution of treatment from other influences on observed events.

The endpoint also highlights the distinction between participant counts and event counts. The primary endpoint concerns participants with treatment-emergent adverse events, whereas a person may experience more than one event. The registry’s serious and other-event categories are presented separately, but their overlap is not reported. Keeping these quantities distinct avoids an unsupported total or an interpretation that the categories represent separate groups of people.

Definitions and time windows are essential to safety interpretation. Here, treatment-emergent events were defined from initial administration through the day of last dose plus 3 days, with the stated timeframe extending up to 28 months 3 days. Results from that window should not be generalized to periods outside it without additional evidence.

The record also demonstrates why statistical detail matters alongside a headline count. The primary endpoint result, confidence intervals, missing-data handling, multiplicity approach, and a comparative analysis are not reported in the information. Without those details, readers can describe the recorded safety observations but cannot assess inferential precision or robustness in the usual ways.

Finally, planning calculations are conceptually separate from observed results. The stated sample-size assumptions are planning illustrations, not observed power and not evidence that the single-group study was designed around a particular comparison. Distinguishing descriptive results, inferential analyses, and planning quantities helps keep conclusions aligned with what the study design and registry record can support.

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 main limitation specific to interpretation is the absence of a concurrent control group. Since all participants received selexipag, the observed adverse-event counts cannot be compared within this study against an alternative treatment or no treatment. The single-group design therefore does not support a randomized estimate of comparative safety or a causal estimate of treatment effect.

Participants were drawn from a previous selexipag study, and the study purpose focused on continued treatment among people described as having derived benefit in the relevant indications. The results therefore concern this follow-up setting and should not automatically be treated as representative of people starting selexipag in other circumstances. More detailed eligibility and recruitment information is not reported in the registry record.

The adverse-event reporting is limited in the registry information. The primary endpoint is defined, but its result is not reported. The reported serious and other adverse-event counts are not accompanied by an explanation of overlap, event recurrence, or individual exposure duration. The causes, timing, and treatment relationship of the 5 deaths are not reported in the registry record.

Statistical uncertainty is difficult to evaluate because confidence intervals and an inferential analysis for the primary endpoint are not reported. The record also does not report missing-data methods, multiplicity handling, detailed follow-up completeness, or the distribution of time on treatment. Those omissions limit assessment of precision and the possible influence of incomplete observation.

Finally, open-label treatment means masking was absent, and safety event identification or classification may be influenced by awareness of treatment in general. The registry information does not show whether such influence occurred in this study. These limitations mean the results are most appropriately treated as descriptive observations from the recorded follow-up period.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of Selexipag in Participants Who Participated in a Previous Selexipag Study — NCT04565990.
• Registry record last updated 2025-05-01; status last verified 2025-04.
• Results first posted 2024-11-20.
• 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

The registry record describes NCT04565990, a completed Phase 3, multicenter, open-label, single-arm, long-term follow-up safety study of selexipag. Enrollment was 43. Participants received selexipag tablets twice daily at a dose strength corresponding to their individual maximum tolerated dose from the parent study. No comparator group was included.

The primary endpoint was the number of participants with treatment-emergent adverse events, defined to include serious and non-serious events from initial treatment through the day of the last dose plus 3 days, for up to 28 months 3 days. The registry lists 7 participants with serious adverse events, 20 with other adverse events, and 5 deaths. The primary endpoint count, event overlap, and details about the deaths are not reported in the registry record.

The results are descriptive safety observations in a follow-up population. The registry record does not report a comparative analysis, confidence intervals, missing-data methods, or a multiplicity strategy. It does not establish that the reported events were caused by selexipag.