Bosentan in Children With Pulmonary Arterial Hypertension Extension Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Bosentan in Children With Pulmonary Arterial Hypertension Extension Study (NCT00319020), a na phase 3 study. The primary outcome is Change From Baseline to End of Study (EOS) in Height for Age..

2. Clinical Question

The study’s stated objective was to assess long-term safety and tolerability of pediatric-formulation bosentan in children with idiopathic or familial pulmonary arterial hypertension who had completed FUTURE 1. The primary outcome listed in the registry focused on growth rather than a comparison of bosentan with another treatment. Height measurements were converted to Z-scores using World Health Organization growth standards, allowing observed height to be expressed relative to the reference population.

Because the study was open-label and single-group, it can describe outcomes recorded during treatment and follow-up, but it does not provide a randomized estimate of how those outcomes compare with an untreated or differently treated group. The registry record does not report the primary endpoint result, so the size and precision of the observed change cannot be assessed from that record.

3. Trial Design

This was a Phase 3, open-label, long-term safety and tolerability extension study of bosentan in children who had completed FUTURE 1. The registry lists a single-group model, no masking, and allocation as not applicable. It reports enrollment of 33 and a completed study, with a start date of 2005-08-23 and primary completion date of 2011-10-28.

Participants received the pediatric formulation of bosentan at 4 mg/kg twice daily through the end of the study. The dose could be down-titrated to 2 mg/kg twice daily if it was not well tolerated. The record does not describe another intervention arm or a concurrent control group. It identifies patients with previous bosentan exposure and bosentan-naive patients in the adverse-event information, but those labels do not create a randomized comparison.

The primary endpoint timeframe ran from baseline in FUTURE 1 through up to 28 days after study treatment discontinuation, whether at end of study or premature treatment discontinuation. The registry describes the average timeframe as 32 months. This combines the original baseline with follow-up during the extension, rather than defining a comparison between independently assigned groups.

The record lists 8 outcome measures and 0 analyses. It does not report a statistical analysis plan, details of any interim analyses, or a sample-size justification. Those details should not be inferred from the Phase 3 designation or from the enrollment listing. The posted design information supports interpretation as a descriptive extension study intended to follow participants over time, not as a randomized efficacy comparison.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: The actual number of patients enrolled in FUTURE 2 (F-2) was 33 because 2 patients did not complete FUTURE 1 (F-1) and one patient completed F-1 but was not enrolled in F-2.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All-treated analysis set (all patients who received at least one dose of study drug in the combined FUTURE 1 / FUTURE 2 trial periods). Missing or incomplete data were treated as missing.

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 Change From Baseline to End of Study (EOS) in Height for Age., measured over From baseline (FUTURE 1) up to 28 days after study treatment discontinuation (EOS or premature study treatment discontinuation), i.e. 32 months in average.

The primary endpoint was change from baseline to end of study in height-for-age Z-score. A Z-score expresses a measured value in standard deviations from the median of a reference population. In this study, height measured at each visit was compared with World Health Organization growth standards. The score was calculated from the participant’s observed height, the reference population median, and the reference population standard deviation.

Using a standardized score can make measurements interpretable relative to a healthy-child growth reference, rather than relying on height in its original measurement units alone. Following the score over time can describe how participants’ height measurements relate to that reference during the study period. The registry specifies the median and full range as the summary statistics for the endpoint.

The endpoint definition does not itself show whether scores increased, decreased, or remained similar. The registry record does not report the median change or its full range, nor does it give a confidence interval for the endpoint. Consequently, the direction, magnitude, and precision of the recorded change cannot be evaluated from the endpoint description alone. The outcome also does not, by itself, establish that any observed growth pattern was caused by bosentan.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry’s primary endpoint uses height-for-age Z-score, a standardized measurement. In general, a Z-score expresses how far an observation lies from a reference population median in units of the reference population’s standard deviation. The record specifies World Health Organization growth standards and gives the calculation in terms of observed height, the reference median, and the reference standard deviation. This transformation places measurements on a common reference scale; it does not make the study a controlled comparison.

What does the median and full range describe?

The endpoint was specified as a median with a full range. In general, the median is the central value after observations are ordered, while the full range gives the lowest and highest observed values. These summaries can describe the center and spread of participant-level changes without assuming that values follow a particular symmetric distribution. The full range is sensitive to extreme observations and does not convey the uncertainty around the median as a confidence interval would.

The registry does not report the median endpoint value or its full range. It also does not report a confidence interval or another inferential estimate for this outcome. As a result, the methods description indicates how the endpoint was intended to be summarized, but does not provide the data needed to judge its magnitude or precision.

What can a confidence interval tell a reader?

In general, a confidence interval gives a range of values compatible with the data and a specified statistical procedure. Its width communicates precision: narrower intervals indicate greater precision under that procedure, while wider intervals indicate less. A confidence interval does not remove bias from an uncontrolled design, and it is not a guarantee that the true effect lies within the interval. No confidence interval for the primary endpoint is reported in the registry record.

Why does the study structure matter?

The registry identifies a single-group, open-label study with no masking and no applicable allocation. General statistical interpretation of a single-group follow-up is descriptive: measurements can be recorded over time, but without a concurrent comparison group, change cannot be attributed uniquely to the intervention. Other explanations for change are not ruled out by calculating a standardized endpoint or by reporting a median.

The safety groupings labeled previous bosentan and bosentan-naive do not alter that design. They are not described as randomized arms, and the record warns that they are not a like-for-like control comparison. Statistical contrasts between such groups would be vulnerable to differences in their histories and characteristics, and the registry does not report an analysis comparing them.

How should the analysis population be read?

The endpoint population is described as the all-treated analysis set: patients who received at least one dose of study drug in the combined FUTURE 1 and FUTURE 2 periods. The record says missing or incomplete data were treated as missing. This tells readers that such observations were not assigned endpoint values in the described population definition. It does not explain how many endpoint measurements were missing, how missingness varied over time, or whether any further handling or sensitivity analyses were used.

In general, missing data can affect an estimate when the availability of measurements is related to participant characteristics or outcomes. Treating a value as missing avoids substituting an unobserved value, but does not by itself eliminate bias from incomplete follow-up. Since counts and patterns of missing endpoint data are not reported, the extent of this issue cannot be assessed.

What is not available about the statistical plan?

The registry lists 0 analyses. It does not report the analysis methods used for the primary endpoint, a hypothesis test, an adjustment strategy, or an interim-analysis plan. It also does not report a sample-size justification. These omissions mean that readers cannot determine from the record whether a particular inferential procedure was prespecified or how any multiplicity was addressed.

In general, an analysis plan helps distinguish planned analyses from exploratory ones and clarifies how outcomes, populations, and missing observations are handled. The absence of that information in this registry record should not be filled by assumptions based on common practice or the study’s phase.

The record includes a general sample-size planning description based on an observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. Such planning illustrations are not observed power, and the record does not identify them as the study’s own design justification. With no reported primary endpoint estimate or inferential analysis, these planning assumptions do not supply evidence about the observed treatment outcome.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 8 outcome measures and 0 analyses. It does not report a multiplicity strategy, such as a hierarchy among outcomes or an adjustment for making multiple statistical tests. Without the analysis methods or results, it is not possible to tell whether formal hypothesis testing was performed or how any multiple-outcome concerns were handled. No inferential plan should be presumed.

The primary endpoint population is described as the all-treated analysis set across the combined FUTURE 1 and FUTURE 2 periods. The registry states that missing or incomplete data were treated as missing. It does not give the number of missing endpoint observations, the reasons for missingness, or the amount of follow-up available for each participant. It also does not describe imputation, sensitivity analyses, or other methods for assessing the consequences of missing data.

In general, missing observations can affect interpretation if follow-up is related to participants’ health, treatment experience, or measured outcomes. Treating incomplete data as missing is a data-handling description, not evidence that missingness has no effect on the result. Because the endpoint result and the pattern of missing data are not reported, the impact cannot be assessed.

The registry record does not report a statistical analysis plan, interim analyses, or a sample-size justification. It lists enrollment of 33, but that listing alone does not establish how enrollment was chosen or whether the study was intended to detect a particular endpoint change. The available record therefore leaves both multiplicity handling and the statistical consequences of missingness unresolved.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% 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 (13.3%), device related infection (9.5%), pulmonary hypertension (9.5%), abdominal pain (6.7%), adenoidectomy (6.7%).

The safety information is grouped by previous bosentan exposure, bosentan-naive status, and a single-arm total. For each grouping, the record gives participants at risk and affected for serious and other adverse events. In the previous-bosentan grouping, 15 were at risk and 9 were affected for serious events; 15 were at risk and 12 were affected for other events. In the bosentan-naive grouping, 21 were at risk and 9 were affected for serious events; 21 were at risk and 16 were affected for other events. The total grouping reports 36 at risk and 18 affected for serious events, and 36 at risk and 26 affected for other events.

The total grouping’s 36 at risk differs from the enrollment listing of 33; the registry does not explain this. The categories should not be combined or reconciled by assumption. The listed serious-event terms include pulmonary arterial hypertension (13.3%), device related infection (9.5%), pulmonary hypertension (9.5%), abdominal pain (6.7%), adenoidectomy (6.7%), and bacteraemia (6.7%). The record does not provide further context for these term percentages in the listed information.

These counts describe events recorded during the study, not proof that treatment caused an event. Deaths are not reported for these groupings. The exposure labels are not randomized comparison arms.

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

The registry record establishes that FUTURE 2 was an open-label, single-group extension study of pediatric-formulation bosentan in children with idiopathic or familial pulmonary arterial hypertension who completed FUTURE 1. It specifies the dose, a possible dose reduction for tolerability, the follow-up timeframe, and a growth-related primary endpoint. These details describe what the study set out to observe.

The record does not report the primary endpoint result. It therefore does not show the median change in height-for-age Z-score or its full range. Without those values, readers cannot determine the observed direction or size of change, or how variable participants’ results were. There is also no reported confidence interval for judging precision.

Even if a before-and-after endpoint estimate were available, the design would not establish that bosentan caused the change. There is no concurrent control group, and the previous-bosentan and bosentan-naive safety labels are not a randomized comparison. Changes during follow-up could reflect multiple influences, and the registry does not report an analysis capable of separating them.

The adverse-event counts provide descriptive safety information, but do not establish causation. The registry does not report deaths in the safety groupings, and it does not explain why the total safety grouping’s at-risk count differs from the enrollment listing. That discrepancy limits straightforward reconciliation of the record’s counts.

Accordingly, the posted information supports understanding the study’s design and the way its primary endpoint was defined. It does not support a numerical conclusion about the growth endpoint, a controlled treatment-effect estimate, or a causal interpretation of the adverse-event listings. Missingness, multiplicity, and the statistical analysis plan are also not described in enough detail to resolve those questions.

13. Why This Trial Matters Statistically

This study illustrates the distinction between an endpoint definition and an endpoint result. A detailed description of how height measurements were converted to Z-scores explains what was intended to be measured, but does not tell readers what the observed change was. Since the registry record does not report the median or full range, the primary result cannot be interpreted quantitatively.

It also illustrates the limits of single-group follow-up. Repeated measurements can describe participants’ outcomes over time, but they do not provide the same evidence as a randomized comparison with a concurrent control. A reference standard helps contextualize height measurements; it is not a control group for estimating the effect of treatment.

The record’s safety categories are another useful statistical distinction. Previous exposure and bosentan-naive status describe group labels in the adverse-event record, but do not make those groups interchangeable or randomized. Counts can be informative as descriptions while remaining insufficient for causal comparisons. The unexplained difference between the listed enrollment and the safety total further emphasizes the need to keep registry entries in their stated context.

Finally, the record shows why interpretation depends on reporting of analysis methods and missing data, not just trial design. The all-treated population and missing-data rule are described, but the amount and pattern of missingness, inferential methods, multiplicity handling, and analysis plan are not reported. The absence of these details constrains conclusions even when an endpoint is clearly defined.

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 primary endpoint result is not reported in the registry record. Although the endpoint is defined as change in height-for-age Z-score, with median and full range as summary statistics, those summaries are absent. There is no reported endpoint confidence interval or other measure of precision.

The single-group, open-label design has no concurrent comparator, and the previous-bosentan and bosentan-naive labels do not form a randomized control comparison. Any descriptive changes over the follow-up period cannot be attributed uniquely to bosentan on the basis of this design alone.

The record says missing or incomplete endpoint data were treated as missing, but does not report the extent, timing, or reasons for missing data. It does not describe sensitivity analyses or explain how incomplete follow-up might affect the endpoint summary. The statistical analysis plan, multiplicity approach, interim-analysis details, and sample-size justification are also not reported.

The registry lists enrollment of 33, while the safety total grouping reports 36 at risk. The record does not explain the difference. Deaths are not reported in the safety groupings. These limitations are specific to the information available in the registry record; they do not establish what may have been reported elsewhere.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Bosentan in Children With Pulmonary Arterial Hypertension Extension Study — NCT00319020.
• Registry record last updated 2025-02-04; status last verified 2025-01.
• Results first posted 2017-03-29.
• 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

This completed Phase 3 study was an open-label, single-group extension of pediatric bosentan in children with idiopathic or familial pulmonary arterial hypertension who completed FUTURE 1. The registry lists enrollment of 33 and identifies long-term safety and tolerability as the main objective. Bosentan was administered at 4 mg/kg twice daily, with possible down-titration to 2 mg/kg twice daily if not well tolerated.

The primary endpoint was change from baseline to end of study in height-for-age Z-score, summarized by median and full range. The registry describes the growth-standard calculation and the all-treated analysis population, but does not report the endpoint result. It lists 8 outcome measures and 0 analyses; the analysis plan, multiplicity handling, and sample-size justification are not reported.

Safety counts are listed by prior exposure grouping and as a single-arm total. The total safety grouping reports 36 at risk, a figure the registry does not reconcile with the enrollment listing. The record does not report deaths in those groupings.