This page separates reported trial information from statistical interpretation. The numerical trial information on this page is restricted to the ClinicalTrials.gov record data. The registry reports results for the registered endpoints, but the ClinicalTrials.gov record does not provide formal statistical analyses with effect estimates, confidence intervals, or p-values.
1. Trial at a Glance
TOMORROW was a completed, randomized, parallel-group, phase 3 study evaluating whether macitentan delays disease progression in children with pulmonary arterial hypertension. The ClinicalTrials.gov record describes two study arms, 165 enrolled participants, and four registered primary endpoints centered on observed steady-state trough plasma concentrations of macitentan and aprocitentan at specified time points and pediatric age or body-weight groups.
| Feature | TOMORROW |
|---|---|
| Trial name | TOMORROW |
| ClinicalTrials.gov identifier | NCT02932410 |
| Condition | Pulmonary Arterial Hypertension |
| Therapeutic area | Pulmonology |
| Phase | Phase 3 |
| Status | Completed |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None |
| Primary purpose | Treatment |
| Enrollment | 165 |
| Interventions | Macitentan (drug); Standard-of-care (other) |
| Lead sponsor | Actelion |
| Sponsor type | Industry |
2. Clinical Question
The registry's brief title describes the clinical objective as assessing whether macitentan delays disease progression in children with pulmonary arterial hypertension. The ClinicalTrials.gov record, however, identify the four registered primary endpoints as observed steady-state trough plasma concentrations rather than a time-to-disease-progression comparison.
Population
Children with pulmonary arterial hypertension enrolled in the TOMORROW study, including specified age groups and a China cohort identified in the registered endpoints and safety data.
Intervention
Macitentan, identified in the registry as macitentan (JNJ-67896062) in the reported safety data.
Comparator
Standard of care, identified in the registry as Standard of Care (SoC; ≥2 Years) and a specified standard-of-care group that crossed over to macitentan.
Primary statistical question
The ClinicalTrials.gov record defines four primary pharmacokinetic outcome measures based on observed steady-state trough plasma concentrations at Week 12 or Week 4 in specified pediatric groups.
3. Trial Design
Macitentan
- Macitentan is the registered drug intervention.
- The reported safety data identify a macitentan group for participants ≥2 years: 26 serious-AE-affected participants among 72 at risk.
- A separate macitentan group for participants <2 years is reported as 4 serious-AE-affected participants among 9 at risk.
Standard of care
- Standard of care is the registered comparator intervention.
- The reported ≥2-year standard-of-care group had 21 serious-AE-affected participants among 75 at risk.
- A separate standard-of-care group with crossover to macitentan is reported as 0 serious-AE-affected participants among 8 at risk.
Study timeline
Study start
The ClinicalTrials.gov record identifies October 24, 2017 as the study start date.
Primary completion
The ClinicalTrials.gov record identifies August 28, 2024 as the primary completion date.
Current trial status in the ClinicalTrials.gov record
The trial is listed as completed, with results posted and 23 outcome measures reported.
4. Registered Primary Endpoints
The four registered primary endpoints are all observed steady-state trough plasma concentration measures. Three are specified by age and time point, while one additionally specifies participants from China aged ≥12 to <18 years.
| Primary endpoint | Time frame | Registry characterization |
|---|---|---|
| Participants ≥2 Years of Age: Observed Steady-State Trough (Pre-dose) Plasma Concentration of Macitentan and Aprocitentan (Active Metabolite) at Week 12 Based on Body Weight | Pre-dose at Week 12 | Observed concentrations reported; planned for specified arms only. |
| Participants ≥2 Years of Age: Observed Steady-State Trough (Pre-dose) Plasma Concentration of Macitentan and Aprocitentan (Active Metabolite) at Week 12 Based on Age Group | Pre-dose at Week 12 | Observed concentrations reported; planned for specified arms only. |
| Participants <2 Years of Age: Observed Steady-State Trough (Pre-dose) Plasma Concentration of Macitentan and Aprocitentan (Active Metabolite) at Week 4 | Pre-dose at Week 4 | Observed concentrations reported; planned for specified arms only. |
| Participants From China With ≥12 to <18 Years of Age: Observed Steady-State Trough (Pre-dose) Plasma Concentration of Macitentan and Aprocitentan (Active Metabolite) at Week 12 | Pre-dose at Week 12 | Observed concentrations reported. |
5. Results
ClinicalTrials.gov reports that results were posted for the trial and that all four registered primary endpoints have results posted. However, no formal statistical analyses were posted. Consequently, the available data do not provide formal statistical comparison estimates, confidence intervals, or p-values for these primary endpoints.
| Primary endpoint | Results posted? | Formal statistical analysis reported? |
|---|---|---|
| ≥2 years, Week 12, based on body weight | Yes | No |
| ≥2 years, Week 12, based on age group | Yes | No |
| <2 years, Week 4 | Yes | No |
| China, ≥12 to <18 years, Week 12 | Yes | No |
What can be concluded from the ClinicalTrials.gov record field
The registry indicates that observed steady-state trough concentrations were reported for the four registered primary endpoints.
No numerical concentration estimates, confidence intervals, p-values, or formal between-group effect estimates are included in the ClinicalTrials.gov record.
The phrase results posted establishes that the registry contains outcome information, but it does not by itself establish that a statistically significant treatment difference was demonstrated. In the ClinicalTrials.gov record, there are no formal statistical-analysis entries to quantify a comparison.
For a concentration endpoint, the clinically relevant question is not simply whether an observed value exists. The analysis should describe the distribution of concentrations and clearly identify the population, time point, age or body-weight grouping, and treatment exposure represented by each observation.
Because no effect estimate or confidence interval is reported, there is no defensible numerical basis here for stating that macitentan produced a particular percentage increase or decrease in concentration relative to standard of care.
Why no conventional treatment-effect table is shown
A conventional randomized efficacy table might report a difference, ratio, hazard ratio, odds ratio, confidence interval, and p-value. The registry-reported TOMORROW data do not provide those quantities for the four primary endpoints. Reconstructing them would require participant-level or adequately summarized concentration data that are not included in the ClinicalTrials.gov record.
6. Planned Analysis
Because the registry reports results but does not provide formal statistical analyses in the ClinicalTrials.gov record, this section explains how endpoints of this type would ordinarily be approached without attributing an unreported method to the TOMORROW registry.
Observed steady-state trough concentration
Steady-state trough plasma concentration is a continuous pharmacokinetic measurement. A descriptive analysis would ordinarily summarize the observed concentration distribution separately by the prespecified population, treatment group, time point, and relevant age or body-weight category.
Concentration data are often right-skewed, making logarithmic summaries useful when the distribution supports that approach. A geometric mean describes the central tendency on the multiplicative scale; it is not the arithmetic mean.
Comparison of treatment groups
If the scientific objective is to compare macitentan exposure with a comparator, a common approach for positive continuous concentration data is analysis on the logarithmic scale followed by presentation of a ratio of geometric means. The exact model, covariates, handling of below-quantification-limit observations, and estimand would need to come from the study's statistical analysis plan or a sufficiently detailed registry result record.
Age and body-weight groupings
The registered endpoint definitions explicitly distinguish analyses based on age group and body weight. These categories should therefore be preserved rather than collapsing all pediatric participants into a single summary. Pediatric pharmacokinetics can vary with maturation and body size, so the grouping structure is part of the scientific interpretation of the endpoint.
Different sampling times
The primary endpoints use pre-dose Week 12 measurements for participants ≥2 years of age and participants in the specified China cohort, while the <2-year endpoint uses pre-dose Week 4. These measurements should not be treated as interchangeable observations from a single time point.
7. Statistical Methodology
Randomization
The trial allocation is identified as randomized. Randomization is important because it establishes a design-based framework for comparing treatment groups, even when the eventual analysis is focused on pharmacokinetic measurements rather than a clinical time-to-event endpoint.
Parallel-group design
The design model is identified as parallel. In a parallel-group trial, participants are assigned to a treatment strategy and analyzed according to that randomized assignment rather than serving as their own randomized crossover comparison.
Continuous pharmacokinetic outcomes
The actual endpoint definitions concern observed plasma concentrations. These are quantitative measurements, so their statistical description should preserve their continuous nature. The registry's inferred endpoint-type field says "Binary," but the wording of the endpoint itself describes concentration measurements. For statistical interpretation, the measurement definition is more informative than an inferred categorical label.
A ratio of 1 would represent equal geometric means. A ratio above 1 would indicate a higher geometric mean concentration in the numerator group, while a ratio below 1 would indicate a lower geometric mean concentration. No such ratio is reported in the registry-reported TOMORROW data.
Age and body-weight analysis
The primary endpoints explicitly organize observations by age group and body weight. A statistically appropriate analysis should preserve those prespecified distinctions rather than treating pediatric age and body size as incidental descriptors.
Time-specific assessment
The registered time frames are precise: pre-dose at Week 12 for three primary endpoints and pre-dose at Week 4 for the endpoint involving participants <2 years of age. A Week 4 concentration and a Week 12 concentration answer different pharmacokinetic questions and should not be combined without an explicit analytical rationale.
8. Safety Results
The ClinicalTrials.gov record reports serious adverse events by arm as the number affected over the number at risk. These data can be presented descriptively, but the ClinicalTrials.gov record does not provide a formal comparative analysis of serious adverse-event risk.
| Reported group | Serious AEs affected / at risk |
|---|---|
| Macitentan (JNJ-67896062) (≥2 Years) | 26/72 |
| Macitentan (JNJ67896062) (<2 Years) | 4/9 |
| Standard of Care (SoC; ≥2 Years) | 21/75 |
| SOC (≥2 Years) Crossover to Macitentan | 0/8 |
| China Single-arm Cohort (≥12 to <18 Years) | 1/8 |
Macitentan ≥2 years
Serious adverse events affected 26 of 72 participants at risk in the reported group.
Standard of care ≥2 years
Serious adverse events affected 21 of 75 participants at risk in the reported group.
Macitentan <2 years
Serious adverse events affected 4 of 9 participants at risk in the reported group.
Crossover group
The reported standard-of-care group that crossed over to macitentan had 0 of 8 participants affected by a serious adverse event.
9. Crossover and Analysis Structure
The ClinicalTrials.gov record explicitly identify a Standard of Care (≥2 Years) Crossover to Macitentan group containing 8 participants at risk. This confirms that crossover is part of the reported trial data.
Crossover creates an important distinction between the treatment assigned at randomization and the treatment actually received later. An analysis that compares participants strictly according to randomized assignment preserves the original randomized comparison, whereas an analysis based on treatment actually received can introduce post-randomization selection effects.
| Design feature | What the ClinicalTrials.gov record establishes | Statistical implication |
|---|---|---|
| Randomization | Allocation is randomized. | Supports comparison based on assigned treatment. |
| Parallel design | Design model is parallel. | Groups are followed as parallel treatment strategies. |
| No masking | Masking is none. | Knowledge of treatment assignment may affect some post-randomization measurements or clinical assessments. |
| Crossover | A standard-of-care group crossing over to macitentan is reported. | Later exposure differs from original assignment and complicates treatment-received interpretations. |
10. Statistical Methods Explained
Why does randomization matter when the primary outcomes are concentrations?
Randomization determines how treatment assignment is generated before subsequent observations occur. Even when the endpoint is pharmacokinetic, randomized allocation provides the design framework for comparing groups. The strength of that comparison still depends on the population analyzed, the measurement process, and the statistical model used.
Why are trough concentrations measured before a dose?
A pre-dose measurement is intended to capture the concentration immediately before the next scheduled dose. At steady state, the trough can be used as a standardized exposure measure because the concentration is observed at a consistent point in the dosing interval.
What does "steady-state trough" mean?
Steady state refers to the condition in which repeated dosing has produced a stable pattern of drug exposure over successive dosing intervals. A trough is the concentration near the end of the dosing interval, immediately before the next dose. The TOMORROW primary endpoints specifically use observed steady-state trough concentrations.
Why might geometric means be useful for concentration data?
Drug concentrations commonly have multiplicative rather than purely additive variation and can be right-skewed. Log transformation can make such distributions more suitable for statistical modeling. Exponentiating the mean on the log scale produces the geometric mean.
What would a ratio of geometric means mean?
A ratio compares the geometric mean concentration in one group with that in another. A ratio of 1 would indicate equal geometric means. A ratio of 1.20 would, in general, describe a 20% higher geometric mean in the numerator group, while a ratio of 0.80 would describe a 20% lower geometric mean. TOMORROW's the ClinicalTrials.gov record do not report such a ratio.
Why should Week 4 and Week 12 results remain separate?
The registered endpoints specify different time frames for different age groups. Combining Week 4 and Week 12 measurements into a single summary would remove a design distinction built into the endpoint definitions and could obscure differences in the exposure state being measured.
Why should the crossover group be treated separately?
A participant who changes treatment after initial assignment no longer has the same subsequent exposure pattern as a participant who remains on the original strategy. This means treatment assignment and treatment actually received become different variables. The distinction is especially important when interpreting later observations.
11. Missing Data and Measurement Considerations
The ClinicalTrials.gov record does not specify a missing-data or imputation method. Therefore, no particular imputation strategy should be attributed to TOMORROW.
Observed versus imputed values
The primary endpoint wording refers to observed steady-state trough concentrations. An observed-value analysis does not automatically imply that missing measurements were imputed.
Below-quantification-limit observations
The ClinicalTrials.gov record does not state how concentrations below a laboratory quantification limit were handled. That rule can materially affect pharmacokinetic summaries and should come from the prespecified analysis documentation.
For pediatric pharmacokinetic data, the completeness and timing of sampling are especially important because a concentration measured outside the intended pre-dose window may not represent the registered endpoint as defined.
12. Multiplicity and Multiple Primary Endpoints
The registry identifies four primary endpoints. The ClinicalTrials.gov record does not provide an alpha-allocation scheme, hierarchical testing procedure, multiplicity adjustment, or formal hypothesis-testing results.
| Multiplicity question | Information in the ClinicalTrials.gov record |
|---|---|
| Number of registered primary endpoints | 4 |
| Multiplicity adjustment reported | Not reported |
| Alpha allocation reported | Not reported |
| Formal p-values in the posted results | None reported |
| Formal confidence intervals in the posted results | None reported |
When multiple primary endpoints are tested formally, the statistical design normally needs to specify how the overall type I error is controlled. Without the relevant prespecified testing procedure, the presence of four primary endpoints should not be interpreted as four independent confirmatory opportunities.
13. Non-Inferiority, Interim Analysis, and Bayesian Methods
The ClinicalTrials.gov record does not report a non-inferiority margin, an interim-analysis strategy, an alpha-spending procedure, or Bayesian methods. These topics therefore cannot be attributed to TOMORROW on the basis of the ClinicalTrials.gov record.
| Design topic | Supported by the ClinicalTrials.gov record? | Interpretation |
|---|---|---|
| Non-inferiority margin | No | No margin is reported. |
| Interim analysis | No | No interim-analysis method is reported. |
| Multiplicity procedure | No | No formal adjustment procedure is reported. |
| Bayesian methods | No | No Bayesian analysis is reported. |
| Randomization | Yes | Allocation is identified as randomized. |
| Crossover | Yes | A standard-of-care crossover-to-macitentan group is reported in the safety data. |
14. Serious Adverse Events: Statistical Perspective
Serious adverse events are presented as affected participants over the number at risk. These proportions can be described as observed event frequencies within the reported groups, but the ClinicalTrials.gov record does not include confidence intervals or a formal between-group hypothesis test.
The registry supplies the numerator and denominator for each reported group. This permits a descriptive proportion, but a formal comparative estimate would require a prespecified statistical analysis and appropriate handling of the different cohort structures.
The <2-year macitentan group contains 9 participants at risk, while the ≥2-year macitentan group contains 72. The much smaller denominator in the younger group means that its observed serious-adverse-event experience is inherently less precise as a description of the broader pediatric population.
15. Limitations
- No formal statistical analyses reported: the ClinicalTrials.gov record contains no formal statistical analyses, so treatment-effect estimates, confidence intervals, and p-values cannot be reported.
- Primary endpoints are concentration-based: the registered primary endpoints concern observed steady-state trough plasma concentrations rather than a reported formal disease-progression comparison.
- Endpoint-type inconsistency: the registry-reported inferred endpoint type is labeled "Binary," while the actual registered endpoint definitions describe quantitative plasma concentrations. The endpoint wording should therefore guide statistical interpretation.
- Different pediatric groups: the primary endpoints distinguish participants ≥2 years, participants <2 years, age groups, body-weight categories, and a specified China cohort. These populations should not be collapsed without justification.
- Different time frames: the registered primary endpoints use pre-dose Week 12 or pre-dose Week 4, so concentration measurements are not all interchangeable.
- Crossover: the ClinicalTrials.gov record identifies a standard-of-care group that crossed over to macitentan. Subsequent exposure therefore differs from initial treatment assignment for some participants.
- Unreported missing-data methods: the ClinicalTrials.gov record does not describe imputation or rules for handling missing or below-quantification-limit concentrations.
- Unreported multiplicity procedure: four primary endpoints are listed, but no multiplicity-control strategy is included in the registry-reported statistical data.
- Safety denominator differences: serious-adverse-event groups have different denominators and cohort structures, limiting direct descriptive comparison across every reported group.
- No formal disease-progression result: although the brief title describes whether macitentan delays disease progression, the ClinicalTrials.gov record does not provide a formal disease-progression treatment comparison.
16. Why This Trial Matters Statistically
TOMORROW is useful as a statistical teaching case because it illustrates an important distinction between a trial's broad clinical question and the specific registered endpoints used to characterize the intervention.
| Concept | How it appears in TOMORROW |
|---|---|
| Randomization | The allocation is randomized. |
| Parallel-group design | The design model is parallel. |
| Unmasked treatment | Masking is listed as none. |
| Pharmacokinetic endpoints | Four registered primary endpoints measure observed steady-state trough plasma concentrations. |
| Age-specific analysis | Primary endpoints distinguish participants ≥2 years and <2 years and include an endpoint based on age group. |
| Body-weight analysis | One primary endpoint specifies analysis based on body weight. |
| Different assessment times | Primary endpoints use pre-dose Week 12 or pre-dose Week 4. |
| Crossover | A standard-of-care group crossing over to macitentan is reported. |
| Multiplicity | Four registered primary endpoints are identified, without a registry-reported multiplicity procedure. |
| Safety denominators | Serious adverse events are reported as affected/at-risk counts across several distinct groups. |
17. Statistical Methods Explained: A Deeper View
Observed concentration is not the same as treatment effect
A concentration observation describes what was measured in a participant at a specified time. A treatment effect compares distributions or model-based summaries between groups. The registry-reported TOMORROW data establish that observed concentrations were reported, but they do not supply the comparative statistical estimates needed to quantify a treatment effect.
Why log transformation is often considered
If concentration measurements are positively skewed, taking logarithms can reduce skewness and make multiplicative differences easier to model. A treatment contrast on the log scale can then be transformed back into a ratio of geometric means. This is a general pharmacokinetic principle, not a claim that this exact model was used in TOMORROW.
Why pediatric subgroup definitions matter
Age and body weight are not merely descriptive labels in pediatric pharmacokinetic research. They can correspond to meaningful differences in drug exposure. TOMORROW explicitly embeds these dimensions in its registered primary endpoints, so the statistical interpretation should preserve those distinctions.
Why a confidence interval would be valuable
A confidence interval would quantify uncertainty around an estimated concentration ratio or difference. It would help distinguish a precise estimate from one driven by a small amount of information. The ClinicalTrials.gov record does not contain such intervals, so no precision statement can be attached to a numerical treatment contrast.
Why a p-value would not measure pharmacokinetic magnitude
Even if a p-value were available, it would address evidence against a specified null hypothesis rather than quantify the magnitude of the exposure difference. A concentration ratio or difference, together with its confidence interval, would be necessary to understand the size and precision of the estimated effect.
18. Interpretation of the Reported Safety Data
The reported serious-adverse-event counts are descriptive observations within several defined cohorts. For example, the ≥2-year macitentan group contains 26 affected participants among 72 at risk, while the ≥2-year standard-of-care group contains 21 among 75 at risk.
These counts do not, by themselves, establish that the difference between groups is statistically significant, clinically important, or causally attributable to treatment. A formal comparison would require a specified estimand, analysis population, and statistical method.
The reported groups have different denominators: 72, 9, 75, 8, and 8. Comparisons involving the smaller cohorts are particularly sensitive to individual observations, and the China single-arm cohort is structurally different from a randomized comparator group.
The presence of a standard-of-care group that crossed over to macitentan means that treatment assignment and later treatment exposure are not necessarily identical. A treatment-received analysis could therefore answer a different question from the original randomized comparison.
19. Clinical Interpretation vs Statistical Interpretation
Statistical interpretation
The ClinicalTrials.gov record describes a randomized, parallel, unmasked phase 3 study with four primary concentration-based endpoints. Formal statistical-analysis entries are not reported, so comparative effect estimates cannot be quantified from the ClinicalTrials.gov record.
Clinical interpretation
The study was designed around macitentan in children with pulmonary arterial hypertension, with primary pharmacokinetic assessments focused on observed steady-state trough concentrations across specified pediatric populations and time points.
20. What the Registry Data Do — and Do Not — Establish
| Question | Supported by the ClinicalTrials.gov record? |
|---|---|
| Was TOMORROW randomized? | Yes. |
| Was the design parallel? | Yes. |
| Was the study masked? | No; masking is listed as none. |
| Was enrollment 165? | Yes. |
| Were four primary endpoints registered? | Yes. |
| Were results posted for the primary endpoints? | Yes. |
| Are formal statistical analyses reported? | No. |
| Are treatment-effect confidence intervals reported? | No. |
| Are treatment-effect p-values reported? | No. |
| Can a formal treatment effect be calculated from the ClinicalTrials.gov record alone? | No. |
| Are serious adverse-event counts by reported group reported? | Yes. |
| Is crossover explicitly represented in the ClinicalTrials.gov record? | Yes. |
21. Sources
- ClinicalTrials.gov: TOMORROW, NCT02932410.
- Linked publication: PubMed record for PMID 41796854: PubMed 41796854.
Continue through the Clinical Biostats learning pathway
Use the trial's design and pharmacokinetic endpoints as a starting point for deeper study of randomization, concentration analysis, pediatric grouping, and clinical-trial statistics.