This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. The ClinicalTrials.gov record identifies a posted primary outcome measure but do not provide a formal statistical-analysis result for that endpoint.
1. Trial at a Glance
ASCEND was a completed randomized phase 3 parallel trial evaluating pirfenidone versus placebo in patients with idiopathic pulmonary fibrosis. The registry reports 555 enrolled participants, 2 arms, quadruple masking, and a binary primary endpoint based on change in percent predicted forced vital capacity from baseline to Week 52.
| Feature | ASCEND |
|---|---|
| Trial name | ASCEND |
| NCT identifier | NCT01366209 |
| Brief title | Efficacy and Safety of Pirfenidone in Patients With Idiopathic Pulmonary Fibrosis (IPF) |
| Therapeutic area | Pulmonology |
| Condition | Idiopathic Pulmonary Fibrosis |
| Phase | Phase 3 |
| Status | Completed |
| Enrollment | 555 |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Primary purpose | Treatment |
| Lead sponsor | Genentech, Inc. |
| Sponsor type | Industry |
2. Clinical Question
The central clinical-trial question was whether pirfenidone differed from placebo on the registered primary endpoint, change in percent predicted forced vital capacity from baseline to Week 52, in patients with idiopathic pulmonary fibrosis.
Population
Patients with idiopathic pulmonary fibrosis (IPF).
Intervention
Pirfenidone.
Comparator
Placebo.
Primary question
How does the randomized comparison of pirfenidone versus placebo perform on change in percent predicted forced vital capacity from baseline to Week 52?
3. Trial Design
Pirfenidone
- Intervention type: drug
- Randomized active-treatment arm
Placebo
- Intervention type: drug
- Randomized comparator arm
The registry describes the allocation as randomized, the design model as parallel, and the masking level as quadruple. These features are important statistically because randomization establishes the basis for a comparative treatment effect, while masking is intended to reduce the influence of treatment knowledge on trial conduct and assessment.
4. Endpoints
The registry identifies 1 primary endpoint and classifies the posted primary endpoint type as binary.
| Endpoint | Registry definition | Time frame | Endpoint type |
|---|---|---|---|
| Primary endpoint | Change in Percent Predicted Forced Vital Capacity (%FVC) From Baseline to Week 52 | 52 weeks | Binary |
5. Primary Endpoint: What Was Measured?
The primary endpoint was Change in Percent Predicted Forced Vital Capacity (%FVC) From Baseline to Week 52, assessed over a 52-week time frame.
The registry's endpoint is expressed as change from baseline to a specified follow-up time. Because the ClinicalTrials.gov record classifies the endpoint as binary, the final analysis would use the prespecified binary outcome definition rather than infer a continuous-scale comparison from the endpoint name alone.
This distinction is important. An endpoint can be based on a continuous measurement while the trial's primary analysis treats the resulting patient-level classification as binary. The statistical question then becomes the comparison of the proportion of patients meeting the prespecified endpoint definition between randomized groups.
6. Statistical Methodology
Binary endpoint analysis
For a binary primary endpoint, a typical randomized comparison would summarize the number and percentage of participants meeting the endpoint definition in each treatment arm and compare those proportions using a prespecified statistical test or regression model.
Depending on the statistical analysis plan, possible measures of treatment effect could include a risk difference, risk ratio, odds ratio, or another prespecified effect measure. The registry-reported ASCEND trial data do not identify which effect measure was used for the primary analysis.
A formal treatment comparison would then quantify the difference between these probabilities according to the prespecified analysis method.
Why randomization matters
Randomization creates the foundation for comparing outcomes between the pirfenidone and placebo groups. In a properly conducted randomized trial, treatment assignment is determined independently of participants' prognostic characteristics. This makes the between-group comparison the central source of evidence about the treatment effect.
Why quadruple masking matters
Quadruple masking means that the trial was registered as having masking at four levels. Masking can reduce opportunities for knowledge of treatment assignment to influence trial behavior, outcome assessment, treatment administration, or other study processes. The ClinicalTrials.gov record does not specify the four masked roles individually, so they should not be inferred.
Week 52 as a fixed assessment time
The registered primary endpoint has a defined 52-week time frame. A fixed-time binary endpoint differs from a time-to-event endpoint: the principal comparison is whether participants meet the endpoint definition at the specified assessment point, rather than modeling the time until an event occurs.
7. Planned Analysis
The registry identifies one posted primary outcome measure: Change in Percent Predicted Forced Vital Capacity (%FVC) From Baseline to Week 52, with a 52-week time frame and a binary endpoint classification.
For a binary endpoint of this type, an appropriate formal analysis would normally compare the proportions meeting the prespecified endpoint definition between the pirfenidone and placebo groups. The choice between a chi-square-type test, Fisher's exact test, logistic regression, or another method would depend on the prespecified statistical analysis plan, sample-size assumptions, covariates, and endpoint definition. Those details are not contained in the ClinicalTrials.gov record.
8. Results Reporting Status
The ClinicalTrials.gov record states Results Posted: Yes and Outcome Measures Posted: 1. However, the ClinicalTrials.gov recordset contains no entries for the primary endpoint and reports 0 analyses omitted for length.
| Result component | Status in the ClinicalTrials.gov record |
|---|---|
| Results posted | Yes |
| Outcome measures posted | 1 |
| Formal statistical analyses reported | None |
| Analyses omitted for length | 0 |
| Primary endpoint | Change in %FVC from baseline to Week 52 |
| Formal treatment-effect estimate reported | Not provided |
| Confidence interval reported | Not provided |
| P-value reported | Not provided |
This distinction prevents a common statistical reporting error: treating the statement that results are posted as equivalent to having a complete formal analysis table available in the ClinicalTrials.gov record. For this page, the latter is absent.
9. Safety
The ClinicalTrials.gov record reports serious adverse events by randomized arm as affected participants divided by participants at risk.
| Safety measure | Affected / at risk |
|---|---|
| Active Arm | 55/278 |
| Placebo Arm | 69/277 |
Active arm
55/278 participants were affected by the reported serious adverse-event measure.
Placebo arm
69/277 participants were affected by the reported serious adverse-event measure.
These are the serious-adverse-event counts posted on ClinicalTrials.gov for the two arms. They should not be converted into an inferred comparative risk estimate here because the task requires reporting the registry-reported numbers exactly and the registry data do not provide a formal statistical analysis of this safety measure.
The serious-adverse-event information provides an important safety description alongside the efficacy endpoint. However, a raw affected/at-risk count does not by itself establish whether the observed difference is statistically significant, clinically important, or causally attributable to treatment. A formal comparison would require the prespecified safety analysis framework and appropriate uncertainty measures.
10. Statistical Methods Explained
Why was randomization important in ASCEND?
Randomization is the mechanism that makes the pirfenidone-versus-placebo comparison a controlled experiment. Rather than allowing treatment assignment to be determined by prognosis or clinician preference, random allocation is intended to balance prognostic factors on average across treatment groups. The primary inference therefore comes from comparing outcomes according to the assigned treatment groups.
What does quadruple masking contribute?
Quadruple masking is a design feature intended to limit knowledge of treatment assignment. When participants and relevant trial personnel do not know assignment, expectations and behavior can have less opportunity to influence study conduct. The ClinicalTrials.gov record identifies the masking level but does not specify the identities of the four masked parties.
Why is the primary endpoint described as binary?
The underlying measurement, percent predicted FVC, is naturally quantitative. But the registry classifies the registered primary endpoint as binary. That indicates that the primary analysis was based on whether participants satisfied a defined binary outcome rather than simply comparing every participant's raw continuous change as the primary endpoint. The ClinicalTrials.gov record does not provide the binary threshold.
How would a binary treatment comparison usually be interpreted?
Suppose the analysis defines a favorable endpoint as occurring or not occurring. The treatment effect can then be expressed through a difference in proportions, ratio of proportions, odds ratio, or another prespecified measure. The choice matters because these measures have different interpretations. The ASCEND ClinicalTrials.gov record do not specify which measure was used.
Why does the 52-week time frame matter?
A binary endpoint assessed at Week 52 asks a fixed-time question: what proportion of randomized participants satisfy the endpoint definition by the specified assessment? That is conceptually different from survival analysis, where the exact time to an event and censoring are modeled.
Why should no p-value be inferred here?
A p-value cannot be reconstructed from the trial name or endpoint description. It requires the observed data, the statistical test or model, and the prespecified hypotheses.
Why should the serious-adverse-event counts be kept separate from efficacy?
Efficacy and safety answer different questions. The primary endpoint concerns the prespecified change in %FVC, whereas serious adverse events describe an important safety outcome. Combining them into a single judgment would conceal the distinct statistical and clinical meanings of the two types of evidence.
11. Randomization and Analysis Interpretation
The registry specifies randomized allocation and a parallel design with 2 arms. This is particularly important for interpreting any eventual primary endpoint comparison because the treatment groups are defined by randomized assignment rather than by treatment actually received.
12. What Cannot Be Concluded From the Supplied Results
Because the registry-reported statistical-analysis section contains no formal analyses, several common trial-results statements would go beyond the available evidence.
- No numerical treatment effect for the primary %FVC endpoint can be reported.
- No confidence interval for the primary endpoint can be reported.
- No p-value for the primary endpoint can be reported.
- No subgroup treatment effects can be reported.
- No formal comparison of serious adverse-event rates can be reported.
- No median outcome or time-to-event result should be inferred because the registry-reported primary endpoint is classified as binary.
- No multiplicity adjustment, interim-analysis boundary, missing-data method, or covariate-adjusted model should be attributed to ASCEND unless it appears in the ClinicalTrials.gov record.
13. Trial Timeline
Trial start
The ASCEND trial began in June 2011 according to the ClinicalTrials.gov record.
Primary completion
The ClinicalTrials.gov record gives February 2014 as the primary completion date.
Primary endpoint time frame
The registered primary endpoint evaluates change in percent predicted FVC from baseline to Week 52.
14. Primary Endpoint: Statistical Interpretation
The primary endpoint concerns change in percent predicted forced vital capacity from baseline to Week 52. Because the registry classifies the endpoint as binary, the primary analysis concerns whether each participant meets the prespecified binary endpoint definition.
The endpoint name alone does not establish the direction or magnitude of a treatment effect. It also does not establish that a particular percentage-point difference, threshold, or responder rate was observed. Those quantities require the actual posted analysis data.
A treatment comparison should normally be accompanied by an uncertainty measure such as a confidence interval. Without the formal analysis in the ClinicalTrials.gov record, the precision of the treatment estimate cannot be evaluated.
Even when a p-value is available, it does not describe the magnitude of a treatment effect. Effect size and uncertainty should be interpreted directly. In this record, no formal p-value is reported, so none is reported.
15. Limitations
- Incomplete statistical-analysis detail: the ClinicalTrials.gov record contains no entries in the statistical analyses field.
- Posted results without formal estimates: the registry data state that results are posted, but the ClinicalTrials.gov record does not include the primary endpoint's treatment estimate, confidence interval, or p-value.
- Binary endpoint definition unavailable: the registry identifies the primary endpoint as binary but does not supply the threshold used to classify participants.
- No subgroup results: the ClinicalTrials.gov record does not contain subgroup analyses, so no subgroup inference is presented.
- No formal safety comparison: serious-adverse-event counts are available by arm, but no inferential safety analysis is reported.
- No missing-data method reported: the ClinicalTrials.gov record does not identify how missing Week 52 measurements were handled.
- No multiplicity information reported: the ClinicalTrials.gov record identifies one primary endpoint but does not provide a multiplicity strategy.
- No interim-analysis information reported: the ClinicalTrials.gov record does not identify whether interim statistical monitoring was used.
- No covariate model reported: the ClinicalTrials.gov record does not identify whether baseline %FVC or other covariates were included in the formal primary analysis.
16. Why This Trial Matters Statistically
ASCEND is useful as a statistical teaching case because it illustrates how a randomized, masked phase 3 trial can reduce bias while using a fixed-time physiologic endpoint. It also illustrates why an analyst must distinguish the underlying clinical measurement from the statistical form of the endpoint.
| Concept | How it appears in ASCEND |
|---|---|
| Randomization | Participants were randomized to 2 parallel treatment arms. |
| Blinding | The registry specifies quadruple masking. |
| Binary endpoint | The primary endpoint is classified as binary. |
| Change from baseline | The primary endpoint evaluates change in percent predicted FVC from baseline to Week 52. |
| Fixed assessment time | The primary endpoint uses a 52-week time frame. |
| Comparator-controlled design | Pirfenidone is compared with placebo. |
| Safety analysis | Serious adverse events are reported as affected participants divided by participants at risk by arm. |
| Evidence hierarchy | The distinction between a posted outcome measure and a registry-reported formal statistical analysis is essential for accurate interpretation. |
17. Related Tutorials
Learn more about the methods used in this trial:
18. Related Statistical Calculators
19. Sources
- ClinicalTrials.gov: ASCEND (NCT01366209).
- PubMed: PMID 39073523.
- PubMed: PMID 37391667.
- PubMed: PMID 35688625.
- PubMed: PMID 33434107.
- PubMed: PMID 31401786.
Continue through Clinical Biostats
Connect randomized-trial design concepts with statistical tutorials and clinical-trial analysis tools.
20. Record Summary
ASCEND was a completed randomized phase 3 parallel trial of pirfenidone versus placebo in idiopathic pulmonary fibrosis, with 555 enrolled participants, 2 treatment arms, and quadruple masking. Its registered primary endpoint was Change in Percent Predicted Forced Vital Capacity (%FVC) From Baseline to Week 52, classified in the ClinicalTrials.gov record as a binary endpoint.
The most important statistical limitation of the ClinicalTrials.gov record is the distinction between results being posted and a formal statistical analysis being available in the ClinicalTrials.gov record. The record states that results were posted and that one outcome measure was posted, but the statistical-analysis dataset contains no formal analysis entries.
The available safety information reports serious adverse events as 55/278 in the active arm and 69/277 in the placebo arm. These counts provide descriptive safety information but, without a registry-reported inferential analysis, do not establish a formal statistical difference.