This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.
1. Trial at a Glance
ORBIT-4 was a randomized, parallel, quadruple-masked phase 3 treatment study evaluating ciprofloxacin dispersion for inhalation against placebo in non-CF bronchiectasis. The registered primary endpoint was time to first pulmonary exacerbation from baseline over 48 weeks.
| Feature | ORBIT-4 |
|---|---|
| Phase | Phase 3 |
| Condition | Non Cystic Fibrosis Bronchiectasis |
| Design | Randomized, parallel-group, quadruple-masked |
| Primary purpose | Treatment |
| Enrollment | 304 |
| Number of arms | 2 |
| Interventions | Ciprofloxacin dispersion for inhalation; placebo |
| Primary endpoint | Time to first pulmonary exacerbation (from baseline) |
| Primary time frame | 48 weeks |
| Lead sponsor | Aradigm Corporation |
| Sponsor type | Industry |
| ClinicalTrials.gov | NCT02104245 |
2. Clinical Question
The central clinical question was whether ciprofloxacin dispersion for inhalation, compared with placebo, affects the time from baseline to the first pulmonary exacerbation in patients with non-CF bronchiectasis during a 48-week observation period.
Population
Patients with Non Cystic Fibrosis Bronchiectasis enrolled in a phase 3 treatment study.
Intervention
Ciprofloxacin dispersion for inhalation.
Comparator
Placebo.
Primary question
Does the intervention alter the time to first pulmonary exacerbation from baseline over 48 weeks?
3. Trial Design
Ciprofloxacin dispersion for inhalation
- Intervention classified as a drug.
- Evaluated in the randomized phase 3 comparison.
Placebo
- Comparator classified as a drug.
- Evaluated in parallel with the intervention.
Study timing
Study start
The registry lists May 28, 2014 as the study start date.
Primary completion
The registry lists August 11, 2016 as the primary completion date.
Current registry status
The ClinicalTrials.gov record identifies the study status as completed.
4. Endpoints
| Endpoint | Registry definition | Time frame |
|---|---|---|
| Primary endpoint | Time to first pulmonary exacerbation (from baseline) | 48 weeks |
The primary endpoint is explicitly a time-to-event endpoint. Unlike a simple binary endpoint that records whether an exacerbation occurred by a fixed date, time-to-first-event analysis uses the timing of the first event and can also accommodate participants whose first event has not occurred by the end of their observable follow-up.
5. Statistical Methodology
Why this is a time-to-event problem
The primary endpoint is measured as the elapsed time from baseline until the first pulmonary exacerbation. That structure is fundamentally different from comparing two proportions at a single fixed assessment. Participants may have their first event at different times, while others may reach the end of the 48-week time frame without a recorded first event.
A time-to-event framework allows the analysis to use the timing of observed events while accounting for right-censored observations. For example, a participant who has not experienced a first pulmonary exacerbation by the end of their observed follow-up has not reported an exact event time; their available follow-up still contributes information to the survival distribution up to the censoring point.
Typical Kaplan-Meier analysis
For a time-to-first-exacerbation endpoint, a standard descriptive approach is the Kaplan-Meier estimator. It estimates the probability of remaining free of the first event over time.
Here, di represents the number of first events occurring at time ti, while ni is the number of participants at risk immediately before that time.
For ORBIT-4, the Kaplan-Meier curve would provide a graphical description of the estimated probability of remaining free of a first pulmonary exacerbation throughout the 48-week time frame. The vertical separation between treatment curves, if present, would describe differences in event-free experience over time rather than simply differences in the percentage experiencing an event at week 48.
Typical treatment comparison
A randomized two-arm time-to-event endpoint is commonly compared using a log-rank test. The log-rank framework evaluates whether the observed event-time distributions differ between the randomized treatment groups over follow-up.
A Cox proportional-hazards model is also commonly used to estimate a hazard ratio. In a simple two-group setting, the hazard ratio compares the estimated instantaneous event rates between groups over the analyzed period under the proportional-hazards model.
A hazard ratio is a relative time-to-event measure. It is not a percentage of participants who experience an exacerbation, not a difference in median time, and not an absolute risk difference.
6. Statistical Methods Explained
What makes the primary endpoint a survival-analysis endpoint?
The defining feature is the inclusion of time in the endpoint. “Time to first pulmonary exacerbation” records when the first event occurs, rather than only whether an event occurred. This makes methods designed for censored time-to-event data appropriate for the endpoint.
Why would Kaplan-Meier estimation be useful?
Kaplan-Meier estimation describes the probability of remaining free of the first pulmonary exacerbation as follow-up progresses. It uses the observed timing of events and retains information from participants whose event time is censored.
What would a log-rank test ask?
A log-rank test would assess whether the overall event-time experience differs between the ciprofloxacin-dispersion-for-inhalation group and the placebo group. It is a comparison of the time-to-event distributions rather than a comparison of means.
What would a hazard ratio mean in this trial?
If a Cox model were used, the hazard ratio would summarize the relative instantaneous rate of first pulmonary exacerbation between the two randomized groups. A value below 1 would correspond to a lower estimated event rate in the intervention group; a value above 1 would correspond to a higher estimated event rate.
Why does censoring matter?
A participant who does not experience a first pulmonary exacerbation during observable follow-up does not necessarily have an event-free time extending indefinitely. Their observation may be censored at the end of follow-up or at another appropriate observation boundary. Time-to-event methods distinguish this situation from an observed event.
Why is a hazard ratio not the same as a risk ratio?
A risk ratio compares probabilities of an event over a specified interval. A hazard ratio instead compares instantaneous event rates within a time-to-event model. The two quantities can differ substantially, especially when event timing varies between treatment groups.
What assumption underlies a standard Cox hazard ratio?
The conventional Cox proportional-hazards interpretation assumes that the relative hazard between groups is reasonably stable over time. If proportional hazards do not hold, a single hazard ratio can compress a changing treatment effect into one summary number and may not fully describe the time pattern of treatment differences.
7. Interpreting the Primary Endpoint Without Posted Results
The registry identifies time to first pulmonary exacerbation from baseline over 48 weeks as the primary endpoint, but no statistical analysis results are posted in the ClinicalTrials.gov record.
What would be measured?
The elapsed time from baseline until the first pulmonary exacerbation, with the registered primary time frame of 48 weeks.
What would be compared?
The time-to-first-event experience between the randomized ciprofloxacin dispersion for inhalation and placebo groups.
Typical effect measure
A hazard ratio from a Cox model could summarize the relative instantaneous rate of first exacerbation between randomized groups.
Typical graphical measure
Kaplan-Meier curves could show the estimated probability of remaining free of a first pulmonary exacerbation throughout follow-up.
Because no numerical endpoint analysis is posted in the registry, there is no reported hazard ratio, confidence interval, p-value, median time to first exacerbation, or event count in the ClinicalTrials.gov record used here.
8. Analysis Population and Randomization
The study was randomized, with 304 participants enrolled across two parallel arms. Randomization is the central design feature that supports a causal comparison between the intervention and placebo groups, because treatment assignment is determined by the study design rather than by the participant's subsequent outcome.
| Design feature | Registry description |
|---|---|
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Number of arms | 2 |
| Enrollment | 304 |
| Primary purpose | Treatment |
For a randomized trial, the primary efficacy comparison would ordinarily be anchored to randomized treatment assignment. That preserves the treatment contrast created by randomization and avoids changing the meaning of the comparison based on treatment received after allocation.
9. Missing Data and Censoring
The time-to-first-event structure creates an important distinction between an observed pulmonary exacerbation and a participant who remains event-free during the period in which their outcome can be observed. The latter contributes censored follow-up rather than an observed event time.
| Observation | Statistical meaning |
|---|---|
| First pulmonary exacerbation observed | An event time is available for the primary endpoint. |
| No first exacerbation observed during follow-up | The participant may contribute right-censored event-free follow-up. |
| Follow-up ends before a first exacerbation | The observed follow-up contributes information up to the censoring time. |
The quality of a time-to-event analysis depends on appropriate handling of censoring. Standard survival methods generally rely on assumptions about the relationship between censoring and the event process. The ClinicalTrials.gov record does not specify a particular missing-data or censoring strategy beyond the definition of the primary endpoint.
10. Statistical Interpretation of a Future Hazard Ratio
If a hazard ratio were reported below 1 for ciprofloxacin dispersion for inhalation versus placebo, it would indicate a lower estimated instantaneous rate of first pulmonary exacerbation in the intervention group under the fitted model.
It would not mean that the same percentage of participants avoided exacerbation, nor would it directly state how many additional weeks an individual participant remained free of exacerbation.
A 95% confidence interval around a hazard ratio would quantify statistical uncertainty around the estimated relative event rate under the specified model and sampling framework. A narrow interval would indicate greater statistical precision than a wide interval, all else equal.
If a p-value were reported, it would address evidence against a specified null hypothesis under the statistical test. It would not measure the magnitude of the treatment effect, the clinical importance of the effect, or the probability that the treatment works.
A hazard ratio alone cannot show the absolute probability of remaining free of a first exacerbation at a clinically relevant time point. Kaplan-Meier estimates at selected times, if reported, would provide an absolute time-specific perspective alongside the relative treatment effect.
11. Proportional-Hazards Assumption
The Cox proportional-hazards model is attractive for a time-to-first-exacerbation endpoint because it provides a compact relative treatment-effect estimate. Its standard interpretation, however, depends on the proportional-hazards assumption: the ratio of the event hazards between treatment groups is assumed to remain reasonably stable over time.
If the treatment effect changes substantially during the 48-week period, a single hazard ratio may not describe the entire treatment pattern adequately. In that situation, the Kaplan-Meier curves and time-specific estimates can provide important complementary information.
Hazard ratio
Summarizes a relative instantaneous event-rate comparison under a proportional-hazards framework.
Kaplan-Meier curve
Shows how estimated event-free probability evolves over time.
Time-specific estimate
Describes the estimated probability of remaining free of a first event at a specified follow-up time.
Median event-free time
When estimable, describes the time at which the estimated event-free probability reaches 50%; it is not equivalent to a hazard ratio.
12. What This Trial's Design Tells Us Statistically
Several design features determine how the primary endpoint should be interpreted. Randomization establishes the treatment comparison. Parallel allocation means the two intervention strategies are evaluated concurrently. Quadruple masking is a design feature intended to reduce the influence of knowledge of treatment assignment on trial conduct and assessment.
| Concept | Application in ORBIT-4 |
|---|---|
| Randomization | Participants were assigned using randomized allocation. |
| Parallel design | Two intervention groups were evaluated in parallel. |
| Masking | The registry describes the trial as quadruple-masked. |
| Time-to-event endpoint | Primary endpoint was time to first pulmonary exacerbation. |
| Fixed primary time frame | The registered time frame was 48 weeks. |
| Phase | Phase 3. |
| Primary purpose | Treatment. |
13. Multiplicity and Interim Analysis
The registry information available for ORBIT-4 identifies one registered primary endpoint: time to first pulmonary exacerbation from baseline, assessed over 48 weeks.
No interim-analysis procedure, alpha-spending strategy, multiplicity adjustment, or endpoint hierarchy is reported in the information available from the ClinicalTrials.gov record.
This distinction matters because statistical significance cannot be interpreted independently of the testing framework. If a trial performs multiple formal looks at accumulating data or multiple confirmatory tests, the type I error structure should be established before interpreting individual p-values. The registry record does not provide such a framework here.
14. Non-Inferiority, Factorial, and Bayesian Methods
The registry description identifies ORBIT-4 as a randomized parallel-group treatment study. It does not identify a non-inferiority margin, factorial design, crossover design, or Bayesian analysis.
| Design or method | Registry information |
|---|---|
| Non-inferiority margin | Not reported in the ClinicalTrials.gov information available here. |
| Factorial design | Not identified; the design model is parallel. |
| Crossover | Not identified in the registry information available here. |
| Bayesian methods | Not reported in the ClinicalTrials.gov information available here. |
| Interim analysis | Not reported in the ClinicalTrials.gov information available here. |
15. Safety
The ClinicalTrials.gov record does not include numerical serious-adverse-event results or other arm-level safety estimates.
Safety is nevertheless conceptually distinct from the primary efficacy endpoint. The primary endpoint asks about time to first pulmonary exacerbation, whereas safety analyses would ordinarily summarize adverse events according to prespecified definitions and observation periods.
Without posted numerical safety analyses in the registry, no arm-level serious adverse event rate or comparative safety estimate can be reported here.
16. Planned Analysis
The registry identifies time to first pulmonary exacerbation (from baseline) over 48 weeks as the primary endpoint. For this type of endpoint, the usual statistical framework would be based on time-to-event methods.
Primary estimand
The treatment comparison concerns the timing of the first pulmonary exacerbation after baseline rather than only whether an event occurred.
Descriptive analysis
Kaplan-Meier estimation would ordinarily describe the probability of remaining free from a first pulmonary exacerbation over time.
Group comparison
A log-rank test is a standard method for comparing time-to-event distributions between randomized groups.
Effect estimate
A Cox proportional-hazards model would ordinarily provide a hazard ratio and confidence interval for the relative event rate.
17. Why the Primary Endpoint Is Statistically Informative
A time-to-first-exacerbation endpoint captures more information than a simple end-of-study binary comparison when participants experience their first event at different times. Two treatment groups could have the same proportion of participants experiencing an exacerbation by 48 weeks while differing meaningfully in when those events occurred. Conversely, a difference in the percentage with an event at 48 weeks does not by itself describe the timing of those events.
This makes the endpoint particularly suited to survival-analysis concepts. Kaplan-Meier estimation describes the evolving event-free probability, while the log-rank test compares the overall observed event-time distributions. A Cox model provides a compact relative-effect summary when its modeling assumptions are reasonable.
The important distinction is that each component answers a different question: design establishes the comparison, follow-up supplies event-time information, Kaplan-Meier describes the event-free distribution, log-rank testing compares groups, and the Cox model summarizes relative event rates.
18. Limitations
- No posted numerical efficacy analysis: the ClinicalTrials.gov record does not provide numerical statistical results for the registered primary endpoint in the information available here.
- No reported event counts: the registry information available here does not provide the number of first pulmonary exacerbations in each randomized arm.
- No reported hazard ratio: a numerical Cox-model treatment effect is not posted in the available registry information.
- No reported confidence interval or p-value: formal inferential statistics for the primary endpoint are not posted in the available registry information.
- No posted Kaplan-Meier estimates: the available registry information does not provide median time to first exacerbation or time-specific event-free probabilities.
- Unspecified censoring details: the available registry information does not specify the complete censoring or missing-data strategy for the primary endpoint.
- Unspecified multiplicity framework: the available registry information does not describe an alpha-allocation or multiplicity-adjustment strategy.
- Unspecified proportional-hazards assessment: the available registry information does not describe an assessment of the proportional-hazards assumption.
These limitations primarily affect the amount of statistical detail that can be used to characterize the treatment comparison. The randomized, parallel, quadruple-masked design and the registered time-to-first-event endpoint remain clearly identifiable features of the study.
19. Why This Trial Matters Statistically
ORBIT-4 is a useful teaching example because its primary endpoint illustrates a central distinction in clinical-trial statistics: an endpoint can be clinically straightforward while requiring a specialized statistical framework because it incorporates time.
| Statistical concept | How it applies to ORBIT-4 |
|---|---|
| Randomization | Creates a randomized comparison between ciprofloxacin dispersion for inhalation and placebo. |
| Parallel-group design | Maintains two treatment groups throughout the comparative design. |
| Time-to-event analysis | The primary endpoint records time to first pulmonary exacerbation from baseline. |
| Censoring | Participants without an observed first event may contribute event-free follow-up up to censoring. |
| Kaplan-Meier estimation | Provides a natural way to describe event-free probability over time. |
| Log-rank comparison | Provides a standard way to compare the event-time distributions of randomized groups. |
| Cox proportional-hazards model | Provides a commonly used relative event-rate estimate through a hazard ratio. |
| Confidence interval | Would quantify uncertainty around a reported treatment-effect estimate. |
| Absolute versus relative effects | Time-specific event-free probabilities complement a relative hazard ratio. |
20. Clinical Interpretation vs Statistical Interpretation
Clinical question
The study asks whether ciprofloxacin dispersion for inhalation changes the time to first pulmonary exacerbation compared with placebo during the 48-week primary time frame.
Statistical question
The analysis asks whether the distribution of time to first pulmonary exacerbation differs between the randomized treatment groups and, if so, how large the relative difference is.
Keeping these questions separate prevents a common interpretive error: treating a statistical test as though it directly measures clinical importance. A p-value addresses evidence under a null hypothesis; an effect estimate describes magnitude; a confidence interval describes precision; and time-specific event-free probabilities describe the absolute experience over follow-up.
21. A Practical Reading Framework
If numerical results for ORBIT-4 are reported in a statistical analysis, a careful reader can work through them in a fixed sequence.
- Start with the event definition. Confirm that the analysis concerns the first pulmonary exacerbation rather than recurrent exacerbations or another endpoint.
- Confirm the time origin. The registered definition specifies baseline as the starting point.
- Check the analysis period. The registered primary time frame is 48 weeks.
- Examine the Kaplan-Meier curves. Look at the estimated probability of remaining free of the first event over time.
- Read the treatment-effect estimate. If a hazard ratio is reported, determine which group is the reference and what event is being modeled.
- Read the confidence interval. Assess the precision of the estimated relative effect.
- Read the p-value separately. Treat it as evidence against a specified null hypothesis rather than as a measure of effect size.
- Assess assumptions. For a Cox analysis, consider whether the proportional-hazards interpretation is appropriate.
- Look for absolute measures. Time-specific event-free probabilities can make the clinical magnitude more transparent than the hazard ratio alone.
22. Sources
- ClinicalTrials.gov: NCT02104245 — ORBIT-4.
- PubMed: PubMed record 32554534.
- PubMed: PubMed record 30658914.
Continue through Clinical Biostats
Explore additional statistical methods, clinical trial analyses, and quantitative tools for understanding randomized clinical research.
23. Record Summary
ORBIT-4 was a completed randomized phase 3, parallel-group, quadruple-masked treatment study enrolling 304 participants with Non Cystic Fibrosis Bronchiectasis. The two-arm comparison evaluated ciprofloxacin dispersion for inhalation against placebo, with time to first pulmonary exacerbation from baseline over 48 weeks as the registered primary endpoint.
Statistically, the defining feature of the trial is its time-to-first-event endpoint. A conventional analysis would describe the event-free distribution with Kaplan-Meier methods, compare randomized groups with a time-to-event test such as the log-rank test, and potentially summarize the relative event rate with a Cox hazard ratio. The interpretation of those measures would depend on censoring, model assumptions, confidence intervals, and the prespecified inferential framework.