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
CLEAR SYNERGY was a randomized, quadruple-masked, phase 3 factorial trial involving 7,264 participants with ST elevation myocardial infarction or non-ST elevation myocardial infarction. The study evaluated colchicine, spironolactone, and the SYNERGY Bioabsorbable Polymer Drug-Eluting Stent within the overall trial design.
| Feature | CLEAR SYNERGY |
|---|---|
| Trial name | CLEAR SYNERGY |
| ClinicalTrials.gov identifier | NCT03048825 |
| Phase | Phase 3 |
| Status | Completed |
| Start date | 2018-02-01 |
| Primary completion date | 2024-08-09 |
| Enrollment | 7,264 |
| Allocation | Randomized |
| Design model | Factorial |
| Masking | Quadruple |
| Primary purpose | Treatment |
| Lead sponsor | Population Health Research Institute |
2. Clinical Question
The trial addresses treatment questions in patients with ST elevation myocardial infarction and non-ST elevation myocardial infarction. Its factorial structure allows treatment comparisons to be evaluated within a common randomized trial framework.
Population
Patients with ST elevation myocardial infarction or non-ST elevation myocardial infarction.
Interventions
Colchicine, spironolactone, and the SYNERGY Bioabsorbable Polymer Drug-Eluting Stent are listed among the study interventions.
Comparators
Colchicine-placebo and spironolactone-placebo are listed as comparator interventions.
Primary questions
The registry specifies primary comparisons involving the SYNERGY Stent, colchicine, and spironolactone, each defined through prespecified clinical event endpoints.
3. Trial Design
A factorial trial evaluates more than one intervention dimension within a common randomized framework. The key statistical advantage is efficiency: multiple treatment questions can be studied in the same enrolled population rather than requiring completely separate trials for every comparison.
Colchicine comparison
- Colchicine
- Colchicine-placebo
- Primary endpoint followed through study completion, with an estimated average of 3 years
Spironolactone comparison
- Spironolactone
- Spironolactone-placebo
- Two co-primary endpoints followed through study completion, with an estimated average of 3 years
4. Trial Interventions
| Intervention | Type | Role in the trial |
|---|---|---|
| Colchicine | Drug | Intervention in the colchicine comparison |
| Colchicine-Placebo | Drug | Comparator in the colchicine comparison |
| Spironolactone | Drug | Intervention in the spironolactone comparison |
| Spironolactone-Placebo | Drug | Comparator in the spironolactone comparison |
| SYNERGY Bioabsorbable Polymer Drug-Eluting Stent | Device | Evaluated through the stent primary endpoint |
The registry describes the study as randomized, quadruple-masked, factorial, and treatment-focused. These design characteristics are important because the statistical interpretation of an endpoint depends not only on the endpoint definition but also on how treatment assignments were randomized and masked.
5. Primary Endpoints
| Primary endpoint | Time frame | Registry definition |
|---|---|---|
| Major Adverse Cardiac Events (MACE) | up to 1 year | Major Adverse Cardiac Events (MACE) for SYNERGY Stent, defined as the composite of death, recurrent target vessel MI, stroke, or ischemia driven target vessel revascularization, compared to performance goal. |
| Composite of cardiovascular death, recurrent myocardial infarction, stroke, or unplanned ischemia driven revascularization | through study completion, an estimated average of 3 years | The first occurrence of cardiovascular death, recurrent myocardial infarction, stroke, or unplanned ischemia driven revascularization in the colchicine comparison. |
| Total composite events of cardiovascular death or new or worsening heart failure (co-primary 1) | through study completion, an estimated average of 3 years | Total composite of cardiovascular death or new or worsening heart failure in the spironolactone comparison. |
| Composite of cardiovascular death, new or worsening heart failure, recurrent myocardial infarction, or stroke (co-primary 2) | through study completion, an estimated average of 3 years | The first occurrence of cardiovascular death, new or worsening heart failure, recurrent myocardial infarction, or stroke in the spironolactone comparison. |
Why these endpoint definitions matter
The primary endpoints combine clinically important events into composite outcomes. A composite endpoint can increase the number of observed events and therefore provide more statistical information than a single component alone. But its interpretation depends on understanding exactly which events are included and whether the analysis concerns the first occurrence or total composite events.
That distinction is particularly important in the spironolactone comparison. The first spironolactone co-primary endpoint is explicitly defined as total composite events, whereas the second is defined as the first occurrence of the listed cardiovascular outcomes. These are different estimands and should not be treated as interchangeable.
6. Statistical Methodology
The ClinicalTrials.gov record identifies the trial as randomized and factorial but does not report posted statistical analyses or numerical endpoint results. Consequently, the appropriate statistical framework is best understood from the endpoint definitions and trial design.
Time-to-event analysis
Three of the four primary endpoint descriptions concern events accumulated through study completion, with an estimated average follow-up of 3 years. For endpoints defined by the time until a first cardiovascular event, a standard analysis would use time-to-event methods that account for differing follow-up durations and right censoring.
Here, T represents the time from the relevant trial origin to the event of interest. Patients who have not experienced the event by their last observed follow-up contribute information up to that time and are then censored.
Kaplan-Meier estimation
For a first-occurrence endpoint, Kaplan-Meier estimation is a standard way to describe the probability of remaining free of the event over time. It can provide estimated event-free probabilities at specified times and a visual description of how event incidence accumulates.
Hazard ratios and Cox models
A Cox proportional-hazards model is commonly used to estimate a relative hazard between randomized treatment groups for a time-to-first-event endpoint. A hazard ratio below 1 indicates a lower estimated instantaneous event rate in one treatment group relative to its comparator under the fitted model.
A hazard ratio is not an absolute risk difference and is not the proportion of patients who avoid the event. Its interpretation also depends on the proportional-hazards assumption and the time period represented by the analysis.
Log-rank testing
For a randomized comparison involving a time-to-first-event endpoint, the log-rank test is commonly used to compare event-time distributions between treatment groups. A stratified version can be used when the randomization or analysis specifies important stratification factors.
Composite endpoint analysis
For a composite such as cardiovascular death, recurrent myocardial infarction, stroke, or unplanned ischemia driven revascularization, the first occurrence of any qualifying component becomes the endpoint event. Statistical interpretation therefore concerns the composite as defined, not automatically each individual component.
7. Planned Analysis
The registry specifies four primary endpoints. For the SYNERGY Stent endpoint, MACE is defined as a composite of death, recurrent target vessel MI, stroke, or ischemia driven target vessel revascularization and is evaluated up to 1 year against a performance goal. A performance-goal analysis differs from an ordinary two-arm randomized comparison because the benchmark is a prespecified external target rather than simply another randomized treatment group.
The colchicine primary endpoint is the first occurrence of cardiovascular death, recurrent myocardial infarction, stroke, or unplanned ischemia driven revascularization through study completion, with an estimated average of 3 years. A conventional analysis would use a time-to-first-event framework, with treatment groups compared using a method such as the log-rank test and an effect measure such as a hazard ratio from a Cox model.
The spironolactone comparison has two co-primary endpoints. The first is the total composite events of cardiovascular death or new or worsening heart failure through study completion, with an estimated average of 3 years. Because this endpoint counts total composite events rather than only the first event, recurrent-event methods may be appropriate depending on the prespecified statistical analysis plan.
The second spironolactone co-primary endpoint is the first occurrence of cardiovascular death, new or worsening heart failure, recurrent myocardial infarction, or stroke through study completion, with an estimated average of 3 years. This endpoint would typically be analyzed as a time-to-first-event outcome.
The registry does not report the statistical analysis plan details needed to identify the exact test statistic, model specification, covariate adjustment, confidence-interval construction, multiplicity procedure, missing-data method, or analysis population for each endpoint. Results have not been posted on ClinicalTrials.gov.
8. Statistical Methods Explained
Why is a factorial design useful here?
A factorial design can evaluate multiple treatment questions within one randomized trial. Instead of requiring a completely separate trial for every intervention comparison, a common population and infrastructure can support more than one treatment evaluation. This can improve efficiency, although the validity of individual treatment-effect estimates depends on the randomization and prespecified analysis structure.
Why are there four primary endpoints?
The registry defines separate primary questions for the SYNERGY Stent, colchicine, and spironolactone comparisons. Spironolactone has two co-primary endpoints. Each endpoint represents a different clinical question, so its definition determines what constitutes an event and what statistical estimand is being targeted.
What is the difference between a first-occurrence and total-event endpoint?
A first-occurrence endpoint counts the first qualifying event after the relevant starting point. Once that event occurs, the patient has experienced the endpoint. A total-event endpoint can count multiple qualifying events from the same participant. The statistical methods for these estimands can therefore differ substantially.
Why does censoring matter?
Not every participant will necessarily experience an endpoint before the end of follow-up. A time-to-event analysis uses information from participants until their last observed event-free time. Treating every participant who has not experienced an event as though the observation continued indefinitely would distort the event-time distribution.
What does a composite endpoint tell us?
A composite endpoint combines several clinically defined events into one outcome. This can increase statistical efficiency by increasing the number of endpoint events. However, the resulting treatment effect applies to the composite as a whole. A statistically important composite result does not automatically imply that every component has the same magnitude or direction of effect.
Why is a performance goal different from a randomized comparator?
The SYNERGY Stent primary endpoint is defined against a performance goal. A performance-goal analysis asks whether the observed outcome meets a prespecified benchmark rather than simply comparing two simultaneously randomized treatment groups. Its interpretation therefore depends critically on how the benchmark was established and incorporated into the statistical design.
Why does a factorial trial require attention to interactions?
When multiple interventions are evaluated in a factorial design, it can be important to determine whether the effect of one intervention depends on the presence of another intervention. Such an interaction is different from the main treatment effect. The registry identifies the factorial design but does not report the interaction-analysis specification.
9. Endpoint-Specific Statistical Considerations
| Endpoint | Statistical structure | Key interpretation issue |
|---|---|---|
| SYNERGY Stent MACE | Composite endpoint up to 1 year compared to a performance goal | Benchmark-based inference rather than a simple randomized two-group comparison |
| Colchicine composite | First occurrence through study completion | Time-to-first-event analysis and censoring are central |
| Spironolactone co-primary 1 | Total composite events through study completion | Repeated events may require methods beyond a conventional time-to-first-event model |
| Spironolactone co-primary 2 | First occurrence through study completion | Time-to-first-event analysis and the co-primary endpoint structure must be considered jointly |
The distinction between these endpoints is not merely semantic. Statistical inference should match the estimand. A method designed for the first event should not automatically be substituted for a total-event analysis, and a performance-goal comparison should not be interpreted as though it were an ordinary randomized comparison.
10. Multiplicity in a Factorial Trial
Multiple primary endpoints create a multiplicity problem because several hypothesis tests can increase the overall probability of obtaining at least one apparently positive result by chance if each test is interpreted independently at the same nominal significance level.
Multiple intervention questions
The factorial design evaluates more than one treatment comparison within the same overall study population.
Multiple primary endpoints
The registry identifies four primary endpoints, including two co-primary endpoints for the spironolactone comparison.
Why hierarchy matters
The inferential meaning of a P-value depends on the prespecified multiplicity strategy, including how multiple endpoints are related.
What the registry does not specify
The available registry information does not report the exact multiplicity adjustment or alpha-allocation procedure.
For a factorial trial, multiplicity can arise from both the number of intervention questions and the number of endpoints. The appropriate family of hypotheses and error-control strategy should therefore be established prospectively rather than selected after observing the results.
11. Non-Inferiority and Performance-Goal Logic
The SYNERGY Stent primary endpoint is explicitly compared to a performance goal. That is statistically distinct from the randomized treatment comparisons involving colchicine and spironolactone.
A performance-goal analysis typically specifies a benchmark representing the level against which the observed MACE outcome is evaluated. The inferential question then concerns whether the observed trial outcome meets the prespecified performance criterion. The registry does not provide the numerical performance goal or the exact inferential procedure.
The benchmark must be defined independently of the observed outcome for the resulting statistical test or confidence interval to have its intended interpretation.
This endpoint should therefore be kept conceptually separate from the colchicine and spironolactone comparisons, where treatment assignment provides a randomized comparator structure.
12. Analysis Populations and Randomization
The registry identifies the allocation as randomized and the masking as quadruple. Randomization is the foundation for causal comparison because treatment assignment is determined by the trial design rather than by investigators' or participants' choice.
For efficacy questions in randomized trials, an intention-to-treat framework is commonly used because it preserves the original treatment assignment. For safety or exposure-related analyses, investigators may use a treated or as-treated population. The precise analysis populations for CLEAR SYNERGY are not specified in the registry information presented here.
| Design feature | Registry information | Statistical importance |
|---|---|---|
| Allocation | Randomized | Supports comparison of treatment groups under the randomized design |
| Masking | Quadruple | Reduces opportunities for knowledge of treatment assignment to influence trial conduct or assessment |
| Design model | Factorial | Allows multiple intervention questions within a common trial structure |
| Primary purpose | Treatment | Frames the trial around evaluation of treatment interventions |
13. Time Frames and Follow-Up
Trial start
The registry lists February 1, 2018 as the study start date.
SYNERGY Stent primary endpoint
MACE is evaluated up to 1 year and compared to a performance goal.
Colchicine and spironolactone follow-up
The colchicine primary endpoint and both spironolactone co-primary endpoints are followed through study completion, with an estimated average of 3 years.
Primary completion
The registry lists August 9, 2024 as the primary completion date.
The different endpoint time frames reflect different statistical questions. A 1-year MACE endpoint is not directly interchangeable with an endpoint accumulated through study completion over an estimated average of 3 years. Comparing estimates across those time frames without accounting for their definitions and follow-up would be misleading.
14. Composite Events and Competing Clinical Outcomes
Cardiovascular trials frequently use composite outcomes because several clinically important events may occur during follow-up. CLEAR SYNERGY includes death, myocardial infarction, stroke, revascularization, and heart-failure outcomes across its primary endpoint definitions.
When an endpoint includes multiple event types, analysts should distinguish between the composite effect and the effects of its individual components. If one component is much more common than the others, the composite estimate may be driven primarily by that component. Conversely, very rare components may contribute little statistical information even when they are clinically serious.
15. Why Masking Matters Statistically
The study is described as quadruple masked. Masking can reduce the risk that knowledge of treatment assignment influences behavior, treatment administration, outcome assessment, or reporting.
Masking is particularly relevant when outcomes involve clinical decisions, such as ischemia driven target vessel revascularization. If the assessment or decision to perform a procedure could be influenced by knowledge of treatment assignment, the endpoint could become susceptible to differential assessment or management.
For a composite endpoint, masking can therefore protect more than one component of the analysis. It helps maintain the credibility of the randomized comparison by limiting opportunities for treatment knowledge to affect the pathway from assignment to observed outcome.
16. Statistical Interpretation of the Primary Questions
The primary endpoint is a composite of death, recurrent target vessel MI, stroke, or ischemia driven target vessel revascularization, assessed up to 1 year against a performance goal. The appropriate interpretation concerns whether the observed composite outcome meets the prespecified benchmark. It should not be treated as though it were automatically a conventional randomized comparison between two treatment groups.
The primary endpoint is the first occurrence of cardiovascular death, recurrent myocardial infarction, stroke, or unplanned ischemia driven revascularization through study completion. A time-to-first-event analysis would preserve information about when events occur and would account for participants with different lengths of follow-up.
The first spironolactone co-primary endpoint counts total composite events of cardiovascular death or new or worsening heart failure. Because total events rather than only the first event are specified, the estimand concerns the overall burden of recurrent composite events rather than simply time until the first qualifying event.
The second spironolactone co-primary endpoint is the first occurrence of cardiovascular death, new or worsening heart failure, recurrent myocardial infarction, or stroke. Its interpretation is therefore based on time to the first qualifying event and differs from the total-event estimand of co-primary 1.
17. What the Trial Design Does — and Does Not — Establish
Randomization supports causal inference for comparisons between randomized treatment groups, provided the trial is conducted and analyzed according to its prespecified design. It does not mean that every measured characteristic will be identical between groups, nor does it remove uncertainty from the estimated treatment effect.
Factorial design adds another layer of interpretation. The main effects of interventions can be estimated efficiently, but if one intervention changes the effect of another, an interaction can be present. The existence of a factorial design therefore makes the specification and interpretation of interaction terms an important statistical consideration.
Quadruple masking can reduce bias associated with knowledge of treatment assignment, but it does not eliminate all possible sources of bias. Loss to follow-up, deviations from assigned treatment, outcome definitions, endpoint adjudication, and subsequent clinical management can still affect the evidence.
18. Limitations
- No posted statistical analyses: the ClinicalTrials.gov record does not report formal posted statistical analyses or numerical results for the primary endpoints.
- Multiplicity: the study contains multiple primary endpoints and multiple intervention questions, making prespecified type I error control an important component of interpretation.
- Factorial interactions: the registry identifies a factorial design but does not report the statistical specification for testing interactions among interventions.
- Performance-goal interpretation: the SYNERGY Stent endpoint is evaluated against a performance goal rather than described solely as a two-group randomized comparison.
- Different estimands: the spironolactone comparison includes both a total-event composite and a first-occurrence composite, which should not be interpreted as equivalent endpoints.
- Composite endpoints: a treatment effect on a composite does not necessarily imply identical effects on each component event.
- Time-frame differences: the stent endpoint is assessed up to 1 year, whereas the colchicine and spironolactone endpoints extend through study completion with an estimated average of 3 years.
- Unreported analysis details: the registry information does not specify the exact covariate adjustment, censoring rules, confidence-interval procedures, missing-data strategy, or analysis populations for the primary endpoints.
19. Why This Trial Matters Statistically
CLEAR SYNERGY is a useful teaching case because it combines a large randomized phase 3 trial with a factorial structure, multiple intervention questions, multiple primary endpoints, composite cardiovascular outcomes, a performance-goal comparison, and both first-event and total-event estimands.
| Concept | How it appears in CLEAR SYNERGY |
|---|---|
| Randomization | Randomized allocation across a phase 3 trial enrolling 7,264 participants |
| Factorial design | Four-arm factorial trial evaluating multiple intervention questions |
| Masking | Quadruple masking |
| Composite endpoint | MACE and several cardiovascular composite endpoints |
| Time-to-event analysis | Colchicine and spironolactone endpoints followed through study completion |
| First-event estimand | Colchicine primary endpoint and spironolactone co-primary 2 |
| Total-event estimand | Spironolactone co-primary 1 |
| Performance goal | SYNERGY Stent MACE endpoint compared to a performance goal |
| Multiplicity | Four registry-defined primary endpoints across multiple intervention questions |
| Longer follow-up | Colchicine and spironolactone endpoints through study completion, estimated average of 3 years |
The most important statistical lesson is that the phrase primary endpoint does not describe a single universal type of analysis. The endpoint's definition determines the appropriate estimand, and the estimand determines the statistical model needed to answer the clinical question.
20. Statistical Methods Explained in Practice
Why not analyze every endpoint with the same test?
Because the endpoints are structurally different. A first-occurrence time-to-event endpoint, a total recurrent-event endpoint, and a performance-goal endpoint represent different statistical questions. Applying one generic method to all of them could answer a different question from the one specified by the trial.
Why does the estimated average follow-up matter?
When an endpoint is followed through study completion, participants may have different amounts of observation time. The estimated average follow-up provides context for interpreting how long participants were expected to contribute information, but a time-to-event analysis still uses each participant's observed follow-up rather than assuming everyone was followed for exactly the average duration.
Why can a composite endpoint improve efficiency?
Combining several clinically relevant events can increase the number of endpoint events. More events can provide greater statistical information and potentially narrower confidence intervals. The trade-off is that the resulting effect estimate summarizes a mixture of event types, which may not have identical clinical importance or treatment effects.
Why is total-event analysis different from Kaplan-Meier analysis?
Kaplan-Meier analysis is naturally suited to the time until a first event. If participants can contribute multiple events, simply stopping observation after the first event discards information about later events. Recurrent-event methods can instead model the total burden of events or the event rate over follow-up.
What would a hazard ratio mean for the first-occurrence endpoints?
For a first-occurrence endpoint, a hazard ratio summarizes the relative instantaneous rate of experiencing the first qualifying event between randomized groups under the fitted survival model. It does not directly represent the difference in cumulative incidence at a particular time point.
Why should interaction be considered in factorial designs?
An interaction asks whether the effect of one intervention changes depending on the level of another intervention. Without an interaction, main effects can often be interpreted more simply. With an important interaction, the overall average effect of one intervention may conceal different effects across levels of the other intervention.
21. Registry Timeline
Study begins
The registry lists the study start date as February 1, 2018.
Randomized factorial evaluation
The trial is classified as a phase 3 randomized, quadruple-masked factorial study with a treatment primary purpose.
SYNERGY Stent endpoint
MACE is defined and evaluated up to 1 year against a performance goal.
Colchicine and spironolactone endpoints
The cardiovascular and heart-failure composite endpoints are followed through study completion, with an estimated average of 3 years.
Primary completion
The registry lists August 9, 2024 as the primary completion date.
22. Sources
- ClinicalTrials.gov: CLEAR SYNERGY, NCT03048825.
- PubMed: 39555823.
- PubMed: 39555814.
- PubMed: 38936755.
- PubMed: 38505729.
- PubMed: 36526898.
23. Related Tutorials
Learn more about the methods used in this trial:
24. Related Calculators
Continue through the Clinical Biostats trial-analysis library
Explore clinical-trial statistical methods, tutorials, and calculators for deeper understanding of randomized study design and analysis.
25. Record Summary
CLEAR SYNERGY is a completed randomized phase 3 factorial trial enrolling 7,264 participants with ST elevation myocardial infarction or non-ST elevation myocardial infarction. Its statistical structure is notable for evaluating multiple intervention questions within one factorial framework, while using different primary endpoint estimands: a performance-goal MACE endpoint for the SYNERGY Stent, a first-occurrence cardiovascular composite for colchicine, and both total-event and first-occurrence composites for spironolactone.
The trial illustrates why statistical analysis must begin with the precise endpoint definition. The distinction between a first event and total events, between a randomized comparator and a performance goal, and between multiple primary hypotheses in a factorial design determines which statistical methods are appropriate and how the resulting estimates should be interpreted.