The ClinicalTrials.gov record describes ASAC as a randomized, parallel-group, quadruple-masked phase 2/3 trial with a prevention primary purpose. The registry lists 466 participants and one registered primary endpoint: disease-free survival after three years of treatment.
1. Trial at a Glance
ASAC evaluates whether acetylsalicylic acid, compared with placebo, affects disease-free survival after resection of liver metastases in colorectal cancer. The registry identifies the study as phase 2/3, randomized, parallel-group, quadruple-masked, and designed with a prevention primary purpose.
| Feature | ASAC |
|---|---|
| Trial acronym | ASAC |
| ClinicalTrials.gov identifier | NCT03326791 |
| Official title | Aspirin in Colorectal Cancer Liver Metastases |
| Phase | 2/3 |
| Status | Active, not recruiting |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Primary purpose | Prevention |
| Enrollment | 466 |
| Number of arms | 2 |
| Lead sponsor | Oslo University Hospital |
2. Clinical Question
The statistical question is whether treatment with acetylsalicylic acid differs from placebo with respect to disease-free survival three years after initiation of treatment following resection of liver metastases in colorectal cancer.
Population
Patients represented by the registry conditions of colorectal cancer, colorectal cancer liver metastases, and liver metastases, in the setting described by the primary endpoint after resection of liver metastases.
Intervention
Acetylsalicylic acid, listed in the registry as a drug intervention.
Comparator
Placebo Oral Tablet, listed in the registry as a drug intervention.
Primary question
Does acetylsalicylic acid, compared with placebo, affect disease-free survival after three years of treatment following resection of liver metastases?
3. Trial Design
Acetylsalicylic acid
- Acetylsalicylic acid
- Drug intervention
- Evaluated after resection of liver metastases
- Primary endpoint assessed at three years after initiation of treatment
Placebo Oral Tablet
- Placebo Oral Tablet
- Drug intervention
- Evaluated after resection of liver metastases
- Primary endpoint assessed at three years after initiation of treatment
The parallel-group structure is statistically important because each participant contributes to one randomized treatment comparison rather than moving between treatment sequences. Randomization provides the basis for comparing disease-free survival between the two assigned groups.
Study chronology
Study start
The registry lists December 15, 2017 as the study start date.
Primary completion
The registry lists February 1, 2022 as the primary completion date.
Active, not recruiting
The ClinicalTrials.gov record currently identifies the study status as active, not recruiting.
4. Primary Endpoint
| Endpoint | Registry definition | Time frame |
|---|---|---|
| Disease Free Survival (DFS) after three years treatment | DFS three years after initiation of treatment with ASA or Placebo after resection of liver metastases. | 3 years |
This endpoint combines two ideas that are central to survival analysis: a defined starting point and the occurrence of a disease-related event or censoring before the specified follow-up horizon. The registry explicitly defines the starting point for this endpoint as initiation of treatment with ASA or placebo after resection of liver metastases.
The phrase after three years treatment identifies a prespecified time frame rather than a request to compare only patients who remain event-free throughout follow-up. Statistically, disease-free survival is ordinarily treated as a time-to-event outcome, allowing information from participants who experience the event at different times during follow-up to contribute to the analysis.
5. Statistical Methodology
The registry identifies disease-free survival after three years of treatment as the primary endpoint but does not post a formal statistical analysis for that endpoint. The ClinicalTrials.gov record therefore establishes the outcome and time frame without specifying a detailed inferential model in the posted statistical analyses.
Here, S(t) represents the probability of remaining free of the event beyond time t. In a clinical trial, participants without an observed event by their last available assessment may be right-censored rather than treated as having experienced the event.
Kaplan-Meier estimation
For a disease-free survival endpoint, a Kaplan-Meier estimator is a standard way to describe the time-to-event distribution. It estimates the probability of remaining event-free over follow-up while accounting for the timing of events and right censoring.
The quantities di and ni represent the number of events and the number at risk immediately before event time ti, respectively.
Between-group comparison
A randomized two-arm disease-free survival comparison is typically evaluated with a log-rank test and described with a hazard ratio from a Cox proportional-hazards model. Those methods are common for randomized time-to-event comparisons, but the registry does not post a statistical analysis specifying that these particular procedures were used for ASAC.
Three-year disease-free survival
The registry's three-year time frame can also be summarized using an absolute disease-free survival probability at three years. Such an estimate answers a different question from a hazard ratio: it describes the estimated proportion remaining disease-free at a particular point in time rather than a relative event rate across the follow-up period.
6. Planned Analysis
The registered primary endpoint is Disease Free Survival (DFS) after three years treatment, with a time frame of 3 years. The registry describes this as DFS three years after initiation of treatment with ASA or placebo after resection of liver metastases.
Primary statistical estimand
Compare disease-free survival between the randomized acetylsalicylic acid and placebo groups after treatment initiation following resection of liver metastases.
For this type of endpoint, a typical analysis would use time-to-event methods rather than simply comparing the proportion of participants who experienced an event. Kaplan-Meier curves can describe the disease-free survival experience over time, while a log-rank test can compare the randomized groups. A Cox model can provide a hazard ratio and confidence interval when its modeling assumptions are appropriate.
A complementary analysis would report the absolute disease-free survival probability at three years for each randomized group, together with a measure of uncertainty. Reporting both an absolute three-year estimate and a relative time-to-event measure helps distinguish the probability of remaining disease-free at a clinically specified time from the relative event rate over follow-up.
7. Statistical Methods Explained
What does disease-free survival measure?
Disease-free survival is a time-to-event endpoint. Instead of asking only whether an event occurred, it considers when the event occurred. That distinction is important because two participants who both experience an event can contribute very different amounts of information if their events occur at different times.
Why use Kaplan-Meier estimation?
Kaplan-Meier estimation is designed for time-to-event data in which some participants may not experience the event during observed follow-up. Those participants can be censored at their last known event-free time, allowing their available follow-up information to contribute without assuming that an event occurred after their last assessment.
What would a hazard ratio mean in this trial?
If a Cox model were used, the hazard ratio would compare the estimated instantaneous disease-free survival event rate between the acetylsalicylic acid and placebo groups over the modeled follow-up. A hazard ratio below 1 would correspond to a lower estimated event hazard in the acetylsalicylic acid group; a hazard ratio above 1 would correspond to a higher estimated event hazard.
A hazard ratio would not mean that a particular percentage of participants benefited. It would also not be identical to the difference between the two groups' three-year disease-free survival probabilities.
Why is three-year DFS different from a simple three-year proportion?
A simple proportion can classify participants according to whether they are event-free at three years, but it does not fully use the timing of events during follow-up. A time-to-event analysis incorporates the event times and censoring structure, while a three-year survival estimate provides a clinically interpretable absolute snapshot at the prespecified time point.
Why does censoring matter?
Participants can have incomplete observed follow-up. In a standard right-censored analysis, a participant contributes information until the last time they are known to be event-free. The validity of the analysis depends on assumptions concerning the censoring mechanism and how follow-up was obtained.
Why would a confidence interval accompany a treatment effect?
A confidence interval quantifies statistical uncertainty around an estimated treatment effect under the specified analysis framework. A narrow interval indicates greater statistical precision than a wide interval. It does not describe the range of outcomes that individual patients will experience.
8. Randomization and Analysis Populations
ASAC is randomized with two parallel arms and an enrollment of 466 participants. Randomization is the key design feature supporting a comparative analysis of disease-free survival between acetylsalicylic acid and placebo.
For a randomized efficacy analysis, the intention-to-treat principle would ordinarily preserve the original treatment assignment when analyzing outcomes. This approach maintains the treatment comparison established by randomization even when participants do not complete treatment exactly as planned.
The registry information summarized here does not specify separate analysis populations, treatment-adherence rules, censoring rules, missing-data procedures, or an as-treated safety population.
9. Censoring and Missing Data
Because disease-free survival is a time-to-event endpoint, incomplete follow-up has a different statistical role from ordinary missing measurements. A participant who has not experienced the endpoint by the last known assessment can contribute observed event-free time before being censored.
Right censoring
A participant contributes information up to the last time their disease-free status is known when an event has not been observed.
Event timing
Time-to-event methods use the timing of disease-free survival events rather than reducing the entire follow-up period to a binary outcome.
Three-year landmark
The registered three-year time frame gives the primary endpoint a fixed clinical horizon for interpretation.
Registry detail
The ClinicalTrials.gov record does not post detailed missing-data or imputation procedures for the primary endpoint.
For survival analysis, conventional multiple imputation of an ordinary continuous outcome is not automatically the appropriate solution to incomplete follow-up. The analysis instead requires an explicit definition of events, censoring, follow-up, and the assumptions under which censored observations remain informative for the survival estimator.
10. Non-Inferiority, Crossover, and Factorial Design
| Design feature | ASAC registry information | Statistical implication |
|---|---|---|
| Non-inferiority margin | Not reported in the registry information summarized here. | The primary endpoint is not presented here as a non-inferiority analysis. |
| Crossover | Not reported in the registry information summarized here. | No crossover adjustment is described. |
| Factorial design | Not reported; the design model is parallel. | The comparison is structured as two parallel intervention groups. |
| Multiplicity | No multiplicity procedure is reported for the primary endpoint in the registry information summarized here. | No additional confirmatory multiplicity framework is described here. |
| Interim analysis | No interim statistical analysis is posted. | The registry does not provide an interim-testing framework for the primary endpoint. |
| Bayesian methods | No Bayesian analysis is posted. | No posterior probability or Bayesian treatment effect is reported. |
These distinctions matter because different trial designs imply different estimands and error-control strategies. ASAC is registered as a randomized parallel-group trial, so the core comparison is between the outcomes of the two randomized groups.
11. Interpretation of the Primary Endpoint
The registered primary endpoint asks whether disease-free survival differs between participants receiving acetylsalicylic acid and those receiving placebo over a three-year time frame following treatment initiation after resection of liver metastases.
If a Cox analysis were reported, the hazard ratio would summarize the relative event hazard between the randomized groups under the model. It would be a relative time-to-event measure, not an absolute three-year disease-free survival difference.
An estimated three-year disease-free survival probability would provide an absolute, time-specific measure of the proportion expected to remain disease-free at the registered three-year horizon under the survival analysis.
If a future hypothesis test were reported, its p-value would address evidence against a specified null hypothesis under the statistical model. It would not measure the magnitude or clinical importance of the treatment effect. Effect estimates and their confidence intervals are needed to describe magnitude and precision.
12. Assumptions Behind a Time-to-Event Analysis
Time-to-event methods are powerful because they preserve information about event timing, but their interpretation depends on assumptions.
- Correct event definition: the disease-free survival event must be defined consistently according to the trial's endpoint rules.
- Accurate event timing: the recorded timing of disease-free survival events affects both the survival curve and any model-based treatment effect.
- Censoring assumptions: Kaplan-Meier and Cox analyses require assumptions about the relationship between censoring and future event risk.
- Proportional hazards: a standard Cox hazard ratio assumes that the relative hazard is appropriately represented by a proportional-hazards structure. If hazards vary materially in a nonproportional way, a single HR may not capture the full treatment-effect pattern.
- Randomization: the randomized allocation provides the basis for causal comparison, provided the analysis follows the randomized groups appropriately.
- Follow-up completeness: informative loss to follow-up can affect the validity of a survival comparison.
The registry information available here does not provide the event definitions beyond the registered DFS description, nor does it post the detailed statistical analysis plan needed to assess the exact implementation of these assumptions.
13. What Would Make the DFS Analysis Statistically Informative?
A complete reported analysis of the primary endpoint would ordinarily allow readers to examine several complementary quantities rather than relying on a single test statistic.
| Quantity | Statistical role |
|---|---|
| Number of randomized participants by arm | Defines the treatment populations underlying the randomized comparison. |
| Number of DFS events | Determines how much event information contributes to the time-to-event analysis. |
| Kaplan-Meier estimates | Describe disease-free survival over time. |
| Three-year DFS | Provides the prespecified absolute time-point measure. |
| Hazard ratio | Provides a relative time-to-event treatment-effect estimate if a Cox model is used. |
| Confidence interval | Describes uncertainty around the treatment-effect estimate. |
| Hypothesis-test result | Assesses evidence against the specified null hypothesis under the chosen testing procedure. |
These quantities answer related but distinct questions. The Kaplan-Meier curve describes the trajectory, the three-year estimate gives an absolute landmark, the hazard ratio gives a relative time-to-event comparison, and the confidence interval communicates statistical precision.
14. Limitations
- No posted primary analysis: the ClinicalTrials.gov record does not post a formal statistical analysis for the registered disease-free survival endpoint.
- No treatment-effect estimate: the registry information does not provide a hazard ratio, risk difference, survival probability, confidence interval, or p-value for the primary endpoint.
- No event counts: the registry information does not report the number of disease-free survival events by treatment arm.
- No arm-specific results: the registry information identifies the two interventions but does not provide comparative efficacy results for them.
- Limited statistical-plan detail: the registry information does not specify the precise survival-analysis model, testing procedure, censoring rules, missing-data strategy, or analysis populations for the primary endpoint.
- No posted safety comparison: arm-specific serious adverse-event results are not reported in the registry information summarized here.
- No subgroup analysis: subgroup estimates or interaction analyses are not reported in the registry information summarized here.
These limitations affect how far the registered endpoint can be interpreted statistically. The design identifies a clear randomized comparison and a clinically defined three-year disease-free survival endpoint, but the posted record does not supply the numerical efficacy analysis needed to quantify the treatment effect.
15. Why This Trial Matters Statistically
ASAC is a useful teaching example because it places a prevention question within a randomized two-arm time-to-event framework. The primary endpoint is not simply a laboratory measurement or a single post-treatment observation. It is disease-free survival measured over a specified three-year horizon after initiation of treatment following resection of liver metastases.
| Concept | How it appears in ASAC |
|---|---|
| Randomization | The registry identifies the allocation as randomized. |
| Parallel-group design | The design model contains two parallel treatment arms. |
| Time-to-event analysis | Disease-free survival is the registered primary endpoint. |
| Three-year landmark | The primary endpoint has a three-year time frame. |
| Kaplan-Meier estimation | A standard descriptive method for disease-free survival over follow-up. |
| Hazard ratio | A typical relative treatment-effect measure for a randomized survival comparison. |
| Confidence interval | A standard way to communicate uncertainty around an estimated treatment effect. |
| Censoring | A central issue whenever participants have incomplete event observation. |
| Masking | The registry identifies the study as quadruple-masked. |
| Prevention purpose | The registry identifies prevention as the primary purpose. |
16. Clinical Question vs Statistical Question
Clinical question
Does acetylsalicylic acid, compared with placebo, affect disease-free survival after resection of liver metastases in colorectal cancer?
Statistical question
Does the disease-free survival distribution differ between the two randomized treatment groups over the registered three-year time frame?
Absolute measure
The three-year disease-free survival probability describes the estimated proportion remaining disease-free at the prespecified time horizon.
Relative measure
A hazard ratio, if used, would compare the estimated instantaneous event hazards between the randomized groups under the fitted model.
Keeping these questions separate is important. A statistically detectable difference does not by itself describe its magnitude, while an effect estimate without an uncertainty interval does not communicate its statistical precision. A rigorous analysis therefore presents the effect estimate, uncertainty, absolute time-point measures, and assumptions together.
17. Record Timeline
Trial begins
The ClinicalTrials.gov record lists December 15, 2017 as the study start date.
Primary completion date
The registry lists February 1, 2022 as the primary completion date.
Active, not recruiting
The current registry status is active, not recruiting.
18. Statistical Perspective on a Three-Year Endpoint
A fixed three-year endpoint is useful because it establishes a common clinical horizon for reporting. However, the underlying disease-free survival data contain more information than a single three-year percentage. Participants may experience events at different times, and some may have incomplete observed follow-up.
For that reason, a complete survival analysis would generally show the estimated disease-free survival function across time and then report the three-year estimate as the prespecified landmark. The curve provides temporal context, while the three-year estimate answers the specific registered endpoint question.
If a hazard ratio is used, it provides another summary of the treatment comparison. Its interpretation should remain tied to the fitted model and its proportional-hazards assumption rather than being treated as a direct translation into an absolute probability difference.
19. Summary of the Statistical Framework
| Element | ASAC |
|---|---|
| Study phase | Phase 2/3 |
| Enrollment | 466 |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Primary purpose | Prevention |
| Intervention | Acetylsalicylic acid |
| Comparator | Placebo Oral Tablet |
| Primary endpoint | Disease Free Survival (DFS) after three years treatment |
| Endpoint time frame | 3 years |
| Posted statistical analyses | None |
The central statistical structure is therefore straightforward: a randomized two-arm comparison with disease-free survival as the primary time-to-event endpoint and a three-year assessment horizon. The registry supplies the design and endpoint definition but does not post the numerical efficacy analysis.
20. Related Clinical Trial Results
21. Sources
- ClinicalTrials.gov: ASAC — NCT03326791.
- PubMed: PubMed record.
Continue through Clinical Biostats
Explore statistical tutorials, calculators, and clinical trial analyses covering study design, survival analysis, and interpretation of clinical evidence.
22. Record Summary
ASAC is a randomized phase 2/3, parallel-group, quadruple-masked trial with a prevention primary purpose and 466 participants. The registry evaluates acetylsalicylic acid against placebo after resection of liver metastases in the setting of colorectal cancer, with disease-free survival three years after initiation of treatment registered as the primary endpoint.
Statistically, the endpoint is naturally framed as a time-to-event outcome. Kaplan-Meier estimation can describe disease-free survival over follow-up, a three-year survival estimate can provide the prespecified absolute landmark, and a Cox model with a hazard ratio can provide a relative treatment-effect summary when its assumptions are appropriate. The registry, however, does not post a formal statistical analysis or numerical result for the primary endpoint.
The study therefore provides a useful framework for understanding how randomized prevention trials with disease-free survival endpoints are designed and analyzed: the randomized treatment comparison establishes the primary contrast, time-to-event methods preserve information about event timing and censoring, and the three-year endpoint supplies a clinically defined horizon for interpretation.