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
AVERT was a randomized phase 2 trial evaluating apixaban for prevention of venous thromboembolism in cancer patients. The study used a parallel-group, quadruple-masked design comparing apixaban with placebo.
| Feature | AVERT |
|---|---|
| Clinical question | Prevention of venous thromboembolism in cancer patients |
| Phase | Phase 2 |
| Enrollment | 575 |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Primary purpose | Prevention |
| Lead sponsor | Ottawa Hospital Research Institute |
2. Clinical Question
The central question of AVERT was whether apixaban could prevent venous thromboembolic events in cancer patients compared with placebo during the registry-defined assessment period.
Population
Patients with cancer and risk of venous thromboembolism.
Intervention
Apixaban.
Comparator
Placebo drug.
Primary question
Does apixaban affect the occurrence of objectively documented venous thromboembolism?
3. Trial Design
Apixaban
Participants received apixaban as the study intervention.
Placebo
Participants received placebo drug as the comparator intervention.
4. Endpoints
| Endpoint | Definition | Time frame |
|---|---|---|
| Primary endpoint | First episode of objectively documented, symptomatic or asymptomatic VTE (DVT and/or PE) | 7 months |
5. Planned Analysis
The primary endpoint is a time-limited clinical event outcome. A statistical analysis of this type of endpoint would typically compare the occurrence of first venous thromboembolism events between randomized groups using methods appropriate for binary or time-to-event outcomes, depending on the prespecified statistical analysis plan.
For a prevention trial, analysis generally focuses on estimating the difference between randomized groups while preserving the benefits of randomization. Methods may include event-rate comparisons, relative measures of treatment effect, confidence intervals, and adjustment for follow-up time when appropriate.
6. Statistical Methodology
Randomization and causal comparison
Randomization creates the basis for comparing apixaban and placebo groups without relying on observational comparisons. The treatment groups are expected to be comparable on average at the time of assignment, allowing differences in outcomes to be attributed to the randomized intervention under the trial assumptions.
Event-based endpoint analysis
The primary endpoint combines objectively documented symptomatic or asymptomatic venous thromboembolic events, including deep vein thrombosis and pulmonary embolism. Event-based outcomes are commonly analyzed by comparing the number and timing of events between randomized groups.
The interpretation depends on the prespecified analysis method, event definitions, follow-up period, and handling of participants without observed events.
Masked randomized design
The quadruple-masked design reduces the potential for knowledge of treatment assignment to influence participant management, outcome assessment, or interpretation of events.
7. Statistical Methods Explained
Why was randomization important in AVERT?
Randomization helps balance measured and unmeasured factors between the apixaban and placebo groups, supporting a causal comparison of the intervention.
Why include asymptomatic VTE events?
The primary endpoint included objectively documented symptomatic or asymptomatic VTE. Including objectively confirmed events can capture clinically relevant thromboembolic outcomes beyond events identified only through symptoms.
What does a prevention endpoint measure?
A prevention endpoint evaluates whether an intervention changes the occurrence of an event before it happens, rather than treating an established disease condition.
Why does follow-up duration matter?
The registry defines the primary endpoint timeframe as 7 months. Event rates and treatment comparisons depend on the period during which outcomes are assessed.
8. Limitations
- No posted statistical analyses: the registry does not report primary endpoint estimates, confidence intervals, or p-values.
- Endpoint interpretation: composite clinical endpoints combine multiple event types, and interpretation depends on the contribution of each component.
- Follow-up period: the primary endpoint timeframe was 7 months, so conclusions are limited to the registry-defined assessment window.
- Phase 2 design: the study phase indicates an earlier-stage randomized evaluation.
- Registry information: ClinicalTrials.gov provides registration information, but complete statistical interpretation generally depends on the full statistical analysis plan and reported results.
9. Why This Trial Matters Statistically
AVERT is a useful teaching example because it illustrates several important principles in randomized prevention trials: event-based outcomes, masked treatment allocation, and the distinction between a registered endpoint and a reported statistical result.
| Concept | How it appears in AVERT |
|---|---|
| Randomization | Randomized allocation of participants |
| Parallel design | Two concurrent intervention groups |
| Masking | Quadruple-masked study design |
| Prevention endpoint | First objectively documented VTE event |
| Clinical event analysis | Assessment of DVT and/or PE events |
10. Sources
- ClinicalTrials.gov: NCT02048865.
- PubMed: Publication record.
- PubMed: Publication record.
- PubMed: Publication record.
- PubMed: Publication record.
- PubMed: Publication record.
Continue through the Clinical Biostats knowledge graph
Explore clinical trial methods, statistical concepts, and analytical workflows used across modern biomedical research.
11. Record Summary
AVERT provides an example of a randomized, quadruple-masked phase 2 prevention trial evaluating an anticoagulant intervention in cancer patients. The registry defines a primary endpoint based on the first objectively documented symptomatic or asymptomatic venous thromboembolic event over 7 months. Interpretation of treatment effects requires the completed statistical analysis, including reported event estimates, uncertainty measures, and prespecified analytical methods.