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
BETonMACE was a randomized, parallel-group, triple-masked phase 3 trial evaluating RVX000222 against placebo in high-risk subjects with type 2 diabetes and coronary artery disease. The primary endpoint was the incidence of first occurrence of adjudication-confirmed narrowly defined MACE over 120 weeks.
| Feature | BETonMACE |
|---|---|
| Phase | Phase 3 |
| Condition | Diabetes Mellitus, Type 2; Coronary Artery Disease |
| Brief title | Effect of RVX000222 on Time to Major Adverse Cardiovascular Events in High-Risk T2DM Subjects With CAD |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Triple |
| Primary purpose | Prevention |
| Enrollment | 2,425 |
| Lead sponsor | Resverlogix Corp |
| Sponsor type | Industry |
| Study period | Started November 2015; primary completion November 30, 2020 |
2. Clinical Question
The central question was whether adding RVX000222 to high-intensity statin therapy changed the time to the first occurrence of adjudication-confirmed narrowly defined MACE compared with high-intensity statin therapy plus placebo over 120 weeks.
Population
High-risk subjects with type 2 diabetes mellitus and coronary artery disease.
Intervention
RVX000222 (apabetalone) in combination with high-intensity statin therapy.
Comparator
Placebo in combination with high-intensity statin therapy.
Primary question
Does RVX000222 alter the time to first adjudication-confirmed narrowly defined MACE during the 120-week assessment period?
3. Trial Design
High-Intensity Statin Therapy + RVX000222
- RVX000222 (apabetalone)
- High-intensity statin therapy
- Randomized treatment assignment
- Triple-masked study design
High-Intensity Statin Therapy + Placebo
- Placebo
- High-intensity statin therapy
- Randomized treatment assignment
- Triple-masked study design
The registry identifies the interventions as apabetalone, placebo, atorvastatin, and rosuvastatin. The statistical comparison reported for the primary endpoint is between high-intensity statin therapy plus RVX000222 and high-intensity statin therapy plus placebo.
4. Primary Endpoint
| Endpoint | Registry definition | Time frame | Statistical analysis |
|---|---|---|---|
| Incidence of First Occurrence of Adjudication-confirmed Narrowly Defined MACE | Incidence of first occurrence of MACE narrowly defined as a single composite endpoint of Cardiovascular (CV) Death (including undetermined cause of death) or Non-fatal Myocardial Infarction (MI) or Stroke. If a subject has multiple events, only the first event contributing to the composite endpoint is counted. | 120 weeks | Stratified log-rank; Cox proportional hazard; hazard ratio |
This is a time-to-event composite endpoint. The outcome is not simply whether a participant ever experienced one of the component events. The registry definition specifies the time to the first qualifying event and counts only the first event contributing to the composite.
5. Secondary and Post-Hoc Endpoints
| Endpoint | Role | Time frame | Effect measure |
|---|---|---|---|
| Incidence of First Occurrence of Adjudication-confirmed Broadly Defined MACE | Secondary | 120 weeks | Hazard ratio |
| Incidence of Hospitalization for Congestive Heart Failure (CHF) | Secondary | 120 weeks | Hazard ratio |
| Incidence of All-cause Mortality | Secondary | 120 weeks | Hazard ratio |
| Incidence of First Occurrence of Adjudication-Confirmed Four-Point MACE | Post hoc | 120 weeks | Hazard ratio |
The registry reports five statistical analyses in total: one primary analysis, three secondary analyses, and one post-hoc analysis. All five use a time-to-event framework with hazard-ratio effect measures.
6. Statistical Methodology
Time-to-event analysis
The primary endpoint is a time-to-event outcome because the registry describes the incidence of the first qualifying MACE over a specified 120-week period. Time-to-event methods use both whether an event occurred and how long each participant was observed before the event or censoring.
For BETonMACE, the relevant event is the first adjudication-confirmed component contributing to the narrowly defined MACE composite. Participants without the event by the relevant end of observation contribute information up to their censoring time.
Stratified log-rank test
The registry reports a stratified log-rank method for the primary analysis and the posted secondary analyses. The log-rank framework compares the observed pattern of events between treatment groups across follow-up, while stratification permits the comparison to account for prespecified or analytically relevant strata.
The registry does not identify the specific stratification variables in the information summarized here. The statistical interpretation therefore should not assign particular baseline variables to the strata without additional documentation.
Cox proportional-hazards model
The primary effect measure is reported as a Cox proportional hazard, normalized here as a hazard ratio. The Cox model expresses the relative instantaneous event rate between the treatment groups while retaining the time-to-event structure of the data.
An HR below 1 indicates a lower estimated instantaneous event rate in the RVX000222 group relative to the placebo group under the fitted model. It is not an absolute risk difference and does not mean that every participant experiences the same proportional change in risk.
Two-sided confidence intervals
The registry-posted confidence intervals are 95% and two-sided. A confidence interval describes statistical uncertainty around the estimated hazard ratio. It is distinct from the observed range of outcomes among individual participants.
Superiority hypothesis
The registered hypothesis type for the posted analyses is superiority. This means the statistical comparison was structured around detecting a difference between the randomized groups rather than demonstrating that the groups were sufficiently similar within a prespecified non-inferiority margin.
7. Primary Result: Narrowly Defined MACE
The primary analysis compared high-intensity statin therapy plus RVX000222 with high-intensity statin therapy plus placebo for the incidence of first occurrence of adjudication-confirmed narrowly defined MACE at 120 weeks.
Primary hazard ratio
95% CI: 0.649–1.044 · P = 0.1070
Stratified log-rank analysis; Cox proportional-hazard effect measure; superiority hypothesis.
| Primary endpoint | RVX000222 + high-intensity statin | Placebo + high-intensity statin | Effect estimate |
|---|---|---|---|
| First occurrence of adjudication-confirmed narrowly defined MACE at 120 weeks | Randomized treatment group | Randomized comparator group | HR 0.823 95% CI 0.649–1.044 P = 0.1070 |
An HR of 0.823 means that the fitted Cox model estimated an instantaneous MACE hazard approximately 17.7% lower in the RVX000222 group than in the placebo group, because 1 − 0.823 = 0.177. This is a relative model-based measure, not an absolute reduction in the probability of MACE.
The 95% CI of 0.649–1.044 expresses uncertainty around the estimated hazard ratio. It spans 1, so the interval includes values corresponding to a lower estimated hazard as well as values close to or above the null value of 1.
The P = 0.1070 value is a measure of compatibility of the observed test statistic with the null hypothesis under the specified statistical framework. It is not a measure of the size, clinical importance, or probability of the treatment effect.
Because the endpoint is time-to-event, the interpretation also depends on censoring and the assumptions underlying the Cox model. In particular, a single hazard ratio is most straightforward when the proportional-hazards assumption is a reasonable approximation over the analyzed period.
8. Secondary Result: Broadly Defined MACE
The registry reports a secondary analysis for the incidence of first occurrence of adjudication-confirmed broadly defined MACE at 120 weeks.
Hazard ratio for broadly defined MACE
95% CI: 0.679–1.061 · P = 0.1495
Stratified log-rank analysis; Cox proportional-hazard effect measure; superiority hypothesis.
The HR of 0.849 corresponds to an estimated instantaneous event rate approximately 15.1% lower in the RVX000222 group relative to the placebo group under the fitted model.
The 95% CI of 0.679–1.061 indicates uncertainty around that estimate and includes 1. The interval therefore does not isolate a single direction of effect at the conventional null value.
The P = 0.1495 value should not be interpreted as the probability that RVX000222 works or does not work. It describes the statistical evidence under the specified hypothesis-testing framework, not the clinical magnitude of the observed hazard ratio.
9. Secondary Result: Hospitalization for Congestive Heart Failure
The registry also reports the incidence of hospitalization for congestive heart failure at 120 weeks.
Hazard ratio for CHF hospitalization
95% CI: 0.38–0.94 · P = 0.03
Stratified log-rank analysis; Cox proportional-hazard effect measure; superiority hypothesis.
An HR of 0.59 corresponds to an estimated instantaneous hospitalization hazard approximately 41% lower in the RVX000222 group than in the placebo group under the Cox model.
The 95% CI of 0.38–0.94 quantifies uncertainty around the estimated hazard ratio and lies below 1. The interval nevertheless does not establish a particular absolute reduction in hospitalization risk because the hazard ratio is a relative time-to-event measure.
The P = 0.03 value provides evidence against the null hypothesis within the reported analysis framework. It should not be interpreted as an effect-size measure, and this endpoint is secondary rather than the registered primary endpoint.
Because multiple outcomes are reported in the trial, interpretation of a secondary P-value also depends on the trial's multiplicity framework. The registry information summarized here does not specify a multiplicity-adjustment strategy for these secondary analyses.
10. Secondary Result: All-Cause Mortality
The incidence of all-cause mortality was also analyzed as a secondary time-to-event endpoint over 120 weeks.
Hazard ratio for all-cause mortality
95% CI: 0.62–1.24 · P = 0.44
Stratified log-rank analysis; Cox proportional-hazard effect measure; superiority hypothesis.
The HR of 0.88 represents an estimated instantaneous mortality hazard approximately 12% lower in the RVX000222 group under the fitted model.
The 95% CI of 0.62–1.24 is relatively broad and spans the null value of 1. The result therefore contains substantial statistical uncertainty about the direction and magnitude of the underlying relative hazard.
The P = 0.44 value is not a statement that the treatment effect is exactly zero. It indicates that the observed result does not provide strong statistical evidence against the null hypothesis under the reported testing framework.
11. Post-Hoc Result: Four-Point MACE
The registry reports a post-hoc analysis of the incidence of first occurrence of adjudication-confirmed four-point MACE at 120 weeks.
Hazard ratio for four-point MACE
95% CI: 0.63–0.98 · P = 0.03
Log-rank analysis; Cox proportional-hazard effect measure; superiority hypothesis.
An HR of 0.78 corresponds to an estimated instantaneous event rate approximately 22% lower in the RVX000222 group relative to placebo under the fitted model.
The 95% CI of 0.63–0.98 lies below 1, while still reflecting uncertainty around the estimated treatment effect. It does not indicate that every patient experienced a 22% reduction in risk, nor does it provide an absolute risk difference.
The P = 0.03 should be read in the context of the post-hoc status of this endpoint. A post-hoc result has a different inferential role from a prespecified primary endpoint, particularly when multiple endpoints or analyses have been examined.
12. Comparing the Time-to-Event Estimates
| Endpoint | Role | HR | 95% CI | P-value |
|---|---|---|---|---|
| Narrowly defined MACE | Primary | 0.823 | 0.649–1.044 | 0.1070 |
| Broadly defined MACE | Secondary | 0.849 | 0.679–1.061 | 0.1495 |
| Hospitalization for CHF | Secondary | 0.59 | 0.38–0.94 | 0.03 |
| All-cause mortality | Secondary | 0.88 | 0.62–1.24 | 0.44 |
| Four-point MACE | Post hoc | 0.78 | 0.63–0.98 | 0.03 |
The hazard ratios are all below 1, but their precision and inferential roles differ. The primary endpoint has an estimate of 0.823 with a confidence interval extending above 1. The secondary CHF analysis has a narrower interval below 1, while all-cause mortality has a wider interval spanning 1. The four-point MACE analysis is explicitly post hoc and therefore should not be interpreted as equivalent to a prespecified primary analysis.
13. Statistical Methods Explained
Why is a time-to-event method appropriate for the primary endpoint?
The primary endpoint records the first qualifying MACE and its timing over 120 weeks. Time-to-event methods use information from participants who experience the event as well as participants who remain event-free until censoring, rather than reducing follow-up to a simple yes/no outcome.
What does the hazard ratio of 0.823 mean?
Under the fitted Cox model, the estimated instantaneous MACE hazard was 0.823 times that in the comparator group. Expressed as a relative reduction, 1 − 0.823 = 0.177, or approximately 17.7%. This is not an absolute risk reduction and does not mean that 17.7% fewer participants necessarily experienced MACE.
Why is the confidence interval more informative than the point estimate alone?
The point estimate 0.823 is one estimate from the observed data. The 95% CI of 0.649–1.044 shows the uncertainty around that estimate under the stated statistical framework. A wider interval indicates less precision than a narrower interval, all else being equal.
What does the log-rank test contribute?
The stratified log-rank test compares the time-to-event experience between the randomized groups across follow-up. It evaluates the pattern of observed events while retaining the ordering of event times and censoring information. The Cox model then supplies the reported hazard-ratio effect estimate.
Why does P = 0.1070 not mean that the treatment effect is zero?
A p-value does not estimate the probability that the null hypothesis is true. A value of 0.1070 summarizes the compatibility of the observed test statistic with the null hypothesis under the specified model and testing procedure. It should be considered alongside the hazard ratio and its confidence interval.
Why should the CHF and four-point MACE results be interpreted differently from the primary MACE result?
The CHF result is a secondary endpoint and the four-point MACE result is post hoc, whereas narrowly defined MACE is the registered primary endpoint. Statistical interpretation depends not only on the numerical estimate but also on the endpoint's prespecified role, the multiplicity structure, and the hypothesis being tested.
14. Reading the Confidence Intervals
The confidence intervals make an important distinction between the estimated treatment effect and the uncertainty around it.
| Endpoint | Point estimate | Lower bound | Upper bound | Relationship to HR = 1 |
|---|---|---|---|---|
| Narrowly defined MACE | 0.823 | 0.649 | 1.044 | Spans 1 |
| Broadly defined MACE | 0.849 | 0.679 | 1.061 | Spans 1 |
| CHF hospitalization | 0.59 | 0.38 | 0.94 | Below 1 |
| All-cause mortality | 0.88 | 0.62 | 1.24 | Spans 1 |
| Four-point MACE | 0.78 | 0.63 | 0.98 | Below 1 |
For a hazard ratio, the null value is 1. An interval extending across 1 means that the data are compatible, within the stated confidence framework, with both a lower and a higher hazard relative to the comparator. An interval entirely below 1 indicates that the values in that confidence interval correspond to a lower estimated hazard in the RVX000222 group.
15. Primary vs Secondary vs Post-Hoc Inference
The statistical hierarchy matters. The primary endpoint was narrowly defined MACE. Broadly defined MACE, CHF hospitalization, and all-cause mortality were secondary endpoints. Four-point MACE was identified as post hoc.
Primary endpoint
Narrowly defined MACE was the registered primary endpoint and therefore the central prespecified efficacy question.
Secondary endpoints
Broadly defined MACE, CHF hospitalization, and all-cause mortality provide additional outcome-specific information beyond the primary endpoint.
Post-hoc endpoint
Four-point MACE was analyzed after the primary/secondary endpoint framework and therefore has a more exploratory inferential role.
Multiplicity
When several hypotheses are examined, the interpretation of individual p-values depends on whether and how the overall type I error was controlled.
16. Safety: Serious Adverse Events
The registry reports serious adverse events by randomized treatment arm using affected participants and the corresponding at-risk populations.
| Treatment group | Serious adverse events | At risk |
|---|---|---|
| High-Intensity Statin Therapy + RVX000222 | 354 | 1,212 |
| High-Intensity Statin Therapy + Placebo | 339 | 1,207 |
The reported serious-adverse-event counts are 354/1,212 in the RVX000222 group and 339/1,207 in the placebo group. These are safety summaries rather than time-to-event efficacy estimates and should be interpreted separately from the MACE analyses.
17. What the Primary Hazard Ratio Does — and Does Not — Mean
The primary HR of 0.823 indicates an estimated instantaneous hazard approximately 17.7% lower for narrowly defined MACE in the RVX000222 group relative to placebo under the fitted Cox model.
This does not mean that 17.7% of participants avoided MACE, that every individual experienced a 17.7% reduction, or that the absolute probability of MACE was reduced by 17.7 percentage points.
The 95% CI of 0.649–1.044 communicates uncertainty around the primary hazard-ratio estimate. The interval includes 1, the null value for a hazard ratio, and therefore encompasses values on both sides of the null.
The primary P = 0.1070 is a hypothesis-testing quantity. It does not quantify the magnitude of the treatment effect, the probability that RVX000222 is effective, or the probability that the observed estimate will repeat in another study.
Time-to-event inference depends on how follow-up is censored and on the assumptions of the statistical model. In particular, the Cox proportional-hazards interpretation treats the hazard ratio as the relevant relative measure over time. If hazards vary substantially in a non-proportional way, a single HR can compress important features of the underlying event curves.
18. A Statistical Reading of the Complete Results
The five posted analyses show a consistent pattern of hazard ratios below 1, but the estimates have different levels of precision and different inferential status.
| Endpoint | What the estimate says | What requires caution |
|---|---|---|
| Narrowly defined MACE | Estimated HR 0.823, corresponding to approximately 17.7% lower estimated instantaneous hazard. | 95% CI 0.649–1.044 spans 1; this is the primary endpoint. |
| Broadly defined MACE | Estimated HR 0.849, corresponding to approximately 15.1% lower estimated instantaneous hazard. | 95% CI 0.679–1.061 spans 1; this is a secondary endpoint. |
| CHF hospitalization | Estimated HR 0.59, corresponding to approximately 41% lower estimated instantaneous hazard. | Secondary endpoint; the registry summary does not specify the multiplicity-adjustment framework. |
| All-cause mortality | Estimated HR 0.88, corresponding to approximately 12% lower estimated instantaneous hazard. | 95% CI 0.62–1.24 spans 1; this is a secondary endpoint. |
| Four-point MACE | Estimated HR 0.78, corresponding to approximately 22% lower estimated instantaneous hazard. | Post-hoc analysis; its inferential role differs from the registered primary endpoint. |
This distinction is important because a set of point estimates should not be treated as though each were an independent confirmatory conclusion. The primary endpoint, secondary endpoints, and post-hoc endpoint occupy different positions in the statistical hierarchy.
19. Important Limitations and Interpretation Issues
- Primary endpoint uncertainty: the 95% confidence interval for the primary MACE hazard ratio is 0.649–1.044 and therefore crosses the null value of 1.
- Multiple analyses: the registry reports five statistical analyses across primary, secondary, and post-hoc endpoints. Interpretation of individual p-values depends on the multiplicity framework.
- Post-hoc analysis: the four-point MACE analysis is identified as post hoc and should not be assigned the same confirmatory status as the registered primary endpoint.
- Hazard-ratio interpretation: a Cox hazard ratio is a relative time-to-event measure and is not equivalent to an absolute risk reduction or relative risk.
- Proportional hazards: the Cox model's standard interpretation relies on the proportional-hazards framework. A single hazard ratio may not fully describe treatment effects if relative hazards change materially over time.
- Composite endpoint: narrowly defined MACE combines cardiovascular death, non-fatal myocardial infarction, and stroke. A composite estimate summarizes the first qualifying component event rather than establishing that treatment effects are identical for every component.
- Registry detail: the statistical information summarized here does not identify the specific stratification variables used for the reported stratified analyses.
- Safety denominators: serious-adverse-event summaries use the reported at-risk populations of 1,212 and 1,207 rather than the overall enrollment count of 2,425.
20. Why This Trial Matters Statistically
BETonMACE is a useful teaching example for understanding how randomized cardiovascular trials translate clinical events into time-to-event estimands. Its statistical structure includes randomized allocation, triple masking, a composite primary endpoint, stratified log-rank testing, Cox hazard ratios, confidence intervals, secondary endpoints, a post-hoc analysis, and separate safety summaries.
| Concept | How it appears in BETonMACE |
|---|---|
| Randomization | Randomized allocation to RVX000222 or placebo in a parallel-group phase 3 design. |
| Blinding | Triple-masked study design. |
| Time-to-event endpoint | First occurrence of adjudication-confirmed MACE over 120 weeks. |
| Composite endpoint | Narrowly defined MACE combines cardiovascular death, non-fatal MI, and stroke. |
| Kaplan-Meier framework | Appropriate framework for describing time-to-first-event distributions when follow-up is censored. |
| Log-rank test | Stratified log-rank analysis was reported for the posted efficacy analyses. |
| Hazard ratio | Cox proportional-hazard effect measure was reported for all five posted analyses. |
| Confidence interval | All posted estimates use two-sided 95% confidence intervals. |
| Superiority | The registered hypothesis type for the posted analyses is superiority. |
| Multiplicity | Primary, secondary, and post-hoc analyses have different inferential roles. |
| Safety analysis | Serious adverse events are reported separately by treatment group and at-risk population. |
21. Related Tutorials
Learn more about the methods used in this trial:
22. Related Calculators
23. Sources
- ClinicalTrials.gov: NCT02586155 — BETonMACE.
- PubMed: Publication indexed under PMID 32219359.
- PubMed: Publication indexed under PMID 34158057.
- PubMed: Publication indexed under PMID 31725015.
- PubMed: Publication indexed under PMID 31520897.
- PubMed: Publication indexed under PMID 30476723.
Continue through Clinical Biostats
Explore statistical tutorials, calculators, and additional clinical trial analyses covering randomized trials, survival analysis, and time-to-event methodology.
24. Record Summary
BETonMACE provides a clear example of a randomized cardiovascular time-to-event analysis in which the primary endpoint is a composite of cardiovascular death, non-fatal myocardial infarction, and stroke. The primary hazard ratio was 0.823 with a two-sided 95% CI of 0.649–1.044 and P = 0.1070. The secondary analyses reported hazard ratios of 0.849 for broadly defined MACE, 0.59 for hospitalization for congestive heart failure, and 0.88 for all-cause mortality. A post-hoc four-point MACE analysis reported an HR of 0.78.
The statistical lesson is that these numbers should be read together with their endpoint definitions, analysis roles, confidence intervals, and time-to-event methodology. A hazard ratio supplies a relative measure of instantaneous event rate; the confidence interval describes uncertainty around that estimate; the p-value addresses a hypothesis-testing question; and the endpoint hierarchy determines how each analysis fits into the overall evidentiary structure of the trial.