ORION-4: Complete Statistical Analysis of Inclisiran in Atherosclerotic Cardiovascular Disease

1. Trial at a Glance

ORION-4 is a randomized phase 3 trial evaluating whether inclisiran affects the occurrence of major adverse cardiovascular events in people with atherosclerotic cardiovascular disease. The trial uses a parallel-group, quadruple-masked design with 16,124 participants assigned to inclisiran or placebo.

2. Clinical Question

The central statistical question is whether randomized assignment to inclisiran changes the time to first major adverse cardiovascular event compared with placebo among people with atherosclerotic cardiovascular disease.

3. Trial Design

The randomized parallel-group structure creates the fundamental comparison for the primary endpoint. Because participants are assigned to one of two arms and followed prospectively, the principal efficacy analysis can be framed around the difference in time-to-first-event distributions between the randomized groups.

4. Masking and Allocation

Randomization and masking address different sources of potential bias. Randomization establishes the comparison structure, while masking can reduce the influence of treatment knowledge on trial conduct and outcome assessment. For a time-to-event endpoint such as MACE, these design features are important because the analysis depends not only on whether an event occurs but also on when it occurs.

5. Primary Endpoint

The endpoint is therefore fundamentally a time-to-first-event outcome. Although the registry labels it as the number of participants with MACE, its definition specifies the time to the first occurrence of one of four component events during the scheduled treatment period.

6. Understanding the MACE Endpoint Statistically

A composite endpoint changes the statistical question from asking whether one particular event occurs to asking whether a participant experiences any first event in a predefined set of clinically relevant events. This can increase the number of observed endpoint events and therefore provide more statistical information than analyzing a less frequent single component alone.

Because the endpoint is defined by the first qualifying occurrence, subsequent events after a participant has already experienced MACE do not create additional first-MACE events for the primary analysis. This is one reason the endpoint should not be interpreted as a count of every cardiovascular event experienced by every participant.

7. Planned Analysis

For a time-to-first-MACE endpoint, the natural statistical framework is a survival-analysis approach. Rather than treating every participant as simply having or not having MACE, the analysis can use the time from the relevant trial origin to the first qualifying event. Participants who have not experienced MACE by their last evaluable follow-up contribute information up to that point and are typically treated as right-censored observations.

A typical analysis for this endpoint would use Kaplan-Meier methods to estimate the event-free probability over time, a log-rank-type comparison to compare the randomized groups, and a Cox proportional-hazards model to estimate a hazard ratio with a confidence interval. Those are standard methods for time-to-first-event cardiovascular endpoints; the registry record does not post a formal statistical analysis specifying that these particular methods were used for ORION-4.

Why time matters

Two treatment groups can have the same eventual proportion of participants experiencing an event while differing substantially in when those events occur. Time-to-event analysis preserves that timing information. It also allows participants who remain event-free at the end of their observed follow-up to contribute partial information rather than being treated as if their eventual outcome were known.

8. Statistical Methodology

Kaplan-Meier estimation

For a time-to-first-MACE endpoint, the Kaplan-Meier estimator provides an estimate of the probability of remaining free of the first qualifying event through each point in time. The method accommodates right censoring, which occurs when a participant's event-free follow-up ends before a MACE event is observed.

Hazard ratio

A Cox proportional-hazards model is commonly used to summarize the relative instantaneous event rate between randomized groups. A hazard ratio below 1 would indicate a lower estimated hazard in the inclisiran group relative to placebo under the fitted model; a hazard ratio above 1 would indicate a higher estimated hazard.

Confidence interval

A confidence interval around a treatment-effect estimate describes statistical uncertainty associated with the estimate under the specified model and sampling framework. A narrow interval indicates greater statistical precision than a wide interval, all else being equal.

Log-rank comparison

A log-rank test is a standard method for comparing time-to-event distributions between randomized groups. It incorporates the ordering of event times and the risk sets at those times rather than reducing the endpoint to a simple binary outcome.

Analysis population

For randomized efficacy comparisons, the intention-to-treat principle is commonly used so that participants are analyzed according to their randomized assignment. The registry information posted on ClinicalTrials.gov for ORION-4 does not specify an analysis population for the primary endpoint beyond the randomized trial structure.

9. Statistical Methods Explained

Why is MACE analyzed as a time-to-event endpoint?

The registry defines MACE in terms of the time to first occurrence of a qualifying cardiovascular event. That wording makes timing part of the endpoint itself. Survival-analysis methods can therefore use both whether an event occurred and when it occurred.

Why not simply compare the percentage of participants with MACE?

A simple proportion ignores differences in follow-up time and event timing. Time-to-event analysis can retain information from participants who remain event-free for different lengths of time and can distinguish early events from later events.

What does a hazard ratio describe?

A hazard ratio compares the estimated instantaneous event rates between two groups within a time-to-event model. For example, an HR of 0.70 would correspond to an estimated hazard 30% lower in the treatment group under the model. It would not mean that exactly 30% fewer participants experience MACE by the end of follow-up.

What does right censoring mean?

A participant is right-censored when the observation period ends before a qualifying MACE is observed. The participant still contributes information through the last time at which they were known to be event-free. Censoring therefore differs from experiencing the event.

Why is the composite endpoint important?

MACE combines coronary heart disease death, myocardial infarction, fatal or non-fatal ischemic stroke, and urgent coronary revascularization into a single first-event endpoint. This creates one primary time-to-event outcome while combining several clinically distinct cardiovascular events.

What assumption matters for a Cox proportional-hazards model?

The conventional Cox model summarized by a single hazard ratio relies on a proportional-hazards interpretation: the relative hazard between groups is assumed to be reasonably stable over the relevant follow-up. If hazards change substantially over time, a single hazard ratio can become less representative of the full pattern of treatment effects, and time-specific measures or alternative models may be informative.

10. Interpreting a Future Primary Result

When a formal primary analysis becomes available, the most informative interpretation should combine the relative treatment estimate with its uncertainty and an absolute description of event-free survival over time.

For a composite endpoint, interpretation should also examine the individual MACE components. A treatment effect on the composite can arise from different patterns across coronary heart disease death, myocardial infarction, ischemic stroke, and urgent coronary revascularization. The clinical meaning of the overall composite therefore depends partly on how its components contribute to the observed result.

11. Composite Endpoint Interpretation

Composite endpoints are efficient statistically but require careful clinical interpretation. The four components in ORION-4 are not interchangeable outcomes: death, myocardial infarction, ischemic stroke, and urgent coronary revascularization can differ in severity, frequency, and clinical consequences.

The registry information currently available does not provide component-specific event counts or treatment-effect estimates. Those details would be important for a complete interpretation of the eventual MACE analysis.

12. Follow-Up and Trial Status

The current registry status is active, not recruiting. Because the primary completion date is listed as October 2026 and no formal statistical analyses are posted in the registry record, the public record does not currently provide a numerical treatment-effect result for the primary MACE endpoint.

13. Multiplicity, Interim Analysis, and Other Design Issues

The registry information identifies the randomized, parallel-group, quadruple-masked design and the single registered primary MACE endpoint. It does not provide a posted statistical analysis plan specifying multiplicity procedures, alpha spending, interim-analysis boundaries, stratification factors, missing-data methods, imputation procedures, or Bayesian methods.

These distinctions matter because statistical interpretation depends on the prespecified analysis plan. A p-value or confidence interval cannot be interpreted independently of the hypothesis, analysis population, multiplicity framework, and model used to obtain it.

14. Crossover and Treatment Switching

The registry information provided for ORION-4 identifies two randomized parallel groups, inclisiran and placebo, but does not report crossover or treatment-switching procedures.

For a randomized time-to-event analysis, treatment switching can complicate interpretation if participants subsequently receive a treatment different from their randomized assignment. An intention-to-treat analysis generally preserves the original randomized comparison, while additional treatment-policy or as-treated analyses may answer different questions. No such additional analysis is posted in the registry information currently available.

15. Missing Data and Censoring

Time-to-event trials typically distinguish between an observed first event and right censoring. A participant who remains free of MACE through the last evaluable observation contributes follow-up information without being classified as having experienced MACE at the censoring time.

The validity of survival analysis depends on appropriate handling and assumptions surrounding censoring. The registry information does not report the prespecified censoring rules or any missing-data or imputation strategy for the primary endpoint.

16. What a Complete MACE Analysis Should Report

A statistically complete presentation of ORION-4's primary endpoint would ideally allow readers to evaluate both the magnitude and precision of the randomized treatment comparison.

These elements answer different questions. The hazard ratio alone does not describe absolute event probability, and a p-value alone does not describe either effect magnitude or precision. A complete statistical interpretation therefore uses the measures together.

17. Limitations

• No posted efficacy results: the ClinicalTrials.gov record does not currently provide numerical treatment-effect estimates, event counts, confidence intervals, or p-values for the primary MACE endpoint.
• Composite endpoint: MACE combines four distinct cardiovascular outcomes, so an overall composite effect should be interpreted alongside its individual components when those results become available.
• Time-to-event interpretation: a hazard ratio, if reported, would summarize a relative event rate under a statistical model and would not directly represent an absolute reduction in the probability of MACE.
• Proportional-hazards assumption: if a Cox model is used, interpretation of a single hazard ratio depends on the adequacy of the proportional-hazards assumption over the relevant follow-up.
• Censoring: survival analysis requires appropriate handling and assumptions concerning participants whose observation ends before MACE is observed.
• Unreported analysis details: the available registry information does not specify the detailed statistical analysis plan, including stratification, interim monitoring, multiplicity, missing-data procedures, or imputation methods.
• Timing of completion: the registry lists primary completion in October 2026, so the current public record does not yet contain the final numerical primary-endpoint analysis.

18. Why This Trial Matters Statistically

ORION-4 provides a useful teaching example of how a large randomized cardiovascular outcomes trial can be structured around a composite time-to-event endpoint. The statistical questions extend beyond whether a difference in the number of participants with MACE is observed: the timing of events, censoring, component composition, relative effect size, absolute risk, and uncertainty all contribute to the interpretation.

19. Statistical Interpretation vs Clinical Interpretation

Keeping these two perspectives separate is important. Statistical evidence describes the magnitude, uncertainty, and compatibility of the observed randomized comparison with the prespecified hypothesis. Clinical interpretation asks what that evidence means in the context of the component outcomes and the population studied.

20. Record Summary

ORION-4 is a randomized phase 3 prevention trial enrolling 16,124 participants with atherosclerotic cardiovascular disease and comparing inclisiran with placebo in a parallel-group, quadruple-masked design. Its registered primary endpoint is the number of participants with a major adverse cardiovascular event, defined as time to first occurrence during the scheduled treatment period of coronary heart disease death, myocardial infarction, fatal or non-fatal ischemic stroke, or urgent coronary revascularization procedure, with a stated time frame of median follow-up of 5-years.

The statistical structure is therefore centered on time-to-first-event methodology. Kaplan-Meier estimation, log-rank comparison, and Cox proportional-hazards modeling are standard analytical frameworks for this type of endpoint, although the available registry information does not specify which formal methods were prespecified for ORION-4. The registry currently contains no posted statistical analyses or numerical primary-endpoint results.

21. Sources

• ClinicalTrials.gov: ORION-4, NCT03705234.