This page separates reported trial results from statistical interpretation. Numerical results are taken from the ClinicalTrials.gov record. The registry provides the official trial record.
1. Trial at a Glance
ORION-11 was a randomized, double-blind, parallel-design phase 3 trial evaluating inclisiran sodium versus placebo in subjects with ASCVD or ASCVD-risk equivalents and elevated LDL-C. The registry reports 1617 enrolled subjects, two treatment arms, two registered primary endpoints, and formal ANCOVA analyses for both primary endpoints.
| Feature | ORION-11 |
|---|---|
| Trial name | ORION-11 |
| Phase | Phase 3 |
| Condition | ASCVD; Risk Factor, Cardiovascular; Elevated Cholesterol |
| Population | Subjects with ASCVD or ASCVD-risk equivalents and elevated low-density lipoprotein cholesterol |
| Design | Randomized, parallel |
| Masking | Double |
| Primary purpose | Treatment |
| Enrollment | 1617 |
| Interventions | Inclisiran Sodium; Placebo |
| Lead sponsor | The Medicines Company |
| Sponsor type | Industry |
| ClinicalTrials.gov | NCT03400800 |
2. Clinical Question
The central question was whether treatment with inclisiran produced a greater reduction in LDL-C than placebo over the registered assessment periods in subjects with ASCVD or ASCVD-risk equivalents and elevated LDL-C.
Population
Subjects with ASCVD or ASCVD-risk equivalents and elevated low-density lipoprotein cholesterol.
Intervention
Inclisiran Sodium.
Comparator
Placebo.
Primary question
Does inclisiran produce a greater reduction in LDL-C than placebo at the registered endpoint time frames?
3. Trial Design
Inclisiran
- Inclisiran Sodium
- Randomized treatment arm
- Primary and secondary efficacy analyses compared with placebo where specified
Placebo
- Placebo
- Randomized comparator arm
- Reference group for the reported treatment-effect estimates
Random allocation was part of the registered trial design.
The registry identifies a parallel-group design.
The registry identifies the trial as double-masked.
The posted statistical analyses use a superiority hypothesis.
Trial timing
Trial start
The registry lists the trial start as 2017-11-01.
Primary completion
The registry lists the primary completion date as 2019-07-31.
Registry status
The current ClinicalTrials.gov record in the ClinicalTrials.gov record identifies ORION-11 as completed.
4. Endpoints
The registry lists two primary endpoints. Both were posted with formal statistical analyses, and both used ANCOVA in the registry analysis records.
| Endpoint | Registry time frame | Analysis |
|---|---|---|
| Percentage Change in LDL-C From Baseline to Day 510 | Baseline, Day 510 | ANCOVA; ITT population |
| Time-adjusted Percent Change in LDL-C Levels From Baseline After Day 90 and up to Day 540 | Baseline, Day 90 to Day 540 | ANCOVA; ITT population |
Secondary endpoints reported in the registry
| Secondary endpoint | Time frame | Effect measure |
|---|---|---|
| Absolute Change In LDL-C From Baseline To Day 510 | Baseline, Day 510 | Median Difference (Final Values) |
| Time-adjusted Absolute Change in LDL-C From Baseline After Day 90 and up to Day 540 | Baseline, Day 90 to Day 540 | Mean Difference (Final Values) |
| Percentage Change in Proprotein Convertase Subtilisin/Kexin Type 9 (PCSK9) From Baseline to Day 510 | Baseline, Day 510 | Mean Difference (Final Values) |
| Percentage Change in Total Cholesterol From Baseline to Day 510 | Baseline, Day 510 | Mean Difference (Final Values) |
| Percentage Change in Apolipoprotein B (ApoB) From Baseline to Day 510 | Baseline, Day 510 | Mean Difference (Final Values) |
| Percentage Change in Non-HDL-C From Baseline to Day 510 | Baseline, Day 510 | Mean Difference (Final Values) |
5. Statistical Methodology
Analysis populations
The registry analyses identify the intention-to-treat (ITT) population for both primary endpoints. Several secondary endpoint analyses are also explicitly identified as using the ITT population.
ANCOVA
All eight posted statistical analyses use ANCOVA as the reported method. The registry normalizes this method to the linear-model family. For the primary analyses, the reported effect measure is a mean difference in final values, and the analysis notes state that the estimate represents the least squares means difference from placebo.
ANCOVA combines a comparison of treatment groups with adjustment for specified covariates. The exact covariates used in the posted ORION-11 analyses are not identified in the registry method fields, so this page does not infer them.
Confidence intervals
Each of the eight posted statistical analyses supplies a 95% two-sided confidence interval. The interval describes statistical uncertainty around the reported treatment-effect estimate under the analysis framework; it is not a range containing the individual responses of 95% of subjects.
Superiority testing
The posted analyses identify superiority as the hypothesis type. In a superiority analysis, the treatment effect is assessed against the null hypothesis of no difference rather than against a prespecified non-inferiority margin.
P-values
The primary endpoint analyses report two-sided p-values of <0.0001. A p-value quantifies the compatibility of the observed data with the specified null hypothesis under the statistical model and testing framework. It does not measure the magnitude of the treatment effect or the probability that the treatment hypothesis is true.
6. Statistical Methods Explained
Why was ANCOVA used?
The registry reports ANCOVA for all eight posted analyses. ANCOVA is a linear-model approach that can compare treatment groups while accounting for prespecified covariates. For a continuous biochemical outcome such as change in LDL-C, the method provides an adjusted treatment contrast rather than relying solely on an unadjusted comparison of raw group means.
What is a least squares means difference?
The primary analyses report a mean difference in final values and state that the estimate represents the least squares means difference from placebo. Least squares means are model-based adjusted means. Their difference therefore represents the estimated treatment contrast after accounting for the covariate structure included in the ANCOVA model.
What does a negative mean difference indicate here?
The primary endpoint estimates are negative: -53.5 and -49.17. Because the endpoint is percentage change in LDL-C, a negative treatment-minus-placebo difference indicates a lower estimated percentage change in LDL-C for inclisiran relative to placebo under the reported model.
What does the 95% confidence interval tell us?
For the first primary endpoint, the 95% CI is -56.66 to -50.35. For the second, it is -51.57 to -46.77. These intervals quantify uncertainty around the respective estimated treatment differences. Their narrowness relative to the estimates indicates that the posted analyses produced fairly precise estimates within their statistical framework.
Why doesn't the p-value measure effect size?
The p-value depends on both the magnitude of an observed difference and the amount of statistical information available. A very small p-value can accompany either a large or small effect depending on sample size and variability. The effect estimate and its confidence interval are therefore essential for understanding the size and precision of the observed treatment contrast.
Why does the ITT population matter?
The ITT population preserves the randomized comparison by analyzing subjects according to their assigned treatment group. This is especially important when interpreting a superiority trial because excluding subjects after randomization can disrupt the balance created by randomization and potentially change the question being answered.
What does double masking contribute statistically?
Double masking can reduce the potential for knowledge of treatment assignment to influence behavior, outcome assessment, or other aspects of trial conduct. The registry identifies ORION-11 as double-masked; the ClinicalTrials.gov record does not provide additional detail about exactly which trial personnel were masked.
7. Primary Results
Percentage Change in LDL-C From Baseline to Day 510
Inclisiran vs placebo
95% CI: -56.66 to -50.35 · P < 0.0001
Effect measure: Mean Difference (Final Values)
Analysis: ANCOVA in the ITT population; the registry analysis notes that the estimate represents the least squares means difference from placebo.
The estimated treatment difference was -53.5 percentage points for the registered endpoint, with inclisiran compared with placebo. Because the outcome is percentage change in LDL-C, the negative sign indicates a lower estimated percentage change in the inclisiran group relative to placebo under the reported ANCOVA model.
This does not mean that every subject experienced a 53.5% reduction, nor does it mean that the individual treatment effect was identical for every subject. The estimate is a model-based group comparison.
The 95% CI of -56.66 to -50.35 expresses uncertainty around the estimated treatment contrast. It does not describe the range of individual patient responses.
The p-value of <0.0001 indicates very strong incompatibility with the specified null hypothesis under the reported testing framework. It does not say that the treatment effect is "53.5% significant," and it does not measure clinical importance.
Because this is an ANCOVA analysis rather than a time-to-event analysis, proportional-hazards assumptions are not relevant to this particular estimate. Interpretation instead depends on the linear-model specification, covariates, outcome definition, analysis population, and handling of missing observations.
Time-adjusted Percent Change in LDL-C Levels From Baseline After Day 90 and up to Day 540
Inclisiran vs placebo
95% CI: -51.57 to -46.77 · P < 0.0001
Effect measure: Mean Difference (Final Values)
Analysis: ANCOVA in the ITT population; the registry analysis notes that the estimate represents the least squares means difference from placebo.
The estimated treatment difference for the time-adjusted percentage-change endpoint was -49.17 percentage points. The endpoint is defined from baseline after Day 90 and up to Day 540, so this estimate summarizes the registered time-adjusted outcome rather than representing a single instantaneous measurement.
The estimate does not mean that LDL-C was reduced by exactly 49.17% in every individual subject. It is the reported model-based treatment contrast between inclisiran and placebo.
The 95% CI of -51.57 to -46.77 provides the statistical precision reported for this estimate. The interval is entirely below zero, which is consistent with a negative treatment difference under the superiority framework.
The p-value of <0.0001 addresses the null hypothesis in the reported analysis; it is not a measure of the magnitude or clinical importance of the treatment effect.
As with the first primary endpoint, the analysis is ANCOVA rather than a survival model. Consequently, proportional-hazards assumptions and censoring rules are not the central interpretive issues for this estimate. The exact covariate specification and missing-data procedures are not provided in the registry analysis fields.
8. Secondary Endpoint Results
The registry contains six posted secondary endpoint analyses. All six use ANCOVA and all report superiority comparisons with 95% two-sided confidence intervals.
| Secondary endpoint | Estimate | 95% CI | P-value |
|---|---|---|---|
| Absolute Change In LDL-C From Baseline To Day 510 | -51.87 | -55.01 to -48.72 | <.0001 |
| Time-adjusted Absolute Change in LDL-C From Baseline After Day 90 and up to Day 540 | -48.94 | -51.39 to -46.48 | <0.0001 |
| Percentage Change in PCSK9 From Baseline to Day 510 | -79.27 | -81.97 to -76.57 | <0.0001 |
| Percentage Change in Total Cholesterol From Baseline to Day 510 | -29.79 | -31.78 to -27.81 | <0.0001 |
| Percentage Change in ApoB From Baseline to Day 510 | -38.94 | -41.21 to -36.67 | <0.0001 |
| Percentage Change in Non-HDL-C From Baseline to Day 510 | -43.32 | -46.04 to -40.60 | <0.0001 |
Absolute change in LDL-C to Day 510
Inclisiran vs placebo
95% CI: -55.01 to -48.72 · P < .0001
Reported effect measure: Median Difference (Final Values); registry method: ANCOVA.
The registry's analysis notes identify the estimate as representing the least squares means difference from placebo despite the posted effect-measure label of "Median Difference (Final Values)." This page preserves that registry wording rather than attempting to reconcile or recompute the estimate.
Time-adjusted absolute change in LDL-C
Inclisiran vs placebo
95% CI: -51.39 to -46.48 · P < 0.0001
Effect measure: Mean Difference (Final Values); analysis: ANCOVA.
Percentage change in PCSK9
Inclisiran vs placebo
95% CI: -81.97 to -76.57 · P < 0.0001
Effect measure: Mean Difference (Final Values); analysis: ANCOVA.
Percentage change in total cholesterol
Inclisiran vs placebo
95% CI: -31.78 to -27.81 · P < 0.0001
Effect measure: Mean Difference (Final Values); analysis: ANCOVA; analysis population: ITT Population.
Percentage change in ApoB
Inclisiran vs placebo
95% CI: -41.21 to -36.67 · P < 0.0001
Effect measure: Mean Difference (Final Values); analysis: ANCOVA; analysis population: ITT Population.
Percentage change in non-HDL-C
Inclisiran vs placebo
95% CI: -46.04 to -40.60 · P < 0.0001
Effect measure: Mean Difference (Final Values); analysis: ANCOVA; analysis population: ITT Population.
9. Understanding the Pattern of Treatment Effects
The posted estimates cover several related lipid measures. The largest numerical percentage-change contrast among the secondary outcomes is for PCSK9 (-79.27), while the other reported percentage-change contrasts include total cholesterol (-29.79), ApoB (-38.94), and non-HDL-C (-43.32).
The graphic is a visual representation of the posted estimates rather than a new statistical analysis. Because these endpoints are measured on different biological scales, their numerical magnitudes should not be treated as directly comparable measures of clinical importance.
10. Safety Results
The ClinicalTrials.gov record reports serious adverse events by randomized arm. The affected and at-risk counts are provided directly in the registry-derived data.
| Safety measure | Inclisiran | Placebo |
|---|---|---|
| Serious adverse events | 181 / 811 | 181 / 804 |
Inclisiran
181 subjects with serious adverse events among 811 at risk.
Placebo
181 subjects with serious adverse events among 804 at risk.
11. Randomization and Blinding
ORION-11 is identified as randomized, parallel, and double-masked. These design features address different sources of bias.
Randomization
Random assignment is intended to create comparable treatment groups in expectation, allowing differences in outcomes to be attributed more credibly to treatment assignment rather than systematic baseline differences.
Double masking
Masking can reduce the possibility that knowledge of treatment assignment influences trial conduct, participant behavior, or assessment.
Parallel design
Each randomized subject belongs to a treatment arm rather than sequentially receiving both randomized interventions in a crossover structure.
Superiority framework
The posted analyses are labeled superiority analyses, so the treatment effect is evaluated as a difference rather than against a non-inferiority margin.
12. What the Primary Estimates Do — and Do Not — Mean
The primary estimate of -53.5 is a between-group treatment contrast for the registered percentage-change endpoint. It is not an individual-level prediction and should not be interpreted as saying that every subject's LDL-C changed by exactly the same amount.
The 95% CI of -56.66 to -50.35 gives a measure of statistical uncertainty around the first primary estimate. The second primary estimate has a 95% CI of -51.57 to -46.77. Neither interval describes the distribution of individual patient responses.
The reported primary p-values are <0.0001. These values indicate strong evidence against the relevant null hypothesis under the reported statistical framework. They do not quantify the size of the treatment effect or its practical importance.
Both primary estimates come from ANCOVA. Consequently, they are conditional on the model specification used by the registry analysis. The ClinicalTrials.gov record does not identify every model covariate or the detailed missing-data strategy, so those elements should not be inferred.
13. Multiplicity and the Two Primary Endpoints
ORION-11 has two registered primary endpoints, and both have formal statistical analyses posted. The ClinicalTrials.gov record identifies superiority as the hypothesis type and report two-sided p-values for both analyses.
| Primary endpoint | Estimate | 95% CI | P-value | Hypothesis |
|---|---|---|---|---|
| Percentage Change in LDL-C From Baseline to Day 510 | -53.5 | -56.66 to -50.35 | <0.0001 | Superiority |
| Time-adjusted Percent Change in LDL-C Levels From Baseline After Day 90 and up to Day 540 | -49.17 | -51.57 to -46.77 | <0.0001 | Superiority |
The presence of two primary endpoints is statistically important because a trial may define a family of confirmatory hypotheses that requires an explicit multiplicity strategy. The ClinicalTrials.gov record does not state whether a multiplicity adjustment, hierarchical testing procedure, or other alpha-allocation rule was used. Therefore, this page does not infer one.
14. Missing Data and Model Assumptions
The registry analysis fields identify ANCOVA and the ITT population but do not specify the complete missing-data or imputation strategy. That distinction matters because repeated lipid measurements and time-adjusted outcomes can involve observations that are unavailable at particular assessment times.
The equation is conceptual rather than a reconstruction of the ORION-11 analysis. The registry's posted analysis identifies the treatment contrast as a least squares means difference from placebo, but the ClinicalTrials.gov record does not contain the full model specification.
For an educational interpretation, it is therefore important to distinguish what is directly documented from what would normally be examined in a statistical analysis plan: covariate specification, treatment-by-covariate interactions if relevant, missing-data assumptions, sensitivity analyses, and the precise estimand being targeted.
15. Time Frames and Estimands
The two primary endpoints are related but are not identical statistical questions. One is defined specifically at Day 510, while the other is defined as a time-adjusted percent change from baseline after Day 90 and up to Day 540.
Day 510 endpoint
Asks about percentage change in LDL-C from baseline to the specific registered Day 510 assessment.
Day 90–540 endpoint
Uses a time-adjusted percentage-change definition over the registered period after Day 90 and up to Day 540.
Why the distinction matters
A single-time-point endpoint and a time-adjusted endpoint summarize treatment effects differently even when they measure the same underlying biomarker.
Interpretation
The two estimates should therefore be viewed as complementary primary analyses rather than treated as duplicate measurements of exactly the same statistical quantity.
16. Statistical Interpretation of the Secondary Outcomes
The secondary analyses extend the statistical story beyond LDL-C to PCSK9, total cholesterol, ApoB, and non-HDL-C. They also include both absolute and time-adjusted absolute LDL-C change.
| Outcome domain | Reported estimate | 95% CI | Statistical method |
|---|---|---|---|
| LDL-C absolute change | -51.87 | -55.01 to -48.72 | ANCOVA |
| LDL-C time-adjusted absolute change | -48.94 | -51.39 to -46.48 | ANCOVA |
| PCSK9 percentage change | -79.27 | -81.97 to -76.57 | ANCOVA |
| Total cholesterol percentage change | -29.79 | -31.78 to -27.81 | ANCOVA |
| ApoB percentage change | -38.94 | -41.21 to -36.67 | ANCOVA |
| Non-HDL-C percentage change | -43.32 | -46.04 to -40.60 | ANCOVA |
Every confidence interval in this table lies below zero, matching the direction of the corresponding point estimate. The statistical interpretation remains model-dependent: these are estimated group differences, not individual treatment effects.
17. Clinical Interpretation vs Statistical Interpretation
Statistical interpretation
The two registered primary analyses report negative ANCOVA treatment differences with 95% confidence intervals entirely below zero and two-sided p-values below 0.0001.
Clinical interpretation
The reported estimates describe differences in LDL-C-related outcomes between randomized treatment groups. Whether a numerical change represents a clinically important benefit requires a separate clinical framework and is not determined by the p-value alone.
18. What the Registry Data Support — and What They Do Not
A strong statistical analysis distinguishes reported evidence from information that is not contained in the available record.
| Topic | What the ClinicalTrials.gov record supports |
|---|---|
| Randomization | ORION-11 was randomized. |
| Blinding | The registry identifies the design as double-masked. |
| Primary endpoints | Two registered primary endpoints are identified, and both have formal analyses. |
| Primary method | ANCOVA was reported for both primary endpoints. |
| Analysis population | The primary analyses used the ITT population. |
| Effect measure | Mean Difference (Final Values), with analysis notes describing least squares means differences from placebo. |
| Confidence intervals | 95% two-sided intervals are reported for both primary analyses and all six secondary analyses. |
| Hypothesis | The analyses are identified as superiority analyses. |
| Serious adverse events | 181/811 for inclisiran and 181/804 for placebo. |
| Detailed baseline characteristics | Not contained in the ClinicalTrials.gov record. |
| Detailed multiplicity strategy | Not contained in the ClinicalTrials.gov record. |
| Detailed missing-data strategy | Not contained in the ClinicalTrials.gov record. |
| Subgroup analyses | Not contained in the ClinicalTrials.gov record. |
| Time-to-event analyses | Not contained in the ClinicalTrials.gov record. |
19. Important Limitations and Interpretation Issues
- Registry-level method detail: the ClinicalTrials.gov record identifies ANCOVA but do not provide the full model specification, including all covariates and other implementation details.
- Multiplicity: two primary endpoints are reported, but the ClinicalTrials.gov record does not specify an alpha-allocation or multiplicity-adjustment strategy.
- Missing-data handling: the ClinicalTrials.gov record does not identify the complete missing-data or imputation approach.
- Effect-measure labeling: one secondary analysis is labeled "Median Difference (Final Values)" while its analysis notes describe a least squares means difference from placebo. This page preserves both pieces of registry information rather than silently changing either.
- Endpoint terminology: the registry identifies the primary endpoint type as Binary even though the endpoint is expressed as a percentage change in LDL-C. This page retains the registry endpoint wording rather than imposing a different classification.
- Secondary endpoints: the statistical significance of multiple secondary outcomes should not automatically be interpreted as multiple independent confirmatory findings without knowing the prespecified multiplicity strategy.
- Safety comparison: serious-adverse-event counts are reported by arm, but the ClinicalTrials.gov record does not include a formal statistical comparison or confidence interval.
- Generalizability: the ClinicalTrials.gov record defines the target population as subjects with ASCVD or ASCVD-risk equivalents and elevated LDL-C. It does not provide a detailed baseline-characteristics table from which broader generalizability claims could be assessed.
20. Why This Trial Matters Statistically
ORION-11 is a useful teaching case because it illustrates how a randomized superiority trial can use ANCOVA to estimate treatment differences for continuous biomarker outcomes while simultaneously requiring careful attention to the endpoint definition, analysis population, confidence intervals, and multiplicity.
| Concept | How it appears in ORION-11 |
|---|---|
| Randomization | Randomized parallel-group design with two treatment arms. |
| Blinding | Double-masked trial. |
| Intention-to-treat analysis | Both primary analyses use the ITT population. |
| ANCOVA | Reported method for all eight posted statistical analyses. |
| Least squares means | Primary analysis notes describe the estimates as least squares means differences from placebo. |
| Confidence intervals | 95% two-sided confidence intervals accompany all eight posted analyses. |
| P-values | Both primary analyses report P < 0.0001; all six secondary analyses also report very small p-values. |
| Superiority | The posted analyses identify superiority as the hypothesis type. |
| Multiple primary endpoints | Two primary endpoints require attention to the overall testing framework. |
| Safety analysis | Serious adverse events are reported as affected subjects over subjects at risk by arm. |
21. A Practical Framework for Reading ORION-11
A reader evaluating the statistical evidence can work through the trial in a structured sequence.
1. Identify the estimand
Start with exactly what was measured: percentage or absolute change from baseline and the specified assessment period.
2. Identify the population
Check whether the reported efficacy analysis is based on the ITT population. For ORION-11, the primary analyses are.
3. Read the effect estimate
Interpret the signed treatment difference before looking at the p-value. The sign establishes the direction of the reported contrast.
4. Read the confidence interval
Use the interval to understand the statistical precision of the treatment estimate rather than treating it as an individual-response range.
5. Then read the p-value
The p-value addresses compatibility with the null hypothesis. It should not replace the effect estimate or confidence interval.
6. Check the testing structure
Because there are two primary endpoints and six secondary analyses, the interpretation of multiple p-values depends on the prespecified multiplicity framework.
22. Statistical Concepts in This Trial
Learn more about the methods used in this trial:
23. Related Statistical Calculators
Apply the same statistical concepts with Clinical Biostats calculators:
24. Sources
- ClinicalTrials.gov: ORION-11 (NCT03400800).
- Linked publication: PubMed record — PMID 41604274.
- Linked publication: PubMed record — PMID 39093262.
- Linked publication: PubMed record — PMID 38757725.
- Linked publication: PubMed record — PMID 38063346.
- Linked publication: PubMed record — PMID 37282500.
Continue through the Clinical Biostats statistical library
Connect this trial's design and analytical methods to deeper tutorials, statistical calculators, and other clinical-trial analyses.
25. Record Summary
ORION-11 provides a clear example of statistical analysis for randomized clinical-trial biomarker endpoints. The trial was a randomized, double-masked, parallel phase 3 study with 1617 enrolled subjects and two treatment arms. Both registered primary endpoints were formally analyzed using ANCOVA in the ITT population, with least squares means differences from placebo reported as the principal treatment contrasts. The first primary endpoint produced an estimate of -53.5 with a 95% CI of -56.66 to -50.35 and P < 0.0001; the second produced an estimate of -49.17 with a 95% CI of -51.57 to -46.77 and P < 0.0001.
The six posted secondary analyses show the same general statistical structure, extending the analysis to absolute LDL-C change, time-adjusted absolute LDL-C change, PCSK9, total cholesterol, ApoB, and non-HDL-C. Serious adverse events were reported as 181/811 in the inclisiran arm and 181/804 in the placebo arm. The principal interpretive issues are therefore not limited to whether p-values are small: the reader should also examine the endpoint definitions, ITT population, model-based effect estimates, confidence intervals, the distinction between primary and secondary endpoints, and the unspecified multiplicity and missing-data details.