The COMPASS Study: A Study of Volanesorsen (Formally ISIS-APOCIIIRx) in Patients With Hypertriglyceridemia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The COMPASS Study: A Study of Volanesorsen (Formally ISIS-APOCIIIRx) in Patients With Hypertriglyceridemia (NCT02300233), a randomized phase 3 study. The primary outcome is Percent Change in Fasting Triglycerides (TG) From Baseline to Month 3.

2. Clinical Question

The central statistical question concerns the magnitude and precision of the treatment-group contrast in a continuous laboratory measure. The posted results describe a markedly more negative mean percent change for Volanesorsen Total than for placebo. The contrast is reported with a confidence interval and a p-value, which together describe the estimated difference and the uncertainty under the fitted analysis.

The question is narrower than whether the study establishes an effect for every dosing schedule, participant subgroup, or outcome. The primary result is reported for a combined volanesorsen group, while the registry also lists separate weekly and biweekly-post-Week-13 regimens in its arm descriptions. A regimen-specific primary comparison is not reported in the registry record. Secondary response outcomes add a different perspective by classifying participants according to triglyceride reductions, but they do not replace the continuous primary endpoint.

3. Trial Design

The COMPASS study was a randomized, parallel-group, placebo-controlled Phase 3 trial of volanesorsen in participants with hypertriglyceridemia. It used double-blind and quadruple masking. The registry lists Ionis Pharmaceuticals, Inc. as sponsor, an enrollment of 114, a start date of 2015-02-05, a primary completion date of 2016-07-27, and a completed status.

The study purpose was to evaluate efficacy and safety of volanesorsen given for 26 weeks. The registry describes placebo given subcutaneously once weekly for 26 weeks; volanesorsen 300 mg given subcutaneously once weekly for 26 weeks; and volanesorsen 300 mg given weekly for 13 weeks, then biweekly for 13 weeks. The primary outcome is reported for placebo and Volanesorsen Total rather than as separate primary estimates for each active regimen. The registry record does not report how participants were allocated between the two active schedules for this combined result.

The primary analysis used the full analysis set, defined as randomized participants who received at least one dose and had a baseline triglyceride assessment. The reported primary group sizes are 38 for placebo and 75 for Volanesorsen Total. These are analysis denominators and should not be substituted for the registry’s overall enrollment figure. The record does not report the number of participants excluded from the full analysis set or the reasons for any exclusions.

The design and endpoint description establish the broad comparison and timing, but they do not supply every operational detail. Allocation concealment procedures, site details, the method for handling treatment discontinuation, and the planned sample-size justification are not reported in the registry record. Randomization and masking reduce some risks of bias in principle, but their presence alone does not resolve questions about missing data, adherence, or how the combined active group should be interpreted.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: 114 participants were randomized, and 113 received study drug. One patient was randomized, but discontinued before dosing, and thus included only in the volanesorsen Total (Not public) column. The study included a ≤ 8-week screening period (including a diet-stabilization period), a 26-week treatment period, and a 13-week post-treatment evaluation period.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The full analysis set (FAS) included all participants who were randomized, received at least one dose of study drug, and had a baseline TG assessment.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: The full analysis set (FAS) included all participants who were randomized, received at least one dose of study drug, and had a baseline triglycerides (TG) assessment.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change in Fasting Triglycerides (TG) From Baseline to Month 3, measured over Baseline to 3 months.

The primary endpoint is percent change in fasting triglycerides from baseline to Month 3. It is a continuous outcome: each participant’s triglyceride measurement contributes information about the size and direction of change, rather than only whether a threshold was met. A percent-change scale expresses change relative to the baseline measurement, making it a way to compare changes across participants whose starting values may differ.

The registry reports least-squares means for the full analysis set, with a 95% confidence interval for each group estimate. It also reports an ANCOVA comparison as the difference in least-squares means. The placebo least-squares mean was -0.9, with a 95% confidence interval from -13.9 to 12.2. For Volanesorsen Total, the estimate was -71.2, with a 95% confidence interval from -79.3 to -63.2. The reported contrast was -70.3, with a 95% confidence interval from -85.4 to -55.3 and p < 0.001.

ANCOVA is commonly used for continuous outcomes to compare groups while accounting for specified covariates. The covariates used in this analysis are not reported in the registry record. The primary endpoint is useful for describing average change across the analysis population, but it does not show how much individual participants varied or how many reached a particular response threshold. The registry separately lists responder outcomes as secondary measures.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a least-squares mean represent?

A least-squares mean is a model-based estimate of a group mean. In an analysis such as ANCOVA, it can account for covariates included in the model, so it is not necessarily identical to an unadjusted arithmetic average. The registry reports least-squares means for placebo and Volanesorsen Total and an ANCOVA difference between those means. The covariates and model specification are not reported in the registry record, limiting independent assessment of the adjustment.

For the primary outcome, a more negative percent-change estimate indicates a larger decrease on the endpoint’s scale. The contrast compares the model-based group means; it is not a comparison of every participant with every other participant. Because the reported active result combines the Volanesorsen Total group, it should not be treated as a separate estimate for each regimen listed in the arm descriptions.

How should the confidence interval be read?

A confidence interval expresses uncertainty around an estimated quantity under a statistical procedure and its assumptions. The 95% confidence interval for the primary contrast runs from -85.4 to -55.3. It indicates that the estimate is not perfectly precise, while giving a range of values compatible with the fitted analysis at the stated confidence level. It does not describe the range of individual participant outcomes, and it does not account for sources of uncertainty that are absent from the model or unreported in the record.

The confidence intervals attached to the arm means describe uncertainty in those estimates; the interval for the difference describes uncertainty in the comparison itself. The latter is the more direct interval for the primary question. A confidence interval that excludes zero is consistent with the small reported p-value for the contrast, but neither statistic by itself establishes clinical importance or freedom from bias.

What does ANCOVA contribute?

ANCOVA compares a continuous outcome between groups using a regression model that can include covariates. In a randomized study, adjustment can improve precision when the model includes relevant baseline information, but conclusions depend on the variables and methods actually used. The registry identifies ANCOVA and the difference in least-squares means; it does not report the covariates, assumptions checks, missing-data method, or sensitivity analyses. Those details are therefore not available for judging whether the model was well specified or how robust the result is to alternative choices.

The displayed group estimates and contrast are sponsor-reported analysis results. No discrepancy between the posted primary result and the sponsor’s reported analysis is flagged in the registry record. The record does not provide participant-level data for an independent reconstruction of ANCOVA.

How do the responder analyses differ?

A responder analysis turns a continuous measurement into a yes-or-no classification using a threshold. The registry lists response at Month 3 based on a fasting triglyceride reduction of at least 40% from baseline, and a separate definition based on reaching a reduction of 150 mg/dL. These definitions do not measure the same thing as the primary average percent change. They can be easier to interpret as proportions of participants meeting a criterion, but they discard information about the size of changes within each response category.

The posted secondary comparison for the at-least-40% definition includes an unadjusted risk difference and a reported logistic-regression odds ratio. These are different effect measures and may differ in magnitude and interpretation. The risk difference compares response proportions directly; an odds ratio compares the odds of response. The registry’s sponsor analysis reports the logistic-regression result, while the secondary posted calculation gives an unadjusted comparison. For the 150 mg/dL definition, the record likewise lists an unadjusted risk difference and a sponsor-reported logistic-regression odds ratio. The results should not be treated as interchangeable estimates.

What do the planning illustrations mean?

The sample-size illustrations are planning examples based on the observed effect, not the trial’s original design justification. They assume a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They describe participants per arm for 80% power and participants per arm for 90% power at specified fractions of the observed effect. These are hypothetical planning quantities, not observed power and not evidence that the trial had a particular probability of detecting its result after the fact.

Such illustrations depend strongly on the assumed effect size. Planning for a fraction of the observed effect is a way to examine how required sample size changes if a future study targets a more modest effect. The observed effect may be an imprecise or optimistic basis for planning, especially when estimates come from a single study. The registry does not report its own sample-size justification, target effect, or original power calculation, so the illustrations cannot be used to infer the actual design assumptions.

Why do analysis details matter?

Model results depend on how the analysis population is formed and what happens when outcome values are missing. The full analysis set requires randomization, at least one dose, and a baseline triglyceride assessment. The record does not state the number of participants with missing Month 3 measurements, how missing outcomes were handled, whether all randomized participants were included in a sensitivity analysis, or whether additional analysis populations were used. These omissions affect how completely the posted estimate can be evaluated.

Multiplicity is another consideration when several outcomes and analyses are examined. The record lists multiple secondary outcomes but does not report a multiplicity adjustment or a prespecified hierarchy for interpreting them. Consequently, unadjusted secondary p-values should be treated as exploratory evidence rather than as confirmatory findings with the same inferential status as the designated primary outcome.

10. Analyses Reported by the Sponsor

The primary result reported by the sponsor is an ANCOVA estimate of the difference in least-squares means: -70.3, with a 95% confidence interval from -85.4 to -55.3 and p < 0.001. This is the direct model-based comparison for the primary endpoint. The registry also reports the group-specific least-squares means, which provide context for the contrast but are not themselves the between-group estimate.

The record does not flag a discrepancy between the posted primary result and the sponsor’s own analysis. The sponsor’s reported analysis therefore remains the relevant primary comparison. A separate independent re-analysis of the primary endpoint is not reported in the registry record. In particular, the registry does not provide the participant-level observations or enough model detail to reproduce the ANCOVA estimate independently.

For the two posted binary secondary outcomes, the registry gives sponsor-reported logistic-regression odds ratios alongside unadjusted comparison estimates. The adjusted odds ratios and the unadjusted risk differences represent different quantities and arise from different analyses; one should not substitute for the other. The unadjusted comparisons are descriptive recalculations from the posted summary counts, while the sponsor’s regression analyses are the reported analyses. They can serve as complementary views, but the sponsor’s analysis takes precedence when a discrepancy is flagged; no such discrepancy is flagged in the record.

None of these comparisons reveals whether all analysis assumptions were satisfied. Covariate specifications, missing-data procedures, and sensitivity analyses are not reported in the registry record, so the relationship between those choices and the estimates cannot be assessed in detail.

11. Multiplicity, Missing Data and Analysis Populations

The primary outcome and the full analysis set are identified in the registry record, but several details needed to assess the analysis are not reported. The record defines the full analysis set as randomized participants who received at least one dose and had a baseline triglyceride assessment. It does not report the number of participants with missing Month 3 measurements, the reasons for missingness, or the method used to handle missing primary outcome data.

Missing-data methods can affect an estimate when participants with unavailable outcomes differ from those with observed outcomes. Without information about the extent and handling of missing data, readers cannot determine whether the analysis relied on complete observations, imputation, or another approach. Sensitivity analyses addressing alternative assumptions about missing outcomes are not reported in the registry record. The record also does not report whether a per-protocol analysis or another analysis population was used.

Several secondary outcomes are listed, and the record includes results for two binary triglyceride response definitions. It does not report a multiplicity adjustment, a prespecified sequence for testing secondary outcomes, or a plan for controlling the overall type I error across multiple comparisons. In the absence of that information, secondary p-values should be interpreted cautiously. They are evidence about individual comparisons under their respective analyses, not proof that the full family of outcomes was evaluated with a controlled error rate.

Interim analyses and stopping rules are not reported in the registry record. The planned sample-size justification and the assumptions behind the original design are also not reported. The planning illustrations based on fractions of the observed effect are not a substitute for those design details; they are retrospective planning examples and not observed power.

The registry identifies ANCOVA for the primary contrast and logistic regression for the sponsor analyses of the posted binary outcomes. It does not report the primary model covariates, diagnostics, or sensitivity checks. These gaps do not invalidate the reported estimates, but they limit the degree to which readers can evaluate robustness and reproduce the analyses from the registry record alone.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pancreatitis acute (5.3%), ulna fracture (4.0%), ileus paralytic (2.6%), pancreatitis relapsing (2.6%), cholelithiasis (2.6%).

The registry reports serious adverse events and other adverse events by listed treatment group. For placebo, 4 of 38 participants were affected by serious adverse events and 31 of 38 were affected by other adverse events. For Volanesorsen 300 mg Weekly, the corresponding counts were 2 of 25 and 24 of 25. For Volanesorsen 300 mg Biweekly, Post Week 13, the counts were 6 of 50 and 49 of 50. The registry reports no deaths in any of these groups.

These are counts of participants affected, not a measure of the total number of adverse-event episodes or their duration. The serious-event terms listed most often include acute pancreatitis at 5.3%, ulna fracture at 4.0%, paralytic ileus at 2.6%, relapsing pancreatitis at 2.6%, cholelithiasis at 2.6%, and pancreas infection at 2.6%. The record does not provide enough detail in these figures to establish whether an event was caused by study treatment.

Safety counts are reported separately for the two volanesorsen schedules, unlike the combined Volanesorsen Total group used for the primary endpoint. Differences in the counts should be read with their group denominators and the study context in mind; they are not, by themselves, adjusted comparisons of risk. The registry record does not report a statistical comparison for these adverse-event counts, nor does it provide event timing or a detailed account of severity beyond the serious-event classification.

13. What the Result Does, and Does Not, Mean

The primary result establishes what the registry’s analysis estimated for fasting triglyceride percent change from baseline to Month 3 in the defined full analysis set. The reported ANCOVA contrast favors a more negative mean change in Volanesorsen Total than in placebo, and the confidence interval and p-value indicate statistical evidence against a zero contrast under the reported analysis. The estimate is a group-level summary, not a prediction of the result for any particular participant.

Statistical evidence for a difference does not by itself establish clinical importance. The registry record does not report a threshold for a clinically important difference, nor does the primary endpoint report how changes relate to later outcomes. The primary result concerns a laboratory measure at Month 3; it should not be expanded into claims about outcomes that were not measured.

The result also does not establish that the two volanesorsen schedules have the same effect. The primary endpoint is reported for Volanesorsen Total, while the registry describes weekly and biweekly-post-Week-13 schedules. Separate primary estimates by schedule are not reported in the registry record. Similarly, an average group contrast does not show the full range of individual responses or identify which participants may have had larger, smaller, or no changes.

Secondary responder findings provide a threshold-based complement to the continuous endpoint, while the listed HDL-C, HOMA-IR, and HbA1c outcomes address other measures. Missing outcome details and multiplicity information limit how strongly those secondary findings can be interpreted. Adverse-event counts offer safety context, but event counts alone do not establish causation or provide a complete benefit-risk assessment.

In general, a randomized comparison can support inference about group differences when allocation, follow-up, and analysis are appropriately handled. The registry record documents randomization and masking, but does not report every detail needed to assess missingness, model assumptions, or analysis robustness. The findings should therefore be understood as the posted statistical results for this trial and endpoint, rather than as a complete account of all possible treatment effects or risks.

14. Why This Trial Matters Statistically

This trial illustrates the value of separating an endpoint estimate from the comparison that answers the principal question. The least-squares means describe each group, while the ANCOVA contrast directly estimates the difference between groups. Its confidence interval conveys uncertainty about that contrast, and the p-value addresses compatibility with a zero difference under the fitted model. Reading these together is more informative than relying on a single significance label.

It also shows why endpoint scale matters. A continuous percent-change outcome preserves information about the size of measured changes, whereas a responder analysis categorizes participants at a threshold. Both can be useful, but they answer different questions and produce effects on different scales. The secondary response findings can provide a consistency check without replacing the primary estimate.

Aggregation is another important statistical issue. The primary result combines active treatment in Volanesorsen Total, while the safety results distinguish the weekly and biweekly-post-Week-13 regimens. Those summaries answer different questions. The combined primary estimate cannot show whether the regimen-specific results are alike, and the separate safety counts do not constitute an adjusted comparison of risk.

Finally, the record demonstrates the limits of a registry summary for reproducibility. It names the analysis methods and analysis population, but leaves the covariates, missing-data procedures, multiplicity approach, and original sample-size justification unreported. Planning illustrations based on observed effects can help frame future design choices, but they are not evidence of observed power and do not reconstruct the trial’s original design rationale. Statistical interpretation depends both on the estimates reported and on knowing how they were produced.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

A key limitation is that the primary result is reported for Volanesorsen Total rather than separately for the two active schedules described in the registry. The combined estimate does not reveal schedule-specific primary effects or whether results were similar across regimens. Conversely, adverse-event counts are given separately by active schedule, so the primary and safety summaries do not use the same grouping structure.

The primary analysis population excludes randomized participants who did not receive at least one dose or did not have a baseline triglyceride assessment. The record defines this population but does not report how many participants failed to meet its criteria. It also does not report the number of missing Month 3 outcome measurements, the missing-data method, or sensitivity analyses. Those omissions limit assessment of the impact of incomplete outcome data.

The primary analysis is identified as ANCOVA, but the registry record does not report model covariates, diagnostics, or sensitivity checks. This makes it difficult to reproduce or evaluate the adjusted estimate fully. The record also does not report a multiplicity strategy for the secondary outcomes, so the secondary results should not be assigned the same confirmatory weight as the designated primary result.

The posted secondary information includes two binary triglyceride outcomes, while outcome-specific results for the other listed measures are not reported in the registry record. The sample-size illustrations are based on the observed effect and are not the trial’s original design justification. The registry does not report that original justification, interim analyses, or stopping rules.

Adverse-event summaries give participant counts by group, but the record does not provide event timing or a statistical comparison in the registry record reported. These counts therefore offer limited descriptive safety context and cannot establish causation. Together, these limitations do not erase the reported primary contrast, but they constrain how broadly it can be generalized or independently evaluated.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: The COMPASS Study: A Study of Volanesorsen (Formally ISIS-APOCIIIRx) in Patients With Hypertriglyceridemia — NCT02300233.
• Registry record last updated 2022-04-13; status last verified 2022-03.
• Results first posted 2022-04-13.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The COMPASS Study was a completed randomized, parallel-group, double-blind, placebo-controlled Phase 3 study of volanesorsen in participants with hypertriglyceridemia. The registry lists an enrollment of 114 and describes subcutaneous placebo and two volanesorsen schedules over 26 weeks.

The primary endpoint was percent change in fasting triglycerides from baseline to Month 3. In the full analysis set, the reported ANCOVA difference in least-squares means between Volanesorsen Total and placebo was -70.3, with a 95% confidence interval from -85.4 to -55.3 and p < 0.001. The result indicates a model-based group difference for this endpoint and analysis population.

The registry also lists secondary triglyceride response measures, HDL-C, HOMA-IR, and HbA1c outcomes, and adverse-event counts. The posted secondary response findings are directionally consistent with the primary result, while multiplicity and missing-data details are not reported in the registry record. The record’s primary estimate should be interpreted alongside those reporting limits and the distinction between the combined efficacy group and regimen-specific safety summaries.