Phase 2 Study of ISIS 681257 (AKCEA-APO(a)-LRx) in Participants With Hyperlipoproteinemia(a) and Cardiovascular Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase 2 Study of ISIS 681257 (AKCEA-APO(a)-LRx) in Participants With Hyperlipoproteinemia(a) and Cardiovascular Disease (NCT03070782), a randomized phase 2 study. The primary outcome is Percent Change From Baseline in Fasting Lipoprotein A [Lp(a)] at the Primary Analysis Time Point.

2. Clinical Question

3. Trial Design

This was a phase 2, randomized, double-blind, placebo-controlled, parallel-group, dose-ranging study sponsored by Akcea Therapeutics. The registry record lists an enrollment of 286 and a completed status. Participants had elevated Lipoprotein(a) and established cardiovascular disease. The stated aims were to assess safety, including tolerability, and to evaluate the efficacy of different ISIS 681257 doses and dosing regimens for reducing plasma Lp(a).

There were five active-treatment cohorts. Cohort A received 20 mg Q4W, Cohort B 40 mg Q4W, and Cohort C 60 mg Q4W. Cohort D received 20 mg Q2W, and Cohort E received 20 mg QW. Injections were subcutaneous. Depending on cohort, treatment could continue for up to 49 weeks, 51 weeks, or 52 weeks, subject to the specified maximum number of doses. Participants in every cohort could instead receive placebo at a dose-matched volume.

The primary analysis time point was Month 6, corresponding to Week 25 for Cohorts A, B, and C and Week 27 for Cohorts D and E. The registry records triple masking but does not report which specific parties were masked. It does not report the randomization procedure, allocation concealment details, or the rationale for the cohort sizes. The overall enrollment and the arm-level analysis populations need not be identical: the primary outcome description defines its analysis population as randomized participants who received at least 1 dose.

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: 286 participants were randomized in a 1:1:1:1:1 ratio to Cohorts A, B, C, D or E. In each cohort, participants were randomized in a 5:1 ratio to receive ISIS 681257 or placebo.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (FAS) included all participants who were randomized and received at least 1 dose of study drug (ISIS 681257 or placebo). FAS represented the practically feasible intent-to-treat (ITT) population as delineated in ICH Guideline E9.

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: Safety Set included all participants who were randomized and received at least 1 dose of study drug (ISIS 681257 or placebo).

Only the first 6 of 7 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change From Baseline in Fasting Lipoprotein A [Lp(a)] at the Primary Analysis Time Point, measured over Baseline and Month 6 (Week 25 for Cohorts A, B and C and Week 27 for Cohorts D and E).

The primary endpoint was percent change from baseline in fasting Lipoprotein(a), or Lp(a), at the primary analysis time point. Lp(a) is the biomarker being targeted in this dose-ranging study, so measuring its change directly addresses the stated objective of assessing whether different doses and schedules reduce plasma Lp(a). The endpoint is a laboratory measure, not a count of cardiovascular events or a direct measure of clinical benefit.

The primary time point was Month 6: Week 25 for Cohorts A, B, and C, and Week 27 for Cohorts D and E. The registry states that the analysis used an ANCOVA model on the log ratio of the time-point Lp(a) value to the baseline value. The estimated log ratio was converted to the original scale, and percent change was calculated from the ratio. The posted arm summaries are geometric-mean percentage changes with 95% confidence intervals.

Percent change expresses the direction and relative size of a biomarker change from baseline. Negative values indicate a decrease from baseline; they do not, by themselves, establish a change in cardiovascular risk. The placebo comparison helps distinguish the observed changes during active treatment from changes observed in the control group over the same analysis period.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint analysis used ANCOVA on the log ratio of the Lp(a) value at the primary time point to the baseline value. In general, ANCOVA estimates a group contrast while accounting for specified covariates. The registry record does not name the covariates in the model, so the exact adjustment cannot be reconstructed from the record. The log ratio compares the time-point measurement with baseline on a multiplicative scale. The registry says the estimate was converted back to the original scale before expressing it as percent change.

The arm-level primary results are geometric-mean percentage changes with 95% confidence intervals. A geometric mean is based on a multiplicative scale and is a natural summary when the analysis is conducted on log-transformed values. It should not be read as the ordinary arithmetic mean of individual percentage changes. The confidence interval describes uncertainty in the estimated arm summary under the model and analysis assumptions; it does not describe the spread of individual participant responses.

The sponsor's primary comparisons are reported as ANCOVA mean differences in percentage change, with 95% confidence intervals and p-values. Each contrast compares one ISIS 681257 cohort with placebo. A difference below zero corresponds to a more negative change in the active group than in placebo under the reported contrast. The confidence interval gives a range of values compatible with the model and observed data at the stated confidence level. The p-value measures how incompatible the observed result is with the null comparison under the statistical model; it is not the probability that the null hypothesis is true, nor a measure of clinical importance.

Some secondary outcomes are binary: whether a participant attained a specified Lp(a) threshold by the analysis time point. For such outcomes, percentages describe the proportion meeting the threshold in each arm. A risk difference compares the proportions by subtraction; a positive difference in these comparisons indicates a higher observed proportion in the active cohort than in placebo. The registry also reports sponsor logistic-regression odds ratios for the threshold comparisons. An odds ratio compares the odds of meeting the threshold, not the probability itself. Odds ratios and risk differences are different effect measures and should not be interpreted as interchangeable quantities.

General statistical context: binary endpoints can be sparse when few participants meet a threshold, and a control group with no observed events makes some unadjusted relative measures difficult or impossible to estimate in the usual way. In those circumstances, confidence intervals can be wide and estimates sensitive to the analysis method. The registry's sponsor-reported logistic analyses and the unadjusted comparisons represented in the posted results are not identical analyses; their estimates need not match exactly.

Randomization supports comparisons between assigned groups, while masking can reduce some sources of bias in treatment delivery and outcome assessment. The record states randomization and triple masking, but does not give the specific masking roles or procedural details. The study used a parallel-group structure, so each participant was assigned to a study group rather than moving through multiple treatment regimens according to a crossover schedule.

Planning estimates based on observed effects are not evidence of achieved or observed power. The stated planning illustration assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; it considers detecting a stated fraction of the observed effect. Such calculations illustrate potential future planning under those assumptions. They do not reveal the trial's original sample-size justification, which is not reported in the registry record, and they do not show the probability that this completed trial detected an effect.

For secondary comparisons, unadjusted p-values do not account for the fact that multiple endpoints and group comparisons may be examined. General statistical practice treats the interpretation of a collection of tests differently from interpretation of a prespecified single test: repeated testing can increase the chance of at least one apparently notable result under null effects. The registry record does not report a multiplicity adjustment or a testing hierarchy, so the degree of control over that issue is not reported.

10. Analyses Reported by the Sponsor

The registry's sponsor-reported primary analyses are ANCOVA comparisons of each ISIS 681257 cohort with placebo. They report mean differences in percent change, 95% confidence intervals, and p-values. The arm-level values are instead geometric-mean percentage changes from baseline with their own confidence intervals. An arm summary and an adjusted between-group contrast answer related but different questions, so their numerical estimates are not expected to be identical.

The reported ANCOVA contrasts are negative for all five active cohorts, and each reported confidence interval excludes zero. The registry does not flag a discrepancy between sponsor analyses and recalculated results. No competing primary analysis is presented in the record that would warrant treating the sponsor's reported analysis as discrepant.

For the binary Lp(a) threshold outcomes, the sponsor reports logistic-regression odds ratios for comparisons of Cohort A and Cohort B with placebo. These model-based odds ratios are distinct from the posted percentages and unadjusted risk-difference calculations. Differences in effect measure and model can produce different numerical values without constituting a contradiction. The sponsor-reported analyses should be read according to their stated model and outcome scale, rather than substituted for the arm percentages.

11. Multiplicity, Missing Data and Analysis Populations

The registry records 13 outcome measures and 35 analyses. It reports five primary ANCOVA comparisons, each contrasting an active cohort with placebo. The record does not report whether these comparisons were handled under a prespecified hierarchy, whether an adjustment for multiple primary comparisons was used, or whether any interim analyses were planned or conducted. Those details are not reported in the registry record, so the family of primary tests cannot be characterized as multiplicity-controlled on the available information.

There are also multiple secondary outcomes and comparisons. The registry record does not report a multiplicity adjustment for secondary analyses. This matters because the interpretation of one test differs from interpretation of many tests considered together: without a stated control strategy, nominal p-values should not be treated as if they carry the same confirmatory force as a single prespecified test. The secondary analyses may still describe patterns in the data, but their evidential strength is more limited when the testing plan is unknown.

The primary analysis population is defined as the full analysis set: participants who were randomized and received at least 1 dose of ISIS 681257 or placebo. The record calls this the practically feasible intent-to-treat population under ICH Guideline E9. It does not report the number excluded from this population, reasons for exclusion, or a separate intention-to-treat analysis including every randomized participant regardless of treatment received.

The endpoint description does not explain how missing baseline or follow-up measurements were handled, whether any imputation was used, or how missingness was assessed. The missing-data approach is not reported in the registry record. The original sample-size justification and any assumptions used to determine enrollment are also not reported. Planning illustrations based on observed effects are distinct from those original design decisions and are not observed power.

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: acute myocardial infarction (4.2%), angina unstable (2.1%), angina pectoris (2.1%), ventricular tachycardia (2.1%), radius fracture (2.1%).

The safety summaries report participants with serious adverse events, participants with other adverse events, and deaths by treatment cohort. Serious adverse events were reported for 7 participants in each of Cohorts A, B, and C, 3 in Cohort D, 4 in Cohort E, and 3 in placebo. Other adverse events were reported for 38, 42, 40, 36, and 42 participants in Cohorts A through E, respectively, and 36 in placebo. The counts describe participants affected, not the total number of event episodes.

Deaths were reported for 1 participant in Cohort C and 1 in Cohort E; the other cohorts, including placebo, had 0 reported deaths. The record lists acute myocardial infarction at 4.2% among the most frequent serious-event terms; angina unstable, angina pectoris, ventricular tachycardia, radius fracture, and road traffic accident are each listed at 2.1%. These terms and frequencies describe reported events and do not establish that study drug caused them.

Safety counts should be read in the context of the number at risk in each arm and the study's duration. They do not by themselves establish comparative safety, and this registry summary does not provide event timing, exposure-adjusted rates, severity details, or causal adjudication. Such details are not reported in the registry record.

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

The primary results show that the measured fasting Lp(a) values decreased from baseline in each active cohort at the specified primary analysis time point, with more negative arm-level estimates than in placebo. The sponsor-reported ANCOVA comparisons also indicate more negative percentage change in each active cohort relative to placebo under the reported model. The results describe a biomarker response over the study period and provide estimates with uncertainty intervals.

They do not establish that cardiovascular events were reduced, that any change persisted beyond the reported follow-up, or that one ISIS 681257 regimen was superior to another. The trial's listed primary endpoint is Lp(a) change, not a clinical event endpoint. No cardiovascular-event treatment effect is reported in the registry record. The results also do not establish how well the findings generalize to people who differ from the enrolled population with elevated Lp(a) and established cardiovascular disease.

The confidence intervals and p-values address statistical uncertainty for specified model-based comparisons, not the clinical importance of a change. The record does not provide a threshold for a clinically meaningful Lp(a) reduction. Nor does an observed change in a biomarker alone prove that changing that biomarker causes a change in disease outcomes.

The secondary threshold outcomes add information about how often participants attained specified Lp(a) levels, but they remain biomarker outcomes. Their interpretation is limited by the multiple endpoints and comparisons and by the absence of a reported multiplicity strategy. Safety counts describe events reported during the study; without detailed attribution and exposure information, they should not be treated as proof that an event was or was not caused by treatment.

14. Why This Trial Matters Statistically

Statistically, this study illustrates why dose-ranging trials report both arm-level summaries and between-group contrasts. The geometric-mean percentage changes describe each cohort's change from baseline, while the ANCOVA estimates compare each active regimen with placebo. Keeping those estimands distinct helps prevent an arm's change from being mistaken for the effect attributable to treatment relative to control.

The study also varies dose and dosing frequency across cohorts. The pattern in the primary estimates provides descriptive information across those regimens, but comparisons against placebo do not by themselves test whether active regimens differ from one another. That distinction is important whenever a dose-ranging result is used to motivate further study: apparent ordering of estimates is not the same as a formal between-dose comparison.

The endpoint is a biomarker, so the findings answer a focused question about Lp(a) change rather than a broader question about cardiovascular outcomes. The registry's reporting of confidence intervals makes the precision of the estimated changes visible, while the absence of reported multiplicity details limits how strongly multiple comparisons can be treated as confirmatory. The study therefore offers a useful example of interpreting model-based contrasts, uncertainty, endpoint scope, and the limits of conclusions from a phase 2 biomarker study without extending those conclusions beyond the outcomes measured.

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.

The registry record does not report the specific covariates included in the primary ANCOVA model, the procedure used to randomize participants, or allocation concealment details. It also does not identify which parties were masked despite recording triple masking. These omissions limit independent assessment of the analysis and trial procedures from the registry record alone.

The five active cohorts vary both dose and schedule. Although each is compared with placebo, the reported primary contrasts do not provide formal comparisons among the active cohorts. Therefore, the pattern of estimates should not be treated as a tested ranking of doses or frequencies. The primary endpoint is a biomarker measured at Month 6, with the week corresponding to that time point differing across cohorts; longer-term persistence of the observed changes is not reported.

The registry record does not report a multiplicity strategy, an interim-analysis plan, a detailed missing-data procedure, or the original sample-size justification. It also does not report numeric results for LDL-C, apoB, or oxidized-phospholipid endpoints in the secondary results described. Safety summaries provide participant counts, but not exposure-adjusted rates, event timing, detailed severity, or causal attribution. These are limitations of the information reported in the registry record and do not establish that the trial did or did not use procedures that are not described there.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Phase 2 Study of ISIS 681257 (AKCEA-APO(a)-LRx) in Participants With Hyperlipoproteinemia(a) and Cardiovascular Disease — NCT03070782.
• Registry record last updated 2020-10-30; status last verified 2020-10.
• Results first posted 2020-10-30.
• 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

This completed phase 2 study randomized 286 participants with hyperlipoproteinemia(a) and established cardiovascular disease to five ISIS 681257 dose and schedule cohorts or dose-matched-volume placebo. It was recorded as double-blind, triple-masked, and parallel-group. The primary outcome was fasting Lp(a) percent change from baseline at Month 6, corresponding to Week 25 or Week 27 depending on cohort.

All five active cohorts had negative arm-level primary estimates, and the reported ANCOVA comparisons with placebo were negative with 95% confidence intervals excluding zero. The result concerns Lp(a) change, not cardiovascular-event reduction. The record does not report multiplicity handling, missing-data procedures, the original sample-size justification, or a formal comparison among active regimens.