A Phase 2b Study of CSL112 in Subjects With Acute Myocardial Infarction.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase 2b Study of CSL112 in Subjects With Acute Myocardial Infarction. (NCT02108262), a randomized phase 2 study. The primary outcome is Percent of Participants With Clinically Important Change in Drug-induced Liver Injury.

2. Clinical Question

This is a safety and tolerability question, not a test of whether CSL112 prevents a later cardiovascular event. The primary endpoint focuses on laboratory changes meeting a threshold and confirmed on repeat measurement. It does not represent every liver-test change, nor does it by itself establish the cause of a change. The registry record also lists time-to-first major adverse cardiovascular event and several measures of apolipoprotein A-I and phosphatidylcholine among the secondary outcomes. Those outcomes address different questions and do not replace the primary safety comparison.

3. Trial Design

The registry describes a randomized, parallel-group, placebo-controlled phase 2b study of CSL112 in participants with acute myocardial infarction. It lists quadruple masking; the brief summary describes the study as double-blind. The registry record does not explain the difference between those descriptions or identify which roles were masked. The stated purpose was to investigate hepatic and renal safety and tolerability of multiple administrations at two dose levels compared with placebo.

The planned regimen was one intravenous infusion each week for 4 consecutive weeks. Placebo was given at the same frequency, volume and duration as either active regimen. The posted primary endpoint data include CSL112 (2 g), CSL112 (6 g) and placebo, with group sizes of 415, 416 and 413, respectively. These endpoint group sizes are distinct from the overall enrollment reported as 1267; the registry record does not explain that difference.

The primary endpoint used the Safety population: randomized participants in the main study or PK/PD substudy who received at least a partial infusion of investigational product. The endpoint timeframe was baseline, before the first infusion, to Day 29. The trial is marked completed, with a start date of 2014-08 and primary completion in 2015-12. Its primary focus was safety and tolerability; secondary outcomes included cardiovascular events and laboratory measures of apoA-I and PC. The registry record does not report a sample-size justification, allocation ratio, details of sequence generation, or a full statistical analysis plan.

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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Safety population (SP) comprised all subjects who were randomized into the main study or PK/PD substudy and received at least a partial infusion of the investigational product.

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.

6. Primary Endpoint

The registry lists the primary outcome as Percent of Participants With Clinically Important Change in Drug-induced Liver Injury, measured over From baseline (before first infusion) to Day 29..

The primary endpoint is the percentage of participants with a clinically important change in drug-induced liver injury from baseline to Day 29. The definition requires either an ALT change greater than 3 times the upper limit of normal or a total-bilirubin change greater than 2 times the upper limit of normal, with the change confirmed on repeat measurement. The endpoint therefore counts participants meeting a threshold-and-confirmation rule, rather than measuring the average change in a liver laboratory value.

A threshold endpoint can make a safety signal interpretable in terms of the frequency of participants crossing a prespecified clinical laboratory boundary. Requiring confirmation can reduce the influence of a single unconfirmed measurement. Those are general properties of this type of endpoint; they do not establish why a particular participant's laboratory result changed. The registry record does not report details about how repeat measurements were timed or adjudicated.

Safety monitoring is relevant when a study administers repeated doses and its stated objective includes hepatic and renal tolerability. However, the primary endpoint addresses only the defined liver-injury measure over the stated timeframe. It does not summarize all liver or kidney outcomes, all adverse events, or longer-term safety. Nor does it measure treatment benefit. Interpretations should stay within the endpoint definition, the recorded follow-up period and the safety population.

7. Results Posted in the Registry

• CSL112 (2 g): 1.0% (4/415) versus 0.0% (0/413) on control; risk difference 1.0 percentage points (95% CI 0.0 to 1.9); unadjusted two-proportion z-test p = 0.045.
• CSL112 (6 g): 0.5% (2/416) versus 0.0% (0/413) on control; risk difference 0.5 percentage points (95% CI -0.2 to 1.1); unadjusted two-proportion z-test p = 0.158.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Percent of Participants Who Experience Bleeding Events

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • CSL112 (2 g): 9.2% (38/415) versus 12.3% (51/413) on control; risk difference -3.2 percentage points (95% CI -7.4 to 1.0), risk ratio 0.74 (0.50–1.10); unadjusted two-proportion z-test p = 0.138.
  • CSL112 (6 g): 9.1% (38/416) versus 12.3% (51/413) on control; risk difference -3.2 percentage points (95% CI -7.4 to 1.0), risk ratio 0.74 (0.50–1.10); unadjusted two-proportion z-test p = 0.135.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CSL112 (2 g) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CSL112 (6 g) and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists a broad set of secondary outcomes, including time-to-first major adverse cardiovascular event, changes in apoA-I and PC concentrations, and bleeding events. These outcomes span different clinical and laboratory questions, populations and timeframes. They should not be treated as interchangeable measures of a single effect or combined into a general claim about benefit or harm.

For bleeding events, the posted comparisons have negative recalculated risk differences for both CSL112 dose groups versus placebo, with confidence intervals that include zero and unadjusted p-values. Those estimates are secondary and do not alter the definition or interpretation of the primary liver-injury endpoint. The registry does not provide sponsor-reported analyses for that secondary comparison. Results for the other listed secondary outcomes are not detailed in the registry record used for this account, so an overall numerical pattern across all secondary outcomes cannot be established.

Secondary findings generally warrant more caution when many outcomes or comparisons are examined. General explanation: testing multiple outcomes increases the chance that some results appear notable by chance, particularly when analyses are unadjusted. The registry record does not report a multiplicity strategy, such as an adjustment or a hierarchy controlling the interpretation of multiple tests. Unadjusted secondary results therefore carry less weight than a prespecified primary analysis and should be viewed as exploratory unless the record establishes otherwise. They can inform questions for further study, but they cannot substitute for the primary safety result or settle the trial's broader clinical questions.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• The Percentage of Participants With a Time-to-first Major Adverse Cardiovascular Event (MACE)
• Change From Baseline in Concentrations of Apolipoprotein A-I (apoA-I) and Phosphatidylcholine (PC) at End of First Infusion for All Participants
• Change From Baseline in Plasma Concentrations of apoA-I and PC at End of Fourth Infusion for All Participants
• Change From Baseline in Plasma Concentrations of apoA-I and PC at End of First Infusion for Participants With Normal Renal Function
• Change From Baseline in Plasma Concentrations of apoA-I and PC at End of Fourth Infusion for Participants With Normal Renal Function
• Change From Baseline in Plasma Concentrations of apoA-I and PC at End of First Infusion for Participants With Mild Renal Impairment
• Change From Baseline in Plasma Concentrations of apoA-I and PC at End of Fourth Infusion for Participants With Mild Renal Impairment
• Change From Baseline in Plasma Cmax for apoA-I and PC After First Infusion for All Participants

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: each person either meets the confirmed threshold definition during the stated interval or does not. A risk is the proportion of participants with the event in a group. A risk difference compares that proportion between groups. In these results, the comparisons are against placebo. General explanation: a risk difference of zero corresponds to equal observed event proportions; a positive value means a higher event proportion in the active group, and a negative value means a lower one. The reported primary risk differences are expressed as proportions in the recalculated comparisons and as differences in event rates (%) in the sponsor analyses.

What does a confidence interval describe?

A confidence interval describes uncertainty around an estimated effect under the assumptions of the method used. It is not a range that guarantees the true effect lies within it, and it does not measure the probability that a treatment works. For a difference, an interval that includes zero is compatible with no difference under the corresponding analysis. An interval that excludes zero can still be imprecise in practical terms, and its interpretation depends on the analysis plan, the endpoint and the study context. General statistical explanation applies; the registry's reported analyses and their stated method remain central.

Why can methods yield different answers?

The sponsor's primary analyses use the Newcombe-Wilson score method for differences in event rates (%). The recalculated results are explicitly unadjusted calculations from the posted summary statistics. Different methods, estimands, handling of small or zero event counts, or analysis specifications can produce different estimates, intervals and p-values. The registry flags a discrepancy, so the sponsor's own analysis takes precedence rather than treating the recalculation as a correction. The record does not report enough detail to identify the reason for the discrepancy or reconcile the methods fully.

General explanation: score-based methods construct intervals using the event proportions and their sampling uncertainty, rather than relying only on a simple normal approximation to the difference. This can be useful for proportions, particularly when event counts are sparse. Here, the placebo group had no recorded primary events, and the active-group event counts were also small. Sparse data make results sensitive to the chosen method and produce uncertainty that should not be concealed by a single p-value.

How should the p-values be read?

A p-value describes how incompatible the observed data are with a specified null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. The recalculated p-values are unadjusted. The reported sponsor p-values are attached to analyses labelled non-inferiority, but the non-inferiority margin and decision rule are not reported in the registry record. Without those details, the p-values cannot independently establish the non-inferiority conclusion.

General explanation: a non-inferiority analysis asks whether the data rule out an unacceptable loss relative to a comparator, using a margin chosen in advance. That question differs from simply testing whether the groups have exactly equal event rates. The margin and analysis population are essential to interpreting such an analysis. The registry record does not provide the margin, so no threshold for an acceptable difference can be inferred from these results alone.

What is the role of the planning illustrations?

The posted sample-size illustrations are based on the observed bleeding-event effect, not on the primary liver-injury endpoint or on the trial's own design. They use a two-sided significance level of 0.05, equal allocation between two groups and a normal approximation; the illustrations show participants per arm for 80% and 90% power at specified fractions of the observed effect. These values are planning scenarios, not observed power, and do not describe the number enrolled in this trial. General explanation: planning calculations estimate how many participants may be needed under stated assumptions to detect a chosen effect. They do not show that a completed study had a particular probability of detecting its actual effect.

The registry reports 63 outcome measures, 61 posted secondary outcomes and 6 analyses. The record does not provide a complete account of model specifications, missing-data methods, interim analyses or multiplicity procedures. Those omissions limit how fully the reported estimates can be evaluated. Results should therefore be interpreted with the population definition, endpoint timeframe, analysis type and discrepancy notice kept in view.

11. Confidence Intervals and Precision

Precision is conveyed by the confidence interval, not by the number of decimal places in an estimate. The recalculated 95% interval for CSL112 (2 g) versus placebo runs from < 0.001 to 0.019; for CSL112 (6 g) versus placebo it runs from -0.002 to 0.011. The sponsor-reported intervals are wider in their displayed scale: -0.1 to 2.5 for the 2 g comparison and -0.5 to 1.7 for the 6 g comparison. The registry identifies a discrepancy, so the sponsor's analysis is the primary reference for interpretation.

General explanation: a wider interval indicates more uncertainty about the effect's magnitude. An interval that includes zero leaves no-difference values compatible with the data under that analysis. For the sponsor analyses, both intervals include zero. The recalculated interval for the 2 g group does not include zero, while its associated p-value is unadjusted; that contrast with the sponsor's result illustrates why the flagged discrepancy matters and why the recalculation should not be privileged.

The small number of observed primary events limits the information available about the difference between groups. A zero count in placebo does not prove that the underlying event risk is zero. Nor does a nonzero count in an active group establish a stable estimate of the underlying risk. Confidence intervals express sampling uncertainty, but they do not account for every possible source of bias, incomplete follow-up or analytic uncertainty. Details about missing data and endpoint adjudication are not reported in the registry record.

12. Analyses Reported by the Sponsor

The registry reports sponsor analyses using the Newcombe-Wilson score method, with a difference in event rates (%) and a 95% confidence interval for each CSL112 dose versus placebo. For CSL112 (2 g), the reported estimate is 1.0, the interval is -0.1 to 2.5 and p = 0.124. For CSL112 (6 g), the estimate is 0.5, the interval is -0.5 to 1.7 and p = 0.499. Both are designated non-inferiority analyses.

The simple recalculations from the posted event counts differ from these sponsor results, particularly for the 2 g comparison. The registry explicitly flags a discrepancy. Accordingly, the sponsor's own analysis takes precedence; the recalculated values are useful only as a transparent unadjusted comparison of the posted summary statistics. The record does not report the non-inferiority margin, the full analysis specification or why the recalculation differs. It is therefore not possible to reconstruct the sponsor's decision criteria from the registry results alone. The sponsor-reported intervals include zero, and the p-values should be understood in the context of the stated non-inferiority analyses rather than as standalone evidence of equivalence or safety.

13. Multiplicity, Missing Data and Analysis Populations

Multiplicity concerns arise when a study evaluates many outcomes, doses, timepoints or subgroup comparisons. The registry lists 63 outcome measures, 61 posted secondary outcomes and 6 analyses. The primary comparisons include two CSL112 doses against placebo, while the secondary outcomes include multiple laboratory measures and other clinical endpoints. The record does not report a multiplicity adjustment, a testing hierarchy or another strategy for controlling the interpretation of multiple comparisons. This means that unadjusted secondary p-values should not be treated as if they were the result of one isolated, prespecified test.

The registry identifies a Safety population comprising randomized participants in the main study or PK/PD substudy who received at least a partial infusion. This is the population definition attached to the primary endpoint. The record does not report how many randomized participants did not receive an infusion, how missing endpoint measurements were handled, or whether any sensitivity analyses assessed missing data. It also does not report the primary endpoint's detailed analysis set beyond that safety-population definition.

The sponsor analyses are labelled non-inferiority, but the margin and full decision rules are not reported in the registry record. The record also does not report a sample-size justification for the primary endpoint, interim analyses, or a complete statistical analysis plan. These omissions do not establish that such procedures were absent; they mean the relevant information is not reported in the registry record. General explanation: missing analysis details make it harder to assess whether the chosen methods and decision rules were specified before outcomes were known, and whether uncertainty from multiple testing or missing measurements was addressed. Interpretation should consequently rely on the stated endpoint, reported sponsor analyses and explicit limits of the record.

14. 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: vascular stent thrombosis (11.1%), cardiac failure (11.1%), pulmonary embolism (11.1%), chest pain (2.2%), angina unstable (1.4%).

The adverse-event summary reports serious adverse events affecting 66 participants in CSL112 (2 g), 54 in CSL112 (6 g) and 55 in placebo. Other adverse events affected 15, 23 and 9 participants, respectively. Deaths are reported for 5 participants in the 2 g group, 4 in the 6 g group and 1 in placebo. The registry also lists a safety lead-in group with 9 participants at risk, 1 with a serious adverse event, 2 with other adverse events and 1 death.

These are descriptive counts, not adjusted comparisons. The record does not establish that an event or death was caused by the assigned study infusion. The listed serious-event terms include vascular stent thrombosis, cardiac failure and pulmonary embolism; the registry reports their percentages but does not give further context in the registry record. Serious adverse events are broader than the primary laboratory endpoint, and the event counts cannot be used as a substitute for the endpoint analysis. Differences in counts should be interpreted with the group sizes and the registry's lack of reported causal attribution in mind.

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

The primary results describe a confirmed laboratory safety endpoint observed from baseline to Day 29 in the defined safety population. The event counts were 4 in CSL112 (2 g), 2 in CSL112 (6 g) and 0 in placebo. The sponsor-reported differences in event rates have confidence intervals that include zero. The recalculated unadjusted estimates do not agree fully with the sponsor's analyses, and the registry flags that discrepancy; the sponsor's own analysis takes precedence.

These data do not establish that CSL112 caused the observed laboratory events, nor do they prove that no risk exists in a group with zero observed events. They also do not establish that the groups are identical. The uncertainty intervals and sparse event counts limit precision. Because the non-inferiority margin and full decision criteria are not reported in the registry record, the non-inferiority labels cannot by themselves be translated into a definitive conclusion about acceptable differences.

The endpoint is limited to a specific liver-injury definition and short timeframe. It does not answer whether treatment changes the risk of cardiovascular events, improves outcomes, or affects every aspect of hepatic or renal safety. Secondary outcomes address some other questions, but the registry does not report multiplicity procedures, and the secondary results should not be used to enlarge the primary conclusion. General explanation: a trial result is best understood as evidence about the endpoint, population, timeframe and analysis actually specified, rather than as a universal claim about treatment effects. The registry record supports a description of the recorded results and their uncertainty, not medical advice or a broad efficacy conclusion.

16. Why This Trial Matters Statistically

This trial illustrates why endpoint definitions matter. A percentage meeting a threshold with repeat confirmation is not the same as an average laboratory change, a serious adverse event count or a cardiovascular outcome. The registry specifies the laboratory thresholds, confirmation requirement, baseline reference and Day 29 endpoint timeframe, allowing the primary question to be stated precisely.

It also illustrates how sparse binary outcomes can make analysis choices consequential. The placebo group had no recorded primary events, while each active group had only a small number. The simple unadjusted recalculations differ from the sponsor's Newcombe-Wilson analyses, and the registry explicitly flags the discrepancy. This makes transparent reporting of the sponsor's method, confidence intervals and analysis purpose particularly important. The sponsor analysis takes precedence, but the missing non-inferiority margin limits independent interpretation of that label.

Finally, the record shows why a result needs context beyond a p-value. Confidence intervals communicate uncertainty; the safety population identifies who contributed to the primary endpoint; secondary outcomes raise multiplicity considerations; and adverse-event counts remain descriptive without causal attribution. The registry does not report several details needed for a fuller audit, including a sample-size justification, missing-data methods and a multiplicity strategy. These are general lessons in reading clinical-trial statistics, not evidence that any unreported procedure was or was not performed.

17. 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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

Several limitations are specific to interpreting this registry record. The primary endpoint has few recorded events and none in placebo, so estimates are sensitive to analysis method. The recalculated unadjusted results differ from the sponsor's reported analyses, and the registry does not explain the discrepancy. The sponsor's analysis should therefore take precedence, while the difference itself remains a constraint on independent reconstruction.

The sponsor analyses are labelled non-inferiority, but the margin and full decision criteria are not reported in the registry record. Without them, the clinical interpretation of the estimates cannot be fully assessed. The record also does not report a primary-endpoint sample-size justification, missing-data handling, interim analyses, multiplicity procedures or a complete statistical analysis plan. These are documentation limits; they do not show that the procedures were absent.

The primary endpoint is restricted to a defined laboratory threshold and the period ending Day 29. It cannot summarize later liver outcomes, renal safety in general, cardiovascular efficacy or all adverse events. The safety population includes participants who received at least a partial infusion, but the record does not explain the difference between the overall enrollment of 1267 and the posted endpoint group sizes. The registry lists quadruple masking while the brief summary says double-blind, without explaining the terminology. Finally, only some secondary results are available in the registry record; a complete pattern across all secondary outcomes cannot be assessed.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: A Phase 2b Study of CSL112 in Subjects With Acute Myocardial Infarction. — NCT02108262.
• Registry record last updated 2021-03-15; status last verified 2021-02.
• Results first posted 2021-03-15.
• 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.

21. Record Summary

This completed randomized, parallel-group, placebo-controlled phase 2b study evaluated multiple weekly infusions of CSL112 at two dose levels in participants with acute myocardial infarction. The primary endpoint was the percentage with a confirmed clinically important liver-injury laboratory change from baseline to Day 29 in the Safety population.

The posted primary event counts were 4 for CSL112 (2 g), 2 for CSL112 (6 g) and 0 for placebo. Sponsor-reported Newcombe-Wilson analyses gave differences in event rates of 1.0 and 0.5, with 95% confidence intervals of -0.1 to 2.5 and -0.5 to 1.7. The registry flags a discrepancy with simple unadjusted recalculations, so the sponsor's own analyses take precedence. The non-inferiority margin, multiplicity strategy and several other analysis details are not reported in the registry record. These findings describe the posted short-term safety endpoint and its uncertainty; they do not establish causation or answer broader efficacy questions.