C1 Esterase Inhibitor (C1INH-nf) for the Treatment of Acute Hereditary Angioedema (HAE) Attacks: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for C1 Esterase Inhibitor (C1INH-nf) for the Treatment of Acute Hereditary Angioedema (HAE) Attacks (NCT00289211), a randomized phase 3 study. The primary outcome is Time to Beginning of Substantial Relief of the Defining Symptom.

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, parallel-group phase 3 study comparing C1INH-nf with matching saline placebo for acute hereditary angioedema attacks. The registry identifies Shire as the sponsor and reports enrollment of 83. The study began on 2005-03-14, had primary completion on 2007-04-13, and is recorded as completed.

Participants assigned to C1INH-nf received 1,000 Units intravenously. A second 1,000 U dose could be given if there was no response 60 minutes after the first dose. Participants assigned to placebo received matching saline intravenously, with a second placebo dose permitted under the same stated condition. Double masking means the study is recorded as blinded to participants and study personnel; further details of who remained masked are not reported in the registry record.

The primary endpoint was time to the beginning of substantial relief of the defining symptom. Participants assessed symptoms every 15 minutes for up to 4 hours after the initial dose, or until substantial relief occurred. The registry describes the primary analysis population as all randomized subjects. It also reports endpoint-specific arm counts of 36 for C1INH-nf and 35 for placebo; the reason these counts differ from total enrollment is not reported in the registry record.

Secondary outcomes included the number of subjects beginning substantial relief, time to complete resolution of the attack, antigenic and functional C1INH serum levels, and complement C4 serum levels. The registry lists 6 outcome measures and 15 analyses. Details of the allocation procedure, sample-size justification, and a fuller statistical analysis plan are not reported in the registry record.

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

Analysis population for the primary outcome, as defined in the registry: Intent-to-treat (ITT) Population (all randomized subjects). Since less than 50% of subjects in the placebo group achieved the endpoint, median time to event was not estimable (NE). Further, the number of censored events in the C1INH-nf and placebo groups precluded estimation of the 95% confidence interval (CI) upper bound for median time to event.

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 Time to Beginning of Substantial Relief of the Defining Symptom, measured over Within 4 hours after initial treatment.

The primary endpoint measured elapsed time from initial treatment to the beginning of substantial relief of the defining symptom. This was not simply the first report of improvement: participants had to show improvement in 3 consecutive assessments, and the event time was assigned to the first assessment in that sequence. Assessments were scheduled every 15 minutes up to 4 hours after treatment, or until substantial relief was achieved.

A time-to-event endpoint can retain information about when an event occurs, rather than recording only whether it occurred by a fixed point. In general, this is useful when the timing of a response is clinically relevant and participants may not all experience the event during the observation period. Participants who do not experience the event during observation can contribute censored follow-up, meaning their exact event time is unknown but their observed follow-up still informs the analysis.

The registry reports median time to substantial relief and a Cox regression hazard ratio. For placebo, fewer than 50% of subjects achieved the endpoint, so the median time was not estimable. The registry also notes that the number of censored events in both groups precluded estimation of a confidence interval for the corresponding time-to-event comparison. The reason for using this endpoint, beyond the stated study objective of evaluating safety and efficacy in acute attacks, is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a time-to-event measure. Each participant's observed time runs from initial treatment to the first assessment in a sequence of 3 consecutive assessments showing improvement in the defining symptom. Participants were assessed every 15 minutes up to 4 hours, or until substantial relief occurred. Because some participants may not reach the event during observation, their records can be censored: the exact event time is unknown, while the participant's observed time without the event remains informative.

What does the median time describe?

In general, the median event time is the time by which half of the analyzed group is estimated to have experienced the event. A median may not be estimable when the observed follow-up does not show that half the group has experienced the event. The registry specifically notes that fewer than 50% of the placebo group achieved the endpoint and states that its median time to event was not estimable. The arm results nevertheless list 4.0 hours for placebo, alongside a 95% confidence interval from 2.0 to 4.0 hours. That presentation and the stated non-estimability are not fully reconciled in the registry record.

How should the confidence intervals be read?

In general, a confidence interval describes the statistical uncertainty around an estimate under the method used. It is not a range containing a known fixed proportion of individual participants' event times. The reported interval for the C1INH-nf median is 1.0 to 4.0 hours, while the listed placebo interval is 2.0 to 4.0 hours. The Cox regression estimate has its own 95% confidence interval, from 1.008 to 4.164. These intervals address different quantities and should not be treated as interchangeable. The registry says that the number of censored events in the two groups precluded estimation of a confidence interval for the relevant time-to-event result; the incomplete wording in the record does not clarify exactly which interval it refers to.

What does the Cox hazard ratio represent?

The reported Cox regression hazard ratio is 2.048, comparing C1INH-nf with placebo. In general, a hazard ratio compares the event rates over follow-up between groups, conditional on the model's structure. It is not a risk ratio, an absolute difference, or a ratio of median times. A value above 1 in this comparison corresponds to a higher modeled rate of beginning substantial relief in the C1INH-nf group. Interpreting this estimate relies on the Cox model and its assumptions, including how the relative event rates behave over time. Whether those assumptions were checked, and how censoring was handled in detail, are not reported in the registry record.

Why can censoring and over-dispersion matter?

In general, censoring occurs when a participant's event time is not observed within the recorded follow-up. Time-to-event methods can use the follow-up information available up to censoring, but a substantial amount of censoring can make estimates less precise or make a median unavailable. The registry explicitly identifies the number of censored events as a limit on estimating a confidence interval. Over-dispersion is a general concern when data vary more than a statistical model expects; it can make standard errors too small if the model does not account for that extra variation. The registry record does not describe an over-dispersion assessment, and the primary endpoint's Cox analysis is not described as a count model.

How do the secondary comparisons differ?

The secondary binary outcome records the number of participants with beginning substantial relief within 4 hours. A recalculated unadjusted risk difference is 0.176, with a 95% confidence interval from -0.058 to 0.410 and p = 0.147. The registry's own analysis of that outcome reports a Cochran-Mantel-Haenszel risk ratio of 1.41, with a 95% confidence interval from 0.87 to 2.29 and p = 0.062. These are different effect measures and analyses, so their point estimates and p-values need not match. The recalculated result is a simple unadjusted comparison from the posted summary statistics; it is not a replacement for the registry's analysis.

The planning illustrations are based on the observed binary effect and stated assumptions: two-sided significance level 0.05, equal allocation, and a normal approximation. They describe participants per arm for 80% or 90% power under the listed fractions of the observed effect. Such values are planning illustrations, not observed power and not evidence about the original study's design. The registry's own sample-size justification is not reported in the registry record.

10. Analyses Reported by the Sponsor

For the secondary outcome counting participants who began substantial relief within 4 hours, the registry reports a Cochran-Mantel-Haenszel risk ratio of 1.41, with a 95% confidence interval from 0.87 to 2.29 and p = 0.062. The recalculated unadjusted analysis expresses the comparison as a risk difference of 0.176, with a 95% confidence interval from -0.058 to 0.410 and p = 0.147. These results use different effect measures and different analysis approaches, so the numerical estimates and p-values are not expected to be identical.

The recalculation is a simple unadjusted analysis of the posted counts. It does not reproduce the stratification or other details that may be part of a Cochran-Mantel-Haenszel analysis; details of those implementation choices are not reported in the registry record. The two analyses are not a direct numerical check of one another because a risk ratio and a risk difference describe effects on different scales. The record does not flag a discrepancy between the sponsor analysis and the recalculation. For the primary time-to-event endpoint, the registry reports a Cox regression analysis, and no corresponding recalculated result is presented.

Neither analysis by itself establishes clinical importance. The confidence intervals and the distinct methods should be considered along with the study design, analysis population, and limitations of the registry record.

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies an intent-to-treat population consisting of all randomized subjects for the primary endpoint. It reports total enrollment of 83, while the primary arm summaries list 36 participants for C1INH-nf and 35 for placebo. The record does not explain the difference between these figures, nor does it describe how participants without an evaluable endpoint were handled. It also does not report reasons for missing observations, withdrawals, or exclusions from particular analyses. These details matter because missing outcome data can affect both the amount of information available and the validity of a comparison.

The primary endpoint was assessed repeatedly up to 4 hours or until substantial relief. The registry notes that fewer than 50% of placebo subjects achieved the endpoint, and that censoring limited estimation of a confidence interval for the time-to-event result. The record does not give a fuller account of censoring patterns, follow-up completeness, or sensitivity analyses addressing missing or censored observations.

The registry lists 6 outcome measures and 15 analyses. Whether the primary endpoint was designated in advance in the same form as reported, which secondary analyses were prespecified, and whether any adjustment for multiple testing was used are not reported in the registry record. In general, testing multiple outcomes can increase the chance of a nominally unusual result when no multiplicity adjustment is used. Without the analysis plan, the confirmatory status of secondary findings cannot be fully assessed.

The allocation method is described as randomized, but the sequence-generation and concealment procedures are not reported in the registry record. The study is described as double-blind, but masking procedures and any assessment of masking are not detailed. Interim analyses, stopping rules, and sample-size justification are also not reported. These omissions do not establish that the procedures were inadequate; they limit what can be verified from the registry record.

12. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 1% reporting threshold.

Most frequent non-serious adverse event terms

The adverse-event figures are reported for separate safety groups: 71 participants at risk in the C1INH-nf group and 12 in the placebo group. Serious adverse events are recorded for 0 participants in each group. Other adverse events are recorded for 2 participants in the C1INH-nf group and 1 participant in the placebo group. No serious adverse-event terms are listed. Death counts are not reported in the registry record.

These figures should be read as descriptive counts, not as a complete comparative safety analysis. The group sizes recorded for the safety data differ from the primary endpoint arm counts, and the record does not explain the difference. It also does not provide event definitions, timing, severity details, or information on how multiple events per participant were handled. The absence of a listed serious adverse event does not establish that no safety risk exists; it means that none is recorded in these specific registry figures. The available record is insufficient to make a broad safety comparison.

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

The primary time-to-event analysis reports a Cox hazard ratio of 2.048 for C1INH-nf compared with placebo, with a 95% confidence interval from 1.008 to 4.164 and p = 0.048. Under the reported analysis, the estimate is consistent with a higher rate of beginning substantial relief in the C1INH-nf group over follow-up. The group median estimates are 2.0 hours for C1INH-nf and 4.0 hours for placebo, but the registry states that the placebo median was not estimable because fewer than 50% of placebo subjects achieved the endpoint. The median entries therefore require particular caution.

The result does not state how an individual participant will respond, and a hazard ratio is not a direct measure of the probability that a participant will experience relief. The confidence interval indicates uncertainty about the size of the relative event-rate estimate. The p-value is not the probability that the result is due to chance or the probability that a treatment effect is absent. Neither statistical significance nor a numerical estimate alone establishes clinical importance.

The secondary binary outcome uses a different analysis and has uncertainty that includes no difference on its reported scales. Detailed numerical results for several other listed secondary outcomes are not reported in the registry record. The registry also does not report whether multiplicity was handled, nor does it provide enough information to assess missing data, censoring patterns, or the full statistical plan.

These results describe the posted analysis of a specific randomized, double-blind, placebo-controlled study of acute attacks. They do not, by themselves, establish how the result applies beyond the participants and circumstances represented in the trial. The registry record is a limited account of study methods and results; it should be interpreted as evidence with stated uncertainty and reporting gaps, not as a complete account of every aspect of the study.

14. Why This Trial Matters Statistically

This trial illustrates why endpoint definition matters. The primary outcome was not simply whether symptoms improved; it required improvement across 3 consecutive assessments, with the event time assigned to the first assessment in that sequence. That operational definition affects which observations count as an event and when the event is considered to have occurred. It also makes the assessment schedule and censoring rules important to interpretation.

The result also demonstrates the distinction between median event time and a hazard ratio. The median describes an estimated point in the event-time distribution, while the Cox hazard ratio compares modeled event rates over follow-up. The registry reports a Cox estimate and arm medians, but also states that the placebo median was not estimable and that censoring limited confidence-interval estimation. Reading these statements together is essential; a single summary statistic does not capture all of the uncertainty in the record.

The secondary binary endpoint provides a further statistical lesson: the same underlying counts can be summarized with different effect measures and analyzed using different methods. A risk difference and a risk ratio answer related but distinct questions, and an unadjusted recalculation does not reproduce a stratified sponsor analysis. Their estimates should not be treated as interchangeable or as competing versions of the same number.

Finally, the registry lists numerous outcomes and analyses, while key information about multiplicity, missing observations, and the original sample-size justification is not reported. Those gaps constrain how strongly secondary findings can support a conclusion. The trial is therefore useful for understanding the roles of time-to-event methods, censoring, and analysis transparency in clinical research.

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 leaves important details about analysis populations unresolved. It describes the primary population as all randomized subjects and reports total enrollment of 83, yet the primary arm counts are 36 and 35. Safety groups are recorded with different at-risk counts. Reasons for these differences, and the handling of participants who did not contribute to particular outcomes, are not reported in the registry record.

The primary endpoint presentation also contains a limitation in interpretation: the registry states that the placebo median was not estimable because fewer than 50% of placebo subjects achieved the endpoint, while listing a placebo median of 4.0 hours and an interval from 2.0 to 4.0 hours. It further notes that censoring limited estimation of a confidence interval, without a complete explanation of the affected estimate in the recorded description. The censoring distribution and any sensitivity analyses are not reported.

The full statistical plan is not reported in the registry record. This includes sample-size justification, interim analyses, prespecification of secondary analyses, and any adjustment for multiple testing. Details on randomization procedures, allocation concealment, and masking are also not reported. These gaps make it difficult to evaluate some potential sources of bias or to distinguish confirmatory analyses from exploratory ones.

Numerical results are not available in the record for several listed secondary outcomes. Safety reporting is limited to counts, with no serious-event terms listed and no death counts reported. The available information therefore supports a focused reading of the reported endpoint results, but not a comprehensive evaluation of every efficacy, safety, or analysis question.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: C1 Esterase Inhibitor (C1INH-nf) for the Treatment of Acute Hereditary Angioedema (HAE) Attacks — NCT00289211.
• Registry record last updated 2021-06-11; status last verified 2021-06.
• Results first posted 2010-06-02.
• 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 registry describes a completed randomized, double-blind, parallel-group phase 3 study of C1INH-nf versus matching saline placebo for acute hereditary angioedema attacks. Total enrollment was 83. The primary endpoint was time to beginning substantial relief of the defining symptom, assessed within 4 hours after initial treatment.

The reported Cox analysis gave a hazard ratio of 2.048, with a 95% confidence interval from 1.008 to 4.164 and p = 0.048. The registry notes that the placebo median time was not estimable and that censoring limited confidence-interval estimation. Secondary outcomes included relief within 4 hours, attack resolution, and serum measures. Important details on missing data, multiplicity, and the original sample-size justification are not reported in the registry record.