A Phase III, Crossover Trial Evaluating the Efficacy and Safety of KVD900 (Sebetralstat) for On-Demand Treatment of Angioedema Attacks in Adolescent and Adult Patients With Hereditary Angioedema (HAE): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase III, Crossover Trial Evaluating the Efficacy and Safety of KVD900 (Sebetralstat) for On-Demand Treatment of Angioedema Attacks in Adolescent and Adult Patients With Hereditary Angioedema (HAE) (NCT05259917), a randomized phase 3 study. The primary outcome is Time to Beginning of Symptom Relief Patient Global Impression of Change (PGI-C).

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, placebo-controlled, Phase III, three-way crossover trial sponsored by KalVista Pharmaceuticals, Ltd. The trial evaluated two dose levels of KVD900, 300 mg and 600 mg, alongside placebo for on-demand treatment of hereditary angioedema attacks. Its status is completed. The registry reports enrollment of 136.

A crossover design evaluates treatments across more than one period, rather than assigning every participant to just one treatment for the entire trial. In this trial, the three interventions were KVD900 300 mg, KVD900 600 mg, and placebo. Randomization and double blinding were part of the recorded design. The registry record does not report the treatment sequences, period lengths, washout details, or the specific procedures used to maintain masking.

The primary efficacy analysis used the full analysis set. This included randomized patients who received trial medication from at least one period for the respective qualifying hereditary angioedema attack. Participants were presented according to randomized treatment. The endpoint results therefore have treatment-specific counts: 87 for KVD900 300 mg, 93 for KVD900 600 mg, and 84 for placebo. Because the trial used a crossover model, these arm counts are not a basis for summing or inferring the number of distinct participants represented by the endpoint results.

The registry lists two secondary outcomes: time to the first incidence of a decrease from baseline in patient global impression of severity at 2 time points in a row, and time to complete hereditary angioedema attack resolution by PGI-S. The record also lists 3 outcome measures and 6 analyses. The dates recorded for trial start and primary completion are 2022-02-22 and 2023-12-31.

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: A total of 136 patients were randomized across the 6 treatment sequences in a 3-way crossover design, of which a total of 110 patients treated at least 1 attack with IMP and therefore were included in the Full Analysis Set (FAS).

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 randomized patients who received trial medication from at least one period for the respective qualifying HAE attack and are presented according to the randomized treatment. The FAS was the population for efficacy analyses.

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.

Only the first 6 of 7 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Time to Beginning of Symptom Relief Patient Global Impression of Change (PGI-C), measured over Within 12 hours of the first investigational medicinal product (IMP) administration..

The primary endpoint measured time from the first investigational medicinal product administration to the beginning of symptom relief, within 12 hours. Relief was defined using the Patient Global Impression of Change (PGI-C): a rating of at least “a little better” at 2 time points in a row. The endpoint is a time-to-event measure because it records how long it takes for a specified event—in this case, the defined patient-reported improvement—to occur.

A time-to-event endpoint can retain information about when an event occurs, rather than reducing the outcome to whether it occurred by a single assessment time. The median is the time by which half of the observed event-time distribution has reached the event, subject to the analysis and censoring rules. A confidence interval describes uncertainty around the estimated median; it is not a range containing most individual participants’ event times.

The registry reports median times of 1.61 hours for KVD900 300 mg, with a 95% confidence interval from 1.28 to 2.27; 1.79 hours for KVD900 600 mg, with a 95% confidence interval from 1.33 to 2.27; and 6.72 hours for placebo, with a reported lower confidence limit of 2.33 and no upper limit shown. The endpoint was designed to capture a patient-reported change rather than an investigator-defined measure alone. Its use gives the analysis a defined, repeat-confirmed threshold for symptom relief. The registry record does not report further details about the validation or interpretation of PGI-C for this trial.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a time-to-event measure. The event was the beginning of symptom relief, defined as a PGI-C rating of at least “a little better” at 2 time points in a row. The clock began at the first investigational medicinal product administration, and the endpoint window was within 12 hours. The analysis reported median time and a 95% confidence interval for each treatment group. A median gives a central point in the estimated event-time distribution. It does not mean that all participants reached relief at that time, and it is not an average of every participant’s time.

Time-to-event methods are useful when the timing of an event matters and when some event times are not observed during the analysis window. In general, a censored observation contributes information up to the point at which observation ends, without being treated as though the event happened at that time. In this trial, attacks without the defined beginning of relief by 12 hours were treated as right-censored. Attacks that received conventional attack treatment before the endpoint event within 12 hours were also treated as right-censored. The registry states that when an endpoint result was non-evaluable within 12 hours, an event, if it occurred, must have occurred more than 12 hours after study drug.

The sponsor used the Gehan score transformation test for the primary efficacy analysis. In general, a score test compares groups using information across the observed event-time distributions, including the ordering of event times and censored observations. It is a different summary from comparing only the reported medians. The record identifies the test used but does not provide further implementation details, such as the exact test statistic, assumptions, or model specification. The reported comparisons were KVD900 300 mg versus placebo and KVD900 600 mg versus placebo.

Each treatment’s median and confidence interval are descriptive summaries of that treatment’s event-time distribution under the analysis. A confidence interval communicates uncertainty in an estimated quantity. A wider interval generally indicates less precision than a narrower interval, while interval interpretation depends on the method and assumptions used to construct it. The placebo median is reported as 6.72 hours, with a lower confidence limit of 2.33 hours; an upper confidence limit is not reported in the registry record. The record does not explain why that limit is unavailable, so it should not be replaced with an assumed value.

The full analysis set was the efficacy population. It included all randomized patients who received trial medication from at least one period for the respective qualifying attack, with results presented according to randomized treatment. This definition makes clear that the endpoint is analyzed for qualifying attacks after medication was received. The record does not report additional details about exclusions, protocol deviations, or the handling of participants with multiple qualifying attacks beyond the endpoint-specific censoring rules.

Because this was a crossover trial, participants could contribute experience under more than one treatment period. In general, crossover designs can support within-participant comparisons, but interpretation can depend on treatment sequence, period effects, and whether effects carry over between periods. The registry record does not report the sequence allocation details, period effects, carryover assessment, or any specific statistical adjustment for these features. The treatment-specific endpoint counts of 87, 93, and 84 should therefore be read as posted group counts, not treated as independent sets of distinct participants or added to obtain a participant total.

The reported p-values are inferential summaries from the sponsor’s named test. A p-value is calculated under a null hypothesis and the test’s assumptions; it indicates how unusual the observed test result would be if that null hypothesis and assumptions held. It does not measure the probability that the null hypothesis is true, and it does not quantify the clinical importance of the result. The reported values were p < 0.001 for KVD900 300 mg versus placebo and p = 0.001 for KVD900 600 mg versus placebo.

The registry record also lists a planning illustration based on the observed effect. Its assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. For time-to-event planning, the reported quantity is a total number of events, while other planning results are expressed as participants per arm. Such illustrations are planning calculations, not observed power and not evidence of the trial’s original sample-size justification. The registry record does not report the trial’s sample-size justification or whether these assumptions were used in its original design.

General statistical consideration: over-dispersion occurs when outcome variability exceeds what a statistical model assumes. If present, it can make standard errors too small and confidence intervals or tests too optimistic unless the analysis accounts for it. The primary endpoint is time-to-event, and the registry identifies the Gehan score transformation test; it does not report an over-dispersion analysis. This general concept should not be mistaken for a reported feature of this trial.

The registry record lists two secondary outcomes and 6 analyses, but it does not report whether the primary and secondary tests were adjusted for multiplicity or how outcomes were prioritized beyond identifying the primary endpoint. When multiple outcomes or comparisons are examined, the chance of at least one apparently unusual result can increase if no adjustment or hierarchy is used. Since the adjustment approach is not reported, no claim about multiplicity control can be made from the recorded information.

10. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary analyses as Gehan score transformation tests: p < 0.001 for KVD900 300 mg versus placebo and p = 0.001 for KVD900 600 mg versus placebo. These are the reported inferential results for the two active-dose comparisons. They accompany arm-specific median estimates and confidence intervals, rather than a reported confidence interval for either treatment contrast.

No separate unadjusted re-calculation of the treatment comparisons is reported in the registry record. The arm summaries alone do not provide all the information required to reproduce a time-to-event test, including event-time data and censoring patterns. A comparison made by simply contrasting median values would not reproduce the sponsor’s Gehan test. The registry’s own analysis should therefore be read as the inferential analysis available for these comparisons, without treating an arithmetic contrast of medians as an equivalent test.

The sponsor’s tests address each KVD900 dose against placebo. They do not report a direct test comparing the two doses. The registry record also does not report whether the p-values were adjusted for multiple comparisons. Accordingly, the posted results support interpretation of the two stated tests on their own terms; they do not supply a reported dose-versus-dose result or a multiplicity-adjusted conclusion.

11. Multiplicity, Missing Data and Analysis Populations

The record lists a primary endpoint, two secondary outcomes, 3 outcome measures, and 6 analyses. The primary endpoint has two reported comparisons with placebo: KVD900 300 mg and KVD900 600 mg. The registry record does not report a multiplicity adjustment, a testing hierarchy, or a procedure specifying how the primary and secondary tests were interpreted together. It is therefore not possible to determine from the record whether the reported p-values account for the number of analyses conducted.

In general, multiplicity matters when several hypotheses are tested. Without a stated adjustment or testing strategy, the chance of at least one small p-value across a set of tests can be greater than for an individual test. This is a general statistical consideration, not evidence that the trial did or did not use a particular procedure. The sponsor’s reported p-values remain the results stated for the specified comparisons, but a reader should not assume that they are multiplicity-adjusted.

The registry describes the primary endpoint’s censoring rules. Attacks without the defined beginning of relief by 12 hours were treated as right-censored, as were attacks receiving conventional attack treatment before the event within 12 hours. For non-evaluable results within 12 hours, the record says that if an event occurred, it must have occurred more than 12 hours after study drug. These are reported analysis rules; the registry does not describe how other missing data were handled, whether sensitivity analyses were performed, or how any other missing secondary-outcome results were treated.

The efficacy population was the full analysis set, defined as randomized patients who received trial medication from at least one period for the respective qualifying attack, analyzed according to randomized treatment. This definition is reported, but the registry does not provide a detailed participant flow or explain all exclusions from each endpoint analysis. The trial is a crossover design, while treatment-specific endpoint counts differ. The record does not report whether or how the analysis accounted for period, sequence, or carryover effects, and no assumption about those analyses should be inferred.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: intervertebral disc protrusion (1.2%), hereditary angioedema (1.1%), anisocoria (1.1%).

The registry reports serious adverse events for each treatment group using a number at risk and a number affected. For KVD900 300 mg, 1 of 86 was affected; for KVD900 600 mg, 2 of 93 were affected; and for placebo, 0 of 83 were affected. Deaths were reported as 0 in each group. The record also lists serious-event terms with percentages: intervertebral disc protrusion, 1.2%; hereditary angioedema, 1.1%; and anisocoria, 1.1%.

These counts describe the posted safety data, not comparative risk estimates. They should not be interpreted as proof that an event was caused by a treatment or that a treatment prevented an event. The registry record does not report a statistical comparison of serious adverse events, the relationship assessment for the listed events, or further detail about event timing and overlap. The “at risk” denominators differ from the efficacy endpoint counts, so the efficacy counts should not be substituted for the safety denominators.

Serious adverse events are a specific safety category and should not be conflated with all adverse events. The record separately lists other affected counts of 5 for KVD900 300 mg, 6 for KVD900 600 mg, and 13 for placebo. These figures are descriptive counts; the registry record does not report a comparative statistical analysis for them.

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

The posted results show that the estimated median time to the defined beginning of symptom relief was 1.61 hours for KVD900 300 mg and 1.79 hours for KVD900 600 mg, compared with 6.72 hours for placebo. The sponsor reported a Gehan score transformation test result of p < 0.001 for the 300 mg comparison with placebo and p = 0.001 for the 600 mg comparison with placebo. These results concern the specified time-to-event endpoint and efficacy population.

The results do not mean that every participant experienced relief at the reported median time. They do not show the distribution of individual experiences in full, and the registry does not report a treatment contrast confidence interval. The tests also do not establish a difference between the two KVD900 doses because no direct dose-versus-dose analysis is reported. A small p-value is not a measure of effect size and is not the probability that the result is due to chance.

These analyses do not answer every clinical or statistical question about treatment. They do not establish results for populations outside the trial’s adolescent and adult hereditary angioedema population, for different attack definitions, or for outcomes not reported in the registry record. The trial evaluated treatment in a crossover design, but treatment sequences, period and carryover assessments, and other associated analysis details are not reported. Such information can matter when interpreting crossover results.

The endpoint rules also shape what the estimates mean. Attacks without beginning of relief by 12 hours and attacks treated with conventional attack treatment before the event were right-censored. The posted estimates therefore reflect the endpoint definition and those censoring rules; they should not be read as uncensored event times for every attack. The record does not report further sensitivity analyses that would show how alternative handling of these situations might affect results.

The safety counts are likewise descriptive. Serious events and deaths were reported by treatment group, but the registry does not report a statistical safety comparison or establish causation for listed events. Taken together, the efficacy and safety figures should be understood as results recorded for this trial, with the stated analysis population, endpoint definition, follow-up window, and reporting limitations.

14. Why This Trial Matters Statistically

This trial raises several useful statistical issues for clinical researchers. First, it uses a time-to-event primary outcome based on a patient-reported threshold. The time to a repeat-confirmed PGI-C response preserves information about timing, while the median supplies a compact summary. Interpreting that summary requires attention to the event definition and to censoring, rather than treating the median as a simple average or as a result experienced by everyone.

Second, the trial’s three-way crossover structure makes the treatment-specific summaries important to read in context. A crossover can involve the same participants under different treatments, so arm counts are not necessarily counts of separate participant groups. The registry identifies the design but does not report treatment sequence, period, or carryover details. Those omissions limit what can be concluded about the implementation and statistical handling of the crossover.

Third, the registry reports a named test and p-values for two active-treatment comparisons with placebo, alongside separate arm medians. This distinction illustrates why a test result and descriptive estimates answer different questions. The p-values summarize the test evidence under its assumptions; the medians describe each group’s event-time distribution. No reported contrast interval or direct comparison between active doses is available, and no multiplicity approach is reported.

Finally, censoring and reporting completeness matter. The primary analysis has explicit rules for attacks that do not reach the event or receive conventional attack treatment before it, but details on additional missing-data handling and sensitivity analyses are not reported. The safety information gives group-specific counts but not a comparative safety analysis. These features show why statistical interpretation should keep the outcome definition, design, analysis population, and limits of the registry record together rather than relying on a single number.

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.

Several trial-specific details that could inform interpretation are not reported in the registry record. Although the study is described as a three-way crossover trial, treatment sequences and period structure are not provided. The record also does not describe an assessment or adjustment for period effects or carryover. These details matter because results in a crossover design can depend on the order and timing of treatments and on whether effects persist into later periods.

The primary endpoint has stated censoring rules, but broader missing-data information is limited. The registry does not report the extent of non-evaluable outcomes, the handling of missing data outside the stated endpoint rules, or sensitivity analyses. It also does not report a detailed participant flow for the endpoint analyses. Arm-specific counts are reported, but the record does not give enough information to reconstruct all exclusions or explain the difference between efficacy and safety denominators.

The registry reports p-values for each active dose versus placebo, but does not report confidence intervals for those comparisons or a direct statistical comparison between the two active doses. It does not state whether multiplicity was controlled across the comparisons and listed secondary outcomes. As a result, the posted statistics do not resolve the dose-versus-dose question or establish the effect of any adjustment for multiple analyses.

The placebo median’s lower confidence limit is reported, but its upper confidence limit is not reported in the registry record. The reason is not given. The registry identifies the Gehan score transformation test without further details about implementation or assumptions. It also does not report the trial’s original sample-size justification. Planning illustrations based on observed effects are not a substitute for that information, and are not observed power.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Phase III, Crossover Trial Evaluating the Efficacy and Safety of KVD900 (Sebetralstat) for On-Demand Treatment of Angioedema Attacks in Adolescent and Adult Patients With Hereditary Angioedema (HAE) — NCT05259917.
• Registry record last updated 2025-05-02; status last verified 2025-04.
• Results first posted 2024-07-19.
• 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

NCT05259917 is a completed, randomized, double-blind, placebo-controlled, Phase III, three-way crossover trial sponsored by KalVista Pharmaceuticals, Ltd. It evaluated KVD900 300 mg and KVD900 600 mg for on-demand treatment of hereditary angioedema attacks in adolescent and adult patients. Enrollment was 136.

The primary endpoint was time to beginning of symptom relief within 12 hours, defined as a PGI-C rating of at least “a little better” at 2 time points in a row. Median times were 1.61 hours for KVD900 300 mg, 1.79 hours for KVD900 600 mg, and 6.72 hours for placebo. The reported Gehan score transformation test results were p < 0.001 and p = 0.001 for the respective dose comparisons with placebo.

The registry also lists two secondary outcomes and reports serious adverse event counts and deaths by treatment group. It does not report a multiplicity procedure, a direct comparison between the active doses, or several crossover and missing-data details relevant to further interpretation.