Study to Assess the Efficacy and Safety of Rilonacept Treatment in Participants With Recurrent Pericarditis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Assess the Efficacy and Safety of Rilonacept Treatment in Participants With Recurrent Pericarditis (NCT03737110), a randomized phase 3 study. The primary outcome is Time to Pericarditis Recurrence in the RW Period.

2. Clinical Question

The study was designed to assess efficacy and safety in recurrent pericarditis. Its primary endpoint was time from randomization to the first recurrence in the randomized withdrawal period, with recurrences confirmed by a Clinical Events Committee for the primary analysis. A recurrence required a return or worsening of typical pericarditis pain together with specified objective evidence. The registry description includes clinical, laboratory, electrocardiographic, and imaging findings among the evidence used to adjudicate events.

This withdrawal design asks about the difference between continuing rilonacept and switching to placebo after the run-in. It does not compare starting rilonacept with never receiving it. The registry reports 86 enrolled participants overall and identifies 30 in the randomized rilonacept group and 31 in the randomized placebo group for the primary analysis. The reported comparison concerns these randomized withdrawal groups and the study’s defined recurrence endpoint, rather than every possible measure of symptoms or disease activity.

3. Trial Design

This was a Phase 3, randomized, triple-masked, parallel-group study with a double-blind, placebo-controlled randomized withdrawal period and an open-label extension. The registry lists recurrent pericarditis as the condition and identifies rilonacept and placebo as the interventions. The study was sponsored by Kiniksa Pharmaceuticals (UK), Ltd.; its status is completed.

The study began on 2019-01-07, and the registry lists 2020-05-29 as the primary completion date. Overall enrollment is reported as 86. That enrollment figure describes the study record as a whole; the primary endpoint analysis specifically used the intent-to-treat population of participants randomized during the randomized withdrawal period. The reported group sizes for that analysis are 30 for rilonacept and 31 for placebo.

The arm descriptions indicate that participants received rilonacept during a single-blind run-in before eligible participants entered randomized withdrawal. In that period, participants were assigned to continue rilonacept or receive placebo. The trial also had an open-label extension. The registry excerpt does not provide complete descriptions of every treatment sequence or the detailed eligibility and transition rules, so those details are not inferred.

The primary endpoint was analyzed as time to first recurrence, using Kaplan–Meier methods and a log-rank comparison. Recurrences confirmed by the Clinical Events Committee were used for the primary analysis. The registry reports additional outcomes on clinical response, pericarditis pain, and symptoms during the randomized withdrawal period, including measurements at Week 16. The record lists 64 outcome measures and 19 analyses, but does not report the full statistical plan or explain the relationship among all listed measures.

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.

Run-in (RI) Period

Reasons for not completing this period:

Randomized Withdrawal (RW) Period

Reasons for not completing this period:

Long-Term Extension (LTE) Period

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat (ITT) Analysis Set: all participants who were randomized in the RW period.

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: All participants entering the double-blind Randomized Withdrawal Treatment Period.

6. Primary Endpoint

The registry lists the primary outcome as Time to Pericarditis Recurrence in the RW Period, measured over RW Period (mean 24.8 weeks).

The primary endpoint measures the time from randomization in the withdrawal period to the first pericarditis recurrence. It is a time-to-event outcome: participants contribute information about when recurrence occurred, rather than only whether recurrence was present at a single visit. The analysis used Kaplan–Meier methods, and the primary recurrence events were confirmed by a Clinical Events Committee.

The registry defines recurrence through the return or worsening of typical pericarditis pain together with objective supporting evidence. Its adjudication criteria refer to pain ratings and C-reactive protein findings, with additional supportive evidence such as fever, a pericardial rub, electrocardiogram changes, or imaging findings. The precise combination of criteria depends on the recorded clinical findings, as described in the registry record.

A time-to-event endpoint is useful when the timing of an event is important and when participants may have different lengths of observation. It can use information from participants who have not had a recorded event by the end of their follow-up, rather than treating them as if an event occurred at that point. This general advantage depends on appropriate follow-up and analysis assumptions; the registry excerpt does not report enough detail to assess all such assumptions.

The randomized withdrawal period is listed as having a mean duration of 24.8 weeks. The primary endpoint therefore addresses recurrence during that study period, not recurrence risk over an unreported longer horizon or outcomes outside the trial’s follow-up.

7. Results Posted in the Registry

• Randomized Withdrawal Period: Rilonacept vs Randomized Withdrawal Period: Placebo: hazard ratio 0.04 (95% CI 0.01–0.18), reported p < 0.001 (as reported in the registry).

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.

Major Secondary Efficacy Endpoint: Percentage of Participants Who Maintained Clinical Response at Week 16 of the RW Period

• 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:
  • Randomized Withdrawal Period: Rilonacept: 81.0% (17/21) versus 20.0% (4/20) on control; risk difference 61.0 percentage points (95% CI 36.7 to 85.2), risk ratio 4.05 (1.64–9.96); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Randomized Withdrawal Period: Rilonacept than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Major Secondary Efficacy Endpoint: Percentage of Days With No or Minimal Pericarditis Pain at Week 16 of the RW Period

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Randomized Withdrawal Period: Rilonacept: mean 98.60 versus 47.40 on control; difference 51.20 (95% CI 30.67 to 71.73), standardized effect d = 1.52; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Randomized Withdrawal Period: Rilonacept than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Major Secondary Efficacy Endpoint: Percentage of Participants With Absent or Minimal Pericarditis Symptoms Based on the Patient Global Impression of Pericarditis Severity (PGIPS) at Week 16 of the RW Period

• 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:
  • Randomized Withdrawal Period: Rilonacept: 81.0% (17/21) versus 25.0% (5/20) on control; risk difference 56.0 percentage points (95% CI 30.6 to 81.3), risk ratio 3.24 (1.47–7.11); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Randomized Withdrawal Period: Rilonacept than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes examine related but distinct features: maintaining clinical response, the amount of time with little or no pericarditis pain, and participant-reported symptom severity. At Week 16, the posted group estimates generally favor rilonacept across these measures. Taken together, that pattern is directionally consistent with the primary time-to-recurrence comparison, while reflecting different definitions and measurement scales.

These outcomes do not duplicate the primary endpoint. A response or symptom assessment at a scheduled time point is not the same as time until a confirmed recurrence. Pain days summarize experience over a period, while a global impression captures a participant’s assessment. Differences in endpoint definitions, observation windows, and analysis methods mean that agreement across measures can be supportive without making them interchangeable.

The registry lists additional secondary outcomes at Week 8 and Week 24, as well as Week 16 outcomes, but the detailed results summarized in the record are not available for every listed measure. The overall pattern should therefore be understood from the posted analyses, not extended to outcomes without reported results.

Secondary analyses also carry less confirmatory weight when the multiplicity strategy is not reported. With many outcomes and analyses, chance alone can produce apparently strong findings if each is judged independently. The record lists 64 outcome measures and 19 analyses but does not state a testing hierarchy or whether multiplicity adjustments were used. For that reason, the primary endpoint remains the central statistical result, with secondary findings serving as context rather than independent proof of every possible effect.

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

• Major Secondary Efficacy Endpoint: Percentage of Participants Who Maintained Clinical Response at Week 16 of the RW Period
• Major Secondary Efficacy Endpoint: Percentage of Days With No or Minimal Pericarditis Pain at Week 16 of the RW Period
• Major Secondary Efficacy Endpoint: Percentage of Participants With Absent or Minimal Pericarditis Symptoms Based on the Patient Global Impression of Pericarditis Severity (PGIPS) at Week 16 of the RW Period
• Percentage of Participants Who Maintained Clinical Response at Week 24 of the RW Period
• Percentage of Participants Who Maintained Clinical Response at Week 8 of the RW Period
• Percentage of Days With No or Minimal Pericarditis Pain at Week 24 of the RW Period
• Percentage of Days With No or Minimal Pericarditis Pain at Week 8 of the RW Period
• Percentage of Participants With Absent or Minimal Pericarditis Symptoms at Week 24 of the RW Period Based on the PGIPS

9. Statistical Interpretation

10. Statistical Methods Explained

The primary analysis concerns time to first recurrence. Kaplan–Meier methods estimate the event-free experience over follow-up while retaining information from participants whose follow-up ends before a recurrence is observed. In general, this approach requires that censoring be handled appropriately and that the relevant follow-up assumptions are reasonable. The registry identifies Kaplan–Meier analysis but does not report the detailed censoring rules or diagnostics.

What does the hazard ratio describe?

A hazard is the instantaneous event rate among participants who have not yet had the event, conditional on remaining under observation and event-free up to that time. A hazard ratio compares this rate between groups over follow-up. It is not the same as a risk ratio, a difference in recurrence percentages, or a ratio of median times. The reported hazard ratio is 0.04, with a 95% confidence interval from 0.01 to 0.18. Its interpretation relies on the time-to-event comparison and should not be translated into an absolute probability without appropriate event-rate information.

The registry identifies a log-rank method for the reported comparison. In general, the log-rank test compares the observed and expected event experience between groups across follow-up, accounting for when events occur. Its p-value evaluates compatibility with a null comparison under the test’s assumptions; it does not measure the probability that the null hypothesis is true or the clinical importance of the result. The sponsor reports p < 0.001.

Why is event adjudication relevant?

The primary analysis used Clinical Events Committee-confirmed recurrences. Adjudication applies the study’s event criteria to clinical information, which can help make event classification more consistent. Here, the registry defines recurrence using pericarditis pain and objective evidence, with additional supportive findings in some combinations. The endpoint is therefore not simply any pain report or any isolated laboratory value. The record does not give the adjudication committee’s operating procedures or report how often events were disputed.

How should secondary analyses be read?

The secondary endpoints include binary outcomes, such as maintaining clinical response or having absent or minimal symptoms, and a continuous outcome, percentage of days with no or minimal pain. For a binary outcome, a percentage is a group-level summary of how often the defined outcome was recorded. A confidence interval describes uncertainty around that estimate. A risk difference expresses an absolute difference in proportions, while an odds ratio compares the odds of the outcome between groups. These measures are not numerically interchangeable and can convey different impressions when outcomes are common.

For the pain outcome, the registry reports least-squares means and a standard error. In general, a least-squares mean is a model-based group estimate, and a standard error describes uncertainty in that estimate. The sponsor reports an ANCOVA comparison. Model-based estimates can account for factors included in the model, but the registry excerpt does not state all covariates or model details. The recalculated, unadjusted comparisons are not substitutes for those sponsor analyses.

Multiplicity matters because interpreting many outcome tests as if each were the only test can increase the chance of at least one apparently noteworthy result. The registry lists 64 outcome measures and 19 analyses. This does not establish how many were prespecified, which were confirmatory, or whether an adjustment was applied. The multiplicity plan is not reported in the registry record. Secondary p-values should therefore be treated as supportive and interpreted in light of the primary endpoint and any prespecified hierarchy, which is not available.

Over-dispersion is a general concern when count outcomes vary more than a simple model assumes, potentially making standard errors too small if the extra variation is ignored. It is not the principal issue for this reported time-to-event endpoint, and the registry does not report an over-dispersion analysis. More broadly, methods should match the outcome type: event timing, binary status, and continuous percentages each require different summaries and assumptions.

11. Confidence Intervals and Precision

The primary hazard ratio estimate is 0.04, and its 95% confidence interval extends from 0.01 to 0.18. The interval expresses sampling uncertainty under the analysis model: values within the interval are reasonably compatible with the observed data at the stated confidence level, subject to the model and study assumptions. It is not a range containing 95% of individual participants’ outcomes, nor does it account for every potential source of bias or uncertainty.

The interval remains below 1, consistent with the reported direction of the comparison. It nevertheless spans a range of relative hazard values, so the point estimate should not be treated as exact. The sponsor-reported p-value, p < 0.001, indicates strong evidence against the null comparison under the log-rank analysis; it does not remove uncertainty about the magnitude of the effect.

For placebo, the median time to recurrence is 8.6 weeks, with a 95% confidence interval from 4.0 to 11.7 weeks. The primary result does not report a rilonacept median or its interval. The interval for the placebo median describes uncertainty in that group’s estimated median, not the uncertainty in a direct difference between group medians.

Secondary endpoints also have intervals, but their interpretation depends on the outcome scale and analysis. For example, intervals for percentages or differences in percentages are expressed on an absolute scale, whereas odds ratios use a relative odds scale. Wide intervals indicate less precision than narrow intervals on the same scale. Precision is affected by the amount and timing of observed information, not just by enrollment. The registry excerpt does not report all details needed to assess the causes of uncertainty, including missing-data handling and follow-up completeness.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size table and curve are planning illustrations based on the observed effect, not the trial’s original design calculation. For the time-to-event endpoint, the entries are the total number of events needed. For the other outcomes, entries are participants per arm. These quantities should not be confused with the randomized group sizes or with the registry’s overall enrollment.

The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is a stated fraction of the observed effect. The time-to-event calculations use the Schoenfeld approach. The trial’s own original sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the planned sample or event count at which the assumed design would have an 80% chance of detecting the specified effect under those planning assumptions, if that effect and the other assumptions were true. The 90% power column uses a higher detection probability and consequently requires a larger planning count in the listed scenarios. These are design probabilities under assumptions, not guarantees about the result.

Rows planning for a smaller fraction of the observed effect require more participants or events than rows planning for a larger fraction. This is because smaller effects are harder to distinguish from random variation. The values are not observed power; they are planning illustrations and should not be read as a retrospective measure of the trial’s success.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

d is the total number of events needed (Schoenfeld's formula) with 1:1 allocation and proportional hazards.

Worked calculation for this trial

1. Hazard ratio HR = 0.040, so ln(HR) = -3.2189.
2. d = 4 × (1.96 + 0.842)² / [ln(0.040)]² = 3.0, rounded up to 4 events in total for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
hr <- 0.04                              # observed hazard ratio
events_needed <- function(power, frac = 1) ceiling(round(4 * (z_a + qnorm(power))^2 / log(hr^frac)^2, 6))   # Schoenfeld, 1:1 allocation
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: events = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(events_needed(0.80, f), big.mark = ","), format(events_needed(0.90, f), big.mark = ",")))

# Simulation: given d total events under proportional hazards, events in the treatment arm are Binomial(d, hr / (1 + hr)).
# The score (log-rank type) statistic is (D1 - d / 2) / sqrt(d / 4).
set.seed(3737110)
d    <- events_needed(0.80)
reps <- 20000
D1 <- rbinom(reps, d, hr / (1 + hr))
z  <- (D1 - d / 2) / sqrt(d / 4)
cat(sprintf("Simulated power with %s events: %.3f (%d replicates)\n", format(d, big.mark = ","), mean(abs(z) > z_a), reps))

Output from running the code

100% of observed effect: events = 4 (80% power), 5 (90% power)
 75% of observed effect: events = 6 (80% power), 8 (90% power)
 50% of observed effect: events = 13 (80% power), 17 (90% power)
Simulated power with 4 events: 0.855 (20000 replicates)

The simulation uses the fact that, with d events in total and proportional hazards, the number of events in the treatment arm is binomial. It is an approximation that conditions on the total number of events, and it should land close to the target power.

14. Analyses Reported by the Sponsor

The primary analysis reported by the sponsor is a log-rank comparison with a hazard ratio of 0.04, a 95% confidence interval from 0.01 to 0.18, and p < 0.001. The recalculated time-to-event display uses the sponsor-reported hazard ratio, as the posted summary statistics do not support an independent reconstruction of the time-to-event analysis.

For secondary outcomes, the registry includes sponsor analyses as well as simple unadjusted recalculations from posted summary statistics. The sponsor’s analyses may use methods such as Cochran–Mantel–Haenszel or ANCOVA, whereas the recalculations use the available group summaries and do not reproduce every feature of those models. Differences between an unadjusted recalculation and an adjusted or stratified sponsor estimate can arise from the analysis method and information used.

There is no discrepancy flagged for the posted results. Where sponsor analyses are available, they remain the primary reported analyses; recalculations are descriptive checks and should not replace the sponsor’s results. The registry excerpt does not include enough model detail to independently validate every adjustment or reproduce the full analysis.

15. Multiplicity, Missing Data and Analysis Populations

The registry record lists 64 outcome measures and 19 analyses. It identifies a primary endpoint and several major secondary endpoints, but does not report the complete prespecified statistical analysis plan, the hierarchy among all endpoints, or whether and how multiplicity was controlled. It is therefore not possible to determine from the record whether every secondary test was intended as confirmatory or how family-wise error was handled.

Multiplicity is a general issue when several hypotheses are tested. If each test is interpreted at its own significance threshold without adjustment, the chance of at least one false-positive finding across the set can increase. An adjustment, gatekeeping strategy, or prespecified hierarchy can address this concern, but none should be attributed to this study without documentation. Secondary p-values should be considered in this context.

The primary analysis population is identified as the intent-to-treat set of all participants randomized in the randomized withdrawal period. The registry does not report the detailed handling of missing observations, withdrawals, censoring, or deviations from assigned treatment. It also does not describe how missing data were handled for the Week 16 secondary endpoints. These omissions limit assessment of whether the analysis populations and missing-data methods were consistent across outcomes.

Interim analyses, stopping rules, and the original sample-size justification are not reported in the registry record. Their absence from this record does not establish that a particular procedure was or was not used; it means the posted information is insufficient to characterize it. Interpretation should therefore distinguish the clearly reported primary result from unreported design and analysis details.

16. 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: squamous cell carcinoma (3.3%), cardiac flutter (3.2%), ileus (3.2%), pyrexia (3.2%), acute myocardial infarction (1.4%).

The safety record presents serious adverse events and other adverse events for the run-in, randomized withdrawal, and long-term extension periods. In the run-in, 1 serious adverse event was reported among 86 participants at risk, and other adverse events were reported for 55. During randomized withdrawal, the rilonacept-only period before bailout lists 1 serious adverse event among 30 at risk; the corresponding placebo-only period before bailout lists 1 among 31. The period summaries that include bailout rilonacept list 1 serious adverse event in the rilonacept group and 3 in the placebo group.

For other adverse events, the record lists 18 in the randomized rilonacept group and 7 in the placebo group before bailout; the summaries including bailout rilonacept list 18 and 13, respectively. The long-term extension lists 6 serious adverse events and 43 other adverse events among 74 at risk. The registry lists zero deaths in each of the reported groups.

These are descriptive counts across periods with differing treatment exposure and observation. They do not by themselves establish that an event was caused by treatment, and overlapping period descriptions should not be combined into a new total. The listed serious-event terms and percentages are registry safety summaries, not comparative estimates of risk. The record does not provide enough context to assess event timing, exposure-adjusted rates, or causal attribution.

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

The primary result indicates that, in this randomized withdrawal study, the reported event-time comparison favored continued rilonacept over placebo for time to a Clinical Events Committee-confirmed pericarditis recurrence. The sponsor-reported hazard ratio is 0.04, with a 95% confidence interval from 0.01 to 0.18. The placebo median time to recurrence is reported as 8.6 weeks, while a rilonacept median is not reported in the primary result.

This finding concerns a specific population and study design: eligible participants entered a withdrawal period after a rilonacept run-in, and the comparison was between continued rilonacept and placebo. It does not directly answer how treatment compares with no prior rilonacept exposure, nor does it establish what would happen outside the trial’s observation period. The randomized withdrawal period is listed as having a mean duration of 24.8 weeks.

The result does not imply that every participant will experience the same outcome, or that recurrence is impossible in the rilonacept group. A hazard ratio is a group-level relative measure, and the posted estimate has uncertainty. Nor does a statistically strong primary result by itself settle the clinical importance of every secondary endpoint or establish the balance of potential benefits and harms.

Safety findings should be read separately from efficacy. The registry lists serious and other adverse events across several study periods, but the counts alone do not establish causation or provide exposure-adjusted comparisons. Overall interpretation also depends on the event definition, analysis population, follow-up, and missing-data methods. Several of those details are not reported in the registry record.

18. Why This Trial Matters Statistically

This study illustrates why the choice of endpoint and design matters in clinical statistics. A withdrawal design compares continued treatment with placebo among eligible participants after a run-in; its result addresses that treatment decision in that study context. It is not equivalent to a trial comparing treatment initiation with no treatment from the outset.

The primary outcome is time to recurrence, so the analysis incorporates when events occur rather than reducing follow-up to a single yes-or-no measurement. Kaplan–Meier estimation and a log-rank comparison are recognizable tools for that question. The reported hazard ratio communicates a relative comparison, while the placebo median provides a group-specific time summary. Neither statistic alone describes every participant’s experience.

The record also makes visible the distinction between primary and secondary evidence. Symptom, pain, and response outcomes can help characterize the pattern of findings, but they address different questions and are subject to the consequences of multiple testing. Without a reported multiplicity strategy, secondary p-values deserve more cautious interpretation than the identified primary analysis.

Finally, the record shows why precision and reporting details belong alongside statistical significance. The primary confidence interval conveys uncertainty around the hazard ratio; the group median is reported for placebo but not rilonacept. Missing-data procedures, interim analyses, and the original sample-size justification are not reported in the registry record. Those limitations do not erase the posted finding, but they define what can and cannot be concluded from the available statistical account.

19. 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 withdrawal design limits the scope of inference: the randomized comparison followed a rilonacept run-in and involved participants eligible for randomized withdrawal. The result therefore pertains to that setting and should not be generalized automatically to people who have not received rilonacept or to different treatment sequences.

The primary result does not report a median time to recurrence for the rilonacept group, while the placebo median and its confidence interval are reported. The registry provides a hazard ratio and interval, but the excerpt does not describe detailed censoring rules, proportional-hazards checks, or follow-up completeness. It is not possible to assess those features from the record alone.

Several details relevant to interpretation are not reported in the registry record: the original sample-size rationale, interim analyses, the complete multiplicity plan, detailed missing-data methods, and the full model specifications for secondary analyses. The registry lists outcome measures beyond those with detailed results in the record. Their results should not be inferred from the outcomes that are reported.

The safety summaries cover different periods, including summaries before and including bailout rilonacept. Those groups and periods should be read as separately reported descriptions, not combined into an overall comparative rate. Counts do not establish treatment causality, and the excerpt does not provide exposure-adjusted adverse-event rates. These reporting limitations warrant caution when comparing safety across periods or interpreting secondary findings.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study to Assess the Efficacy and Safety of Rilonacept Treatment in Participants With Recurrent Pericarditis — NCT03737110.
• Registry record last updated 2023-09-06; status last verified 2023-08.
• Results first posted 2021-07-13.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

The registry record describes a completed Phase 3, randomized, triple-masked, parallel-group study of rilonacept and placebo in recurrent pericarditis. It included a single-blind rilonacept run-in, a double-blind randomized withdrawal period, and an open-label extension. Overall enrollment was 86; the primary analysis used the intent-to-treat set randomized in the withdrawal period.

The primary endpoint was time to the first Clinical Events Committee-confirmed pericarditis recurrence during randomized withdrawal. The sponsor-reported comparison was a hazard ratio of 0.04, with a 95% confidence interval from 0.01 to 0.18 and p < 0.001. The placebo median time to recurrence was 8.6 weeks, with a 95% confidence interval from 4.0 to 11.7 weeks; a rilonacept median is not reported.

Secondary outcomes addressed response, pain, and symptoms. The registry also reports adverse-event summaries across study periods. Multiplicity procedures, missing-data methods, interim analyses, and the original sample-size justification are not reported in the registry record.