A Study of Baricitinib (LY3009104) in Adult and Pediatric Japanese Participants With NNS/CANDLE, SAVI, and AGS: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Baricitinib (LY3009104) in Adult and Pediatric Japanese Participants With NNS/CANDLE, SAVI, and AGS (NCT04517253), a non-randomized phase 2, phase 3 study. The primary outcome is Change From Baseline in Mean Daily Diary Scores in Participants With CANDLE (Primary Treatment Period).

2. Clinical Question

3. Trial Design

The registry describes a Phase 2/3, multicenter, open-label study in adult and pediatric Japanese participants with NNS/CANDLE, SAVI, or AGS. Its allocation is nonrandomized and its model is single-group. There was no masking and no control arm; consequently, the design does not support a randomized, like-for-like treatment comparison.

Enrollment was 10. The registry lists baricitinib as the intervention and describes condition-specific optimized dosage ranges. CANDLE participants were given a final daily dose ranging from 8 mg to 12 mg, with gradual escalation from 8 mg to 10 mg and 12 mg. SAVI participants had a range of 6 mg to 12 mg, with gradual escalation from 6 mg to 8 mg, 10 mg, and 12 mg. For AGS, the described range was 6 mg to 8 mg daily, with gradual escalation from 6 mg to 8 mg for 24 weeks. The registry record does not give a complete account of every dosing detail for each condition.

The study started on 2020-10-27, and primary completion was recorded as 2023-04-04. Its status is terminated. The reason for termination is not reported in the registry record, and termination alone does not establish why the study stopped or what its results were.

The primary outcome is a CANDLE-specific diary measure assessed at baseline and 20 weeks. Secondary outcomes include diary-score changes for CANDLE, SAVI, and AGS, symptom-specific diary scores, and corticosteroid-dose decreases. The record lists 25 outcome measures overall. It does not report numerical efficacy estimates or analyses for these outcomes.

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: Participants who met all eligibility criteria underwent an 8-week dose adjustment period and received an optimized dosage of baricitinib during the primary treatment period (12 weeks \[week 8 to week 20\] for CANDLE and 24 weeks \[week 8 to week 32\] for SAVI or AGS participants), followed by a maintenance period of up to 191.1 weeks for CANDLE participants, and up to 202.9 weeks for SAVI participants, and up to 206.1 weeks for AGS participants.

Primary Treatment Period

Reasons for not completing this period:

Maintenance Period

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: CANDLE: All enrolled participants who received at least one dose of study drug.

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 enrolled participants who received at least one dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Mean Daily Diary Scores in Participants With CANDLE (Primary Treatment Period), measured over Baseline, 20 weeks.

The primary endpoint is change from baseline in the mean daily diary score for participants with CANDLE during the primary treatment period. The recorded timeframe is baseline and 20 weeks. The diary asks the participant or caregiver to rate fever, rash, musculoskeletal pain, headache, and fatigue. Each symptom is scored from 0 to 4: 0 means no symptoms, 1 mild, 2 moderate, 3 more severe, and 4 severe, equivalent to the worst symptoms. The mean daily score ranges from 0 to 4, with higher scores indicating more severe symptoms. A total score was not used.

As a general statistical concept, repeated diary ratings can describe symptom experience over time and can make within-participant change the quantity of interest. A mean gives a summary of the recorded scores; it does not show how every symptom or every participant changed. Interpretation also depends on how diary observations were collected and how incomplete records were handled, details not reported in the registry record.

The endpoint is specifically for CANDLE. It should not be read as a common primary endpoint across all three conditions. Although the registry names this measure and its assessment times, it does not report the baseline mean, the 20-week mean, the estimated change, variability, or a statistical test. The magnitude and precision of change therefore cannot be assessed from the posted results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record supplies limited information about the planned design and outcome definitions, but no efficacy analysis results. Statistical concepts below are general explanations unless explicitly tied to the recorded study details.

What does a within-group change estimate mean?

The primary endpoint is a change in the mean daily diary score from baseline to 20 weeks in CANDLE participants. In general, a within-group change compares a measure at two times among the same participants or within the defined analysis population. It describes how the measure differs across those times. It does not, by itself, estimate what would have happened over the same period without the intervention.

A comparison group can help distinguish intervention-associated change from other time-related influences, when the comparison is appropriately designed and analyzed. This study is described as single-group and nonrandomized, with no control arm. The primary endpoint therefore does not create a randomized contrast. Any numerical change, if reported, would need to be interpreted in light of the design and the possibility of other explanations for change.

What does a mean and standard deviation describe?

The primary endpoint specifies a mean, and its dispersion is identified as the standard deviation. In general, a mean summarizes the central level of a continuous measure across observations. A standard deviation describes how dispersed individual measurements are around the mean. It is not the same as the uncertainty around an estimated mean or change. A confidence interval is commonly used to express estimation uncertainty; no confidence interval for the primary endpoint is reported in the registry record.

The diary score is an ordinal rating scale for each symptom, while the endpoint is described as a mean daily score. A mean can condense repeated ratings, but it can obscure differences among symptoms, participants, and days. The recorded scale runs from 0 to 4, with higher scores indicating more severe symptoms. The total score was not utilized. The registry does not report the rule used to calculate the mean across diary entries or how incomplete diary days affected it.

Why does the design affect the interpretation?

In general, random assignment helps balance measured and unmeasured prognostic factors across treatment groups on average. This study did not use random allocation and had a single-group model. It therefore cannot provide a randomized between-group estimate. Open-label conduct also means that treatment masking was absent. In general, knowledge of treatment can influence subjective reporting or other study behavior, particularly for diary-based measures. This is a design consideration, not evidence that reporting was biased in this study.

The study included participants with three conditions, and the primary endpoint is restricted to CANDLE. Results from one condition would not automatically describe the other conditions, whose disease context and secondary diary outcomes are distinct. The registry record does not report a pooled efficacy analysis or a disease-specific efficacy estimate.

What can be concluded from the statistical record?

The record lists 25 outcome measures but reports zero analyses. No primary-endpoint estimate, standard error, standard deviation value, confidence interval, or p-value is reported. A p-value, in general, summarizes how compatible data are with a specified statistical model under a null hypothesis; it does not measure the size or clinical importance of an effect. Because no p-value or estimate is reported, neither statistical evidence nor precision can be assessed for the primary endpoint.

The record describes a sample-size planning illustration as based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. For time-to-event outcomes, that illustration refers to a total number of events; for other outcomes, it refers to participants per arm. These are planning assumptions, not observed study findings. The study itself is single-group, so an equal-allocation two-arm illustration is not a description of its actual design. The registry does not report the trial's sample-size justification or numerical planning results.

As a general concept, over-dispersion occurs when variability is greater than a statistical model assumes. It can make standard errors and confidence intervals too small if ignored. The primary endpoint is described as a mean with standard deviation, but the registry record does not report an analysis model or diagnostics, so no claim about over-dispersion or its handling is possible.

Small samples generally yield imprecise estimates and provide limited information about uncommon events. This is a general statistical limitation; the registry record does not provide an efficacy estimate or confidence interval with which to quantify precision. The absence of reported analysis results must not be mistaken for a finding of no change, or for proof of a particular effect.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("CANDLE")
n   <- c(5)
m   <- c(-0.217)
sp  <- c(0.586)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  CANDLE  5  -0.217  0.586  -0.945  0.511

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 25 outcome measures, spanning diary-score changes, symptom-specific diary measures, and corticosteroid-dose decreases. Multiple outcomes create a general statistical issue: when many hypotheses are tested, the chance of at least one apparently notable result can increase unless the analysis plan accounts for multiplicity. The registry record does not report a multiplicity adjustment, a hierarchy of outcome tests, or which secondary analyses were intended to be confirmatory. It also reports zero analyses, so no outcome-specific test results are available to interpret.

Missing-data handling is not reported in the registry record. For a diary endpoint, incomplete entries could affect which days or participants contribute to a mean. In general, an analysis should make clear how missing observations were handled and what assumptions were made about them. No such method, imputation strategy, or sensitivity analysis is reported for this study. The registry identifies the primary population as all enrolled CANDLE participants who received at least one dose of study drug, but does not report the number included in the endpoint analysis or how participants with incomplete diaries were treated.

Analysis-population details for the secondary outcomes are also not reported in the registry record. It is therefore unclear from the posted results which participants contributed to each secondary outcome or whether the analysis populations differed by condition or assessment period. These omissions limit reproducibility and make it difficult to judge whether the reported outcome definitions and analysis populations align. No conclusion about the direction or strength of efficacy evidence can be drawn from the outcome list alone.

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: atypical mycobacterial infection (33.3%), bronchopulmonary aspergillosis (33.3%), haemorrhage intracranial (33.3%), pancytopenia (20.0%), acute coronary syndrome (20.0%).

The registry reports adverse-event counts separately for CANDLE, SAVI, and AGS. For serious adverse events, the recorded risk and affected counts are 5 and 3 for CANDLE, 3 and 1 for SAVI, and 2 and 0 for AGS. For other adverse events, the risk and affected counts are 5 and 5 for CANDLE, 3 and 3 for SAVI, and 2 and 2 for AGS. The record reports deaths of 0 in CANDLE, 1 in SAVI, and 0 in AGS.

These counts describe participants at risk and those affected in the registry's adverse-event record. They do not establish that an event was caused by baricitinib. The groups are small, and the counts should be read as descriptive rather than as stable estimates of event frequency. The record does not report additional detail about event timing, duration, recurrence, or relationship to study treatment in these summaries.

The listed serious-event terms include atypical mycobacterial infection, bronchopulmonary aspergillosis, haemorrhage intracranial, pancytopenia, acute coronary syndrome, and nausea, with percentages stated in the registry. The denominator and context for those term-specific percentages are not reported in the registry record. The single reported death in SAVI is a count, not evidence about its cause. The record does not report a causal attribution for that death.

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

The record establishes that a Phase 2/3, multicenter, open-label, nonrandomized, single-group study of baricitinib was conducted in Japanese adult and pediatric participants with NNS/CANDLE, SAVI, or AGS. It describes condition-specific dosing and defines a CANDLE primary endpoint based on mean daily diary scores from baseline to 20 weeks. It also reports adverse-event counts by condition and a terminated study status.

Those details do not establish the amount or direction of change in the primary endpoint. Numerical CANDLE diary results, uncertainty estimates, and statistical tests are not reported in the registry record. The registry also does not report numerical efficacy analyses for the secondary outcomes. Without these results, readers cannot assess the observed magnitude or precision of change, or determine whether the data support a particular efficacy interpretation.

Even if a within-group change were available, general interpretation would need to account for the absence of a control group. A change over time is not, on its own, a randomized estimate of the effect attributable to treatment. The open-label design and participant- or caregiver-rated diaries are relevant context, but they do not show that the results were biased; the registry does not report evidence to make that determination.

The safety counts show that adverse events were recorded in each condition group, including serious adverse events and one death in the SAVI group. These observations do not establish causation, and they do not answer the efficacy question. Nor does the terminated status identify the reason the study ended or determine what its results mean.

In short, the registry supports a description of the study's design, outcome definitions, and reported safety counts. It does not support a quantitative conclusion about the primary CANDLE endpoint or about treatment effects across CANDLE, SAVI, and AGS. Those efficacy results and their statistical uncertainty are not reported in the registry record.

14. Why This Trial Matters Statistically

This trial is statistically notable because it brings together three rare conditions in a small, single-group study while defining a condition-specific primary endpoint. The registry records enrollment of 10, but the primary endpoint concerns only CANDLE participants who received at least one dose. The efficacy question is therefore narrower than the overall study population and should not be generalized across conditions without condition-specific evidence.

The endpoint uses daily participant or caregiver ratings of symptoms, summarized as a mean score. Such measures can capture changes in reported symptom experience, but their interpretation depends on how repeated ratings are collected, summarized, and handled when missing. The registry gives the symptom scale and assessment timeframe, yet does not report the numerical endpoint results or the detailed analysis method.

The design also matters statistically. A nonrandomized, open-label, single-group study can describe outcomes over time, but it does not supply a randomized control comparison. That distinction is important when interpreting change: observed change and treatment-attributable effect are not interchangeable quantities. The registry's stated planning illustration uses equal allocation between two arms, while the actual study is described as single-group; the planning assumptions should not be confused with the trial's design.

Finally, the record lists many outcome measures but no analyses. This limits assessment of multiplicity, missing-data handling, and precision, as well as evaluation of whether secondary outcomes are consistent with the primary question. The study therefore illustrates why registry outcome definitions and numerical results are both needed for statistical interpretation. The record offers useful design and safety information, but does not report the efficacy analyses needed to quantify what changed.

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 primary limitation for interpreting efficacy is that the registry record does not report numerical results or analyses. The record names the CANDLE diary endpoint and its timeframe, but does not give an estimated change, baseline and follow-up values, a standard deviation value, a confidence interval, or a p-value. The absence of these results prevents assessment of effect size and precision.

The design is nonrandomized, open-label, and single-group, with no control arm. As a general design limitation, change over time in one group cannot isolate the intervention's contribution from other influences. This is especially relevant to a participant- or caregiver-rated diary measure. The registry record does not report additional design features that would address these concerns.

Enrollment was 10 across CANDLE, SAVI, and AGS, while the primary endpoint is CANDLE-specific. The registry does not report the primary endpoint analysis count or condition-specific efficacy results. The adverse-event groups identify participants at risk by condition, but those safety denominators should not be assumed to be efficacy analysis populations.

The status is terminated, but the reason is not reported in the registry record. The record does not report the sample-size justification, detailed statistical analysis plan, multiplicity approach, missing-data strategy, or interim analyses. It lists 25 outcome measures and zero analyses, leaving the planned and performed analysis choices unclear. Termination and incomplete statistical reporting limit what can be concluded, but they do not establish a particular efficacy result or explain why the study ended.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of Baricitinib (LY3009104) in Adult and Pediatric Japanese Participants With NNS/CANDLE, SAVI, and AGS — NCT04517253.
• Registry record last updated 2025-07-10; status last verified 2025-06.
• Results first posted 2022-12-21.
• 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

NCT04517253 is a Phase 2/3, multicenter, open-label, nonrandomized, single-group study of baricitinib in Japanese adult and pediatric participants with NNS/CANDLE, SAVI, or AGS. Enrollment was 10. The study started on 2020-10-27, primary completion was 2023-04-04, and the recorded status is terminated.

The primary outcome is change from baseline in mean daily diary score among CANDLE participants during the primary treatment period, with a baseline and 20-week timeframe. The registry describes the symptom scale but reports no numerical efficacy analyses. It lists 25 outcome measures and zero analyses.

Adverse-event counts are reported by condition, including one death in the SAVI group. The reason for termination, efficacy estimates, analysis methods, missing-data handling, and multiplicity approach are not reported in the registry record.