Assess the Long Term Efficacy and Safety of Ruxolitinib Cream in Participants With Vitiligo: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Assess the Long Term Efficacy and Safety of Ruxolitinib Cream in Participants With Vitiligo (NCT04530344), a randomized phase 3 study. The primary outcome is Time to Relapse (Defined as <F-VASI75).

2. Clinical Question

3. Trial Design

This completed phase 3 study used a randomized, double-blind, parallel-group design and a vehicle control. It enrolled 458 participants overall. Its treatment-extension period ran from Week 52 to Week 104, with Week 52 described as the first visit of the extension study and Week 104 as the end of treatment in that study. The registry lists a study start date of 2020-09-24 and a primary completion date of 2022-11-14.

The trial selected participants from two parent studies, NCT04052425 and NCT04057573. Participants had been using ruxolitinib cream twice daily for 28 to 52 weeks before entering this extension, depending on their initial randomization in the parent study. In Cohort A, participants who achieved at least 90% improvement in facial F-VASI at Week 52 were randomized 1:1 to ruxolitinib cream or vehicle cream. This is a randomized-withdrawal comparison among participants who had already reached a substantial response threshold, rather than a comparison in an unselected population starting treatment.

Participants who did not achieve the F-VASI90 threshold at Week 52 entered Cohort B and continued ruxolitinib cream. The primary outcome population is described as the randomized participants who met the F-VASI90 threshold. The displayed primary analysis includes 56 participants assigned to vehicle and 55 assigned to ruxolitinib. The registry reports no primary outcome values for Cohort B. The study also lists 25 outcome measures and two analyses, but the registry record does not report a full statistical analysis plan or the original sample-size justification.

The design has an important interpretive boundary: the randomized comparison concerns maintenance of response after an initial response, not initial response among all participants with vitiligo. The double-blind, vehicle-controlled allocation supports a comparison between the two Cohort A assignments, but the result should be understood within that selected group and the extension period described in the 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.

Before assignment: This randomized withdrawal and treatment-extension study (extension of treatment received in 2 parent studies: NCT04052425 or NCT04057573) was comprised of a 52-week extension treatment period and a 4-week safety follow-up period, starting 4 weeks (30 days) after the last application of study treatment or the last study visit.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat Long-term Extension (ITT-Ext) Population: randomized participants who achieved ≥90% improvement from Baseline in the F-VASI score (≥F-VASI90) at Week 52. Treatment groups were defined according to the treatment assignment at randomization regardless of the actual study medication the participant received. Week 52 was the first visit of this Treatment Extension study (visits were named to reflect continuation from the parent studies). Only participants in Cohort A were analyzed.

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 Relapse (Defined as <F-VASI75), measured over from Week 52 (first visit of this Treatment Extension study) to Week 104 (end of treatment in this Treatment Extension study).

The primary endpoint is time to relapse, with relapse defined as the loss of 75% improvement from baseline in facial F-VASI response. The registry further defines relapse as improvement in F-VASI from baseline falling below 75%. Baseline is Day 1 of the parent study. The time window begins at Week 52, the first extension-study visit, and ends at Week 104, the end of treatment in the extension study.

This endpoint asks not only whether relapse occurred, but also when it occurred during follow-up. In general, a time-to-event outcome preserves information about the timing of an event. It can also accommodate participants whose observed follow-up ends before an event, provided the analysis handles such censoring appropriately; the registry record does not describe the censoring rules in detail. The endpoint is clinically and statistically aligned with the study’s stated focus on duration and maintenance of response after withdrawal or continuation.

The registry reports a median-time outcome type for the primary measure, but does not provide an arm-specific median value in the posted primary results. Instead, the reported comparison is a hazard ratio from a log-rank analysis. The endpoint definition and the reported comparison therefore describe related but distinct summaries: the definition specifies the event and follow-up window, while the hazard ratio summarizes the relative event rate over the analysis period under the model’s comparison coding.

7. Results Posted in the Registry

• Cohort A: Vehicle Cream BID vs Cohort A: Ruxolitinib 1.5% Cream BID: hazard ratio 0.42 (95% CI 0.18–0.99), reported p = 0.041 (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.

Time to Loss of Adequate Response

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor (Log Rank); it was not recalculated.
• Result:
  • Cohort A: Vehicle Cream BID vs Cohort A: Ruxolitinib 1.5% Cream BID: hazard ratio 0.32 (95% CI 0.17–0.61), reported p < 0.001 (as reported in the registry).
• Interpretation:
  • The 95% confidence interval excludes no difference (a ratio of 1), so the hazard of the event is statistically lower for Cohort A: Vehicle Cream BID vs Cohort A: Ruxolitinib 1.5% Cream BID than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry lists a broad set of secondary outcomes, including time to loss of adequate response, thresholds of improvement in facial F-VASI, mean and change-from-baseline F-VASI measures, and a total-body VASI improvement threshold. It reports 24 secondary outcome measures. The posted results include a time-to-loss-of-adequate-response comparison with a hazard ratio below 1 and a reported p-value below 0.05. The registry does not provide enough numerical results in the record to characterize an overall pattern across the other secondary outcomes.

The time-to-loss-of-adequate-response analysis is conceptually related to the primary endpoint because both concern loss or maintenance of response during the extension period. A secondary finding in a similar direction may add context to the primary comparison, but it is not equivalent to an independent replication: the measures are assessed within the same study and responder population, and may be statistically related. The registry’s unclear hazard-ratio group coding also limits a definitive treatment-direction interpretation.

Multiple outcomes create an interpretive challenge. General explanation: when many outcomes are examined, the chance of at least one apparently notable result due to random variation can increase unless the analysis accounts for the multiplicity or clearly distinguishes confirmatory from exploratory outcomes. The registry record does not report whether multiplicity adjustment was used or how the secondary outcomes were prioritized. For that reason, unadjusted secondary results generally carry less confirmatory weight than a prespecified primary result, and should be read as supportive or exploratory rather than as a collection of separate definitive tests.

The outcome list documents what the study assessed, but does not by itself establish that each measure produced a favorable or statistically notable result. Where numerical estimates are not reported in the registry record, no direction, magnitude, or precision can be inferred. The secondary outcome set therefore provides a framework for understanding the study’s broader assessment of response, while the posted numerical evidence remains incomplete for judging consistency across that set.

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

• Time to Loss of Adequate Response
• Percentage of Participants Achieving a ≥50% Improvement From Baseline in the Face Vitiligo Area Scoring Index (F-VASI50) Score During the Extension Treatment Period
• Percentage of Participants Achieving a ≥75% Improvement From Baseline in the F-VASI (F-VASI75) Score During the Extension Treatment Period
• Percentage of Participants Achieving a ≥90% Improvement From Baseline in the F-VASI (F-VASI90) Score During the Extension Treatment Period
• Mean F-VASI Scores During the Extension Treatment Period
• Change From Baseline in F-VASI Scores During the Extension Treatment Period
• Percent Change From Baseline in F-VASI Scores During the Extension Treatment Period
• Percentage of Participants Achieving a ≥50% Improvement From Baseline in the Total Body Vitiligo Area Scoring Index (T-VASI50) Score During the Extension Treatment Period

9. Statistical Interpretation

10. Statistical Methods Explained

The primary comparison uses a log-rank method and reports a hazard ratio. These methods are commonly used for time-to-event outcomes, such as time from a defined starting point to relapse. General explanation: time-to-event methods use information about when events occur, rather than reducing follow-up to a simple count of participants with or without an event. They are most informative when the event definition, follow-up, and handling of participants without an observed event are clearly described.

What does a hazard ratio summarize?

In general, a hazard is the instantaneous rate at which an event occurs among participants who have not yet experienced it, at a particular point in follow-up. A hazard ratio compares those rates between groups over time. A value of 1 represents equal hazards under the comparison coding; a value below 1 represents a lower hazard for the group in the numerator relative to the reference group, if the group coding is known. The registry reports 0.422 for the vehicle-versus-ruxolitinib comparison, but does not explicitly state which group is the numerator or reference. The estimate should therefore be retained as reported without assigning its direction to a specific treatment arm.

A hazard ratio is not a ratio of median times and is not an absolute probability difference. It does not say how many additional participants remain relapse-free at a particular time. Interpreting it also depends on the analysis and follow-up structure. General explanation: if the relative hazard changes over time, a single hazard ratio may not capture every feature of the time pattern. The registry record does not report an assessment of that assumption, so no conclusion about it can be drawn from the posted results.

What does the log-rank method do?

In general, the log-rank test compares observed event patterns between groups across follow-up, taking account of the timing of events and participants who are still under observation without an event at each point. It is a standard approach for comparing time-to-event distributions. The registry identifies the log-rank method for the primary hazard-ratio analysis. It does not report further details about how the analysis handled ties, censoring, stratification, or other potential design features. Those details are not reported in the registry record.

The registry also labels the analysis as superiority. That label identifies the analysis type, but it does not replace the estimate and its uncertainty. A result judged against a statistical threshold may still be imprecise, and a statistical comparison does not by itself describe the magnitude of an absolute difference. The reported hazard ratio and confidence interval should be considered together with the study population and endpoint definition.

Why is the median relevant?

For a time-to-event outcome, the median event time is the point at which the estimated event distribution reaches 50%, if that point is observed during follow-up. General explanation: a median gives a time-scale summary that many readers can interpret directly, while a hazard ratio gives a relative comparison of event rates over follow-up. They answer different questions and should not be substituted for one another. Although the primary outcome is classified as a median-time measure, the arm-specific primary medians are not reported in the registry record. The primary result available for interpretation is the hazard ratio and its interval.

The secondary time-to-loss-of-adequate-response result does include a median for the vehicle group, with a confidence interval, while no median value is reported for the ruxolitinib group. Missing values should not be interpreted as a particular event pattern or as proof that a median was not reached; the registry does not explain why those values are absent. The secondary comparison also reports a hazard ratio, so the same distinctions between a relative rate summary and a median time apply.

What does the confidence interval add?

General explanation: a confidence interval expresses the statistical uncertainty around an estimate under the analysis assumptions. It is not a range containing a known fixed proportion of individual participant outcomes. The primary 95% interval extends from 0.180 to 0.990. Because the interval is close to 1 at its upper end, the estimate is compatible with contrasts ranging from more pronounced to quite modest on the hazard-ratio scale. The interval does not show the absolute risks or the number of relapses in each group.

General explanation: the width of an interval is influenced by the amount of information available, including the number and timing of events, not simply the total enrollment. In time-to-event analysis, participants may contribute differing amounts of follow-up, and participants without an observed event may contribute information until their follow-up ends. The registry does not report the primary event count or detailed follow-up distribution, so precision cannot be connected to those components from the record alone.

What is not specified?

The registry identifies a randomized, double-blind, parallel design, the primary population, the endpoint, and the log-rank method. It does not report the trial’s original sample-size justification, full statistical analysis plan, planned interim analyses, multiplicity adjustment, missing-data rules, detailed censoring rules, or the hazard-ratio group coding. These omissions limit the ability to reproduce or fully audit the primary analysis from the registry record. They should be treated as unreported details, not filled in by assumptions about common practice.

The primary population is described as randomized participants meeting the F-VASI90 criterion at Week 52, with groups defined by randomized assignment regardless of the actual study medication. This gives a stated basis for the analysis population, but the registry record does not supply a fuller estimand description or details about handling treatment discontinuation and other post-randomization events. General explanation: clear analysis-population and estimand definitions help readers understand exactly which treatment contrast is being estimated and which participant experiences are included.

11. Confidence Intervals and Precision

The primary hazard-ratio estimate is 0.422, with a reported 95% confidence interval from 0.180 to 0.990. The interval is entirely below 1, but its upper endpoint is close to 1. This supports a statistically distinguishable contrast under the reported analysis while also showing that the estimate’s precision is limited enough to include values near the null on the hazard-ratio scale. The reported p-value is below 0.05, consistent with the registry’s superiority label.

General explanation: a confidence interval describes the uncertainty around an estimated quantity under the statistical method and assumptions. It is not a guarantee that the true effect lies within the particular interval, nor does it describe the range of individual participants’ outcomes. Here, the interval concerns the relative hazard, not an absolute probability of relapse or a difference in median time. The registry does not report the primary arm-specific medians, event counts, or detailed censoring information needed to add those other perspectives.

The comparison involves 56 participants in the vehicle group and 55 in the ruxolitinib group in the primary outcome display. Total enrollment in the overall study was 458, but the primary analysis was restricted to randomized participants who met the response criterion at Week 52. Enrollment across the entire study should not be mistaken for the amount of information available for this particular comparison. In general, time-to-event precision depends substantially on observed events and follow-up as well as group size; those details are not reported for the primary result.

The interval should therefore be read as a range of plausible relative-effect estimates under the sponsor’s analysis, not as a measure of clinical importance. The registry does not report absolute relapse risks by group, so the confidence interval cannot be translated into an absolute benefit or harm from the posted information alone. The unreported group coding also constrains a treatment-specific directional reading.

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 planning figures for the primary time-to-event comparison are total event counts, not participants per arm. They are planning illustrations based on the observed effect, not the original trial’s design calculation. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; each planning scenario uses the stated fraction of the observed effect. The registry does not report the trial’s own original sample-size assumptions or justification.

At 100% of the observed effect, the planning illustration gives 43 total events for 80% power and 57 total events for 90% power. At 75% of the observed effect, it gives 75 total events for 80% power and 101 total events for 90% power. At 50% of the observed effect, it gives 169 total events for 80% power and 226 total events for 90% power. These are total events across the two groups, not the number of events required in each group.

In plain language, 80% power means that, under the stated assumptions and if the assumed effect were true, a study with the planned information would have an 80% chance of detecting an effect at the chosen significance threshold. The 90% power column describes a higher chance of detection under those same assumptions. A smaller effect is harder to distinguish from random variation, so the planning illustration requires more events and, in practice, may require more participants or longer follow-up to observe them. The event totals are not equivalent to enrollment totals, because not every participant necessarily experiences the event during observation.

These figures are planning scenarios, not observed power. They do not indicate the probability that this completed trial’s conclusion is correct, and they do not show that the trial actually accrued the listed event counts. The registry record does not report the primary event count needed to compare the study’s observed information with these illustrations. The study’s own prospective assumptions remain unreported.

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 = 0.422, so ln(HR) = -0.8627.
2. d = 4 × (1.96 + 0.842)² / [ln(0.422)]² = 42.2, rounded up to 43 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 <- 0.422                              # 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(4530344)
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 = 43 (80% power), 57 (90% power)
 75% of observed effect: events = 75 (80% power), 101 (90% power)
 50% of observed effect: events = 169 (80% power), 226 (90% power)
Simulated power with 43 events: 0.818 (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 registry reports the sponsor’s log-rank analysis as a hazard ratio of 0.422, with a 95% confidence interval of 0.180 to 0.990 and a reported p-value below 0.05. The comparison is labeled vehicle cream versus ruxolitinib 1.5% cream, and the registry labels the analysis as superiority. The same reported estimate and interval appear in the posted analysis summary; there is no discrepancy flagged between the listed comparison and the reported analysis.

No separate recalculated estimate is needed to replace that sponsor result in the registry record. The sponsor’s reported analysis is the primary source for the posted inferential result, and the comparison details are consistent across the record. The record does not specify the hazard-ratio numerator and reference group, however, so the treatment direction should not be inferred solely from the order in which groups are named.

The sponsor’s result remains an analysis of the selected Week 52 responders and the stated time-to-relapse endpoint. The statistical estimate does not supply unreported arm-specific medians, event counts, absolute risks, or details of missing-data and censoring procedures. Those aspects cannot be recovered from the reported hazard ratio and should not be treated as if they were available.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists 25 outcome measures and 24 posted secondary outcome measures, alongside two analyses. It identifies a primary endpoint and reports secondary outcomes spanning time to loss of response and several facial or total-body VASI measures. The multiplicity strategy is not reported in the registry record. In particular, the record does not say whether the study controlled the family-wise error rate, used another adjustment, applied a hierarchical testing strategy, or treated secondary outcomes as exploratory.

General explanation: testing multiple endpoints can increase the chance of obtaining at least one small p-value by chance alone. A prespecified hierarchy or multiplicity adjustment can help control how strongly a family of results supports confirmatory conclusions. Without a reported strategy, the secondary p-values should not be read as if each came from a single isolated test. The primary result can be interpreted as the registry’s designated primary analysis, but its place in a complete testing plan cannot be independently assessed from the record.

The primary population is defined as randomized participants who achieved at least 90% improvement from baseline in F-VASI at Week 52. The registry says treatment groups were defined according to assignment at randomization regardless of the actual study medication, but the available population description is incomplete. A fuller estimand, including how treatment discontinuation, rescue treatment, or other post-randomization events were handled, is not reported in the registry record. No assumptions about those procedures should be made.

The registry does not report the missing-data approach, censoring rules for the time-to-event analysis, or sensitivity analyses. General explanation: these decisions matter because participants may leave follow-up or finish observation without a recorded relapse, and the method used to represent their information can affect the estimate and its uncertainty. The record also does not report whether interim analyses were planned or performed. Their presence or absence should not be inferred.

These reporting gaps do not show that the analyses were conducted improperly; they limit what can be evaluated from the registry record. They mean that readers cannot fully assess the multiplicity control, robustness to missing data, or detailed analysis population beyond the definitions that are explicitly posted.

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: acute respiratory failure (0.2%), angina pectoris (0.2%), bipolar i disorder (0.2%), covid-19 pneumonia (0.2%), cholelithiasis (0.2%).

The registry reports serious adverse events for 0 of 58 participants in the vehicle group and 12 of 423 participants in the ruxolitinib group. The total is reported as 12 of 458 participants. Other adverse events are reported for 6 of 58 in the vehicle group, 55 of 423 in the ruxolitinib group, and 61 of 458 overall. Deaths are reported as 0 in both groups and 0 overall.

These are counts and denominators from the registry record, not adjusted comparisons. The safety denominators differ from the primary outcome analysis groups, which included 56 and 55 participants. The record does not explain the difference between safety and efficacy populations, and the group denominators do not add to the reported overall denominator. The counts should therefore be read as posted rather than reconciled through assumptions.

The registry lists Acute respiratory failure, Angina pectoris, Bipolar I disorder, COVID-19 pneumonia, Cholelithiasis, and Cystocele among the serious adverse event terms, each reported as 0.2%. The record does not provide enough context in this summary to assess timing, duration, relatedness, or comparative risk for these terms. These figures describe reported safety events; they do not establish whether an event was caused by study treatment.

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

The registry’s primary result is a reported hazard-ratio comparison for time to relapse among participants who had achieved at least 90% improvement in facial F-VASI by Week 52. The estimated hazard ratio is 0.422, with a 95% confidence interval of 0.180 to 0.990. The interval’s upper endpoint is near 1, and the reported p-value is below 0.05. These statistics describe the sponsor’s comparison on a relative time-to-event scale.

The result does not provide a direct estimate of absolute relapse probability, the number of participants who relapsed in each group, or the difference between group medians. Primary arm-specific median times are not reported in the registry record. The hazard ratio also cannot be assigned confidently to a particular treatment direction from the group order alone, because the registry does not clearly state the numerator and reference coding.

The population matters. This was a randomized withdrawal comparison among participants who had already reached a substantial response threshold after treatment in parent studies. The result does not answer how often participants starting treatment achieve that response, and it should not be generalized automatically to participants who did not meet the threshold or to all people with vitiligo. Cohort B participants continued ruxolitinib and are not part of the posted primary comparison.

The statistical result is not a medical recommendation and does not establish an individual participant’s likely outcome. It does not answer questions about treatment choice, duration beyond the extension period, or outcomes not reported in the registry. The study assesses a defined response-maintenance question over the Week 52 to Week 104 extension window; conclusions beyond that scope are not supported by this comparison alone.

Finally, statistical significance is not the same as certainty or clinical importance. The confidence interval conveys uncertainty, and important analysis details—including the original sample-size justification, multiplicity handling, missing-data procedures, censoring rules, and hazard-ratio coding—are not reported in the registry record. The posted result is informative within its stated population and endpoint, but those limitations should remain part of its interpretation.

18. Why This Trial Matters Statistically

This trial is statistically distinctive because it uses a randomized-withdrawal design in participants who had already reached a defined response threshold. That design asks a maintenance question: how the time-to-relapse outcome compares after assignment to vehicle withdrawal or continued active treatment among selected responders. It is not the same estimand as a trial of initial treatment in an unselected population. Recognizing that distinction is essential before interpreting the reported hazard ratio.

The time-to-event endpoint also makes the analysis more than a comparison of simple outcome proportions. The log-rank method and hazard ratio use the timing of relapse across follow-up, while the median time offers a different possible summary. In this record, the primary median values are not reported, so the relative hazard is the main posted comparative statistic. The confidence interval helps convey how much uncertainty remains in that estimate.

The trial illustrates why registry reporting details matter. A hazard ratio requires clear group coding for a treatment-specific directional interpretation. Evaluation of a time-to-event result also benefits from event counts, censoring rules, and follow-up details. Multiplicity and missing-data procedures affect how confidently secondary findings can be interpreted. Several of these details are not reported in the registry record, making a careful, bounded interpretation more appropriate than a broad claim.

Finally, the planning illustrations distinguish event requirements from participant enrollment. For time-to-event designs, the quantity driving information is often the number of events rather than total enrollment alone. The planning figures are hypothetical and based on the observed effect; they do not document the study’s original design rationale or its realized information. Keeping that distinction clear helps prevent planning calculations from being mistaken for evidence about the completed trial’s power.

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 primary comparison is restricted to Cohort A participants who achieved at least 90% improvement in facial F-VASI by Week 52. This selection is central to the randomized-withdrawal question, but it limits the result to responders who reached that threshold after treatment in parent studies. It does not describe the experience of all enrolled participants or those who did not meet the response criterion. Cohort B continued ruxolitinib and has no reported primary comparison in the registry record.

The posted primary result includes a hazard ratio and confidence interval, but no arm-specific primary median, event count, or absolute relapse estimate. The hazard-ratio group coding is also not explicit. These omissions restrict interpretation of the result’s treatment direction and absolute magnitude. Details about censoring, missing data, treatment discontinuation, and sensitivity analyses are not reported in the registry record.

The registry lists a large set of outcome measures, but the multiplicity approach is not reported. Consequently, the confirmatory status of the secondary findings cannot be fully assessed. Numerical findings for many listed secondary outcomes are not available in the record, so an overall consistency pattern cannot be established from the outcome list alone.

The safety denominators differ from the primary efficacy group sizes, and the registry does not explain those populations or reconcile the group and total denominators. The reported safety counts do not include enough information to evaluate event timing, duration, relatedness, or adjusted comparisons. The record also does not report the original sample-size justification, planned interim analyses, or a complete statistical analysis plan. These limitations concern the scope and interpretability of the posted information; they do not by themselves establish that the study procedures were inadequate.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Assess the Long Term Efficacy and Safety of Ruxolitinib Cream in Participants With Vitiligo — NCT04530344.
• Registry record last updated 2025-08-14; status last verified 2025-08.
• Results first posted 2023-07-19.
• 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.

23. Record Summary

NCT04530344 is a completed phase 3, randomized, double-blind, vehicle-controlled treatment-extension study in vitiligo. It enrolled 458 participants and examined response maintenance after treatment in two parent studies. In Cohort A, participants who achieved at least 90% improvement in facial F-VASI at Week 52 were randomized 1:1 to vehicle cream or continued ruxolitinib 1.5% cream twice daily. Cohort B participants who did not reach that threshold continued ruxolitinib.

The primary endpoint was time to relapse, defined as facial F-VASI improvement falling below 75% from baseline during the Week 52 to Week 104 extension period. The registry reports a log-rank hazard ratio of 0.422 with a 95% confidence interval of 0.180 to 0.990 and labels the analysis as superiority. Primary arm-specific median times and several analysis details are not reported in the registry record. The result applies to the defined responder population and should be interpreted with the stated endpoint, uncertainty, and reporting limitations in mind.