Open Label Extension Trial of PDE4 Inhibition With Roflumilast for the Management of Atopic Dermatitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Open Label Extension Trial of PDE4 Inhibition With Roflumilast for the Management of Atopic Dermatitis (NCT04804605), a na phase 3 study. The primary outcome is Number of Participants With ≥1 Treatment-emergent Adverse Event (TEAE).

2. Clinical Question

The study was a Phase 3, multicenter, open-label extension for long-term safety in atopic dermatitis. Its stated purpose and primary endpoint concern safety rather than a randomized comparison of symptom or disease-control outcomes. The registry lists secondary measures involving investigator-assessed disease status, vIGA-AD success, itch scores in a specified age group, and EASI change. Those measures may describe clinical status over time, but the registry record does not report numerical results for them.

The extension population was drawn from people who completed or rolled over from parent studies, rather than from a newly randomized sample. This context matters: participants who enter an extension may differ from the wider population with atopic dermatitis, and their prior study treatment histories are reflected in the cohort labels. The record describes the population and treatment schedules, but it does not report enough information to determine how representative the enrolled participants were or how outcomes varied by prior treatment history.

3. Trial Design

This was a completed Phase 3, multicenter, open-label extension study of roflumilast cream for people with atopic dermatitis. The registry lists a planned enrollment of 1220 and dates the study from 2021-02-25 to primary completion on 2024-05-28. The intervention is ARQ-151 cream at 0.15% or 0.05%, applied once daily for up to 52 weeks. The record identifies atopic dermatitis eczema as the condition.

The design is single-group, with allocation listed as not applicable and masking listed as none. Consequently, the extension does not provide a concurrent randomized control group for estimating a treatment contrast. Its cohort labels refer to prior parent-study sequences: ARQ-151-311/312 participants who continued roflumilast or moved from vehicle to roflumilast, and ARQ-151-315 participants with corresponding sequence labels and cream strengths. These histories describe how participants arrived in the extension; they should not be read as randomized arms within this study.

The brief description specifies that participants with mild to moderate atopic dermatitis applied the cream once daily for up to 52 weeks. The 0.15% cohort includes completers of ARQ-151-311 or ARQ-151-312 aged at least 6 years and ARQ-151-315 rollovers who turned 6 years of age during the study. The 0.05% cohort includes ARQ-151-315 rollovers aged 2 to 5 years. The registry record does not report further eligibility details or a detailed account of how participants were assigned between strengths.

For interpretation, the central distinction is between monitoring events during continued treatment and comparing treatment strategies. The design supports descriptive reporting of events among treated participants in this extension. It does not, by itself, separate events associated with treatment from events that would have occurred without treatment, nor does it establish a causal difference between the cohort labels.

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 with atopic dermatitis were enrolled at 153 sites in the United States (US), Canada, and Poland.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All treated participants are included.

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: One participant discontinued from the VC/RC Group prior to dosing.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With ≥1 Treatment-emergent Adverse Event (TEAE), measured over Up to 52 weeks.

The primary endpoint is the number of participants with at least one treatment-emergent adverse event (TEAE) during up to 52 weeks. A participant is counted if they experienced one or more qualifying events during the stated period. This is a participant count, not a count of all event episodes; a person with multiple events may still contribute once to this endpoint. The registry describes the analysis population as all treated participants.

Safety studies commonly use an endpoint of this kind to describe how many treated people experienced an event after treatment began. It offers a straightforward summary of the proportion of the treated group affected when a denominator and complete outcome data are available. It does not, without a suitable comparator and supporting definitions, show whether treatment caused an event or how event risk compares with another treatment.

The primary endpoint title and definition are registered, but an overall numerical result for this endpoint is not reported in the registry record. The separate safety figures list serious and other event counts by prior-study sequence group, but they do not substitute for a reported overall primary-endpoint result. The registry also does not provide a detailed TEAE definition, event attribution rules, or a breakdown of the primary endpoint by severity and timing.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registered design and outcome types determine what statistical conclusions are reasonable. This is a single-group, open-label extension, so the main results are best read as descriptions of outcomes among participants who received study treatment. General statistical principles can clarify the scope of those descriptions, but the statistical analysis plan for this trial is not reported in the registry record.

What does a participant count measure?

A count of participants with at least one event gives the number of people who experienced the specified outcome, rather than the total number of event episodes. This distinction matters because one participant may experience several events. For the primary endpoint, the registry explicitly describes a count of participants with at least one TEAE and says all treated participants are included. The numerical result for that endpoint is not reported in the registry record.

The safety figures separately list the number at risk and the number affected for serious events and other events within each parent-study sequence group. Such figures are descriptive summaries. They can indicate how many people in each listed group were recorded as affected, but the counts alone do not account for differences in participant characteristics, follow-up, or prior treatment history. The record does not report a statistical comparison of these groups.

Why does the design limit comparisons?

Randomization helps make comparison groups similar, on average, with respect to factors that could influence outcomes. This extension has no randomized allocation and no masking. The group names reflect sequences from earlier studies, not a randomized assignment made within this extension. Comparing their event counts as if they were interchangeable trial arms could therefore confuse differences in participant history or cohort composition with treatment differences. The posted data support description of the groups, not a causal treatment contrast.

Open-label treatment means participants and study personnel were not masked to treatment assignment. In general, knowledge of treatment can influence reporting, assessment, and decisions about whether an event is recorded. The registry entry does not report whether independent adjudication or other procedures were used to reduce such influences. This is relevant when interpreting event recording and subjective assessments, but the size or direction of any influence cannot be determined from the record.

What does uncertainty add?

In general, a confidence interval describes the range of effect estimates that are reasonably compatible with the data and the statistical model used. It communicates precision as well as the point estimate. The registry record does not report confidence intervals for the primary safety endpoint or the group-level safety counts. Without interval estimates, the precision of those counts or any comparison cannot be assessed from the entry. No p-value for the primary endpoint is reported.

In general, inferential comparisons also depend on a defined estimand, a specified analysis population, and assumptions about the data. The registry identifies all treated participants as the primary endpoint population, but it does not report a detailed analysis plan, a model, or a method for handling incomplete follow-up. The absence of those details should not be filled in by assuming a conventional approach.

How should the secondary outcomes be viewed?

The registry lists five outcome measures in total and names secondary outcomes involving vIGA-AD status or success, WI-NRS change in participants at least 12 years of age in the parent study, and EASI percent change. These outcomes use different scales and populations. In general, a change score describes movement from a baseline value, whereas a status measure describes whether a participant meets a defined category at an assessment. Their interpretation depends on assessment timing, missing observations, and the analysis population. Numerical results and further details for these analyses are not reported in the registry record.

The record does not report a statistical analysis plan, model specifications, covariate adjustment, interim analyses, or confidence intervals for these outcomes. No particular method should therefore be attributed to the study. The stated endpoint definitions and reported safety counts are the available basis for statistical interpretation.

What do the planning assumptions mean?

General sample-size calculations ask how many participants would be needed to detect a specified effect under chosen assumptions. The registry describes a planning illustration based on an observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation; for non-time-to-event outcomes, the stated convention is participants per arm. Such planning values are not observed trial outcomes and are not observed power. The registry record does not report numerical planning results in the information, and the stated illustration is not a substitute for the trial's own design rationale. The trial's own sample-size justification is not reported in the registry record.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("ARQ-151-311/312: RC/RC 0.15% Group", "ARQ-151-311/312: VC/RC 0.15% Group", "ARQ-151-315: RC/RC 0.05% or 0.15% Group", "ARQ-151-315: VC/RC 0.05% or 0.15% Group")
n      <- c(439, 218, 382, 180)
events <- c(151, 90, 189, 91)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  ARQ-151-311/312: RC/RC 0.15% Group  439  151  34.4  30.0  39.0  30.1  39.0  30.1  39.0
ARM  ARQ-151-311/312: VC/RC 0.15% Group  218  90  41.3  34.7  48.1  35.0  47.9  35.0  47.9
ARM  ARQ-151-315: RC/RC 0.05% or 0.15% Group  382  189  49.5  44.4  54.6  44.5  54.5  44.5  54.5
ARM  ARQ-151-315: VC/RC 0.05% or 0.15% Group  180  91  50.6  43.0  58.1  43.3  57.8  43.3  57.8

11. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures, including the primary safety endpoint and secondary measures of investigator-assessed disease status, itch, and EASI change. It does not report a multiplicity strategy for interpreting multiple outcomes or repeated assessments. When several outcomes or time points are examined, general statistical practice distinguishes prespecified primary questions from additional analyses and considers the possibility of chance findings across repeated tests. The registry record does not state how that issue was handled in this trial.

Missing-data methods are not reported in the registry record. The primary endpoint population is described as all treated participants, but the entry does not explain how participants with incomplete follow-up were handled, whether event status was ascertained after discontinuation, or how missing secondary outcome measurements were treated. These details can affect interpretation, particularly in an extension lasting up to 52 weeks, where observation time may vary. No assumption about imputation, censoring, or another missing-data method is warranted.

The registry record also does not report a detailed statistical analysis plan, interim analyses, model specifications, or adjustments for baseline characteristics. It does not explain whether analyses were conducted by parent-study sequence, treatment strength, or another subgroup beyond the listed safety groups. The absence of those details limits assessment of how outcome definitions and analysis decisions were fixed before results were observed.

The entry states that all treated participants are included for the primary endpoint, but it does not give a complete account of analysis populations for the secondary measures. One secondary itch outcome is explicitly limited to participants at least 12 years of age in the parent study. Further population rules are not reported in the registry record. These omissions matter because the population contributing to each analysis determines what the estimate describes.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: vomiting (0.6%), gastroenteritis (0.6%), pneumonia (0.6%), eczema infected (0.6%), skin bacterial infection (0.6%).

The registry lists serious-event and other-event counts separately for four groups labelled by parent-study sequence and cream strength. For each group, the number at risk and number affected are stated. The serious-event counts are 5 of 439 in the ARQ-151-311/312 RC/RC 0.15% group, 3 of 218 in the ARQ-151-311/312 VC/RC 0.15% group, 12 of 382 in the ARQ-151-315 RC/RC 0.05% or 0.15% group, and 3 of 180 in the ARQ-151-315 VC/RC 0.05% or 0.15% group. The record lists no deaths in these groups.

For other events, the registry lists 18 of 439, 12 of 218, 82 of 382, and 40 of 180, respectively, in the same group order. These are descriptive counts, not randomized comparisons. The cohort labels reflect prior study sequences, and the record does not report a statistical comparison that would establish a difference between them. The counts also do not establish that treatment caused an event.

The most frequent serious-event terms listed are vomiting, gastroenteritis, pneumonia, eczema infected, skin bacterial infection, and viral upper respiratory tract infection, each reported as 0.6%. The registry record does not give further event-level context or a comparative rate for these terms. They should be interpreted as reported safety descriptions, not as proof of causation.

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

The registry record establishes that this was a completed, open-label extension study of roflumilast cream in people with atopic dermatitis, with treatment applied once daily for up to 52 weeks. It identifies a primary safety endpoint based on participants with at least one TEAE and provides serious-event and other-event counts for four groups defined by parent-study sequence. It also identifies several secondary outcomes intended to describe disease status, itch, and EASI change.

The record does not report a numerical result for the overall primary endpoint. The group-level safety counts therefore do not provide a complete numerical answer to the registered primary question. The entry also does not report numerical results for the listed secondary outcomes, so it cannot establish the magnitude or precision of changes in investigator-assessed signs, itch, or EASI.

Because the study is single-group, unmasked, and not randomized between the listed cohorts, the event counts should not be treated as evidence that one sequence or cream strength is safer than another. Differences in counts can reflect characteristics of the groups, prior study experience, follow-up, or other factors as well as any treatment-related differences. The registry does not report analyses that separate these explanations.

Likewise, the presence of an event during treatment does not alone establish that treatment caused it. The listed serious-event terms describe events recorded in the study; they do not establish attribution. The registry record does not provide a concurrent untreated or active-control group for estimating what would have occurred under another strategy.

These limits do not make the safety descriptions meaningless. They define their scope: the entry records a long-term extension and some event counts among treated participants, while leaving important results and analysis details unreported. It should not be used to infer comparative effectiveness, causal safety differences, or the outcomes of people outside the described extension population.

14. Why This Trial Matters Statistically

Extension studies address a different statistical question from short-term randomized trials. They can describe what is recorded during continued treatment over a longer period, including events that may be uncommon or occur after extended exposure. For this study, treatment continued once daily for up to 52 weeks, and the registered primary endpoint focuses on participants with at least one TEAE. That makes clear reporting of the population, follow-up, and event definitions especially important.

The record also illustrates why design labels matter. Prior-study sequence groups can look like parallel arms when set out side by side, but they are not a randomized comparison within this extension. A valid interpretation must distinguish group descriptions from treatment-effect estimates. Reporting counts alongside their group histories helps readers recognize that these cohorts may not be directly comparable.

Another statistical issue is completeness. The registry identifies multiple secondary measures, but numerical results for them are not reported in the registry record. It also does not report the analysis plan, missing-data methods, uncertainty intervals, or multiplicity approach. Those details are important for assessing not only the size of an estimate but also its precision and susceptibility to analytic choices.

The safety figures and serious-event terms offer descriptive information, while the missing primary endpoint result and analysis details constrain conclusions. This distinction is useful in clinical research: a registry can document a study's purpose, design, and selected results without furnishing everything needed to judge comparative effects or statistical uncertainty. Clear limits on inference protect against reading descriptive counts as stronger evidence than the design supports.

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 design has no randomized comparator and is open label. The groups in the safety listing are defined by parent-study treatment sequences, so they are not like-for-like control groups in this extension. Comparisons among their counts may be affected by differences in prior study treatment or cohort composition. The record does not report an analysis that addresses those differences.

The overall numerical result for the primary endpoint is not reported in the registry record, despite the endpoint being specified as the number of treated participants with at least one TEAE. The record lists some safety counts by group, but does not provide confidence intervals or statistical comparisons for those counts. It therefore does not show the precision of the reported safety summaries or support comparative inference.

The registry also does not report numerical results for the named secondary outcomes. Details of the statistical analysis plan, missing-data handling, interim analyses, multiplicity procedures, and the trial's sample-size justification are not reported in the registry record. The entry does not give detailed event definitions, follow-up completeness, or event-level context for the serious-event terms.

The extension population consisted of completers or rollovers from named parent studies, with age and treatment-strength descriptions that differ across cohorts. Results from such a selected population may not describe people with atopic dermatitis who did not enter or continue from those parent studies. The registry record does not provide information to assess how representative the extension participants were of a broader population.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Open Label Extension Trial of PDE4 Inhibition With Roflumilast for the Management of Atopic Dermatitis — NCT04804605.
• Registry record last updated 2025-12-23; status last verified 2025-12.
• Results first posted 2025-12-23.
• 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

NCT04804605 is a completed Phase 3, multicenter, open-label extension study of ARQ-151 cream, identified as roflumilast cream, in participants with atopic dermatitis. The registry lists an enrollment of 1220, a start date of 2021-02-25, and primary completion on 2024-05-28. Participants applied 0.15% or 0.05% cream once daily for up to 52 weeks, according to the described parent-study and age cohorts.

The primary endpoint counts treated participants with at least one TEAE. The registry record does not report its overall numerical result. It lists serious and other event counts by four parent-study sequence groups, and names six serious-event terms at 0.6% each. No deaths are listed in those groups. The record also identifies secondary outcomes for vIGA-AD, WI-NRS, and EASI, but numerical results for these outcomes are not reported in the registry record.

The single-group, unmasked design supports descriptive interpretation of the extension data, not randomized comparisons between the listed groups.