Study of Dupilumab Administered to Adult Patients With Moderate-to-Severe Atopic Dermatitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Dupilumab Administered to Adult Patients With Moderate-to-Severe Atopic Dermatitis (NCT01859988), a randomized phase 2 study. The primary outcome is Percent Change in Eczema Area and Severity Index Score (EASI) From Baseline to Week 16.

2. Clinical Question

3. Trial Design

This completed Phase 2 study evaluated multiple dupilumab dose regimens against placebo in adults with moderate-to-severe atopic dermatitis. The registry lists randomized allocation, a parallel-group model, quadruple masking, and an enrollment of 380. Regeneron Pharmaceuticals is listed as the sponsor. The recorded start was 2013-05 and primary completion was 2014-05.

There were five dupilumab schedules: 300 mg weekly; 300 mg every 2 weeks; 200 mg every 2 weeks; 300 mg every 4 weeks; and 100 mg every 4 weeks. Each schedule included a loading dose on Day 1, followed by injections through Week 15. Placebo injections were used on weeks when dupilumab was not scheduled for the less frequent regimens. The comparator group received placebo injections on the weekly schedule.

The primary endpoint was percent change in EASI score from baseline to Week 16. The registry also lists outcomes concerning investigator global assessment, pruritus scores, EASI changes and response thresholds, and SCORAD. The record lists 14 outcome measures and 5 analyses. The posted primary comparisons are sponsor ANCOVA analyses of each dupilumab regimen against placebo.

The registry record does not report the original sample-size justification, the intended statistical analysis plan in full, or whether interim analyses were planned or conducted. The design features establish how the groups were organized and masked, but they do not by themselves establish the allocation concealment details or the handling of all protocol deviations.

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: Randomization was stratified by disease severity (moderate Investigator's global assessment \[IGA\] = 3 versus severe IGA = 4 atopic dermatitis) and region (Japan versus rest of world). Assignment to arms was done centrally in 1:1:1:1:1:1 ratio for Dupilumab (300 mg qw; 300 mg q2w; 200 mg q2w; 300 mg q4w and 100 mg q4w) and Placebo.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set (FAS) that included all randomized participants who received at least 1 dose of study drug. Here, number of participants analyzed = participants with EASI score assessment at specified time-point. Efficacy data was set to missing after use of rescue medication. Missing values imputed by last observation carried forward (LOCF).

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: Baseline population included safety analysis set (SAF) that included all randomized participants who received any investigational product and analyzed as treated. One participant randomized into Dupilumab 200 mg q2w arm was not treated and hence, excluded from the baseline population.

Only the first 6 of 7 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change in Eczema Area and Severity Index Score (EASI) From Baseline to Week 16, measured over Baseline to Week 16.

The primary endpoint is percent change in EASI score from baseline to Week 16. EASI assesses the extent and severity of atopic dermatitis by measuring erythema, infiltration, excoriation, and lichenification across the head, trunk, upper extremities, and lower extremities. Its score ranges from 0 to 72, with higher scores reflecting worse severity. A negative percent change represents a reduction from baseline.

EASI is useful as a continuous endpoint because it captures change across the scale rather than reducing participants to a responder category. The group mean describes the average change, while the spread describes how much individual outcomes vary around that average. Neither alone conveys the full pattern of individual responses, but together they help characterize the magnitude and variability of change.

The primary outcome is assessed over the period from baseline to Week 16. The registry says the analysis population included randomized participants who received at least one dose of study drug, and that the number analyzed at each time point was based on participants with an EASI assessment at that time. Efficacy data were set to missing after rescue medication use. The remaining detail in the registry's missing-value description is incomplete, so the specific imputation approach is not reported in the registry record.

7. Results Posted in the Registry

• Dupilumab 300 mg qw: mean -75.50 versus -20.20 on control; difference -55.30 (95% CI -68.70 to -41.90), standardized effect d = -1.46; z-test p = <0.001.
• Dupilumab 300 mg q2w: mean -70.50 versus -20.20 on control; difference -50.30 (95% CI -64.76 to -35.84), standardized effect d = -1.23; z-test p = <0.001.
• Dupilumab 200 mg q2w: mean -67.40 versus -20.20 on control; difference -47.20 (95% CI -61.33 to -33.07), standardized effect d = -1.19; z-test p = <0.001.
• Dupilumab 300 mg q4w: mean -64.90 versus -20.20 on control; difference -44.70 (95% CI -59.40 to -30.00), standardized effect d = -1.07; z-test p = <0.001.

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.

Percentage of Participants Who Achieved Investigator's Global Assessment (IGA) Response at Week 16

• 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:
  • Dupilumab 300 mg qw: 33.3% (21/63) versus 1.6% (1/61) on control; risk difference 31.7 percentage points (95% CI 19.6 to 43.8), risk ratio 20.33 (2.82–146.53); unadjusted two-proportion z-test p = <0.001.
  • Dupilumab 300 mg q2w: 29.7% (19/64) versus 1.6% (1/61) on control; risk difference 28.0 percentage points (95% CI 16.4 to 39.7), risk ratio 18.11 (2.50–131.17); 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 Dupilumab 300 mg qw than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Dupilumab 300 mg q2w than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Who Achieved IGA Score Reduction of ≥2 at Week 16

• 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:
  • Dupilumab 300 mg qw: 50.8% (32/63) versus 9.8% (6/61) on control; risk difference 41.0 percentage points (95% CI 26.5 to 55.4), risk ratio 5.16 (2.33–11.47); unadjusted two-proportion z-test p = <0.001.
  • Dupilumab 300 mg q2w: 46.9% (30/64) versus 9.8% (6/61) on control; risk difference 37.0 percentage points (95% CI 22.7 to 51.4), risk ratio 4.77 (2.13–10.64); 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 Dupilumab 300 mg qw than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Dupilumab 300 mg q2w than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percent Change in Peak Weekly Averaged Pruritus Numerical Rating Scores (NRS) From Baseline to Week 16

• 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 deviations.
• Result:
  • Dupilumab 300 mg qw: mean -52.85 versus -0.43 on control; difference -52.42 (95% CI -65.02 to -39.82), standardized effect d = -1.50; z-test p = <0.001.
  • Dupilumab 300 mg q2w: mean -46.22 versus -0.43 on control; difference -45.79 (95% CI -58.44 to -33.14), standardized effect d = -1.30; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Dupilumab 300 mg qw than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Dupilumab 300 mg q2w than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes cover several related aspects of atopic dermatitis: global assessment response, changes in itch scores, changes in EASI and SCORAD, and EASI response thresholds. The posted results generally follow the primary endpoint's broad pattern, with more favorable average changes or response proportions in the dupilumab groups than in placebo. The available secondary summaries do not all compare every dose regimen, and their outcome scales and analysis populations differ, so they should not be treated as interchangeable replications of the primary endpoint.

These measures add context. Continuous scores describe average change, while responder outcomes describe the share of participants meeting a defined threshold. A consistent direction across related measures can make the overall pattern easier to interpret, but it does not eliminate uncertainty about the individual estimates or establish that one regimen is better than another. The registry record does not report sponsor analyses for the posted secondary results.

Unadjusted secondary results carry less inferential weight when many outcomes and comparisons are examined. In general, testing many hypotheses increases the chance that some results will look notable by chance unless the analysis plan addresses multiplicity. The registry record does not report whether or how such adjustment was made for these outcomes. Secondary findings are therefore best read as supportive descriptions of the outcome pattern, not as a separate set of definitive confirmatory conclusions.

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

• Percentage of Participants Who Achieved Investigator's Global Assessment (IGA) Response at Week 16
• Percentage of Participants Who Achieved IGA Score Reduction of ≥2 at Week 16
• Percent Change in Peak Weekly Averaged Pruritus Numerical Rating Scores (NRS) From Baseline to Week 16
• Absolute Change in Peak Weekly Averaged Pruritus NRS From Baseline to Week 16
• Absolute Change in EASI Score From Baseline to Week 16
• Percent Change in SCORing Atopic Dermatitis (SCORAD) Scores From Baseline to Week 16
• Absolute Change in SCORAD Scores From Baseline to Week 16
• Percentage of Participants Who Achieved 50%, 75% and 90% Reduction From Baseline in EASI Score (EASI-50, EASI-75 and EASI-90 Respectively) at Week 16

9. Statistical Interpretation

10. Statistical Methods Explained

Several statistical summaries describe different aspects of the primary outcome. The arm mean is the average percent change in EASI score among participants analyzed in that group. A negative value indicates a reduction from baseline. The standard deviation describes the spread of participant outcomes around the group mean; it is not the uncertainty in the estimated mean itself. Larger standard deviations indicate more variability among observed outcomes.

What does the ANCOVA comparison represent?

The sponsor reports analysis of covariance (ANCOVA) comparisons as least-squares mean differences between each dupilumab regimen and placebo. In general, ANCOVA estimates a group contrast while accounting for covariates included in the model. The registry record does not identify the covariates or provide the complete model specification, so the adjustment cannot be reconstructed from the reported summary alone. The reported LS mean differences and their confidence intervals are the sponsor's analyses and should be distinguished from the unadjusted recalculations based on arm summaries.

The registry's primary outcome parameter is a least-squares mean, and the reported sponsor comparisons are LS mean differences. The posted arm-level results also include means and standard deviations. These summaries are related but are not necessarily identical: adjusted model estimates may differ from crude arm means because they are based on the specified analysis model.

What do the recalculated comparisons add?

The simple recalculations compare the posted mean changes in each dupilumab group with the posted placebo mean. They are unadjusted comparisons derived from summary statistics, rather than participant-level data or the sponsor's ANCOVA model. Their role is to provide a transparent descriptive cross-check, not to replace the trial analysis. The recalculated estimates are presented with confidence intervals and unadjusted p-values. Because the primary endpoint recalculation uses an approximate pooled standard deviation, it should be read as an approximation based on the available summaries.

A confidence interval describes a range of values compatible with the estimate under the model and assumptions used. Its width reflects statistical precision: a narrower interval indicates less uncertainty about the estimated contrast than a wider interval. It does not show the range of individual participant responses, and it is not a guarantee that the true value lies in that particular interval.

Why can over-dispersion matter?

In general, over-dispersion means that outcome variability is greater than a model assumes. If that extra variability is ignored, standard errors can be too small and confidence intervals too narrow, making results appear more precise than they are. The primary endpoint is continuous and the posted standard deviations show variation, but the registry record does not report a specific over-dispersion assessment. It would therefore be inappropriate to assert that over-dispersion was present or that it was handled in a particular way in the sponsor's model.

For binary secondary outcomes, proportions and their intervals summarize responder frequency. A risk difference is the difference between group proportions, while relative measures compare them as a ratio. These measures answer different questions and can be unstable when an outcome is uncommon. No sponsor analyses are listed for the secondary results described in the record, so any simple comparisons or planning illustrations for them should not be mistaken for adjusted trial analyses.

How should p-values be read?

A p-value measures how incompatible the observed data are with a specified null hypothesis under the analysis assumptions. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. The sponsor reports p-values for the primary comparisons, while the recalculations give unadjusted p-values. The registry record does not report the full testing hierarchy or multiplicity procedure, so the p-values do not resolve questions about the overall error rate across outcomes and dose comparisons.

The analysis population is described as a full analysis set including randomized participants who received at least one dose. At a given time point, the registry counts those with an EASI assessment. Efficacy data were set to missing after rescue medication use. The rest of the registry's missing-data statement is incomplete, and the method used to handle other missing values is not reported in the registry record. These details matter because analysis populations and missing-data methods affect the interpretation of model estimates and their uncertainty.

11. Confidence Intervals and Precision

The sponsor's primary ANCOVA estimates are accompanied by 95% confidence intervals. For dupilumab 300 mg weekly, the reported LS mean difference is -55.7, with a 95% confidence interval from -68.9 to -42.4. For 300 mg every 2 weeks, it is -50.1, with an interval from -63.3 to -37.0. For 200 mg every 2 weeks, it is -47.4, with an interval from -60.6 to -34.1. For 300 mg every 4 weeks, it is -45.4, with an interval from -58.5 to -32.3. For 100 mg every 4 weeks, it is -26.8, with an interval from -39.8 to -13.7.

In general, a confidence interval gives a range of estimates reasonably compatible with the observed data under the model and assumptions used. The intervals for these sponsor comparisons are entirely negative, so they indicate that the estimated group contrasts favor a more negative EASI percent change in each dupilumab group than in placebo within the fitted analyses. Interval width still matters: it conveys uncertainty about the estimated contrast, not the range of likely participant-level changes.

The recalculated unadjusted intervals are broadly similar in direction to the reported sponsor intervals, but they are not identical analyses. Recalculation uses posted summary statistics and an approximate pooled standard deviation; the sponsor used ANCOVA. The registry record does not give enough model detail to reproduce the ANCOVA or explain every difference between the estimates. Precision should therefore be judged primarily from the sponsor's reported intervals for the trial's primary comparisons, with the recalculated estimates serving as a separate summary-based check.

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 planning figures are illustrations based on the observed effect, not the trial's original design justification. For this continuous endpoint, the figures mean participants per arm; the time-to-event convention of total events does not apply. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each scenario uses the stated fraction of the observed effect as the effect to detect.

The 80% power column gives the estimated participants per arm for a hypothetical study to have an 80% chance of detecting the specified effect under those assumptions, if that effect is present. The 90% power column gives the corresponding planning figure for a 90% chance. These are design calculations, not observed outcomes. They are not observed power.

Rows that plan for 100% of the observed effect use the full observed contrast as the target. Rows planning for 75% or 50% use smaller target effects. Smaller effects are harder to distinguish from random variation, so they require more participants per arm under the stated assumptions. These values should not be read as recommended trial sizes: they rely on effects observed in this study, and future results may differ. The original trial's sample-size assumptions and justification are not reported in the registry record. A curve based on these calculations should be interpreted in the same way: it illustrates the relationship between the assumed effect and the required planning sample, rather than describing the actual power of this completed study.

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

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -55.300; pooled standard deviation σ = 37.758.
2. n = 2 × (1.96 + 0.842)² × 37.758² / -55.300² = 7.3, rounded up to 8 participants per arm 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
delta <- -55.3                      # observed difference in means
sigma <- 37.7576621363135                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(1859988)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 8 (80% power), 10 (90% power)
 75% of observed effect: n per arm = 14 (80% power), 18 (90% power)
 50% of observed effect: n per arm = 30 (80% power), 40 (90% power)
power.t.test: 8.4 
Simulated power at n = 8 per arm: 0.778 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The sponsor reports ANCOVA comparisons of each dupilumab regimen against placebo, expressed as LS mean differences with 95% confidence intervals and p-values. The simple recalculations use the posted arm means and standard deviations to estimate unadjusted mean differences. They do not reproduce the sponsor's adjusted model, which cannot be fully reconstructed because the registry record does not report its covariates or complete specification.

The two approaches tell a consistent directional story: all sponsor estimates and recalculated estimates are negative, and both sets describe greater average EASI reduction in each dupilumab group than in placebo. The numerical estimates are close but not identical. That is expected when an adjusted model estimate is compared with a summary-statistic calculation that does not adjust for covariates. The recalculations also use an approximate pooled standard deviation for the primary endpoint.

The registry lists no discrepancy flag. The sponsor's ANCOVA results remain the trial's reported analyses; the recalculations are supplemental unadjusted checks and should not be treated as a correction or replacement. The registry record does not report enough information to determine precisely why individual estimates differ.

15. Multiplicity, Missing Data and Analysis Populations

The trial has multiple dose regimens and multiple outcome measures. The primary results include five comparisons of dupilumab regimens with placebo, while the registry lists secondary measures concerning global assessment, pruritus, EASI, and SCORAD. In general, testing multiple hypotheses raises the chance of at least one false-positive result if each test is interpreted independently. A prespecified testing hierarchy or multiplicity adjustment can address this issue, but the registry record does not report whether either was used for the primary or secondary analyses.

The sponsor reports p-values and confidence intervals for the primary ANCOVA comparisons. These are the reported analyses, but the record does not explain whether the p-values were adjusted across dose comparisons, identify a gatekeeping approach, or describe how secondary endpoints were handled statistically. The unadjusted recalculated p-values do not account for multiple testing. The record therefore supports describing the reported estimates and intervals, but not claiming a particular family-wise error rate or confirmatory status for the full set of outcomes.

The primary analysis population is described as the full analysis set, including randomized participants who received at least one dose of study drug. The number analyzed at each time point is the number with an EASI assessment at that time. Efficacy data were set to missing after rescue medication use. The registry's statement about what happened to other missing values is truncated, and the specific imputation method is not reported in the registry record. The record also does not report sensitivity analyses for missing data or a complete account of how missing assessments affected the ANCOVA.

These gaps matter because the analysis population, rescue-medication rule, missing-data assumptions, and multiplicity approach all influence how confidently the reported estimates can be generalized beyond the observed data. Their absence from the registry record does not establish that the trial lacked such procedures; it means their details cannot be assessed from this record.

16. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

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: dermatitis atopic (6.2%), tachycardia (1.6%), hip dysplasia (1.6%), anaphylactic shock (1.6%), osteonecrosis (1.6%).

The registry reports serious adverse event counts by treatment group, with the number at risk shown for each group. The serious adverse event counts are 1, 2, 1, 3, and 5 across the five dupilumab regimens, and 4 in the placebo group. These are event counts as reported; the registry summary does not provide enough detail to determine the nature, timing, relatedness, or overlap of events within participants.

The record lists serious adverse event terms including atopic dermatitis, tachycardia, hip dysplasia, anaphylactic shock, osteonecrosis, and respiratory failure. The listed percentages are not clearly mapped to individual treatment groups in the summary, so they should not be used to infer a group-specific rate. Death counts are not reported in the registry record.

Serious adverse events are distinct from the broader category of other adverse events. The counts do not establish causation, and differences in counts should be interpreted cautiously in this trial context. The registry record does not provide a complete narrative safety analysis or enough information to assess comparative risk beyond the listed outcomes.

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

The results describe average changes and reported comparisons over the trial's baseline-to-Week 16 period. In the primary endpoint, the posted group means show reductions in EASI score in both the dupilumab groups and the placebo group. Sponsor ANCOVA estimates indicate more negative mean changes for each dupilumab regimen than for placebo in the reported comparisons. These statements describe the registry results; they do not determine what will happen for an individual.

The findings do not establish that every participant experienced improvement, because group means summarize outcomes and the standard deviations show variation. Nor do the primary results establish the relative merits of the five dupilumab schedules: the sponsor's listed comparisons are each against placebo, not direct comparisons between dupilumab regimens. The numerical ordering of group means should not be treated as a tested ranking.

The record also does not establish durability beyond Week 16, outcomes beyond those reported, or the balance of benefits and harms over longer follow-up. Safety counts are descriptive and do not establish that an event was caused by a treatment. The absence of complete details about multiplicity, missing-data methods, and the ANCOVA specification limits how fully the analysis can be evaluated.

In general, trial estimates are evidence about average outcomes under the study's conditions, not guarantees or individualized predictions. The record supports describing the direction and precision of the posted comparisons while keeping their population, time point, analysis assumptions, and safety reporting in view. It does not by itself provide a basis for medical decisions.

18. Why This Trial Matters Statistically

Statistically, this trial is informative because it evaluates several dose schedules within a randomized, masked, placebo-controlled parallel-group design and reports a continuous primary outcome. Comparing mean changes preserves more information than converting every outcome to a responder/non-responder label. The sponsor's ANCOVA estimates also illustrate how an adjusted trial analysis can differ from an unadjusted comparison reconstructed from group summaries.

The results show why effect size and precision should be read together. The reported estimates describe the size and direction of group contrasts, while confidence intervals show uncertainty around those estimates. The posted standard deviations add a separate view of variation among participant outcomes. None of these summaries alone captures every aspect of the outcome distribution.

The record also highlights an important interpretive issue in dose-ranging studies: several regimens and endpoints create a multiple-testing context. Without the full multiplicity strategy and statistical analysis plan, the reported p-values cannot be interpreted as if they were the only tests performed. Likewise, incomplete information about missing data limits independent assessment of how post-baseline assessments entered the analyses.

Finally, planning calculations based on observed effects are sensitive to the effect assumed for a future study. Showing scenarios for smaller fractions of the observed effect makes clear why sample-size requirements change as the target effect changes. Those illustrations are useful for understanding design trade-offs, but they do not recover the trial's original planning rationale or its observed 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 registry record does not report the complete statistical analysis plan, including the ANCOVA covariates, detailed model specification, multiplicity strategy, or original sample-size justification. The sponsor's adjusted analyses can be quoted, but they cannot be fully reproduced from the information recorded. The simple recalculations are unadjusted and use summary statistics, with an approximate pooled standard deviation for the primary endpoint.

The missing-data description is incomplete. The record states that efficacy data were set to missing after rescue medication use, but the following text is truncated. The specific approach to other missing values, sensitivity analyses, and the influence of missing assessments are not reported in the registry record. The full analysis set also excludes randomized participants who did not receive at least one dose, according to the stated definition.

The study's primary endpoint is assessed at Week 16, which limits what can be said about persistence beyond that time point. The posted sponsor comparisons are against placebo; they do not directly test differences among the dupilumab schedules. The record lists many secondary outcomes but does not report sponsor analyses for the secondary results, and it does not establish a multiplicity adjustment for them.

Safety reporting is summarized as counts and selected serious-event terms. The record does not provide a complete assessment of event timing, relatedness, or treatment-group mapping for the listed terms. Death counts are not reported in the registry record. These limitations concern what can be assessed from the registry record and do not establish that additional analyses or documentation do not exist elsewhere.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of Dupilumab Administered to Adult Patients With Moderate-to-Severe Atopic Dermatitis — NCT01859988.
• Registry record last updated 2017-08-28; status last verified 2017-07.
• Results first posted 2017-08-28.
• 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

This completed Phase 2 randomized, double-blind, placebo-controlled, parallel-group study compared five dupilumab dose schedules with placebo in adults with moderate-to-severe atopic dermatitis. The primary outcome was percent change in EASI score from baseline to Week 16. The sponsor reported ANCOVA LS mean differences favoring more negative EASI changes in each dupilumab group relative to placebo, with 95% confidence intervals and p-values listed for each comparison.

The registry also reports secondary outcomes spanning global assessment, itch, EASI, and SCORAD, and provides serious adverse event counts by group. The record does not report the complete statistical plan, the multiplicity procedure, or the full missing-data method. Summary-based recalculations are unadjusted checks and do not replace the sponsor's analyses.