Effect of Benralizumab in Atopic Dermatitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of Benralizumab in Atopic Dermatitis (NCT03563066), a randomized phase 2 study. The primary outcome is Number of Allergen-induced Eosinophils in the Skin.

2. Clinical Question

The trial asks whether benralizumab changes the number of eosinophils found in skin after an allergen challenge in people with moderate-to-severe atopic dermatitis. Eosinophils are inflammatory cells, and the study rationale describes atopic dermatitis as involving inflamed skin and elevated eosinophil levels. The primary endpoint is a skin-cell measurement rather than a direct measure of symptoms or overall disease severity.

The registry also identifies an allergen-induced late phase cutaneous response as a secondary outcome. The central statistical question is whether the observed arm-specific results are sufficiently precise to support a clear estimate of the between-arm effect. The registry record reports means and standard deviations for the primary endpoint, but it does not report a comparative estimate or inferential analysis. The result can therefore be described at the level of the posted arm summaries, not as a demonstrated treatment effect.

3. Trial Design

This was a phase 2, randomized, parallel-group trial with placebo control. The registry classifies masking as quadruple. Its brief summary describes the study as double-blind, so the record uses both descriptions of masking. The sponsor was McMaster University, and the trial status is completed.

The registry lists 20 participants as enrolled. For the primary endpoint, the benralizumab arm has 9 participants and the placebo-control arm has 11. The reported intervention was a fixed 30 mg dose of benralizumab; the summary describes subcutaneous administration and three doses every 4 weeks. The placebo arm was described as identical in form. The registry does not report further details about allocation procedures, concealment, adherence, or treatment discontinuation.

The listed start date was 2018-09-04, and primary completion was 2021-09-28. The primary endpoint was measured at Day 65, 24 hours after intradermal allergen challenge. The outcome description says histological examination was used to assess allergen-induced eosinophils in skin and that intradermal saline challenge served as a control. The available record does not further explain the timing or analytical role of the saline challenge.

The primary endpoint is continuous, expressed as cells/mm2, and reported as a mean with standard deviation. The full analysis set is named as its population. The registry lists two outcome measures and one secondary outcome, but no analyses are listed. It does not report a statistical analysis plan, a sample-size justification, interim analyses, or details about how the randomized allocation was implemented. Those omissions limit what can be concluded about the planned analysis, even though the trial design and descriptive arm results are recorded.

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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Full analysis set

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: Total patients included in study

6. Primary Endpoint

The registry lists the primary outcome as Number of Allergen-induced Eosinophils in the Skin, measured over Day 65, 24 hours post-intradermal allergen challenge.

The primary endpoint is the number of allergen-induced eosinophils in skin, measured in cells/mm2 by histological examination. The stated assessment time was Day 65, 24 hours after intradermal allergen challenge. The registry says the primary objective was to evaluate benralizumab’s effect on this cell count compared with placebo and that intradermal saline challenge was used as a control.

This endpoint measures a biological feature of the challenged skin. It is relevant to the study rationale because the registry describes atopic dermatitis as involving inflamed skin and elevated eosinophils, and it proposes that benralizumab may attenuate eosinophilic skin inflammation. That rationale explains the choice of a cell-based endpoint; it does not establish what the observed results mean for symptoms, disease severity, or longer-term outcomes.

The reported summary is a mean and standard deviation for each arm. The benralizumab arm had a mean of 11.95 cells/mm2 and a standard deviation of 28.51, based on 9 participants. The placebo-control arm had a mean of 71.15 cells/mm2 and a standard deviation of 130.1, based on 11 participants. The endpoint’s stated population was the full analysis set.

A group mean describes the average observed value, while the standard deviation describes variability among observations. Neither quantity alone gives the precision of the between-arm comparison. The registry record does not report a confidence interval or a statistical test for the primary comparison, so the endpoint summaries do not establish how precisely the arm difference is estimated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous measurement. A continuous endpoint can take values across a numeric scale, rather than being limited to categories such as event or no event. Here the unit is cells/mm2, and the registry reports the mean and standard deviation for each arm. These summaries describe the observed endpoint distribution in compact form, but they do not by themselves quantify uncertainty about the difference between treatment groups.

What does a mean describe?

The mean is the arithmetic average of the recorded measurements within an arm. It provides a measure of the group’s central level, but it can be influenced by unusually high or low observations. In this trial, the benralizumab arm mean was 11.95 cells/mm2, and the placebo-control arm mean was 71.15 cells/mm2. These values are descriptive summaries, not an adjusted treatment effect. The registry does not report an adjustment model or a separately estimated contrast.

What does the standard deviation describe?

The standard deviation describes how dispersed the observations are around their arm mean. A larger standard deviation indicates greater variation among the observed measurements. The posted standard deviations are 28.51 for benralizumab and 130.1 for placebo control. The placebo-control value is large relative to its mean, suggesting that its measurements vary substantially around that average. This pattern also cautions against reading the mean as a complete description of the observed data. The registry does not report the individual measurements or additional distribution summaries that would show the form of this variability.

Why is an uncertainty interval useful?

In general, a confidence interval gives a range of values compatible with the data and a specified statistical procedure. It communicates uncertainty in an estimate more directly than a point estimate alone. For a randomized comparison, an interval around the between-arm effect would help readers assess both the direction and precision of that estimated contrast. No confidence interval for the primary comparison is reported in the registry record. The arm-specific standard deviations are not substitutes for such an interval because they describe participant-level spread, not uncertainty in the between-arm effect.

What can a significance test add?

In general, a significance test evaluates how compatible the observed data are with a specified null hypothesis under a statistical model. A p-value does not measure the size or clinical importance of an effect, and it does not replace an effect estimate with uncertainty. The registry record does not report a p-value or another test for the primary comparison. Since no analysis is listed, statistical significance cannot be assessed from the posted endpoint summaries. It would not be appropriate to infer a test result from the means and standard deviations without a documented analysis.

Why can variability affect analysis?

In general, over-dispersion means that data vary more than a chosen model expects. For count outcomes, extra variability can make a simple model’s uncertainty estimates unreliable if the model assumes less spread than the data contain. This endpoint is a count of cells expressed per area, and the reported standard deviations are large compared with the arm means, especially for placebo control. Those summaries raise a reason to examine the distribution and model assumptions, but they do not establish which statistical model would be appropriate. The analysis plan and model are not reported in the registry record.

Randomization is intended to support a fair comparison by assigning participants to study arms using a chance process. A parallel-group design compares outcomes from separate groups assigned to benralizumab or placebo. Masking can reduce the influence of expectations on aspects of trial conduct and assessment. The registry classifies masking as quadruple, while the brief summary calls the study double-blind. The record does not specify which roles were masked or describe allocation concealment, so more detailed claims about implementation cannot be made.

The study summary describes three doses of a fixed 30 mg dose administered subcutaneously every 4 weeks. The primary endpoint was assessed at Day 65, 24 hours after allergen challenge. These registry entries describe the intervention and measurement timing; they do not establish the analysis model, handling of repeated measurements, or exact relationship between the saline control and primary comparison. Such details are not reported in the registry record.

The registry lists an illustrative sample-size approach based on the observed effect, using a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. In general, such calculations can illustrate how planning requirements vary with an assumed effect size and design conditions. They are not observed power and do not show the probability that the trial’s result is true. The sample-size assumptions are explicitly distinguished from the trial’s own design, and a trial-specific sample-size justification is not reported in the registry record.

The main methodological limit is therefore not merely the size of the observed means. The registry supplies descriptive summaries but no analysis estimating the comparison or quantifying its uncertainty. The distributions, model assumptions, missing-data approach, and pre-specified analysis are also not reported. A full interpretation would require those elements alongside the posted endpoint values.

10. Confidence Intervals for Each Arm

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

R code and output

arm <- c("Benralizumab", "Placebo Control")
n   <- c(9, 11)
m   <- c(11.95, 71.15)
sp  <- c(28.51, 130.1)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Benralizumab  9  11.950  28.510  -9.965  33.865
CARM  Placebo Control  11  71.150  130.100  -16.252  158.552

11. Multiplicity, Missing Data and Analysis Populations

The registry names one primary endpoint and one secondary outcome, the allergen-induced late phase cutaneous response. It also states that there are two outcome measures, but lists zero analyses. A family of multiple outcomes can create a multiplicity issue: when several hypotheses are tested, the chance of at least one apparently unusual result can increase. In general, investigators may address this through a pre-specified testing hierarchy or an adjustment procedure. The registry record does not report whether such a plan was used, whether the primary endpoint was tested separately, or whether any multiplicity adjustment was planned.

The primary endpoint population is identified as the full analysis set. The record does not define that set or explain how participants were included in it. It does not report whether any endpoint values were missing, how missing observations were handled, or whether sensitivity analyses were conducted. The endpoint arm counts are 9 for benralizumab and 11 for placebo control, but these counts do not by themselves explain whether every enrolled participant contributed an observed value or how exclusions were handled.

In general, missing-data methods matter because conclusions can depend on why measurements are absent and on the assumptions used to address them. No imputation method, missingness assumption, or complete-case rule is described in the registry record. The same absence of analysis information prevents assessment of whether the posted descriptive summaries represent the originally intended analysis population. The registry does not report an analysis plan or enough detail to determine how multiplicity and missing data were addressed.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The posted safety figures list zero serious adverse events in the benralizumab group, with 9 participants at risk, and zero in the placebo-control group, with 11 participants at risk. The record also lists zero other adverse events in each group. No deaths are listed in either group. No serious adverse event terms are listed.

These are event counts in a small trial, not proof that an event cannot occur. In general, observing no events does not establish that the underlying risk is absent; the ability to detect uncommon events depends on the amount and duration of observation. The registry record does not give further detail about event definitions, the observation window for safety, or how adverse events were collected. The figures should therefore be read as the counts posted for this trial, not as a complete characterization of safety or a comparison establishing that the arms differ.

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

The result does show a descriptive difference in the posted primary endpoint means: the mean allergen-induced eosinophil count was 11.95 cells/mm2 in the benralizumab arm and 71.15 cells/mm2 in the placebo-control arm. The corresponding standard deviations were 28.51 and 130.1, with 9 and 11 participants, respectively. This is the direct numerical content of the registry’s primary-endpoint summaries.

The result does not show a reported estimate of the between-arm effect with a confidence interval or a p-value. It therefore does not establish how precisely the difference is estimated, whether an inferential test met any threshold, or how robust the comparison is to analysis choices. No statistical analysis is listed in the registry. A lower observed mean in one arm should not be converted into a claim of efficacy or lack of efficacy without the missing inferential context.

The endpoint is a measurement of allergen-induced eosinophils in skin at a specified time after challenge. It does not directly measure symptoms, day-to-day disease burden, quality of life, or long-term control. The registry’s rationale connects eosinophils to inflammation in atopic dermatitis, but a biological measure is not interchangeable with all clinical outcomes. The secondary outcome named in the record is the allergen-induced late phase cutaneous response; its result is not included among the reported primary endpoint summaries.

The trial is randomized and placebo-controlled, which are important design features for a treatment comparison. However, the record does not describe the analysis plan, missing-data handling, or sample-size justification. The posted arm summaries alone cannot show whether these issues influenced the result. Similarly, the safety figures record no serious adverse events, other adverse events, or deaths in either group, but these counts do not prove that adverse events are impossible or provide a broad safety assessment.

A careful conclusion is limited to the observed endpoint summaries and the information the record reports. The arm means differ descriptively, and variability is substantial. The comparative uncertainty and clinical implications are not reported in the registry record. This distinction preserves the value of the observed data without treating descriptive results as a definitive conclusion.

14. Why This Trial Matters Statistically

This trial illustrates why a clinical result needs more than an arm mean to support statistical interpretation. The posted primary outcome is biologically focused: it measures allergen-induced eosinophils in skin, in keeping with the stated interest in eosinophilic inflammation in atopic dermatitis. The registry reports an average and standard deviation for each randomized arm, giving readers a direct description of the observed measurements.

At the same time, the standard deviations are large, particularly in the placebo-control arm, and the endpoint groups are small. These characteristics make it especially important to distinguish the observed group summaries from an estimate of the comparative effect and its uncertainty. Without a confidence interval or a documented test, readers cannot assess the precision of the contrast from the registry record. The comparison is a useful example of why reporting an effect estimate with uncertainty is more informative than reporting group averages alone.

The trial also highlights the difference between a mechanistic endpoint and a patient-centered outcome. A cell count can address a focused biological question, but it does not by itself describe symptom experience or overall disease course. The registry names a secondary cutaneous-response outcome, but does not post its result in the information summarized for this record. The scope of the primary endpoint should therefore remain clear in interpretation.

Finally, the record distinguishes trial design information from analysis information. Randomization, placebo control, masking, and a stated endpoint provide context, but do not reveal how the primary comparison was modeled, how missing data were handled, or whether multiplicity was considered. Those details are not reported. This case demonstrates the statistical importance of transparent analysis reporting alongside descriptive outcome data.

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 registry record contains limited information for evaluating the primary comparison. It reports arm-specific means and standard deviations, but no between-arm effect estimate, confidence interval, p-value, or listed analysis. Consequently, the precision and inferential status of the observed difference cannot be assessed from the record.

The primary endpoint groups include 9 participants in the benralizumab arm and 11 in the placebo-control arm. The reported standard deviations indicate considerable variability in the measurements, especially in the placebo-control arm. The record does not include individual values or additional distribution summaries, so the influence of extreme observations and the shape of the outcome distribution cannot be evaluated.

The registry identifies the full analysis set as the primary population but does not define its membership. Missing endpoint data, exclusions, adherence, and analysis-population handling are not reported. The record also does not describe the statistical model, the role of the intradermal saline challenge in the analysis, or any adjustment for multiplicity. A sample-size justification and details of allocation procedures are not reported in the registry record.

The primary outcome is a histological cell count after allergen challenge at a single stated assessment time. Its relation to symptoms, overall atopic dermatitis severity, or longer-term outcomes is not established by these summaries. The registry names a secondary late phase cutaneous response, but its results are not reported in the endpoint information. These limitations constrain both statistical interpretation and the scope of any clinical inference.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Effect of Benralizumab in Atopic Dermatitis — NCT03563066.
• Registry record last updated 2025-08-22; status last verified 2021-10.
• Results first posted 2025-08-22.
• 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

NCT03563066 is a completed phase 2 randomized, parallel-group, placebo-controlled trial sponsored by McMaster University. The registry classifies masking as quadruple. Enrollment was 20. The primary endpoint was the number of allergen-induced eosinophils in skin, measured in cells/mm2 at Day 65, 24 hours after intradermal allergen challenge, with the full analysis set identified as the population.

The benralizumab arm had 9 participants, a mean of 11.95, and a standard deviation of 28.51. The placebo-control arm had 11 participants, a mean of 71.15, and a standard deviation of 130.1. No primary comparative analysis is listed. One secondary outcome is named: allergen-induced late phase cutaneous response.

The registry lists zero serious adverse events, other adverse events, and deaths in both groups. It does not report a statistical analysis plan, missing-data methods, or sample-size justification. The posted results are therefore descriptive, with no reported inferential estimate for the primary comparison.