This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Evaluation of Dupilumab's Effects on Airway Inflammation in Patients With Asthma (NCT02573233), a randomized phase 2 study. The primary outcome is Change From Baseline in Eosinophils Cells Count in the Bronchial Submucosa at Week 12.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02573233 |
| Phase | Phase 2 |
| Condition | Asthma |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 42 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Sanofi |
| Results posted | Yes |
2. Clinical Question
Population
Adults with persistent asthma were eligible for this exploratory study. The primary analysis was narrower than the overall enrolled population: it included randomized participants who had baseline and Week 12/end-of-treatment bronchoscopies and adequate biopsies at both time points. The number analyzed under that definition is not reported in the registry record. The record lists enrollment as 42.
Intervention
Dupilumab was added to stable inhaled corticosteroid/long-acting beta-agonist therapy. The regimen began with two subcutaneous injections on Day 1 for a total loading dose of 600 mg, followed by a single 300 mg injection every two weeks from Week 2 to Week 14. Reliever medication was also described in the arm record.
Comparator
The comparator was placebo for dupilumab, added to stable inhaled corticosteroid/long-acting beta-agonist therapy. Its injection schedule began with two injections on Day 1, followed by a single injection every two weeks from Week 2 to Week 14. The registry describes reliever medication in this arm as well.
Primary question
Among participants with evaluable bronchial biopsies at baseline and Week 12/end of treatment, did change in eosinophil cell count in the bronchial submucosa differ between the dupilumab and placebo groups? The registry reports a Rank ANCOVA comparison but does not report a between-group effect estimate or confidence interval for that analysis.
The clinical question focuses on an airway-tissue measure rather than a symptom score or an exacerbation outcome. The stated primary objective was to evaluate dupilumab, compared with placebo, on airway inflammation in participants with persistent asthma. The study also had secondary objectives addressing safety, tolerability, and immunogenicity. Those objectives are distinct: the primary endpoint concerns a tissue-based inflammatory measurement, while the other outcomes cover additional biological and safety observations.
The treatment contrast is between dupilumab and placebo on top of stable background inhaled therapy. It is not a comparison of dupilumab with no asthma treatment. The registry names several inhaled corticosteroid/long-acting beta-agonist products among the interventions, but does not report how background products were distributed or whether their use differed between arms. The effect being assessed is therefore the addition of the study injection regimen in the context of the recorded background therapy.
The narrow primary analysis population matters when interpreting the question. A participant needed suitable biopsies at both time points to enter the pharmacodynamic analysis. This means the primary result pertains to participants with paired, adequate tissue samples, not automatically to everyone enrolled or randomized. The exact number in the primary analysis population is not reported in the registry record.
3. Trial Design
This was a Phase 2, randomized, double-masked, parallel-group, placebo-controlled study of dupilumab in adults with persistent asthma. The registry lists an enrollment of 42 and the study status as completed. The recorded study start was 2016-01-27, and primary completion was 2018-01-03. The sponsor was Sanofi.
Participants were assigned to dupilumab or placebo injections while continuing stable inhaled corticosteroid/long-acting beta-agonist therapy. The dupilumab regimen was a 600 mg loading dose delivered as two subcutaneous injections on Day 1, followed by 300 mg as a single injection every two weeks from Week 2 to Week 14. The placebo arm followed the corresponding injection schedule. The record also mentions salbutamol/albuterol or levo-salbutamol/levalbuterol as reliever medication.
Randomization and double masking are design features intended, in general, to reduce systematic differences between groups and the influence of treatment knowledge on trial conduct or outcome assessment. The registry does not report the randomization sequence method, allocation concealment details, or how masking was maintained. It also does not report the planned allocation ratio. The parallel-group structure means the arms were followed as separate groups rather than participants crossing from one treatment to the other.
The primary endpoint was assessed from bronchial biopsies at baseline and Week 12/end of treatment. The analysis population was defined by completion of bronchoscopies at both time points and adequate biopsies at both. Separate FeNO outcomes were recorded at Week 12 and over Week 6 through Week 12. The registry lists five secondary outcomes, including FeNO, antidrug antibodies, pharmacokinetics, and treatment-emergent adverse events.
The record identifies the study as exploratory. It does not report a sample-size justification, a prespecified target for the primary analysis population, an interim-analysis plan, or detailed statistical methods beyond the named analyses. Those omissions limit what can be concluded about the planning assumptions and the complete analysis strategy. The enrollment figure should not be treated as the number contributing to every endpoint, because the endpoint-specific analyzed populations differ or may differ.
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 were randomized in 1:1 ratio to receive dupilumab 300 mg every 2 weeks (q2w) and placebo q2w by using Interactive Voice/Web Response System (IVRS). Randomization was stratified by inhaled corticosteroids (ICS) dose (medium and high) and region (North America and Europe).
Overall Study
| Stage | Placebo | Dupilumab |
|---|---|---|
| STARTED | 22 | 20 |
| COMPLETED | 22 | 20 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: Analysis was performed on pharmacodynamic (PD) population which consisted of all randomized participants who underwent baseline and Week12/end of treatment (EOT) bronchoscopies and have adequate biopsies for analysis at both baseline and EOT. Here, overall number of participants analyzed = participants with available data for this outcome measure.
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 all randomized participants.
| Characteristic | Unit | Placebo | Dupilumab | Total |
|---|---|---|---|---|
| Participants analysed | 22 | 20 | 42 | |
| Age, Continuous | years | 41.0 (SD 10.3) | 45.5 (SD 10.6) | 43.1 (SD 10.6) |
| Sex: Female, Male: Female | Participants | 8 | 13 | 21 |
| Sex: Female, Male: Male | Participants | 14 | 7 | 21 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 1 | 1 | 2 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 21 | 19 | 40 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Asian | Participants | 0 | 1 | 1 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Black or African American | Participants | 3 | 3 | 6 |
| Race (NIH/OMB): White | Participants | 19 | 16 | 35 |
| Race (NIH/OMB): More than one race | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Baseline Eosinophils Count in Bronchial Tissue | cells/mm^2 | 12.97 (9.91 to 21.24) | 34.96 (10.07 to 263.75) | 20.73 (9.99 to 68.87) |
| Baseline Mucin-Stained Area in The Bronchial Tissue Specimen | cells/mm^2 | 561.12 (SD 289.00) | 520.57 (SD 511.69) | 542.33 (SD 402.61) |
| Baseline Mast Cells Count (Chymase Positive) in Bronchial Tissue | cells/mm^2 | 74.36 (SD 58.63) | 79.17 (SD 68.07) | 76.59 (SD 62.42) |
| Baseline Mast Cells Count (Tryptase Positive) in Bronchial Tissue | cells/mm^2 | 80.10 (SD 64.92) | 105.53 (SD 104.68) | 91.88 (SD 85.49) |
| Baseline Total T-Lymphocytes Count in Bronchial Tissue | cells/mm^2 | 301.09 (121.01 to 500.43) | 153.33 (83.81 to 192.40) | 181.50 (105.08 to 392.13) |
| Baseline T-Helper Lymphocytes Count in Bronchial Tissue | cells/mm^2 | 237.10 (190.17 to 511.48) | 200.85 (102.49 to 398.08) | 215.05 (147.29 to 467.90) |
| Baseline Fractional exhaled nitric oxide (FeNO) | parts per billion (ppb) | 41.2 (SD 31.5) | 36.4 (SD 22.8) | 38.9 (SD 27.5) |
6. Primary Endpoint
The registry lists the primary outcome as Change From Baseline in Eosinophils Cells Count in the Bronchial Submucosa at Week 12, measured over Baseline, Week 12.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Inter-Quartile Range |
| Unit | cells/mm^2 |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was change from baseline in eosinophil cell count in the bronchial submucosa at Week 12. Eosinophils were counted in thin sections of bronchial tissue using quantitative immunohistochemistry, with results expressed as cells per square millimeter. The recorded time points were baseline and Week 12.
This endpoint uses a tissue measurement to assess airway inflammation, which aligns with the study’s stated primary objective. In general, measuring the same biological feature before and after treatment allows the analysis to focus on change during the study rather than only on the final measurement. A biopsy-based measure can offer information about the tissue itself; it also requires invasive sampling and depends on having adequate specimens at both time points. The registry’s analysis-population definition reflects that requirement.
The endpoint is summarized using a median and an interquartile range. In general, the median describes the middle of the observed values, while the interquartile range describes the spread of the central portion of those values. These summaries do not describe the same feature as a mean and standard deviation, and they can be useful when values are skewed or include influential observations. The registry does not state why this particular summary was selected.
The recorded arm values were 5.8 for placebo and -6.04 for dupilumab, with the endpoint identified as change from baseline. The sponsor’s reported comparison used Rank ANCOVA. The registry does not provide a between-group estimate or confidence interval for that primary analysis, so its direction and precision cannot be fully assessed from the reported comparison alone.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The primary endpoint describes change in bronchial-submucosal eosinophil count, a tissue-based measure aligned with the trial’s airway-inflammation objective. The registry reports a Rank ANCOVA comparison between placebo and dupilumab, with a p-value of 0.840. It does not report the corresponding between-group effect estimate or confidence interval. Consequently, the primary analysis does not give a registry reader a direct estimate of the size and precision of the treatment contrast. The group-level summaries alone do not replace that missing contrast estimate.
The reported primary p-value does not indicate statistically persuasive evidence of a between-group difference under the named analysis. That statement is about the analysis reported in the registry; it is not proof that the treatments are identical, nor does it establish that a clinically important difference is absent. Without a reported effect estimate and confidence interval, the range of treatment differences compatible with the primary analysis cannot be judged from the record.
The FeNO outcomes provide a different view of airway inflammation. For both reported FeNO summaries, the dupilumab group’s mean change was negative while the placebo group’s mean change was positive. The sponsor’s repeated-measures analyses report between-group differences with 90% confidence intervals that do not cross zero. These outcomes are secondary, however, and should not be treated as if they were the primary tissue endpoint. They also involve a different measurement, time structure, and analysis population.
The posted FeNO group summaries are accompanied by unadjusted recalculations based on the summary statistics. These recalculations estimate mean differences and confidence intervals but are not substitutes for the sponsor’s modeled analyses. The sponsor used MMRM, which accounts for repeated measurements under a model; the simple recalculation is based on the posted summary values. The two approaches can yield different estimates because they do not necessarily use the same information or assumptions.
Precision differs by outcome. A confidence interval communicates uncertainty around an estimated effect: narrower intervals indicate greater precision than wider intervals, while an interval spanning no difference leaves a wider range of effects compatible with the data. The primary comparison lacks a reported interval. The FeNO sponsor analyses do include 90% intervals, but their interpretation must account for the secondary status of these endpoints and the absence of a reported multiplicity strategy.
Results for the primary and secondary outcomes should therefore be read as a mixed set of evidence, not collapsed into a single conclusion. The primary analysis does not provide a reported treatment-effect interval, and its p-value is not statistically persuasive. The FeNO outcomes show a different pattern in their sponsor analyses, but the registry does not report whether those analyses were adjusted for testing multiple outcomes or time points. This distinction affects how much confirmatory weight the secondary findings can carry.
Neither an outcome-specific p-value nor a group-level change establishes how a treatment will perform outside this study. The registry describes an exploratory Phase 2 trial with a biopsy-defined primary population. It does not report details needed to judge the generalizability of that population, the complete handling of missing data, or the rationale for the study’s size. Interpretation should remain focused on the measurements and analyses actually recorded.
9. Statistical Methods Explained
The statistical methods answer different questions and summarize data in different ways. The primary outcome is a median change in a tissue count, analyzed with Rank ANCOVA. The FeNO outcomes are means and standard deviations, with sponsor comparisons from MMRM. In general, the choice of method reflects the outcome’s scale, distribution, and timing. The registry does not provide enough detail to reconstruct every model specification.
What does Rank ANCOVA mean?
ANCOVA is a general approach for comparing groups while accounting for baseline information. A rank-based version works with the ordering of observations rather than treating the measured values as normally distributed raw values. In general, ranks can make an analysis less dependent on strong distributional assumptions when data are asymmetric or have extreme values. The registry names Rank ANCOVA for the primary comparison but does not report its covariates, exact implementation, or the resulting adjusted treatment-effect estimate.
The primary endpoint is expressed as a median and interquartile range, which describes the central value and spread without assuming that the group distribution is symmetric. This differs from a mean and standard deviation, which summarize an arithmetic average and dispersion around it. Neither summary, by itself, supplies a between-group estimate with uncertainty. A treatment comparison and its confidence interval are needed to communicate the contrast and its precision.
What does MMRM contribute?
MMRM means a mixed model for repeated measures. In general, it analyzes measurements collected at multiple times while accounting for the relationship among measurements from the same participant. It can estimate group differences over time using the repeated observations rather than reducing all time points to one value. The exact fixed effects, covariance structure, baseline adjustment, and missing-data assumptions used for this trial are not reported in the registry record.
The sponsor’s FeNO analyses report least-squares mean differences with 90% confidence intervals. In general, a least-squares mean is a model-based group estimate, and its difference is conditional on the model specification. A confidence interval gives a range of values compatible with the estimate and its stated model. A 90% interval is not directly interchangeable with an interval at another confidence level. The registry reports the confidence level and bounds for each FeNO comparison, but does not provide the full model details.
How should recalculated comparisons be read?
The FeNO recalculations are simple, unadjusted comparisons derived from the posted group summaries. They are useful as a check on the broad contrast between group means and as an illustration of uncertainty based on those summaries. They do not reproduce the sponsor’s MMRM, and they should not be mistaken for a separate prespecified trial analysis. Their p-values are unadjusted, and the registry does not identify them as multiplicity-adjusted.
The recalculated FeNO comparisons include 95% confidence intervals and unadjusted p-values. In general, a confidence interval for a mean difference describes uncertainty about the group contrast under the calculation’s assumptions; it is not a probability statement that the particular interval contains the true effect. The recalculations assume the summary statistics adequately represent the analyzed groups. They cannot account for features of the original repeated-measures model that are not encoded in those summaries.
Why does over-dispersion matter in general?
Over-dispersion is a general issue in analyses of counts when the observed variability is greater than a model’s assumed variability. If ignored, uncertainty can be understated and inferential results can become too optimistic. It is not reported that the primary endpoint was analyzed with a count model, and the registry does not identify over-dispersion as an issue in this trial. The primary analysis is instead named as Rank ANCOVA, so over-dispersion should not be used to reinterpret its result.
How do planning illustrations differ from observed power?
The planning illustrations are calculated from observed FeNO effects and are not the trial’s design justification. Their assumptions are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation; each illustration detects a stated fraction of the observed effect. For 100% of the observed effect, the illustrations give 19 participants per arm for 80% power and 26 for 90% power for the Week 12 FeNO outcome; for the Week 6 through Week 12 average, the corresponding figures are 16 and 21.
For 75% of the observed effect, the Week 12 FeNO illustration gives 34 participants per arm for 80% power and 46 for 90% power; the Week 6 through Week 12 average gives 28 and 38. For 50% of the observed effect, the Week 12 illustration gives 76 and 102 participants per arm for 80% and 90% power; the average-change illustration gives 63 and 84.
These are planning illustrations, not observed power and not evidence that the study was designed around those effects. The registry does not report its own sample-size justification or the assumptions used to choose the enrollment target. Planning calculations based on observed effects can be optimistic because observed effects are uncertain, particularly in exploratory studies. They should be read as hypothetical planning scenarios rather than as a measure of the trial’s success.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Rank ANCOVA | not reported | 0.840 |
The sponsor’s primary analysis is a Rank ANCOVA comparison, with a reported p-value of 0.840. The registry does not report a treatment-effect estimate or confidence interval for that comparison. As a result, there is no primary sponsor estimate and interval against which to compare a simple recalculation. The group summaries are not enough to recreate the rank-based analysis.
For FeNO, the sponsor reports MMRM least-squares mean differences. For change at Week 12, the reported difference was -22.4 ppb, with a 90% confidence interval from -32.9 to -11.9 and a reported p-value of 0.001. For the average change from Week 6 through Week 12, the reported difference was -22.0 ppb, with a 90% confidence interval from -31.3 to -12.8 and a reported p-value of 0.001.
The posted-summary recalculations give mean differences of -19.0 ppb and -19.5 ppb for those respective FeNO outcomes. They are unadjusted calculations rather than reproductions of the MMRM. Differences between the recalculated and sponsor estimates are therefore not, by themselves, evidence of an error: the methods use different information and modeling assumptions. The registry does not flag a discrepancy. The sponsor’s reported analysis is the appropriate reference for what the registry identifies as the trial’s own analysis; the recalculations provide a separate, simpler view of the posted summaries.
11. Multiplicity, Missing Data and Analysis Populations
The registry record does not report a multiplicity-adjustment strategy. This matters because the trial includes a primary endpoint and several secondary outcomes, and the FeNO outcomes involve more than one time summary. In general, testing multiple hypotheses increases the chance of at least one apparently positive result arising by chance. Adjustments or a prespecified hierarchy can help control that risk, but the record does not say whether either approach was used. The unadjusted recalculations should therefore be interpreted as exploratory comparisons, not as multiplicity-controlled evidence.
The primary analysis population required a baseline and Week 12/end-of-treatment bronchoscopy and adequate biopsies at both visits. This makes availability and quality of tissue samples central to inclusion in the primary analysis. The number analyzed under the stated criteria is not reported in the registry record. The record also does not describe how participants without an adequate paired biopsy were handled, whether missing outcomes were imputed, or what assumptions were used about missingness.
For the repeated-measures FeNO outcomes, the sponsor used MMRM. Such models can use repeated observations when some measurements are missing, but their interpretation depends on model specification and assumptions about the missing data. The registry does not report the details needed to assess those assumptions, the amount or pattern of missing data, or any sensitivity analyses.
The registry does not report an interim-analysis plan, the complete statistical analysis plan, or a sample-size justification. It also does not state whether the reported outcomes and analyses were selected before results were known. These omissions do not demonstrate that a particular method was or was not used; they limit what a reader can verify from the registry record. In general, prespecification, transparent missing-data methods, and an explicit multiplicity plan help distinguish confirmatory conclusions from exploratory findings.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Placebo | 0 | 22 | 0.0% |
| Dupilumab | 1 | 20 | 5.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Placebo | 0 of 22 |
| Dupilumab | 0 of 20 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Placebo | 11 of 22 |
| Dupilumab | 14 of 20 |
Most frequent serious adverse event terms
| Event term | Organ system | Placebo | Dupilumab |
|---|---|---|---|
| Neutropenia | Blood and lymphatic system disorders | 0 of 22 | 1 of 20 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Placebo | Dupilumab |
|---|---|---|---|
| Headache | Nervous system disorders | 3 of 22 | 5 of 20 |
| Nasopharyngitis | Infections and infestations | 4 of 22 | 2 of 20 |
| Injection Site Erythema | General disorders | 2 of 22 | 3 of 20 |
| Nausea | Gastrointestinal disorders | 1 of 22 | 3 of 20 |
| Upper Respiratory Tract Infection | Infections and infestations | 1 of 22 | 3 of 20 |
| Injection Site Inflammation | General disorders | 2 of 22 | 1 of 20 |
| Injection Site Pruritus | General disorders | 1 of 22 | 2 of 20 |
| Accidental Overdose | Injury, poisoning and procedural complications | 1 of 22 | 2 of 20 |
| Back Pain | Musculoskeletal and connective tissue disorders | 0 of 22 | 3 of 20 |
| Asthma | Respiratory, thoracic and mediastinal disorders | 1 of 22 | 2 of 20 |
Serious events with the highest arm-level rate: neutropenia (5.0%).
The registry reports serious adverse events in the safety groups: 0 of 22 participants in the placebo group and 1 of 20 in the dupilumab group were affected. It reports 0 deaths in each group. Neutropenia is listed among the top serious adverse event terms at 5.0%; the record does not associate that term with a particular group in the information.
These counts describe observed events in the recorded safety groups. They do not establish that an event was caused by the assigned treatment, and the small group counts mean that the figures alone cannot precisely characterize uncommon harms. Zero reported deaths or serious events in a group does not establish that the risk is absent. The registry also lists treatment-emergent adverse events as a secondary outcome, but the serious-event figures should not be confused with a complete account of all adverse events, their severity, timing, or relationship to treatment.
Safety comparisons require attention to the number and type of participants observed, the duration of observation, and how events were collected and classified. Those details are not reported sufficiently to support broader conclusions from these counts. The figures are descriptive registry results, not individualized estimates of risk.
13. What the Result Does, and Does Not, Mean
The primary result addresses a specific question: whether change in bronchial-submucosal eosinophil count differed between the placebo and dupilumab groups among participants with paired, adequate biopsy samples. The registry reports a Rank ANCOVA p-value but not the treatment-effect estimate or its confidence interval. It therefore does not show how large the primary contrast may have been or how precisely it was estimated. The reported primary analysis does not provide statistically persuasive evidence of a difference, but this is not evidence that the groups were identical.
The FeNO results add secondary evidence about a separate airway-inflammation measure. Their sponsor analyses show modeled differences between groups, while the simple recalculations use only the posted summaries. Those results do not replace the primary biopsy endpoint, and unadjusted secondary analyses are less conclusive when the multiplicity plan is not reported. The different results can coexist because the endpoints, populations, time structures, and methods differ.
The study’s design also shapes what the findings mean. It was an exploratory Phase 2 parallel-group trial in adults with persistent asthma, with treatment added to stable inhaled therapy. The primary analysis required paired biopsy data. The reported findings should not automatically be generalized to people outside that study population or to outcomes that were not measured or reported.
In general, a statistically significant result does not by itself establish clinical importance, and a nonsignificant result does not prove no effect. Clinical importance depends on the size and relevance of an effect, the precision of the estimate, the outcome being measured, and the intended use of the treatment. The registry does not report a threshold for a clinically meaningful change in bronchial eosinophil count.
The recorded outcomes are measurements from a clinical trial, not a basis for individual treatment decisions. The registry information supports a cautious statistical reading: the primary comparison is incompletely quantified, FeNO analyses are secondary, and safety counts are descriptive. Missing details about the analysis plan, missing data, and multiplicity further constrain the strength of conclusions that can be drawn from the record.
14. Why This Trial Matters Statistically
This trial is statistically notable because its primary endpoint depends on paired tissue samples and a specialized analysis. The endpoint is not simply a routinely collected symptom measure: it is a quantitative immunohistochemical count in bronchial biopsy sections. Requiring adequate specimens at both baseline and Week 12/end of treatment creates an analysis population defined by data availability as well as randomization. Reporting the exact analyzed population and the handling of unavailable or inadequate specimens is therefore important for interpreting the result.
The registry also illustrates why an analysis method and a numerical result are not enough unless the effect estimate and its uncertainty are reported. Rank ANCOVA is named for the primary comparison, and a p-value is available, but the corresponding between-group estimate and interval are not. Without them, readers cannot judge the magnitude or precision of the primary contrast from the reported analysis. This limits the information available even when a test result is recorded.
Another statistical point is the distinction between model-based analyses and simple recalculations. The sponsor’s FeNO results come from MMRM, while the recalculations use summary statistics. Comparing them requires recognizing that they answer related but not identical analytical questions. A recalculation can help orient a reader to the posted group summaries, but it should not be presented as an exact reconstruction of the trial’s model.
Finally, the study includes multiple secondary outcomes, while its multiplicity strategy and sample-size justification are not reported. Those gaps matter when judging how much weight to give secondary findings and how the trial was planned. The record is useful for showing the design and reported outcome pattern, while also illustrating the importance of full estimates, uncertainty intervals, analysis-population counts, and prespecified statistical methods.
15. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The primary comparison is difficult to interpret fully because the registry reports no between-group effect estimate or confidence interval for the Rank ANCOVA analysis. The group-level values do not supply the missing contrast or its precision. The count of participants included in the biopsy-defined primary population is also not reported in the registry record, despite the stated eligibility requirements for that analysis.
Several methodological details are not reported: the randomization procedure, allocation concealment, model specification, complete statistical analysis plan, multiplicity strategy, and sample-size justification. The record also does not explain how missing or inadequate biopsies were handled, or how missing repeated FeNO observations were addressed. These omissions prevent a full assessment of potential bias and the assumptions underlying the analyses.
The primary outcome is based on bronchoscopic biopsies, and the analysis includes only participants with adequate samples at both time points. Such a restriction can make the analyzed population differ from the full randomized population. The registry does not provide enough information to determine the extent or consequences of that difference. Endpoint-specific counts are available for some outcomes, but they should not be assumed to apply to the primary analysis.
The study is described as exploratory and Phase 2. Its enrollment figure and observed effects cannot establish that the study was adequately powered for the primary endpoint; the registry’s sample-size justification is not reported. Planning illustrations based on observed FeNO effects are not the original design rationale and are not observed power.
Finally, the primary tissue endpoint and secondary FeNO outcomes are distinct measures, and the record’s primary and secondary analyses do not carry equal inferential weight. The absence of a reported multiplicity plan limits interpretation of the secondary findings. Safety event counts are also descriptive, and the registry entries do not support precise conclusions about uncommon events or causal attribution.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Evaluation of Dupilumab's Effects on Airway Inflammation in Patients With Asthma — NCT02573233.
- Registry record last updated 2022-04-04; status last verified 2022-03.
- Results first posted 2019-01-28.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
NCT02573233 is a completed Phase 2 randomized, double-masked, parallel-group, placebo-controlled study sponsored by Sanofi. It enrolled 42 adults with persistent asthma. The recorded start date was 2016-01-27, and primary completion was 2018-01-03.
Dupilumab or placebo was added to stable inhaled corticosteroid/long-acting beta-agonist therapy. The primary endpoint was change from baseline in bronchial-submucosal eosinophil count at Week 12, analyzed in a biopsy-defined pharmacodynamic population. The sponsor reported a Rank ANCOVA p-value but no primary between-group estimate or confidence interval.
The registry also reports two FeNO comparisons from MMRM analyses, along with listed secondary outcomes for antidrug antibodies, pharmacokinetics, and treatment-emergent adverse events. It records 0 serious adverse events in the placebo group and 1 in the dupilumab group, with 0 deaths in either group. Multiplicity, missing-data handling, and the sample-size justification are not reported in the registry record.