Dose Ranging Study of Amlitelimab in Adult Participants With Moderate-to-severe Asthma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Dose Ranging Study of Amlitelimab in Adult Participants With Moderate-to-severe Asthma (NCT05421598), a randomized phase 2 study. The primary outcome is Annualized Rate of Severe Asthma Exacerbation Events Over 48 Weeks.

2. Clinical Question

3. Trial Design

This was a global, multicenter, randomized, double-blind, placebo-controlled, parallel-group, dose-ranging Phase 2 study of add-on amlitelimab in adults with moderate-to-severe asthma. The registry lists 437 participants as enrolled and the study status as completed. The study began on 2022-06-30, and the primary completion date was 2024-10-11.

Participants were allocated among four arms: placebo and three amlitelimab regimens. The active regimens used loading doses of 125 mg, 250 mg, or 500 mg, followed by maintenance doses of 62.5 mg, 125 mg, or 250 mg, respectively. Active treatment was administered subcutaneously every four weeks through Week 20 and every twelve weeks starting at Week 24 until Week 48. Placebo followed the same schedule. The registry describes masking as quadruple and the study model as parallel.

The study purpose was to assess efficacy, safety, and tolerability. The randomized treatment period was up to approximately 60 weeks. Per-participant duration was up to approximately 76 weeks for participants not entering the LTS study and up to approximately 64 weeks for participants entering it. The scheduled number of visits was 13. Further detail about the meaning of LTS is not reported in the registry record.

The primary endpoint covered Baseline (Day 1) to Week 48 and was evaluated in the intent-to-treat population. The record lists 25 outcome measures and 12 analyses. It also lists secondary measures involving spirometry, asthma control, asthma-related quality of life, and time to first severe exacerbation. Further detail about randomization procedures, the planned sample-size justification, and the complete analysis plan is not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: A total of 437 participants were randomized in a ratio of 2:1:2:2 to one of the following groups: placebo, amlitelimab 62.5 milligrams (mg) with 125 mg loading dose, amlitelimab 125 mg with 250 mg loading dose, or amlitelimab 250 mg with 500 mg loading dose. Randomization was stratified by region, screening blood eosinophil count (\<300 cells per \[/\] microliter \[mcL\] and \>=300 cells/mcL), and number of severe asthma exacerbations in the previous 12 months (=1 or \>1 exacerbations).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-treat (ITT) population included all randomized participants.

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: Randomized population included all participants from screened population who had been allocated to a randomized treatment by interactive voice/web response system regardless of whether the treatment was received or not.

6. Primary Endpoint

The registry lists the primary outcome as Annualized Rate of Severe Asthma Exacerbation Events Over 48 Weeks, measured over Baseline (Day 1) to Week 48.

The primary endpoint was the annualized rate of severe asthma exacerbation events from Baseline (Day 1) to Week 48, measured in exacerbations per participant-year. A severe event was defined as worsening asthma requiring systemic corticosteroids for at least 3 days; for a stable maintenance regimen of oral corticosteroids for asthma, it could instead be a doubling of the dose for 3 or more days. Hospitalization or an emergency room visit because of asthma requiring systemic corticosteroids also met the definition.

The annualized rate was calculated as the total number of severe events divided by total observation duration. Rates account for the time participants were observed, which is useful when follow-up duration differs. The registry says that this rate was estimated using negative binomial regression. The endpoint therefore summarizes recurrent events over follow-up, rather than only whether a participant experienced at least one event.

In general, an annualized event rate provides a rate-scale summary of event burden, while the rate ratio compares that rate between an active-treatment group and placebo. The registry reports the endpoint for the intent-to-treat population, defined as all randomized participants. This statistical endpoint does not by itself describe every aspect of asthma control, lung function, quality of life, or safety; those topics are addressed by other recorded outcomes.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a count-based endpoint expressed as an annualized rate. Each participant may contribute more than one severe exacerbation event, and observation time contributes to the denominator. In general, analysis of recurrent event counts must account for both the number of events and the amount of time observed. A rate comparison can be more informative than comparing raw event totals when follow-up time varies.

Why negative binomial regression?

The registry identifies negative binomial regression as the method used for the primary endpoint. In general, count data often vary more across participants than a simple Poisson model assumes; this is called over-dispersion. A negative binomial model can accommodate that additional variability. The registry does not report the model specification, covariates, handling of observation time beyond the endpoint definition, or any diagnostics, so those details are not inferred.

The reported parameter is a rate ratio. In general, a rate ratio compares the modeled event rate in one group with the modeled event rate in another. A value of one corresponds to equal modeled rates; values below one indicate a lower estimated rate in the numerator group, and values above one indicate a higher estimated rate. The registry's confidence intervals convey uncertainty around the estimated ratios, while the reported p-values describe evidence against the comparison's null hypothesis under the stated model. A p-value does not measure the probability that a treatment effect is real, nor does it describe the size or clinical importance of an effect.

How to read the arm estimates and comparisons

The arm-level event rates are accompanied by 95% confidence intervals. The primary comparison results are model-based negative binomial estimates. Because one set of values is an arm-level rate and the other is a regression-based relative comparison, the arm estimates should not be substituted for the reported rate ratios. Each comparison uses the placebo arm as the reference and has its own estimate and uncertainty interval.

The record identifies all randomized participants as the intent-to-treat population for the primary outcome. In general, this population definition preserves the randomized grouping for analysis. The registry does not provide the number of participants included in the model for each comparison or details of any exclusions, missing observations, or sensitivity analyses. Handling of missing data is not reported in the registry record.

Recalculated continuous-outcome summaries

Some posted continuous-outcome comparisons are labeled recalculated. These are simple, unadjusted re-calculations from the posted summary statistics, rather than the sponsor's adjusted analyses. The re-calculated intervals use approximate standard deviations inferred from the reported standard errors, and the associated p-values are unadjusted. In general, an unadjusted comparison does not incorporate the covariate adjustment or other modeling choices that may be part of an ANCOVA analysis. Thus it can differ from the registry's reported estimate, interval, or p-value without indicating that either arithmetic procedure is erroneous.

The reported outcome summaries use least-squares means and standard errors for the continuous secondary endpoints. The registry identifies ANCOVA for the available sponsor comparisons. Least-squares means are model-adjusted means; the exact covariates and model details are not reported in the registry record. The recalculated mean differences are useful as transparent checks based on the posted summaries, but they are not a substitute for the sponsor's model-based results.

Planning illustrations are not observed power

Planning illustrations use the observed effect as a reference and show participants per arm under stated assumptions: two-sided significance level 0.05, equal allocation between the two arms, and a normal approximation. For example, the FEV1 comparison using 100% of the observed effect gives participants per arm for 80% power of 269 and participants per arm for 90% power of 360. Other rows use 75% or 50% of the observed effect. These are hypothetical planning calculations, not the trial's actual enrollment target, not the trial's sample-size justification, and not observed power. The trial's own sample-size justification is not reported in the registry record.

The registry reports 12 analyses, but details of any interim analyses, multiplicity adjustment, model checks, or prespecified hierarchy are not reported in the registry record. Those omissions limit how fully the statistical plan can be reconstructed from the posted record.

10. Analyses Reported by the Sponsor

The sponsor's registry analyses are the primary inferential results for the posted comparisons. For the primary endpoint, the sponsor reports negative binomial regression rate ratios and 95% confidence intervals for each amlitelimab dose group versus placebo. For the available continuous secondary comparisons, the sponsor reports ANCOVA results. These model-based estimates should be distinguished from recalculated comparisons based on arm-level summary statistics.

The recalculated continuous-outcome contrasts are unadjusted and use approximate standard deviations derived from the reported standard errors. Their intervals and p-values can therefore differ from the sponsor's adjusted ANCOVA results. The registry record does not identify a discrepancy for these calculations, but methodological differences remain important when interpreting the two sets of results. The sponsor's analyses are the relevant results for understanding what the registry reports as the trial's own comparisons; the recalculations serve as a separate, approximate check and should not replace those analyses.

For the primary endpoint, the listed comparisons are sponsor-reported negative binomial analyses rather than recalculations from summary means. The registry does not describe enough model detail to independently reproduce those estimates. No additional sponsor interpretation, such as an explanation of model choices or a discussion of clinical importance, is reported in the registry record.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 25 outcome measures and 12 analyses, including the primary exacerbation-rate endpoint and multiple secondary measures. When many outcomes, doses, and follow-up assessments are tested, the chance of obtaining one or more small p-values by chance can increase. In general, a multiplicity strategy may control this risk through methods such as a prespecified testing order or adjusted significance thresholds. Whether any such approach was planned or used is not reported in the registry record.

The record identifies the intent-to-treat population for the primary endpoint as all randomized participants. It does not specify the analysis population for each secondary endpoint, the number of participants contributing to each model, or whether any analysis used a different population. The registry record also does not report the handling of missing data, including assumptions about missing outcome measurements, withdrawal, or incomplete follow-up. These details matter because different methods for incomplete data can affect estimates and their uncertainty.

Further details about interim analyses, early stopping rules, covariate adjustment, model diagnostics, and sensitivity analyses are not reported in the registry record. Their absence from the posted record does not establish that they were not conducted; it means that their role cannot be assessed from this record. The primary endpoint description names negative binomial regression, and the available sponsor analyses of continuous outcomes name ANCOVA, but full model specifications are not reported.

The sample-size planning justification is also not reported in the registry record. The separate planning illustrations based on observed effects are hypothetical and do not establish the original target or the trial's operating characteristics. Because the testing hierarchy, missing-data approach, and planning rationale are unavailable, interpretation should focus on the reported estimates and intervals while recognizing that the full statistical plan cannot be reconstructed.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: asthma (3.1%), influenza (1.6%), pneumonia (1.6%), pyelonephritis acute (1.6%), acute left ventricular failure (1.6%).

The registry reports participants with serious adverse events by randomized group. The listed counts are 11 of 127 in the placebo group, 6 of 61 in the amlitelimab 62.5 mg group, 10 of 125 in the 125 mg group, and 12 of 124 in the 250 mg group. Deaths are listed as 0, 1, 2, and 0, respectively. These are descriptive counts; the registry does not provide a comparative statistical analysis of these figures in the information.

The most frequently listed serious-event terms include asthma (3.1%), influenza (1.6%), pneumonia (1.6%), acute pyelonephritis (1.6%), acute left ventricular failure (1.6%), and renovascular hypertension (1.6%). Such term-level percentages do not by themselves establish a cause or show how events were distributed across every arm. The registry also reports other adverse events by arm, but these counts should not be treated as equivalent to serious adverse events.

Safety counts should be interpreted in the context of the number at risk, the duration of observation, and the registry's event definitions. The record does not provide enough information to assess event relatedness, timing, exposure-adjusted rates, or comparative uncertainty. Descriptive differences alone do not establish that one regimen is safer or less safe than another.

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

The results describe estimates from a randomized, placebo-controlled Phase 2 trial of three amlitelimab dose regimens in adults with moderate-to-severe asthma. For the primary endpoint, the registry reports model-based annualized severe-exacerbation rates and rate ratios over 48 weeks. The reported estimates differ across dose comparisons, and their confidence intervals communicate differing levels of precision. These findings are a statistical description of this study, not a guarantee of what would be observed in another population or setting.

A rate ratio and its confidence interval address the relative event rate under the regression model. They do not directly measure an individual participant's experience, establish a mechanism, or indicate the clinical importance of a difference. Clinical importance thresholds are not reported in the registry record. Nor does a p-value quantify the size of an effect or the probability that a result is true.

The secondary outcomes add information about lung function, asthma control, and quality of life. They are different endpoints, with distinct scales and analyses, and should not be combined informally into a single claim. The registry record does not report whether multiplicity was controlled, how missing data were handled, or a complete analysis hierarchy. Those unknowns limit how strongly secondary comparisons can be interpreted.

The safety information is also descriptive. Serious adverse event counts and listed event terms are relevant to the trial's safety record, but the posted figures alone do not establish causation or provide a complete comparative safety assessment. Overall, the registry supports a careful reading of the reported estimates, intervals, and methods; it does not, by itself, settle questions about long-term outcomes, performance in other groups, or clinical decision-making.

14. Why This Trial Matters Statistically

This trial is statistically informative because it evaluates an event-rate endpoint rather than only a single change score. Severe asthma exacerbations can recur during follow-up, and the registry's annualized rate captures the total event count relative to observation time. Negative binomial regression is the stated method, a model commonly suited to over-dispersed count data. The reported rate ratios therefore provide a comparison on a scale suited to recurrent event rates.

The dose-ranging design also places several active regimens alongside a placebo comparator under a shared schedule. That structure allows the registry to report separate dose-versus-placebo estimates, but it also creates multiple comparisons. The reported outcome is not a simple monotonic pattern across the listed dose estimates, so the comparison requires attention to uncertainty and model-based estimates rather than a conclusion based on dose ordering alone.

The study also illustrates why a primary endpoint should remain distinct from secondary outcome signals. Spirometry, asthma-control scores, and quality-of-life scores measure different aspects of disease experience. Their estimates can complement the event-rate analysis, but multiple secondary comparisons need cautious interpretation, especially when the adjustment strategy is not reported.

Finally, the record demonstrates the value of separating sponsor analyses from unadjusted recalculations. A calculation based on summary statistics can help make assumptions transparent, but it does not reproduce a covariate-adjusted analysis or replace a model fitted to participant-level data. Recognizing that distinction helps readers evaluate what the registry's estimates say, how precisely they are reported, and what remains uncertain.

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.

Important limitations arise from the level of statistical detail available in the registry record. The primary method is identified as negative binomial regression, but the model specification, covariates, handling of observation time, and diagnostics are not reported. It is therefore not possible to assess the full modeling strategy or independently reproduce the primary rate ratios from the posted summaries.

The registry does not report the sample-size justification, intended effect used for planning, multiplicity procedure, or analysis hierarchy. It also does not report the approach to missing data, sensitivity analyses, or detailed participant flow for each endpoint. These omissions constrain evaluation of the precision the study was designed to achieve and the robustness of results to analytical choices.

The secondary outcomes span several domains and follow-up times. The record lists many measures, while the reported sponsor comparisons available for interpretation do not cover every dose-outcome pairing. A reader should not assume that an unlisted comparison was negative, not conducted, or not estimable unless the registry explicitly says so. The record notes that values are missing or not estimable for a two-arm comparison for the count-other endpoint.

Some continuous-outcome recalculations rely on approximate standard deviations derived from reported standard errors and are unadjusted. They can be useful as rough checks, but they do not match the sponsor's ANCOVA framework. Safety reporting is similarly limited by the descriptive counts presented: event relatedness, timing, exposure-adjusted comparisons, and detailed arm-level term distributions are not reported in the registry record. These limitations do not invalidate the posted results, but they narrow the conclusions that can be drawn from them.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Dose Ranging Study of Amlitelimab in Adult Participants With Moderate-to-severe Asthma — NCT05421598.
• Registry record last updated 2026-03-30; status last verified 2026-02.
• Results first posted 2026-03-30.
• 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

NCT05421598 is a completed, randomized, quadruple-masked, parallel-group Phase 2 dose-ranging trial sponsored by Sanofi. It enrolled 437 adults with moderate-to-severe asthma and compared three subcutaneous amlitelimab regimens with matching placebo as add-on therapy. The primary outcome was the annualized rate of severe asthma exacerbation events from Baseline (Day 1) to Week 48, analyzed using negative binomial regression in the intent-to-treat population.

The registry reports three dose-versus-placebo rate ratios, with one comparison's 95% confidence interval excluding one and the other two including one. Secondary measures addressed lung function, asthma control, quality of life, and exacerbation timing. The record also reports serious adverse event counts and deaths by group. Full details of the sample-size justification, multiplicity strategy, missing-data handling, and complete model specifications are not reported in the registry record.