A Study of GSK3511294 (Depemokimab) in Participants With Severe Asthma With an Eosinophilic Phenotype: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of GSK3511294 (Depemokimab) in Participants With Severe Asthma With an Eosinophilic Phenotype (NCT04718103), a randomized phase 3 study. The primary outcome is Annualized Rate of Clinically Significant Exacerbations up to 52 Weeks.

2. Clinical Question

The question is about the additional effect of assignment to the study intervention when both groups remained on their existing maintenance care. The primary endpoint expresses exacerbation frequency as a rate per participant per year. The registry reports a negative binomial analysis comparing the groups. The outcome is therefore a rate comparison over follow-up, not a comparison of the proportion of participants who had any exacerbation. The result should be interpreted as a group-level estimate over the stated timeframe; it does not describe what happened to every participant individually.

3. Trial Design

This was a Phase 3, randomized, double-blind, parallel-group, placebo-controlled, multicentre study sponsored by GlaxoSmithKline. Its official title describes a 52-week study of adjunctive GSK3511294 in adults and adolescents with severe uncontrolled asthma and an eosinophilic phenotype. The registry lists the condition as asthma and the study status as completed. The study start was 2021-02-04 and primary completion was 2024-04-11.

Participants were allocated to GSK3511294 or placebo, with masking recorded as double. The active group received a 100 milligram subcutaneous injection at week 0 and week 26. The placebo group received placebo injections on the same schedule. Both groups were to continue their existing baseline maintenance asthma standard of care throughout the study. The comparison therefore concerns adding the assigned injection to that ongoing care, rather than replacing it.

The registry reports enrollment of 397. For the primary analysis, the full analysis set included randomized participants who received at least one dose of study intervention, except participants from 2 sites excluded because of a GCP violation. Participants were analyzed according to the intervention allocated at randomization. The registry does not report further detail about the reasons for site exclusion or its effect on the analysis.

The primary endpoint was the annualized rate of clinically significant exacerbations up to Week 52. The record also lists secondary measures covering respiratory quality of life, asthma control, lung function, nighttime and daily symptoms, and exacerbations requiring hospitalization and/or an emergency department visit. The posted results include six secondary outcomes. Further details of randomization procedures, the allocation ratio, and the masking procedures are 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 397 were randomized in the study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set population included all randomized participants who received at least one dose of study intervention excluding participants from 2 sites with GCP violation. Participants were analyzed according to the intervention they were allocated at randomization.

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: Full analysis set population included all randomized participants who received at least one dose of study intervention excluding participants from 2 sites with GCP violation. Participants were analyzed according to the intervention they were allocated at randomization.

6. Primary Endpoint

The registry lists the primary outcome as Annualized Rate of Clinically Significant Exacerbations up to 52 Weeks, measured over Up to Week 52.

The primary endpoint is the annualized rate of clinically significant exacerbations up to Week 52, measured as exacerbations per participant per year. The registry defines a clinically significant exacerbation as worsening asthma requiring systemic corticosteroids and/or hospitalization and/or an emergency department visit. For participants not on maintenance systemic corticosteroids, intravenous or oral steroids for at least 3 days or a single intramuscular corticosteroid dose was required. For participants already taking maintenance systemic corticosteroids, the definition required at least double the existing maintenance dose for at least 3 days.

Exacerbations recorded in the electronic case report form were treated as verified events for the primary analysis. Events separated by less than 7 days were treated as a continuation of the same exacerbation, rather than as separate events. This operational definition matters because the measured rate depends on what qualifies as an event and how closely spaced episodes are handled.

General statistical context: an exacerbation rate can represent recurrent events over follow-up, rather than only whether a participant experienced any event. Annualizing the rate places the frequency on a per-year scale, allowing the groups’ event rates to be compared over the study timeframe. Such a summary does not show the full distribution of events among participants or the timing of every event. The registry does not report additional details about event adjudication beyond the verification and definition rules stated above.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

How should a rate ratio be read?

The primary outcome is an event rate, and the reported comparison is a rate ratio. A rate ratio compares the estimated event rate in one group with that in the reference group. A value of 1 corresponds to equal rates; a value below 1 indicates a lower estimated rate in the intervention group under the model, while a value above 1 indicates a higher estimated rate. Here, the registry reports a rate ratio of 0.52 for GSK3511294 relative to placebo. This interpretation concerns the modeled rate, not the probability that a particular participant will have an event.

The registry reports a negative binomial distribution for the primary analysis. General statistical explanation: event counts can vary more across participants than a basic count model assumes. A negative binomial model can allow this extra variation, often called over-dispersion. That flexibility is useful when some participants may have many events while others have few or none. The registry does not report the model’s detailed specification, covariates, or diagnostic checks, so those aspects cannot be assessed from the record.

What does the confidence interval add?

A confidence interval describes the precision of an estimate under the analysis method and its assumptions. The primary rate-ratio interval is 0.36 to 0.73 at the 95% confidence level. Because the interval is below 1, the reported model’s estimated comparison favors a lower event rate for GSK3511294 relative to placebo. The interval also shows that the estimate is not exact: values across the interval remain compatible with the analysis. A confidence interval is not a range of individual participant outcomes, and it does not account for limitations that are not represented by the model.

The p-value reported for the primary comparison is p < 0.001. General statistical explanation: a p-value measures how unusual the observed result, or a more extreme result, would be if the null hypothesis and the statistical model were true. It is not a measure of effect size, clinical importance, or the probability that the treatment has no effect. The p-value should be read together with the estimated rate ratio, its interval, the endpoint definition, and the analysis population.

How do the continuous secondary outcomes differ?

SGRQ total score, ACQ-5 score, and pre-bronchodilator FEV1 are continuous measures. The registry reports least-square mean changes from baseline and uses mixed models repeated measures for their group comparisons. General statistical explanation: an MMRM can use repeated measurements over time and model how measurements within a participant relate to each other. The reported least-square means are model-based estimates, not necessarily simple averages of the observed values. Details about the model’s covariance structure, covariates, and handling of missing measurements are not reported in the registry record.

The recalculated mean differences for these secondary outcomes are unadjusted calculations from posted summary statistics. They use approximations and do not reproduce the sponsor’s MMRM. The recalculated confidence intervals and p-values therefore serve as a separate, simplified check, not as a replacement for the sponsor’s analysis. For example, the posted MMRM p-values are 0.200 for SGRQ, 0.333 for ACQ-5, and 0.267 for FEV1; the recalculated unadjusted p-values are 0.197, 0.325, and 0.26, respectively. Differences between these approaches can arise because the MMRM uses a model and repeated-measures information that the summary-statistic recalculation does not reproduce.

What do the planning illustrations mean?

The planning illustrations use the observed effect as a reference and show hypothetical participants per arm for 80% power and for 90% power under stated assumptions. Those assumptions are a two-sided significance level of 0.05, equal allocation, a normal approximation, and an effect to detect defined as a stated fraction of the observed effect. General statistical explanation: planning calculations ask how much information a future study might need to detect a chosen effect under specified assumptions. They do not show the power achieved by the completed trial, and the figures are not observed participant counts or a record of the original trial’s design justification.

Planning illustrations based on an observed effect are especially dependent on that estimate. If the observed effect differs from the effect a future study aims to detect, the planning requirements will also differ. The registry’s own sample-size justification and assumptions are not reported in the registry record. The illustrations should therefore be used as conditional planning examples, not as evidence that the completed study was adequately or inadequately powered.

What analysis population is represented?

The primary analysis population included randomized participants who received at least one dose, excluding participants from 2 sites because of a GCP violation. Participants were analyzed according to their randomized assignment. This defines the population behind the primary reported analysis and should be kept distinct from enrollment. The registry does not report detailed participant flow, the number of participants with outcome data at each assessment, or reasons for missing observations. The safety denominators are reported separately and should not be assumed to match the primary analysis population.

10. Analyses Reported by the Sponsor

The primary estimate is the registry’s reported negative binomial analysis: a rate ratio of 0.52, with a 95% confidence interval from 0.36 to 0.73 and p < 0.001. No recalculated primary comparison is reported, so there is no unadjusted primary estimate to set alongside the sponsor’s model-based result.

For the posted continuous secondary outcomes, the recalculated estimates are simple unadjusted comparisons from summary statistics, while the registry’s own results use mixed models repeated measures. The recalculations are approximate and do not incorporate the sponsor’s full repeated-measures modeling. Their purpose is to provide a transparent, simplified comparison, not to supersede the analyses reported by the study sponsor.

The posted MMRM estimates are close to the recalculated mean differences, and the corresponding confidence intervals and p-values are also broadly similar. Small differences are expected when different methods use different information and assumptions. The record does not flag a discrepancy between the analyses. In interpreting the trial, the sponsor’s reported analyses remain the primary results; the recalculated values should be treated as secondary checks, not as corrected results or independent confirmation.

11. Multiplicity, Missing Data and Analysis Populations

The registry record reports the primary endpoint and several secondary outcomes, but a multiplicity strategy is not reported in the registry record. In particular, the record does not state whether secondary hypotheses were tested in a prespecified sequence, whether p-values were adjusted for multiple comparisons, or which outcomes were intended to support confirmatory claims. This limits how strongly individual secondary p-values can be interpreted when considered together.

General statistical explanation: testing multiple outcomes can increase the chance of at least one small p-value even when no true differences exist. A prespecified hierarchy or adjustment can control that risk, but the record does not describe such a method. The primary result and the secondary findings therefore have different interpretive roles: the primary comparison addresses the stated main question, while the secondary results provide additional outcome-specific information whose confirmatory status is unclear.

Missing-data methods are also not reported in the registry record. The SGRQ, ACQ-5, and FEV1 outcomes use mixed models repeated measures, but the registry does not give the model details, assumptions about missingness, or sensitivity analyses. The record does not state how missing primary outcome data were handled, or whether any participants were censored or otherwise excluded from particular endpoint analyses. Consequently, the extent to which missing observations could affect the estimates cannot be evaluated from the posted information.

The primary analysis population is described as randomized participants who received at least one dose, excluding participants from 2 sites because of a GCP violation, and analyzed according to randomized assignment. The registry does not report detailed participant flow or explain the site-level issue further. The reported safety denominators differ by group and are presented separately; they should not be treated as the primary efficacy population. The record does not report interim analyses, a sample-size justification, or further statistical plan details. Those omissions prevent a full assessment of the planned testing strategy and the assumptions underlying all analyses.

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 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: asthma (4.7%), pneumonia (1.6%), coronary artery disease (0.8%), phimosis (0.8%), anaphylactic reaction (0.8%).

The registry reports serious adverse events affecting 19 of 251 participants at risk in the GSK3511294 group and 13 of 129 participants at risk in the placebo group. Other adverse events affected 138 of 251 participants at risk in the GSK3511294 group and 80 of 129 participants at risk in the placebo group. Deaths were reported as 0 in each group. These counts describe the posted safety summaries; they are not, by themselves, a complete comparison of safety risk.

The most frequently listed serious adverse event term was asthma, reported as 4.7%. Other listed terms were pneumonia at 1.6%, coronary artery disease at 0.8%, phimosis at 0.8%, anaphylactic reaction at 0.8%, and sepsis at 0.8%. The record does not provide further context about event timing, severity, relatedness to intervention, or whether participants experienced more than one event. The listed terms should not be added together or treated as mutually exclusive categories.

General statistical explanation: adverse event counts need interpretation alongside exposure, follow-up, event definitions, and uncertainty. The registry summary does not establish that an event was caused by the assigned intervention, and the counts alone do not establish comparative safety. Further details about safety analysis methods and event attribution are not reported in the registry record.

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

The primary result means that, in the registry’s reported negative binomial analysis of the defined full analysis set, the estimated annualized rate of clinically significant exacerbations was lower in the GSK3511294 group than in the placebo group through Week 52. The reported rate ratio was 0.52, with a 95% confidence interval from 0.36 to 0.73 and p < 0.001. The interval indicates statistical uncertainty around the estimate, while remaining below the value corresponding to equal rates.

This is a comparison of group-level rates under the study conditions. It does not mean that every participant in the GSK3511294 group had fewer exacerbations, that no participant in the placebo group had a lower rate, or that the result predicts an individual outcome. It also does not describe all possible asthma outcomes: the primary endpoint counts events meeting a specific definition, including rules for steroid use, hospitalization or emergency department visits, and closely spaced events.

The result does not establish that the same estimate would apply to people outside the study population, under different care conditions, or beyond the reported timeframe. The intervention was studied as an addition to existing baseline maintenance asthma standard of care, and the endpoint covered up to Week 52. General statistical context: statistical evidence and clinical importance are related but distinct questions. The registry provides an estimate and its uncertainty, but it does not report a threshold for a clinically important difference or enough context to determine how the result translates into individual experience.

The secondary outcomes provide additional measures, but their posted comparisons do not show the same clear statistical pattern as the primary result, and detailed results for all listed secondary outcomes are not reported in the registry record. Safety summaries also require cautious interpretation: counts of adverse events do not by themselves establish causation or settle the balance of benefits and harms. The result should be understood as evidence from this particular randomized comparison, with the limits created by the available registry information.

14. Why This Trial Matters Statistically

This trial illustrates why endpoint choice and statistical model matter in clinical research. The primary outcome counts clinically significant exacerbations over follow-up and expresses them as an annualized rate. Because participants may experience recurrent events, the analysis addresses more than whether someone experienced any event. The negative binomial method is suited to count data that may show greater variability than a simpler count model assumes, although the registry does not report enough model detail to assess its specification.

The randomized, double-blind, placebo-controlled, parallel-group design creates a structured comparison of an adjunctive intervention with placebo while participants continue existing maintenance care. The primary result is reported with an effect estimate, confidence interval, and p-value, enabling readers to consider both the direction and precision of the group comparison rather than relying on a significance label alone.

The record also demonstrates the distinct roles of primary and secondary outcomes. The primary exacerbation-rate comparison has a confidence interval below 1, while the available secondary comparisons for continuous measures have intervals that include no difference. Those outcomes address different domains and should not be collapsed into a single claim. The lack of reported multiplicity procedures and detail for some secondary outcomes makes restraint in interpreting those comparisons especially important.

Finally, the recalculations for continuous outcomes show why a transparent summary-statistic check can be informative but cannot replace a trial’s model-based analysis. The two approaches are close, yet they do not use identical assumptions or information. The planning illustrations likewise serve a different purpose from the trial results: they are hypothetical planning aids based on observed effects, not observed power or evidence about the original design. Together, these features make the study a useful example of reading estimates, uncertainty, analysis populations, and limitations as parts of one statistical interpretation.

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 gives the primary effect estimate and analysis method but does not report the negative binomial model specification, covariates, diagnostic checks, or sensitivity analyses. The precision and validity of the primary estimate therefore cannot be fully assessed from the registry description alone. The endpoint definition is detailed, but the record does not describe additional event adjudication procedures beyond the stated verification rule.

The primary analysis excluded participants from 2 sites because of a GCP violation. The record does not explain the nature of the violation, the number of affected participants, or how the exclusion may have influenced the analysis. It also does not provide detailed participant flow, missing outcome counts, or methods for handling missing primary data. These omissions make it difficult to assess how complete follow-up was and whether missingness could have affected the estimated results.

For secondary endpoints, the registry identifies MMRM for the reported continuous outcomes but does not give details about model structure, missing-data assumptions, or sensitivity analyses. Detailed numerical results for all listed secondary outcomes are not reported in the registry record. Multiplicity handling is also not reported, limiting interpretation of secondary p-values as confirmatory evidence.

The safety summary reports counts and selected serious adverse event terms, but the record does not provide event-level detail, relatedness assessments, timing, or methods for comparing safety between groups. No causal conclusion should be drawn from these counts alone. Further, the planning illustrations use observed effects and stated hypothetical assumptions; they do not document the original sample-size justification or the study’s achieved power. These trial-specific reporting gaps constrain how completely the efficacy, safety, and design can be evaluated from the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of GSK3511294 (Depemokimab) in Participants With Severe Asthma With an Eosinophilic Phenotype — NCT04718103.
• Registry record last updated 2024-11-29; status last verified 2024-11.
• Results first posted 2024-11-29.
• 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

NCT04718103 is a completed Phase 3 randomized, double-blind, placebo-controlled, parallel-group, multicentre study of GSK3511294 as adjunctive therapy for adults and adolescents with severe uncontrolled asthma and an eosinophilic phenotype. Participants received GSK3511294 or placebo by subcutaneous injection once every 26 weeks, while continuing baseline maintenance asthma standard of care. The registry reports enrollment of 397.

The primary endpoint was the annualized rate of clinically significant exacerbations up to Week 52. The reported negative binomial comparison gave a rate ratio of 0.52, with a 95% confidence interval from 0.36 to 0.73 and p < 0.001. The registry also lists six secondary outcomes, with detailed comparison estimates posted for three continuous measures. Their sponsor-reported confidence intervals include no difference between groups.

The primary analysis population excluded participants from 2 sites because of a GCP violation. The registry does not report the multiplicity strategy, detailed missing-data methods, or additional model specifications. Safety counts and selected serious adverse event terms are posted, but they do not by themselves establish causation or comparative safety.