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 Study of Efficacy and Safety of QAW039 in Patients With Severe Asthma Inadequately Controlled With Standard of Care Asthma Treatment. (NCT02563067), a randomized phase 3 study. The primary outcome is Rate of Moderate-to-severe Asthma Exacerbations During the 52-week Treatment Period in High Eosinophils Subpopulation.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02563067 |
| Phase | Phase 3 |
| Condition | Asthma |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 877 |
| Arms in the primary outcome | 3 |
| Lead sponsor | Novartis Pharmaceuticals |
| Results posted | Yes |
2. Clinical Question
Population
People with inadequately controlled severe asthma receiving GINA 2016 steps 4 and 5 standard-of-care asthma therapy. The primary endpoint focused on the Full Analysis Set in the high-eosinophil subpopulation, defined by a baseline blood eosinophil count of at least 250 cells/μL.
Intervention
QAW039 once daily at either 150 mg or 450 mg, added to existing asthma therapy.
Comparator
Placebo once daily, added to existing asthma therapy.
Primary question
During the 52-week treatment period, how did the rate of moderate-to-severe asthma exacerbations in the high-eosinophil subpopulation compare between each QAW039 dose and placebo?
The registry describes two study populations: people with inadequately controlled severe asthma and a high-eosinophil subpopulation, and the overall study population. Inadequate control was defined as partly controlled or uncontrolled asthma under GINA 2016. The primary endpoint was specified for the high-eosinophil subpopulation rather than the full overall study population. That distinction matters when interpreting the estimates: they address the population defined for this endpoint, not automatically every person with severe asthma.
The clinical question is comparative. It concerns whether the estimated exacerbation rate differed between each dose group and placebo while participants continued their existing therapy. The registry describes the study aim as assessing efficacy and safety; the posted primary results quantify the exacerbation rate and the two rate-ratio comparisons. The registry record does not report a result for every possible clinical subgroup or a conclusion about all people with asthma.
3. Trial Design
This was a Phase 3, randomized, double-blind, parallel-group, placebo-controlled study sponsored by Novartis Pharmaceuticals. The registry reports enrollment of 877 and a study period from 2015-12-03 to primary completion on 2019-07-04. The status is completed. The official title describes a 52-week study of QAW039 added to existing asthma therapy in people with uncontrolled severe asthma.
Participants were assigned to QAW039 150 mg once daily, QAW039 450 mg once daily, or placebo once daily. Each was added to GINA 2016 steps 4 and 5 standard-of-care asthma therapy. Randomization is intended to make treatment groups comparable on average, while masking can reduce the influence of treatment knowledge on care and outcome assessment. These are general features of the design; the registry record does not report the specific randomization procedure or exactly who was masked.
The primary outcome concerned the Full Analysis Set in the high-eosinophil subpopulation, defined at baseline by blood eosinophils of at least 250 cells/μL. The endpoint covered the 52-week treatment period. The arm-level analysis counts posted for that endpoint are 198 for QAW039 150 mg, 196 for QAW039 450 mg, and 193 for placebo. Those endpoint analysis counts should not be confused with the overall enrollment figure or the safety counts, which use different arm-level figures.
The registry lists nine outcome measures and 16 analyses. It also lists secondary outcomes covering asthma-related quality of life, asthma control, and lung function in both the high-eosinophil subpopulation and the overall population. The detailed protocol, rationale for the planned enrollment, and statistical analysis plan 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: The study included a Screening period of up to 2 weeks and a Placebo Run-in period of 2 to 6 weeks, during which eligibility for randomization was determined
Overall Study
| Stage | QAW039 150 mg | QAW039 450 mg | Placebo |
|---|---|---|---|
| STARTED | 296 | 294 | 287 |
| FAS/SAF: High Eosinophils Subpopulation | 198 | 196 | 193 |
| FAS/SAF: Overall Population | 296 | 293 | 287 |
| COMPLETED | 277 | 268 | 267 |
| NOT COMPLETED | 19 | 26 | 20 |
Reasons for not completing this period:
| Reason | QAW039 150 mg | QAW039 450 mg | Placebo |
|---|---|---|---|
| Subject/guardian decision | 14 | 20 | 16 |
| Physician Decision | 0 | 1 | 3 |
| Lost to Follow-up | 1 | 2 | 0 |
| Technical problems | 0 | 1 | 0 |
| Lack of Efficacy | 1 | 0 | 0 |
| Death | 0 | 1 | 1 |
| Adverse Event | 2 | 1 | 0 |
| Death post Treatment/safety period | 1 | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (FAS) in high eosinophils subpopulation (count ≥ 250 cells/μL)
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.
| Characteristic | Unit | QAW039 150 mg | QAW039 450 mg | Placebo | Total |
|---|---|---|---|---|---|
| Participants analysed | 296 | 294 | 287 | 877 | |
| Age, Continuous | years | 49.8 (SD 14.60) | 49.6 (SD 15.25) | 50.9 (SD 14.16) | 50.1 (SD 14.68) |
| Sex: Female, Male: Female | Participants | 165 | 178 | 177 | 520 |
| Sex: Female, Male: Male | Participants | 131 | 116 | 110 | 357 |
| Race/Ethnicity, Customized: Caucasian | Participants | 234 | 235 | 227 | 696 |
| Race/Ethnicity, Customized: Black | Participants | 6 | 4 | 1 | 11 |
| Race/Ethnicity, Customized: Asian | Participants | 44 | 41 | 41 | 126 |
| Race/Ethnicity, Customized: Native American | Participants | 2 | 2 | 4 | 8 |
| Race/Ethnicity, Customized: Unknown | Participants | 2 | 2 | 3 | 7 |
| Race/Ethnicity, Customized: Other | Participants | 8 | 10 | 11 | 29 |
6. Primary Endpoint
The registry lists the primary outcome as Rate of Moderate-to-severe Asthma Exacerbations During the 52-week Treatment Period in High Eosinophils Subpopulation, measured over 52 weeks.
| Element | Registry entry |
|---|---|
| Result type | Least Squares Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | Events/year |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is the rate of moderate-to-severe asthma exacerbations during the 52-week treatment period in the high-eosinophil subpopulation. It is expressed in events per year and reported as least-squares means with 95% confidence intervals. This is a rate outcome: it summarizes exacerbation events over the treatment period rather than simply reporting whether each participant had any exacerbation.
The registry defines a severe exacerbation as rescue systemic corticosteroid treatment for at least 3 days together with hospitalization, or such treatment together with an emergency department visit lasting more than 24 hours, or death due to asthma. A moderate exacerbation is rescue systemic corticosteroid treatment for at least 3 days as an outpatient or during an emergency department visit lasting 24 hours or less. The high-eosinophil group is defined by a baseline blood eosinophil count of at least 250 cells/μL.
Exacerbation rate is used to capture repeated events over time, which can be clinically relevant when a participant may experience more than one exacerbation during follow-up. As a general statistical point, a rate analysis accounts for event frequency over exposure time, rather than treating every participant as having only a yes-or-no outcome. The posted outcome does not by itself describe the severity or timing of every individual event. Detailed event-level information and the handling of incomplete follow-up are not reported in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
In the high-eosinophil subpopulation, the posted least-squares mean exacerbation rate was 0.73 events/year for QAW039 150 mg, with a 95% confidence interval from 0.57 to 0.94. For QAW039 450 mg, it was 0.76 events/year, with a 95% confidence interval from 0.59 to 0.99. For placebo, it was 1.06 events/year, with a 95% confidence interval from 0.84 to 1.33. These are estimated rates, not counts of events or proportions of participants experiencing an exacerbation.
The registry reports negative binomial regression comparisons against placebo. For QAW039 150 mg, the rate ratio was 0.69, with a 95% confidence interval from 0.50 to 0.96 and p = 0.059. For QAW039 450 mg, the rate ratio was 0.72, with a 95% confidence interval from 0.52 to 1.01 and p = 0.114. A rate ratio below 1 indicates a lower estimated rate in the QAW039 group relative to placebo under this model. The confidence intervals express uncertainty around each estimated ratio; the interval for the 450 mg comparison includes 1, while the interval for the 150 mg comparison does not.
The p-values and confidence intervals are not identical summaries and should be read together. A p-value describes how compatible the data are with a specified null hypothesis under the model and its assumptions; it is not the probability that the treatment has no effect. A confidence interval conveys a range of values compatible with the data and model at its stated confidence level. The registry record does not report the decision threshold, any multiplicity adjustment, or the sponsor's rationale for the apparent relationship between the 150 mg confidence interval and its p-value. The posted result therefore does not justify resolving that tension by selecting only one statistic.
These findings are confined to the primary endpoint and the defined high-eosinophil population. They do not establish what the rate comparison would be in the overall population, nor do they describe every secondary outcome. The estimates are model-based and depend on the negative binomial regression analysis. The registry record does not provide the model specification beyond its name, or detailed information about covariates, exposure time, missing observations, or sensitivity analyses. Those omissions limit independent assessment of the analysis and its assumptions.
The arm-level mean rates and the rate-ratio comparisons answer related but different questions. The former describe estimated rates within each arm; the latter compare each active-treatment arm with placebo under the stated model. Neither set of estimates alone provides a complete account of individual participant experiences, the distribution of events, or treatment safety. Statistical interpretation should remain tied to the endpoint, population, and analysis reported.
9. Statistical Methods Explained
Why model an event rate?
The primary outcome counts asthma exacerbations across a treatment period and expresses the result as an annualized rate. A participant can contribute more than one event, so a rate model is suited to comparing event frequency rather than only classifying participants as having or not having an event. The registry reports use of a negative binomial regression model for the rate-ratio comparisons. In general, negative binomial models are used for count data when variability is greater than a simple Poisson model would allow; this is called over-dispersion. Whether and how over-dispersion was assessed in this trial is not reported in the registry record.
The registry also reports least-squares mean rates with 95% confidence intervals. Least-squares means are model-based estimates, not necessarily raw arithmetic averages. The record does not report enough detail to describe their exact adjustment or calculation. The posted arm-level estimates are 0.73, 0.76, and 1.06 events/year for QAW039 150 mg, QAW039 450 mg, and placebo, respectively. Their confidence intervals describe uncertainty in the estimated rates under the analysis.
What does a rate ratio describe?
A rate ratio compares the estimated event rate in one arm with the rate in another arm. The placebo group is the comparator for both active-treatment analyses. A ratio of 1 would represent equal estimated rates; a value below 1 represents a lower estimated rate in the numerator arm. This is a general interpretation of the measure, not a statement about the cause of any observed difference. The reported comparisons are 0.69 for QAW039 150 mg versus placebo and 0.72 for QAW039 450 mg versus placebo. Their confidence intervals are 0.50 to 0.96 and 0.52 to 1.01, respectively.
How should uncertainty be read?
A 95% confidence interval is a measure of uncertainty around an estimate. In general, under repeated use of the same valid procedure, intervals constructed this way would cover the underlying parameter at the stated rate. It is not a range that guarantees where the true value lies in this particular study, nor does it describe the range of individual participant outcomes. The width of an interval reflects both the data and the analysis method. A p-value, in contrast, summarizes how unusual data at least as incompatible with a null hypothesis would be if that hypothesis and the model assumptions were true. Neither measure alone establishes clinical importance.
The reported p-values for the primary comparisons are 0.059 and 0.114. The registry record does not state the prespecified significance threshold or explain how the p-values relate to the corresponding confidence intervals. The sample-size planning description elsewhere in the record uses a two-sided significance level of 0.05, but that planning illustration is not enough to establish the formal testing rules for the primary analysis.
Why can recalculations differ?
For secondary continuous outcomes, the registry posts least-squares mean changes and standard errors. The unadjusted recalculations use those posted summary statistics to approximate a mean difference, confidence interval, and p-value. They do not reproduce the sponsor's ANCOVA model, which may use participant-level data and specified covariates. Standard errors are used to approximate variability; the posted arm-level summaries are not a substitute for the full dataset. The recalculated results are therefore descriptive checks rather than replacements for the sponsor's model-based analyses.
For example, a confidence interval formed from summary statistics can differ slightly from a sponsor's interval because of rounding, model adjustment, or other analysis details. The registry record does not provide enough detail to determine the precise source of each difference. Where the sponsor's analysis is reported, it remains the relevant result of that stated analysis; a simple unadjusted recalculation answers a different, more limited question.
What do the planning illustrations mean?
The sample-size illustrations are based on observed secondary-outcome effects, not on the trial's design or its observed power. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The illustrations show participants per arm for 80% power and participants per arm for 90% power when planning for 100%, 75%, or 50% of the observed effect. These are hypothetical planning calculations. They are not estimates of the power achieved in this completed trial, and they do not show that a particular future study would have the stated power under different assumptions.
Observed-effect planning can be unstable because an estimate from one study is uncertain. A smaller assumed effect requires a larger planning sample under the stated calculation framework. This is a general statistical relationship; the illustrations do not establish a clinically meaningful target effect. The registry record does not report the sponsor's original sample-size justification, assumptions about attrition, or any planned interim analyses.
What is not specified?
The registry identifies the primary analysis population as the Full Analysis Set in the high-eosinophil subpopulation. It does not report the exact analysis-set rules, treatment of participants who discontinued, missing-data procedures, model covariates, or sensitivity analyses. These details can affect how robustly the result represents the randomized groups and how dependent it is on assumptions. The registry record also does not report whether over-dispersion was assessed or whether the primary comparisons were adjusted for multiple testing. These are limitations of the available statistical description, not proof that the trial did or did not use particular procedures.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| negative binomial regression model | Rate Ratio 0.69 | 0.50 to 0.96 | 0.059 |
| negative binomial regression model | Rate Ratio 0.72 | 0.52 to 1.01 | 0.114 |
The primary results are reported as negative binomial regression rate ratios, with 95% confidence intervals and p-values. There is no recalculated primary comparison in the posted secondary-outcome summaries, so the primary interpretation rests on the sponsor's reported model-based results: a rate ratio of 0.69 for QAW039 150 mg versus placebo and 0.72 for QAW039 450 mg versus placebo.
For the secondary continuous outcomes, the sponsor reports ANCOVA estimates. The recalculations are simple unadjusted comparisons based on the posted summary statistics and use approximations; they are not a reproduction of ANCOVA. For example, the AQLQ+12 recalculated mean differences are 0.15 and 0.13, while the sponsor's reported mean differences are also 0.15 and 0.13 after rounding. The recalculated intervals and p-values need not match the model-based results. The registry reports sponsor p-values of 0.369 and 0.824 for those comparisons, while the unadjusted recalculated p-values are 0.085 and 0.138.
Such differences do not establish that either analysis is erroneous: they address results using different inputs and methods. The recalculations are best treated as approximate, unadjusted checks, while the sponsor's reported analysis describes the stated ANCOVA model. The registry record does not provide the detailed model specification needed to reconcile all differences.
11. Multiplicity, Missing Data and Analysis Populations
The registry lists multiple outcome measures and analyses, including two primary rate-ratio comparisons and secondary measures across more than one population. When several hypotheses are tested, the chance of at least one apparently unusual result can increase. This is the general multiplicity problem. A prespecified testing hierarchy or adjustment can address it, but the registry record does not report whether such a strategy was used, which tests were included, or how conclusions were prioritized.
The primary endpoint is defined for the Full Analysis Set in the high-eosinophil subpopulation. The record does not describe the operational rules for inclusion in that set, such as how participants were handled after treatment discontinuation or major protocol deviations. It also does not state whether the efficacy analysis followed randomized assignment regardless of treatment received. These details are not reported in the registry record and should not be inferred from the population label alone.
Missing outcome information can affect estimates, particularly when participants do not complete follow-up or have incomplete event reporting. The registry record does not describe the amount or pattern of missing data, imputation or other missing-data methods, assumptions about missingness, or sensitivity analyses. For a rate endpoint, exposure time and follow-up also matter; the registry identifies a 52-week treatment period but does not provide enough detail to assess how incomplete observation was handled in the rate model.
The record names negative binomial regression for the primary comparisons, but does not report the full model specification, covariates, dispersion assessment, or model diagnostics. The secondary outcomes are reported with ANCOVA, but the complete covariate structure and missing-data method are not reported. Nor are interim analyses or the full statistical plan reported in the registry record.
These omissions limit the ability to assess how the reported results would change under alternative analysis assumptions. They do not show that an appropriate method was absent; they mean that the registry record does not give enough information to evaluate those decisions. The estimates should therefore be interpreted as the posted analyses, with uncertainty about unreported methods kept explicit.
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 |
|---|---|---|---|
| QAW039 150 mg | 20 | 296 | 6.8% |
| QAW039 450 mg | 20 | 293 | 6.8% |
| Placebo | 18 | 287 | 6.3% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| QAW039 150 mg | 0 of 296 |
| QAW039 450 mg | 1 of 293 |
| Placebo | 1 of 287 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 4% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| QAW039 150 mg | 181 of 296 |
| QAW039 450 mg | 183 of 293 |
| Placebo | 180 of 287 |
Most frequent serious adverse event terms
| Event term | Organ system | QAW039 150 mg | QAW039 450 mg | Placebo |
|---|---|---|---|---|
| Asthma | Respiratory, thoracic and mediastinal disorders | 8 of 296 | 6 of 293 | 6 of 287 |
| Pneumonia | Infections and infestations | 2 of 296 | 4 of 293 | 3 of 287 |
| Appendicitis | Infections and infestations | 1 of 296 | 0 of 293 | 1 of 287 |
| Bronchitis | Infections and infestations | 0 of 296 | 1 of 293 | 1 of 287 |
| Gastroenteritis | Infections and infestations | 0 of 296 | 1 of 293 | 1 of 287 |
| Nasal polyps | Respiratory, thoracic and mediastinal disorders | 0 of 296 | 2 of 293 | 0 of 287 |
| Anaemia | Blood and lymphatic system disorders | 0 of 296 | 0 of 293 | 1 of 287 |
| Arteriosclerosis coronary artery | Cardiac disorders | 0 of 296 | 0 of 293 | 1 of 287 |
| Ulcerative keratitis | Eye disorders | 1 of 296 | 0 of 293 | 0 of 287 |
| Abdominal discomfort | Gastrointestinal disorders | 0 of 296 | 0 of 293 | 1 of 287 |
Most frequent non-serious adverse event terms
| Event term | Organ system | QAW039 150 mg | QAW039 450 mg | Placebo |
|---|---|---|---|---|
| Asthma | Respiratory, thoracic and mediastinal disorders | 129 of 296 | 119 of 293 | 145 of 287 |
| Nasopharyngitis | Infections and infestations | 33 of 296 | 42 of 293 | 26 of 287 |
| Bronchitis | Infections and infestations | 37 of 296 | 30 of 293 | 32 of 287 |
| Viral upper respiratory tract infection | Infections and infestations | 20 of 296 | 28 of 293 | 24 of 287 |
| Headache | Nervous system disorders | 19 of 296 | 21 of 293 | 9 of 287 |
| Influenza | Infections and infestations | 17 of 296 | 18 of 293 | 13 of 287 |
| Upper respiratory tract infection | Infections and infestations | 18 of 296 | 16 of 293 | 14 of 287 |
| Rhinitis | Infections and infestations | 10 of 296 | 15 of 293 | 13 of 287 |
| Upper respiratory tract infection bacterial | Infections and infestations | 10 of 296 | 8 of 293 | 14 of 287 |
| Rhinitis allergic | Respiratory, thoracic and mediastinal disorders | 13 of 296 | 8 of 293 | 9 of 287 |
Serious events with the highest arm-level rate: asthma (2.7%), pneumonia (1.4%), nasal polyps (0.7%), anaemia (0.3%), arteriosclerosis coronary artery (0.3%).
The safety figures are reported by treatment arm, with 20 participants affected by serious adverse events in each QAW039 arm and 18 in the placebo arm. The corresponding at-risk figures are 296, 293, and 287. The registry also reports other adverse events affecting 181, 183, and 180 participants, respectively. These safety figures use arm counts different from the primary endpoint analysis counts, which are for the high-eosinophil subpopulation. The registry record does not explain the safety population rules in detail.
The listed serious adverse event terms include asthma at 2.7%, pneumonia at 1.4%, nasal polyps at 0.7%, and anaemia, arteriosclerosis coronary artery, and abdominal discomfort at 0.3% each. These terms describe reported events, not proof that study treatment caused them. The registry records deaths of 0 in the QAW039 150 mg arm, 1 in the QAW039 450 mg arm, and 1 in the placebo arm. The record does not state that these deaths were due to asthma or attribute them to treatment.
Counts and term percentages should be read as safety reporting, not as a comparative causal analysis. The registry does not provide detailed event timing, exposure-adjusted rates, or a full account of seriousness and relatedness assessments. A simple comparison of these figures cannot establish that one regimen is safer.
13. What the Result Does, and Does Not, Mean
The primary results show lower estimated exacerbation rates in both QAW039 arms than in the placebo arm within the high-eosinophil subpopulation. The reported rate ratios are 0.69 for QAW039 150 mg and 0.72 for QAW039 450 mg. Their confidence intervals and p-values indicate uncertainty, and the registry does not provide enough statistical-plan detail to establish how the sponsor intended to interpret every test. These estimates describe the posted analysis; they are not a guarantee of what any individual would experience.
The endpoint concerns moderate-to-severe exacerbations over a specified treatment period. It does not capture every aspect of asthma control, quality of life, lung function, adverse events, or long-term outcomes. Secondary measures add context but address different questions, and their unadjusted recalculations are not confirmatory evidence. The primary result applies to the defined high-eosinophil population; it should not automatically be generalized to the overall study population or other groups.
A confidence interval is not a forecast for an individual participant, and a p-value is not the probability that a result is true or false. The registry record also does not provide detailed missing-data procedures, the full model specification, or a multiplicity strategy. Those limits matter when evaluating how robustly the posted estimates support broader conclusions.
Accordingly, the statistical result is a comparison of model-estimated event rates in a particular randomized trial population. It does not by itself establish the size or importance of a benefit for every patient, resolve safety questions, or replace a complete assessment of the trial protocol and analysis plan. No medical advice follows from these statistics alone.
14. Why This Trial Matters Statistically
This trial illustrates why the outcome definition and analysis population matter as much as the treatment labels. The primary endpoint is a recurrent-event rate, not a proportion with at least one exacerbation, and it is restricted to a baseline-defined high-eosinophil subpopulation. Reading the result as though it represented all participants or all people with severe asthma would change the question the data answer.
It also shows why a single summary number is not enough. Arm-level estimated rates, rate ratios, confidence intervals, and p-values each communicate different aspects of the result. The rate ratios compare groups; the intervals display uncertainty; and the p-values summarize compatibility with a null hypothesis under a model. Their interpretation depends on the statistical plan and assumptions, several of which are not reported in the registry record.
The trial also includes multiple secondary outcomes across populations. These offer a broader view of measured domains, but they raise multiplicity concerns and do not replace the primary endpoint. Unadjusted recalculations from summary statistics are useful for orientation, while remaining distinct from the sponsor's model-based analyses.
Finally, the safety counts show why efficacy and safety must be read separately and with attention to the populations and denominators used. A registry record can support a careful account of posted estimates while still leaving important questions about missing data, model details, and analysis decisions unanswered. Distinguishing what the estimates show from what is not reported is central to sound clinical-trial 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 limited detail about the methods behind the primary endpoint. It names negative binomial regression but does not report the full model specification, covariates, diagnostics, or the handling of incomplete follow-up. For a recurrent-event rate, the exposure and observation time used in the analysis can affect interpretation, but those details are not reported in the registry record.
The primary endpoint is focused on the high-eosinophil subpopulation, defined by a baseline blood eosinophil count of at least 250 cells/μL. The results should not be treated as a direct estimate for the overall population. Although secondary outcomes are listed for both populations, the registry record does not provide the complete analysis plan or a full set of detailed results across all outcomes in this account.
The record also does not state whether the primary comparisons were adjusted for multiple testing, how missing data were handled, or what sensitivity analyses were conducted. The reported primary confidence intervals and p-values have a relationship that the registry record does not explain. These are reasons for caution when drawing inferential conclusions from the abbreviated record, not evidence of a specific flaw.
The secondary recalculations are based on rounded summary statistics and approximate standard deviations. They are unadjusted and do not reproduce the sponsor's ANCOVA analyses. They therefore cannot resolve differences in model-based results or account for participant-level data. The safety summaries also lack detail on event timing, exposure, and causality assessments. These limitations constrain interpretation beyond the posted estimates.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Study of Efficacy and Safety of QAW039 in Patients With Severe Asthma Inadequately Controlled With Standard of Care Asthma Treatment. — NCT02563067.
- Registry record last updated 2021-05-03; status last verified 2021-04.
- Results first posted 2021-05-03.
- 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
NCT02563067 is a completed Phase 3 randomized, double-blind, parallel-group, placebo-controlled study sponsored by Novartis Pharmaceuticals. It enrolled 877 people with inadequately controlled severe asthma and assessed QAW039 150 mg once daily and QAW039 450 mg once daily, each added to existing therapy, against placebo once daily.
The primary endpoint was the rate of moderate-to-severe asthma exacerbations during the 52-week treatment period in the high-eosinophil subpopulation. The registry reports lower estimated rates in both QAW039 arms than placebo, with rate ratios of 0.69 and 0.72. The confidence intervals and p-values convey uncertainty, and key analysis details, including multiplicity and missing-data methods, are not reported in the registry record.
Secondary measures addressed quality of life, asthma control, and lung function. Safety summaries report serious adverse events, other adverse events, and deaths by arm. The posted findings are specific to their outcome definitions and analysis populations and should be interpreted alongside the registry's statistical detail and limitations.