Study to Compare QVM149 and Free Triple Combination of Salmeterol/Fluticasone + Tiotropium: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Compare QVM149 and Free Triple Combination of Salmeterol/Fluticasone + Tiotropium (NCT03158311), a randomized phase 3 study. The primary outcome is Change From Baseline in Asthma Quality of Life Questionnaire (AQLQ) Total Score.

2. Clinical Question

The comparison concerns two distinct QVM149 dose groups rather than a single pooled QVM149 group. The registry record lists change in AQLQ total score at Baseline and Week 24 as the primary outcome. It also posts a Week 16 AQLQ analysis and other secondary outcomes, including lung-function measures and asthma-control scores. Those additional results address related but different questions; they do not replace the primary Week 24 outcome.

3. Trial Design

This was a Phase 3, randomized, parallel-group active-controlled study in asthma. The official title describes a multicenter, partially-blinded, open-label study with a 24-week treatment period and a non-inferiority objective. The registry’s design fields list randomized allocation, a parallel model, and single masking. The difference in masking descriptions does not establish which participants or study personnel were masked; that detail is not reported in the registry record.

The registry record lists 1426 participants enrolled. The three treatment groups were QVM149 150/50/80 μg, QVM149 150/50/160 μg, and salmeterol/fluticasone plus tiotropium. The planned treatment duration was 24 weeks, followed by a 7-day follow-up period. The record lists the study as completed, with a start date of 2018-02-05 and a primary completion date of 2019-07-12.

The primary endpoint was measured at baseline and Week 24. The registry reports sponsor analyses using a Mixed Model for Repeated Measures (MMRM), with separate non-inferiority comparisons for the two QVM149 groups against the active comparator. The record also includes descriptive arm-level endpoint results and unadjusted recalculations based on posted summary statistics. These are different analyses: the sponsor’s model-based comparisons are the trial’s reported analyses, while the recalculations are not a substitute for them.

The registry record does not report the original sample-size justification, the planned non-inferiority margin, or the detailed statistical analysis plan. It also does not report the model’s covariates or covariance structure. Those omissions limit how fully the design and the sponsor’s non-inferiority decision can be reconstructed from registry results alone.

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 1917 participants were screened. A 2 week run-in phase preceded the randomized treatment phase of the study. 1821 participants entered the run-in phase. Those participants who met the inclusion criteria entered the randomized treatment phase. Participants were randomized with a randomization ratio of 1:1:1 to three treatment arms.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analysis population included FAS with evaluable AQLQ data at both timepoints. No imputation was used for missing questions or missing total AQLQ score for this analysis. The analysis was performed on observed data only.

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.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Asthma Quality of Life Questionnaire (AQLQ) Total Score, measured over Baseline and Week 24.

The primary endpoint was change from baseline in Asthma Quality of Life Questionnaire (AQLQ) total score at Week 24. The AQLQ is a 32-item asthma-specific questionnaire intended to measure functional impairments important to people with asthma. It asks about the preceding two weeks. Each item uses a 7-point scale, from 1, “totally limited/problems all the time,” to 7, “not at all limited/no problems.” The overall score is the mean of the 32 items and covers symptoms, activity limitation, emotional function, and emotional stimuli.

Higher scores indicate better health-related quality of life. A change from baseline therefore describes the direction and extent of change in the questionnaire score during treatment, rather than an objective lung-function measurement. Using a patient-reported measure is relevant to a question about asthma-related functioning and experience; the questionnaire captures several aspects of daily impact in a single overall score.

The primary analysis population consisted of participants in the Full Analysis Set with evaluable AQLQ data at both timepoints. The record says no imputation was used for missing questions or missing total AQLQ scores in this analysis, which was performed on observed data only. The sponsor’s reported comparisons use MMRM least-square means. The model’s missing-data assumptions and further details of the analysis population are not reported in the registry record.

7. Results Posted in the Registry

• QVM149 150/50/160 μg: mean 0.83 versus 0.71 on control; difference 0.11 (95% CI -0.08 to 0.30), standardized effect d = 0.08; z-test p = 0.255.
• Salmeterol/Fluticasone 50/500 μg Plus Tiotropium 5 μg: mean 0.75 versus 0.71 on control; difference 0.04 (95% CI -0.15 to 0.23), standardized effect d = 0.03; z-test p = 0.699.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline in AQLQ Total Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • QVM149 150/50/160 μg: mean 0.76 versus 0.69 on control; difference 0.07 (95% CI -0.12 to 0.25), standardized effect d = 0.05; z-test p = 0.502.
  • Salmeterol/Fluticasone 50/500 μg Plus Tiotropium 5 μg: mean 0.67 versus 0.69 on control; difference -0.02 (95% CI -0.21 to 0.17), standardized effect d = -0.01; z-test p = 0.862.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between QVM149 150/50/160 μg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Salmeterol/Fluticasone 50/500 μg Plus Tiotropium 5 μg and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Patients Achieving the Minimally Clinically Important Difference (MCID) Decrease From Baseline ACQ-7 ≥ 0.5

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • QVM149 150/50/160 μg: 85.2% (387/454) versus 87.9% (393/447) on control; risk difference -2.7 percentage points (95% CI -7.1 to 1.8), risk ratio 0.97 (0.92–1.02); unadjusted two-proportion z-test p = 0.239.
  • Salmeterol/Fluticasone 50/500 μg Plus Tiotropium 5 μg: 83.9% (375/447) versus 87.9% (393/447) on control; risk difference -4.0 percentage points (95% CI -8.6 to 0.5), risk ratio 0.95 (0.90–1.01); unadjusted two-proportion z-test p = 0.084.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between QVM149 150/50/160 μg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Salmeterol/Fluticasone 50/500 μg Plus Tiotropium 5 μg and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Patients Achieving the Minimally Clinically Important Difference (MCID) Change From Baseline AQLQ ≥ 0.5

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • QVM149 150/50/160 μg: 73.3% (333/454) versus 71.6% (318/444) on control; risk difference 1.7 percentage points (95% CI -4.1 to 7.6), risk ratio 1.02 (0.94–1.11); unadjusted two-proportion z-test p = 0.562.
  • Salmeterol/Fluticasone 50/500 μg Plus Tiotropium 5 μg: 67.8% (299/441) versus 71.6% (318/444) on control; risk difference -3.8 percentage points (95% CI -9.9 to 2.2), risk ratio 0.95 (0.87–1.03); unadjusted two-proportion z-test p = 0.216.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between QVM149 150/50/160 μg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Salmeterol/Fluticasone 50/500 μg Plus Tiotropium 5 μg and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover related domains, including AQLQ at Week 16, asthma control, lung function, and responder definitions based on MCID thresholds. Taken as a group, these results do not form one consistent independent confirmation of the primary outcome. Sponsor-reported comparisons include both non-significant results and an isolated nominally significant logistic-regression result for an AQLQ responder outcome. That pattern calls for restraint rather than treating any single secondary finding as decisive.

The secondary AQLQ measurement at Week 16 is related to the primary Week 24 AQLQ endpoint, but it is a different timepoint and analysis. The responder outcomes also differ from analysis of the continuous AQLQ score: they classify participants according to whether they meet a threshold. Measures of asthma control and lung function address other aspects of the condition. Agreement across these outcomes could add context, but they do not have the same meaning or precision as the primary endpoint.

Unadjusted secondary results carry less weight when multiple outcomes and comparisons are examined. A nominal p-value does not account for the broader set of tests unless a multiplicity procedure is used, and that procedure is not reported in the registry record. The record lists 8 outcome measures and 30 analyses; it does not state which secondary analyses were prespecified or whether their interpretation was adjusted for multiple testing. Secondary findings are therefore most appropriately read as exploratory or supportive evidence, with the primary endpoint and sponsor’s stated analysis remaining central.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change From Baseline in Trough Forced Expiratory Volume in 1 Second (FEV1)
• Change From Baseline in Asthma Control Questionnaire (ACQ-7) Total Score
• Change From Baseline in AQLQ Total Score
• Percentage of Patients Achieving the Minimally Clinically Important Difference (MCID) Decrease From Baseline ACQ-7 ≥ 0.5
• Percentage of Patients Achieving the Minimally Clinically Important Difference (MCID) Change From Baseline AQLQ ≥ 0.5
• Change From Baseline in Forced Vital Capacity (FVC)
• Change From Baseline in Forced Expiratory Flow Between 25% and 75% of Forced Vital Capacity (FEF25-75)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a score, and the posted arm summaries are least-squares means with standard errors. A mean describes the central value of the outcome, while a standard error describes the estimated sampling uncertainty around that mean. The sponsor’s reported method is a Mixed Model for Repeated Measures (MMRM), which is designed for analyses that use outcome measurements collected at multiple timepoints. The registry record names the method but does not report its covariates, covariance structure, or other model specifications.

What does a least-squares mean represent?

A least-squares mean is a model-based estimate of an arm’s mean outcome. It can differ from a simple arithmetic mean because it is obtained from the fitted model and may reflect the model’s adjustment structure. The registry lists least-squares means for the primary outcome and reports MMRM contrasts between treatment groups. Since the model details are not reported, the reported estimates cannot be fully reconstructed from the registry record’s summary values alone.

How should a treatment contrast be read?

A contrast expresses the estimated difference between groups under the analysis being used. Its sign depends on the order in which groups are subtracted. The sponsor’s reported analyses list the groups and the contrast value, but an incomplete confidence interval and unreported margin make it important not to infer the full non-inferiority decision rule from the estimate alone. In a non-inferiority analysis, the question is whether the data rule out a loss larger than a prespecified acceptable margin. The margin is not reported in the registry record, so that threshold cannot be checked.

What do the recalculations add?

The recalculated comparisons are simple, unadjusted comparisons based on posted arm-level means and standard errors. They are not MMRM results and do not reproduce the trial’s sponsor analysis. The recalculation uses an approximate standard deviation derived from the posted summaries, as indicated by the record. Such a comparison can offer a transparent reference point, but it cannot account for model features or repeated measurements that are not represented in the summary statistics. The record flags a discrepancy between the recalculation and sponsor analysis; the sponsor’s analysis takes precedence.

A confidence interval gives a range of estimates compatible with the data under the analysis and its assumptions. For a difference, an interval that includes zero indicates that a zero difference is compatible with the data at that interval’s confidence level. That observation is not a test of non-inferiority: non-inferiority depends on comparison with a specified margin, not simply on whether the interval includes zero. The sponsor’s primary results are reported with 97.5% confidence intervals, but only the lower bounds are posted. The unadjusted recalculation intervals are available in the statistical results; the recalculated intervals for the continuous comparisons include zero.

A p-value measures how unusual data at least as incompatible with a stated null hypothesis would be if that null and the analysis assumptions were true. It is not the probability that a treatment is effective, nor does it quantify the size or importance of a difference. The sponsor’s p-values are attached to the reported non-inferiority analyses; the recalculated p-values come from a different, unadjusted comparison. They should not be treated as interchangeable evidence.

How are responder outcomes analyzed?

Some secondary outcomes count participants meeting a responder definition, such as the specified minimally clinically important difference (MCID) threshold. These are binary outcomes: each participant either meets the definition or does not. The registry reports logistic regression odds ratios for some secondary comparisons. An odds ratio compares the odds of meeting the outcome across groups; it is not the same quantity as a difference in group means or a risk difference. The registry also posts unadjusted recalculated risk differences for these outcomes, which are distinct from the sponsor’s logistic-regression estimates.

For any odds ratio or risk difference, precision and interpretation depend on the confidence interval and the analysis assumptions. An interval spanning the null value is compatible with no difference under that analysis. A nominal p-value below a conventional threshold does not, on its own, address the number of outcomes and comparisons examined. The registry lists 8 outcome measures and 30 analyses, but the multiplicity strategy is not reported in the registry record.

These methods answer related but separate questions. The MMRM provides the sponsor’s modeled analysis of repeated continuous measurements; the summary-statistic recalculations provide unadjusted comparisons; and logistic regression addresses responder outcomes. Differences in methods, populations, and outcome definitions mean that estimates and p-values should be interpreted within their own analyses rather than combined as if they were one test.

11. Confidence Intervals and Precision

Precision is reflected in the range of values compatible with an estimate, commonly expressed as a confidence interval. A narrower interval indicates less statistical uncertainty under the analysis assumptions; a wider interval indicates greater uncertainty. The interval does not show the probability that a particular value is true. Its interpretation also depends on the analysis method and the population included.

For the sponsor’s primary MMRM comparisons, the registry reports 97.5% confidence intervals but posts only the lower bound: -0.139 for the 150/50/80 μg QVM149 comparison and -0.027 for the 150/50/160 μg comparison. The upper bounds are not reported in the registry record. This leaves the displayed sponsor intervals incomplete, so their full ranges cannot be assessed from the registry results. The record also does not give the non-inferiority margin, which prevents direct comparison of the full interval with the threshold required for that decision.

The unadjusted recalculations report mean differences of 0.112 for the higher-dose comparison and 0.038 for the lower-dose comparison. Both recalculated confidence intervals include zero, and their unadjusted p-values are 0.255 and 0.699, respectively. These intervals reflect the simple summary-statistic recalculations, not the sponsor’s MMRM. The record flags a discrepancy, and the sponsor’s own analysis takes precedence.

Arm-level standard errors for the primary endpoint are 0.07, 0.069, and 0.069, respectively. These quantify uncertainty around the posted arm-level least-squares means; they are not confidence intervals for the treatment contrasts. The distinction matters because uncertainty in a group mean and uncertainty in a difference between groups are different quantities. Overall, the posted results support a more limited precision assessment than a complete sponsor confidence interval and margin would allow.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures are planning illustrations based on the observed effect, not the trial’s original design calculation. For these continuous and binary outcomes, the figures are participants per arm, not total events. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning scenario uses the stated fraction of the observed effect to detect. The trial’s own original sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the number of participants per arm estimated to provide an 80% chance of detecting the specified effect under those planning assumptions if that effect is present. The 90% power column gives the corresponding estimate for a higher chance of detection. These are prospective planning concepts, not descriptions of what happened in the completed trial.

Rows for 100%, 75%, and 50% of the observed effect illustrate how the planning estimate changes as the target effect becomes smaller. A smaller effect is harder to distinguish from random variation, so the planning calculation requires more participants per arm. The displayed figures depend on the effect observed in this trial and should not be treated as universal requirements for future research. They are also based on simplified assumptions that may not match the sponsor’s actual design or analysis.

These figures are not observed power. Observed results should be interpreted using the estimates, confidence intervals, and analysis methods, rather than a post hoc calculation of power. For time-to-event outcomes, the stated convention would be total events, but the primary and secondary outcomes described in this record are not time-to-event outcomes.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 0.112; pooled standard deviation σ = 1.465.
2. n = 2 × (1.96 + 0.842)² × 1.465² / 0.112² = 2,686.5, rounded up to 2,687 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 0.112                      # observed difference in means
sigma <- 1.46518304261457                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3158311)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 2,687 (80% power), 3,597 (90% power)
 75% of observed effect: n per arm = 4,776 (80% power), 6,394 (90% power)
 50% of observed effect: n per arm = 10,746 (80% power), 14,386 (90% power)
power.t.test: 2687.5 
Simulated power at n = 2,687 per arm: 0.802 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry posts sponsor-reported MMRM non-inferiority analyses for both QVM149 dose groups against salmeterol/fluticasone plus tiotropium. The reported contrasts are -0.038 for QVM149 150/50/80 μg and 0.073 for QVM149 150/50/160 μg. Each analysis has a reported p-value of <0.001 and a 97.5% confidence interval lower bound. The upper bounds are not reported. The registry does not report the non-inferiority margin, so the complete criterion cannot be independently evaluated from the posted information.

The recalculated comparisons use posted summary statistics and are unadjusted. They give different estimates and p-values from the sponsor analysis: the higher-dose comparison has a recalculated difference of 0.112 with p = 0.255, and the lower-dose comparison has a difference of 0.038 with p = 0.699. These calculations do not reproduce the MMRM and should not be used to replace it. The registry record flags a discrepancy between the sponsor analyses and the recalculations. In that circumstance, the sponsor’s own analysis takes precedence; the recalculations are best understood as a separate, less fully adjusted reference analysis.

15. Multiplicity, Missing Data and Analysis Populations

The registry record lists 8 outcome measures and 30 analyses. It does not report a multiplicity strategy describing how the chance of false-positive findings across multiple outcomes, dose comparisons, timepoints, and analysis types was controlled. The record also does not identify which secondary comparisons were designated as confirmatory or whether any testing sequence was prespecified. Without those details, nominal p-values for secondary analyses should not be read as if each were the sole test in the trial.

The primary outcome analysis population included the Full Analysis Set with evaluable AQLQ data at both baseline and Week 24. The record states that no imputation was used for missing questions or missing total AQLQ scores and that the analysis used observed data only. This describes the handling of missing AQLQ information for that analysis, but it does not explain why data were missing or whether missingness differed between groups. The assumptions used by the sponsor’s MMRM for incomplete repeated measurements are not reported in the registry record.

Missing-data handling for the secondary outcomes is not reported in enough detail to establish whether the same population or approach was used throughout. The arm-level sample sizes also vary across posted analyses, but the registry does not explain the reasons for those differences. They should not be interpreted as evidence of a particular cause of missingness or exclusion.

The detailed statistical analysis plan, the original sample-size justification, the non-inferiority margin, and any interim-analysis plan are not reported in the registry record. This limits assessment of prespecification and the relationship between planned and posted analyses. Those details matter especially for non-inferiority conclusions and for interpreting numerous secondary comparisons; their absence does not establish that no such plans existed, only that they are not reported in the registry record.

16. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumonia (1.1%), asthma (0.8%), intentional product misuse (0.4%), atrioventricular block second degree (0.2%), gastrointestinal haemorrhage (0.2%).

The registry reports serious adverse events for participants at risk in each treatment group. Serious adverse events affected 14 of 474 participants in the QVM149 150/50/80 μg group, 18 of 476 in the QVM149 150/50/160 μg group, and 19 of 475 in the salmeterol/fluticasone plus tiotropium group. The registry also reports other adverse events affecting 245 of 474, 246 of 476, and 241 of 475 participants, respectively. These are counts as posted; they do not by themselves establish that an event was caused by treatment.

The registry lists 0 deaths in each QVM149 group and 1 in the active-comparator group. It also lists serious-event terms and associated percentages, including pneumonia at 1.1% and asthma at 0.8%. These summaries do not provide a full account of event timing, duration, severity, or individual clinical context. Comparisons of safety counts should be interpreted descriptively, with attention to the numbers at risk and to the fact that the registry summaries do not establish causation.

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

The primary result addresses a specific statistical question: whether each QVM149 dose met the sponsor’s non-inferiority analysis against the free triple combination for change in AQLQ total score at Week 24. The registry reports sponsor MMRM analyses designated as non-inferiority, with p < 0.001 for both comparisons. It does not report the non-inferiority margin or the upper confidence bounds, so the complete decision criterion cannot be independently reconstructed from the registry record.

These results do not establish that the treatments are identical. A non-inferiority analysis asks whether a loss larger than an agreed margin can be excluded under a specified method; it is not a test of exact equality. The estimate, full confidence interval, margin, and analysis assumptions all matter. The record’s unadjusted recalculations differ from the sponsor analyses, and the registry flags that discrepancy. The sponsor’s own analysis takes precedence, while the recalculations remain a separate summary-statistic comparison.

The posted outcome is a patient-reported quality-of-life score. It should not be treated as interchangeable with lung function, asthma control, or safety outcomes. A result on one measure does not answer every question about the treatments, and statistical significance alone does not establish the practical importance of a change. The registry record does not report a threshold for interpreting the primary endpoint’s treatment contrast as clinically important.

The study record also does not supply enough information to independently verify all design and analysis choices, including the original sample-size justification, missing-data assumptions for the sponsor model, and multiplicity approach. The secondary results may add context but should not be elevated over the primary comparison, especially when multiple analyses are posted and adjustment is not described. These findings are statistical summaries of this trial’s posted data, not individualized medical guidance.

18. Why This Trial Matters Statistically

This trial illustrates why a non-inferiority result needs more than a small p-value to be independently interpreted. The stated question depends on a prespecified margin, an appropriate analysis, and a confidence interval that can be compared with that margin. The registry identifies the sponsor’s analyses as non-inferiority comparisons and posts their lower bounds, but the margin and upper bounds are not reported. That makes the distinction between a sponsor-reported conclusion and an independently assessable statistical decision especially important.

It also demonstrates how different analytic approaches can produce different summaries. The sponsor used MMRM for the primary endpoint, while the recalculations use unadjusted arm-level summary statistics. The latter are simpler to inspect but do not reproduce the sponsor’s model. When a discrepancy is flagged, the sponsor’s analysis takes precedence; readers should not treat a recalculated estimate as a correction to the trial analysis.

The outcome structure adds another statistical lesson. A continuous quality-of-life score, a responder threshold, asthma-control measures, and lung-function measures are related but not interchangeable. Looking across them can help characterize the evidence, but multiple tests increase the need for a clear prespecified analysis plan and multiplicity strategy. Neither is reported in sufficient detail in the registry record.

Finally, precision and missing-data handling shape what conclusions can be drawn. The primary analysis used observed AQLQ data from participants evaluable at both timepoints, while the sponsor model’s handling of incomplete repeated measurements is not described. Transparent reporting of margins, full confidence intervals, analysis populations, and missing-data assumptions would make the non-inferiority result easier to evaluate independently.

19. 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.

Several limitations are specific to the posted record. The primary endpoint is reported with incomplete sponsor confidence intervals: lower bounds are available, but upper bounds are not. The non-inferiority margin is also not reported. As a result, the full interval-to-margin comparison that underpins a non-inferiority decision cannot be independently checked using the registry record alone.

The record describes the trial as partially blinded in the official title, as open-label active controlled in that title, and as single-masked in the design field. It does not explain who was masked or reconcile those descriptions. Because AQLQ is patient-reported, clarity about masking would help readers assess the context in which responses were collected, without implying that any particular bias occurred.

The sponsor’s MMRM specifications, including covariates and covariance structure, are not reported. The primary endpoint analysis used observed data without imputation, while the model’s assumptions for incomplete repeated measurements are not described. The record also does not explain why analysis sample sizes vary across endpoints. These gaps limit reproducibility and interpretation, but do not establish that the analyses were inappropriate.

The registry flags a discrepancy between sponsor analyses and unadjusted recalculations. The recalculations are approximate summary-statistic comparisons and are not substitutes for the sponsor model. The original sample-size justification, multiplicity strategy, and detailed statistical analysis plan are also not reported. Finally, the treatment period was 24 weeks with a planned 7-day follow-up; the posted results therefore address the recorded study period, not outcomes beyond it.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study to Compare QVM149 and Free Triple Combination of Salmeterol/Fluticasone + Tiotropium — NCT03158311.
• Registry record last updated 2021-10-08; status last verified 2021-10.
• Results first posted 2020-10-14.
• 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.

23. Record Summary

NCT03158311 was a completed Phase 3 randomized parallel-group trial in asthma comparing two QVM149 dose groups with salmeterol/fluticasone plus tiotropium. The record lists 1426 participants enrolled and a planned 24-week treatment period followed by a 7-day follow-up. The primary outcome was change from baseline in AQLQ total score at Week 24.

The sponsor reported MMRM non-inferiority analyses for both QVM149 dose comparisons, each with p < 0.001. The registry does not report the non-inferiority margin or upper confidence bounds. Unadjusted recalculations differ from those sponsor analyses, and the record flags a discrepancy; the sponsor’s own analysis takes precedence. Important details, including the original sample-size justification, multiplicity approach, and some missing-data assumptions, are not reported in the registry record.