Omalizumab in Patients With Moderate to Severe Persistent Allergic Asthma Not Adequately Controlled Despite GINA (2009) Step 4 Therapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Omalizumab in Patients With Moderate to Severe Persistent Allergic Asthma Not Adequately Controlled Despite GINA (2009) Step 4 Therapy (NCT01202903), a randomized phase 3 study. The primary outcome is Change From Baseline in Mean Morning Peak Expiratory Flow (PEF) Following the 24-week Treatment Period.

2. Clinical Question

3. Trial Design

This was a Phase III, randomized, double-blind, placebo-controlled, parallel-group, multicenter study. The registry record describes a 24-week treatment period and an overall enrollment of 616. The study was listed as completed. The recorded start was 2010-09 and primary completion was 2013-10. Novartis Pharmaceuticals is identified as the sponsor.

The intended population was Chinese patients aged 18 to 75 with moderate to severe persistent allergic asthma that remained inadequately controlled despite treatment according to GINA (2009) Step 4 therapy. The two groups received omalizumab or placebo. The record says placebo used the same mixture of inactive excipients, in quality and quantity, as the drug product. It describes subcutaneous administration; the omalizumab arm description is truncated, so the full dosing schedule for that arm is not reported in the registry record.

Randomization and double masking help limit the influence of treatment selection and expectations on comparisons. A parallel-group design means that participants were followed in their assigned groups rather than switching between treatments. The registry record does not report the randomization method, allocation concealment details, the number of study centers, or the original sample-size justification.

The primary analysis population is described as the FAS, consisting of randomized patients who received at least one dose of study drug. For the primary endpoint, there were 306 participants in the omalizumab group and 302 in the placebo group. These endpoint counts should be distinguished from enrollment and from the safety denominators, which are different. The registry record’s wording about the exclusions from the FAS is not fully clear; additional detail about the handling of these exclusions is not reported in the registry record.

4. Participant Flow and Accounting

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

Before assignment: One additional patient did not receive study drug for greater than 60 days and was also removed from the FAS. 4 patients initially randomized to the placebo group received omalizumab at one or more administration and were considered part of the omalizumab group for the Safety Set.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The FAS consisted of all randomized patients who received at least one dose of study drug. 7 randomized patients did not receive study drug and were excluded along with 1 other patient who did not receive study drug for more than 60 days

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: The Safety Set consisted of all patients who received at least one dose of study medication whether or not being randomized. Patients were analyzed according to the treatment they actually received. If patients switched between placebo and omalizumab during the study, they were analyzed according to omalizumab.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Mean Morning Peak Expiratory Flow (PEF) Following the 24-week Treatment Period, measured over Baseline, 24 weeks.

The primary endpoint was change from baseline in mean morning peak expiratory flow (PEF) following the 24-week treatment period. PEF is a measure of how quickly a person can exhale air. In this study it was recorded in L/min using a meter distributed to participants. The instructions called for morning and evening measurements each day during screening and treatment. Morning measurement was to occur within 15 minutes after waking, and evening measurement approximately 12 hours later. Participants were encouraged to measure before using a long-acting beta agonist or rescue medication. The highest of three values was recorded as the daily personal best, which contributed to the mean morning PEF.

Change from baseline provides a way to compare how the endpoint shifted over the treatment period rather than comparing only the final measurement. Morning PEF is a repeated, objective lung-function measure, and the daily recordings can represent variation across days more fully than a single reading. These are general reasons such measures may be used; they do not establish what the study found.

The registry specifies that the arm estimates are least-squares means and describes an ANCOVA model for the primary endpoint. The model included treatment, stratification group, dosing schedule, gender, center grouping, smoking status and baseline mean morning PEF as covariates. The posted arm estimates were 27.03 L/min for omalizumab and 18.18 L/min for placebo. The comparison estimate and its uncertainty are interpreted separately from these arm-specific adjusted means.

7. Results Posted in the Registry

• Omalizumab: mean 27.03 versus 18.18 on control; difference 8.85 (95% CI -49.29 to 66.99), standardized effect d = 0.02; z-test p = 0.765.

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 Mean Evening PEF (L/Min) Following 24-week Treatment

• 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:
  • Omalizumab: mean 22.96 versus 15.53 on control; difference 7.43 (95% CI -48.69 to 63.55), standardized effect d = 0.02; z-test p = 0.795.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Omalizumab and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in % Predicted FEV1 Following 24-week Treatment

• 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:
  • Omalizumab: mean 19.11 versus 14.99 on control; difference 4.12 (95% CI -11.51 to 19.75), standardized effect d = 0.04; z-test p = 0.605.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Omalizumab and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in ACQ Score Following 24-week Treatment

• 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:
  • Omalizumab: mean -0.51 versus -0.34 on control; difference -0.17 (95% CI -0.42 to 0.08), standardized effect d = -0.13; z-test p = 0.189.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Omalizumab and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover several dimensions: evening PEF, predicted FEV1, asthma quality of life, asthma control, symptom scores, and global evaluations of treatment effectiveness. These measures complement the primary morning PEF endpoint by describing different lung-function and patient-reported or investigator-assessed outcomes. They are related to the same broad study question, but they are not interchangeable measures of a single underlying quantity.

Among the secondary results with numerical comparisons posted, the point estimates do not all point in the same direction. The evening PEF and FEV1-related estimates are positive, while the ACQ estimate is negative. Their intervals include zero. The registry lists additional secondary outcome topics, but numerical results for those listed outcomes are not reported in the registry record. This incomplete numerical reporting makes it difficult to assess the overall pattern across every planned secondary measure.

Secondary endpoints can help describe whether findings are consistent across different kinds of measurement, but they should not overrule the primary endpoint based on isolated comparisons. Multiple outcomes create more opportunities for apparently notable results to arise by chance when each is examined separately. Unless the analysis plan specifies how multiplicity is controlled or how outcomes are prioritized, unadjusted secondary results are best treated as exploratory rather than definitive.

The registry record does not report multiplicity adjustments, a hierarchical testing strategy, or the prespecified status of each secondary comparison. It also does not provide reported analysis results for the secondary comparisons. As a result, these estimates can provide context, but their evidential weight is limited. They do not establish a consistent treatment effect across outcomes, nor do they settle the uncertainty around the primary comparison.

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

• Change From Baseline in Mean Evening PEF (L/Min) Following 24-week Treatment
• Change From Baseline in % Predicted FEV1 Following 24-week Treatment
• Change From Baseline in AQLQ Score Following 24-week Treatment
• Percentage of Patients Achieving at Least a 0.5, 1.0 or 1.5 Point Improvement From Baseline in AQLQ Overall Score Following 24-week Treatment
• Change From Baseline in ACQ Score Following 24-week Treatment
• Percentage of Patients With at Least a 0.5 or 0.75 Point Improvement in the ACQ Score at Week 24
• Change From Baseline in Asthma Symptom Scores Following 24-week Treatment
• Percentage of Participants With Investigator and Patient Global Evaluation of Treatment Effectiveness (GETE) at Weeks 16 and 24

9. Statistical Interpretation

10. Statistical Methods Explained

The registry describes an ANCOVA model for the primary endpoint and reports least-squares means for each group. The comparison and planning figures also include simple unadjusted recalculations from the posted summary statistics. These are different kinds of results: the least-squares means are model-based arm estimates, whereas the recalculated comparison is an approximate unadjusted contrast. The registry record does not report a sponsor-reported numerical treatment contrast for the primary endpoint.

What does ANCOVA do?

In general, analysis of covariance (ANCOVA) compares groups while accounting for selected baseline or design variables. For this trial, the record says the model included treatment, stratification group, dosing schedule, gender, center grouping, smoking status and baseline mean morning PEF. The baseline value is especially relevant because the endpoint is change from baseline. Adjustment can improve the precision of a comparison when the included variables are informative and the model is appropriate. It does not guarantee that all sources of bias or uncertainty have been removed.

A least-squares mean is an adjusted mean estimated from a model, rather than simply the arithmetic mean of the observed values. Its interpretation depends on the model and covariates used. The registry provides the covariate list but does not report the model’s full specification, how missing endpoint measurements were handled, or diagnostics assessing model assumptions. Those details are not reported in the registry record.

How should the recalculated comparison be read?

The recalculated mean difference is the omalizumab estimate minus the placebo estimate, as represented in the posted comparison. Its value is 8.85 L/min. This is a simple, unadjusted comparison derived from posted summary statistics, not an independent re-analysis of participant-level data. The confidence interval and unadjusted p-value should therefore be read as approximate summaries rather than as a replacement for a sponsor-reported ANCOVA contrast.

For a continuous outcome, a mean difference describes the contrast in average values between groups, in the endpoint’s units. A positive estimate indicates that the omalizumab arm’s estimate is higher than the placebo arm’s estimate under the stated contrast direction. The estimate alone does not show how much uncertainty surrounds it; that is the role of the confidence interval and standard error.

Why does the confidence interval matter?

In general, a confidence interval describes a range of estimates compatible with the observed data under the method’s assumptions. A narrower interval indicates greater precision; a wider interval indicates more uncertainty. The primary interval, -49.29 to 66.99 L/min, includes zero and a broad range of possible differences. It should not be interpreted as the probability that the true effect falls within that particular interval. Nor does an interval containing zero prove there is no effect.

The registry reports standard errors for the arm estimates, not arm-level confidence limits. The posted comparison’s pooled standard deviation is marked as approximate, and the recalculated comparison uses that approximation. This matters because the uncertainty calculations depend on the available summary data and assumptions. The registry’s adjusted model may use information differently from a simple unadjusted calculation.

What does over-dispersion mean in general?

For continuous outcomes, variability among individuals affects how precisely a group mean and a difference between means can be estimated. Greater variability generally makes it harder to distinguish a group contrast from background variation at a fixed sample size. In other settings, over-dispersion refers to observed variability exceeding what a simpler statistical model expects; a model that ignores it can understate uncertainty. The registry record does not report a specific over-dispersion assessment for this outcome, so no such finding can be attributed to this trial.

What can the p-value say?

In general, a p-value evaluates how unusual data at least as inconsistent with a null hypothesis would be if that hypothesis and the test assumptions were true. The unadjusted p-value is 0.765. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. The record does not report a multiplicity-adjusted p-value for the primary endpoint or for the secondary endpoints.

The primary endpoint description identifies a FAS analysis population: randomized patients who received at least one dose of study drug. It also reports endpoint counts of 306 and 302 in the two arms. The registry does not give a full account of participant flow, missing outcome data, imputation, or sensitivity analyses. These omissions limit assessment of how robust the model-based estimates are to exclusions and incomplete observations.

11. Confidence Intervals and Precision

The primary comparison has a mean difference of 8.85 L/min and a 95% confidence interval from -49.29 to 66.99. The interval is broad and includes zero. That span indicates that the posted summary statistics do not pin down the size or direction of the group difference with high precision. The point estimate is one estimate among a range of values compatible with the data and calculation method.

Confidence intervals are useful because they communicate uncertainty in the estimate, not just whether a hypothesis test crosses a conventional threshold. In general, a confidence interval is constructed so that a procedure would cover the true value at its stated rate over repeated use under the model assumptions. It is not a probability statement about the fixed true value after the interval has been calculated. The interval also does not account for every possible bias, missing-data problem or model error.

The comparison interval is an unadjusted recalculation from posted summary statistics. The endpoint arm estimates themselves are least-squares means from an ANCOVA model. Since the record does not provide the sponsor’s own numerical contrast and interval, the recalculated interval should not be mistaken for the confidence interval from that model. The registry does not flag a discrepancy, but that does not make the two analysis approaches identical.

The arm standard errors are 20.86 L/min for omalizumab and 21.09 L/min for placebo. Standard errors describe the estimated sampling uncertainty of the arm means; they are not measures of the spread of individual participant values. The recalculated comparison’s pooled standard deviation is marked as approximate, which adds a further reason to read the interval as a summary derived from available statistics rather than a participant-level model result.

A wide interval does not establish that all values it covers are equally plausible or clinically meaningful. The registry record does not identify a threshold for clinical importance, so the practical significance of particular differences cannot be judged from the interval alone.

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 planning figures are participants per arm, not total events. They are planning illustrations calculated from the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms and a normal approximation. The calculations consider detecting the observed effect and stated fractions of that effect. These assumptions describe the illustration, not the trial’s original design; the trial’s own sample-size justification and original design assumptions are not reported in the registry record.

The 80% power column gives the estimated number of participants per arm for a design with an 80% chance of detecting the specified effect under the planning assumptions, if that effect and the other assumptions were true. The 90% power column uses a higher target chance of detection and consequently requires more participants per arm for the same assumed effect. These values are design calculations, not counts of participants actually enrolled or analyzed.

For the primary endpoint, planning for 100% of the observed effect gives 26,806 participants per arm for 80% power and 35,885 participants per arm for 90% power. Planning for 75% of the observed effect gives 47,654 participants per arm for 80% power and 63,795 participants per arm for 90% power. Planning for 50% of the observed effect gives 107,222 participants per arm for 80% power and 143,539 participants per arm for 90% power. The smaller the effect a study is designed to detect, the more participants it generally needs, because a smaller contrast is harder to distinguish from random variation.

The figures are not observed power. They should not be used to infer whether the completed study was adequately powered based on its observed result. They are illustrations tied to the observed effect, which is itself uncertain, and to the stated simplified assumptions. The registry record does not report the assumptions used to plan the original trial.

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 = 8.850; pooled standard deviation σ = 365.708.
2. n = 2 × (1.96 + 0.842)² × 365.708² / 8.850² = 26,805.3, rounded up to 26,806 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 <- 8.85                      # observed difference in means
sigma <- 365.707962477181                     # 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(1202903)
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 = 26,806 (80% power), 35,885 (90% power)
 75% of observed effect: n per arm = 47,654 (80% power), 63,795 (90% power)
 50% of observed effect: n per arm = 107,222 (80% power), 143,539 (90% power)
power.t.test: 26806.2 
Simulated power at n = 26,806 per arm: 0.803 (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. Multiplicity, Missing Data and Analysis Populations

The primary and secondary outcomes involve multiple opportunities to compare groups. In general, when many hypotheses are tested, the chance of obtaining at least one apparently unusual result by chance can increase. Statistical plans often address this through a hierarchy of outcomes, an adjustment to the testing procedure, or a clear designation of analyses as exploratory. The registry record does not report a multiplicity strategy, adjusted secondary p-values, or a hierarchy for interpreting the endpoints. The posted unadjusted comparisons should therefore be read without assuming that family-wise error or another multiplicity criterion has been controlled.

The registry lists a primary outcome and several secondary outcome measures. It reports no analyses in the analysis listing, and no reported analyses are included with the posted secondary summaries. This means the reader cannot confirm from the registry record how the sponsor’s planned inferential procedure handled the endpoint family. A lack of reported adjustment is not evidence that no adjustment was performed; the method is simply not reported in the registry record.

Missing data are another important issue for continuous endpoints. The record gives the primary analysis population definition and arm counts, but does not report the amount or pattern of missing PEF measurements, the method used to handle missing observations, or sensitivity analyses under alternative assumptions. It also does not report whether any participants were excluded after randomization for reasons related to outcome availability. Without those details, the effect of missing data on the estimates cannot be assessed.

The FAS is defined as all randomized patients who received at least one dose of study drug. The record states that 7 randomized patients did not receive study drug and were excluded, along with 1 other patient described as not receiving study drug for more than 60 days. The wording of the latter exclusion is unclear. Further detail about participant flow and the rationale for the exclusions is not reported in the registry record. The FAS definition also differs from a strict intention-to-treat population that includes every randomized participant, so the distinction matters when considering the analysis population.

The registry describes ANCOVA covariates for the primary outcome, but does not report all model details, diagnostics, or sensitivity analyses. These omissions leave uncertainty about robustness to model specification, incomplete measurements and analysis-population choices.

15. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 1% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: asthma (2.3%), pneumonia (0.7%), upper respiratory tract infection (0.3%), multiple fractures (0.3%), patella fracture (0.3%).

The safety summary reports serious adverse events in 8 of 310 participants at risk in the omalizumab group and 11 of 299 participants at risk in the placebo group. Other adverse events are reported for 95 of 310 and 90 of 299 participants, respectively. These denominators differ from the primary endpoint analysis counts, reflecting that the safety and efficacy summaries use different populations or reporting bases. The registry record does not explain the difference in detail.

The listed serious adverse-event terms include Asthma (2.3%), Pneumonia (0.7%), Upper respiratory tract infection (0.3%), Multiple fractures (0.3%), Patella fracture (0.3%) and Pregnancy (0.3%). These are reported terms and percentages, not a complete description of all clinical details, timing, severity or causal assessments. The registry record does not report deaths.

Serious adverse events and other adverse events are distinct reporting categories. Their counts should not be added or treated as a direct measure of comparative risk without more information about event definitions, follow-up, attribution and participant-level overlap. The registry summary alone does not establish that an event was caused by the study treatment.

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

The primary result describes a comparison of mean morning PEF change over 24 weeks in the registry-defined analysis population. The arm estimates are 27.03 L/min for omalizumab and 18.18 L/min for placebo, and the unadjusted recalculated mean difference is 8.85 L/min. Its 95% confidence interval, -49.29 to 66.99, includes zero and is wide. The data therefore leave substantial uncertainty about the size and direction of the between-group difference.

That uncertainty should not be translated into a claim of equivalence or a claim that there is no effect. A non-small p-value does not prove the groups have identical outcomes. Conversely, a point estimate in one direction does not by itself establish a reliable difference. The interval and the limitations of the calculation matter when describing what the result can support.

The outcome is a physiological measure, not a direct measure of every aspect of asthma experience. The secondary outcomes include measures of lung function, asthma control, quality of life, symptoms and global evaluations. The record does not provide numerical results for all of the listed secondary outcomes, and the available secondary comparisons are unadjusted. They offer context but do not remove the uncertainty in the primary result.

The trial studied a defined population: Chinese patients aged 18 to 75 with moderate to severe persistent allergic asthma inadequately controlled despite GINA (2009) Step 4 therapy. Results from this population should not automatically be generalized to people outside those eligibility characteristics or to other treatment settings. The record also does not report enough information about missing data or participant flow to fully assess how the analysis population relates to everyone randomized.

These results are a statistical description of the posted trial data, not individualized medical guidance. They do not establish what an individual should do, and the registry record alone cannot resolve clinical importance, long-term outcomes or effects beyond the 24-week treatment period.

17. Why This Trial Matters Statistically

This trial illustrates why a point estimate should be read together with its uncertainty. The primary mean difference is 8.85 L/min, but the interval extends from -49.29 to 66.99. A reader who focused only on the point estimate would miss how much uncertainty remains; a reader who focused only on the p-value would miss the range of differences compatible with the recalculated comparison.

It also shows why analysis methods and reporting provenance matter. The registry gives least-squares means and describes an ANCOVA model, but the numerical comparison is an unadjusted recalculation rather than a reported sponsor contrast. Those results answer related but not identical statistical questions. Clear reporting of the model-based treatment contrast, its uncertainty, and the handling of incomplete data would allow a more direct assessment.

The outcome set spans physiological measures and participant-relevant measures, but the registry record does not report numerical results for every listed secondary outcome or describe a multiplicity plan. A broad set of outcomes can be informative, yet it also increases the importance of prespecification and careful interpretation. Secondary findings should be read in relation to the primary endpoint and the analysis plan, rather than selected in isolation.

Finally, the study population and follow-up period define the scope of the evidence. The registry describes a specific asthma population and a 24-week treatment period. Statistical conclusions should remain within that scope and acknowledge the limits of the available reporting.

18. 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 main limitation for statistical interpretation is that the registry reports adjusted arm estimates and an ANCOVA description but does not provide the sponsor’s numerical adjusted treatment contrast, confidence interval or p-value. The displayed contrast is an unadjusted recalculation from summary statistics, with an approximate pooled standard deviation. It cannot reproduce a participant-level ANCOVA analysis.

The record does not report the original sample-size justification, the intended power for the trial, or the assumptions used in its design. The planning illustrations based on the observed effect do not fill that gap. Nor does the registry report a multiplicity strategy, so the status of unadjusted secondary comparisons is uncertain.

Missing-data handling is not described. The amount and pattern of missing PEF data, the method for incomplete measurements, and sensitivity analyses are not reported in the registry record. The FAS exclusion wording is also unclear, and additional participant-flow information is not available in the record. These omissions make it difficult to assess potential effects of exclusions or missing outcomes.

There is also a labeling issue in the secondary outcome descriptions: the predicted FEV1 outcome title refers to percentage predicted FEV1, while its listed unit is L/min. The registry record does not explain this discrepancy. The complete numerical results for all listed secondary outcomes are not reported. Follow-up information beyond the 24-week treatment period is not included in the posted results, and deaths are not reported in the safety summary.

These reporting limitations do not show that a particular analysis was inappropriate or that a particular result is wrong. They identify details that are needed to judge the estimates more fully.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Omalizumab in Patients With Moderate to Severe Persistent Allergic Asthma Not Adequately Controlled Despite GINA (2009) Step 4 Therapy — NCT01202903.
• Registry record last updated 2015-03-23; status last verified 2015-03.
• Results first posted 2015-03-23.
• 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.

22. Record Summary

NCT01202903 is a completed Phase III, randomized, double-blind, parallel-group, placebo-controlled, multicenter study sponsored by Novartis Pharmaceuticals. It enrolled 616 Chinese patients aged 18 to 75 with moderate to severe persistent allergic asthma that was not adequately controlled despite GINA (2009) Step 4 therapy. The treatment period was 24 weeks.

The primary endpoint was change from baseline in mean morning PEF. The registry reports least-squares mean changes of 27.03 L/min for omalizumab and 18.18 L/min for placebo. The unadjusted recalculated mean difference is 8.85 L/min, with a 95% confidence interval from -49.29 to 66.99 and p = 0.765. This interval is wide and includes zero.

The registry describes the primary ANCOVA covariates but does not report the sponsor’s numerical adjusted contrast, a multiplicity approach, or missing-data methods. Numerical results are not reported for all listed secondary outcomes. The safety summary reports serious adverse events in 8 of 310 participants at risk in the omalizumab group and 11 of 299 in the placebo group; deaths are not reported in the registry record.