A Study to Evaluate the Dosing of AMG 827 for Subjects With Inadequately Controlled Asthma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Dosing of AMG 827 for Subjects With Inadequately Controlled Asthma (NCT01199289), a randomized phase 2 study. The primary outcome is Change From Baseline in Asthma Control Questionnaire (ACQ) Composite Scores to Week 12.

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, placebo-controlled phase 2 study with a parallel-group design. The registry reports enrollment of 315 participants and a completed status. The study start date was 2010-10-04, and primary completion was 2011-12-21. Amgen is listed as the sponsor.

Participants were assigned to placebo or one of three AMG 827 dose groups: 140 mg, 210 mg, or 280 mg. The registry describes the injections as subcutaneous and gives the same administration schedule for all four groups: Day 1 and weeks 1, 2, 4, 6, 8, and 10. Double masking means the study is recorded as double-blind, but the registry record does not report further details about which roles were masked or how masking was maintained.

The primary outcome was change from baseline in ACQ composite score at Week 12. The outcome population was not simply everyone enrolled: it was defined as randomized participants who received at least 1 dose, were not excluded due to GCP violations, and had non-missing baseline and post-baseline data. The registry does not report the original sample-size justification or the trial's original statistical design assumptions. It also does not report an interim-analysis plan in the information available in the registry record.

Placebo provides a concurrent reference for interpreting the dose groups. Randomization supports comparison of assigned groups, while blinding is intended to limit certain influences on study conduct and assessment. Those design features do not by themselves determine whether a result is precise or whether a difference is clinically important; those questions also depend on the outcome estimates, uncertainty, and analysis methods.

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 washout period for specific asthma medications was conducted after informed consent and prior to run-in period. After completing screening and meeting all eligibility criteria, all eligible participants underwent run-in visits for 4 weeks prior to randomization.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized participants who received at least 1 dose of study treatment who were not excluded due to GCP violations, and had non-missing baseline and post-baseline data.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Full Analysis Set (FAS): All randomized participants who received at least 1 dose of study treatment and were not excluded from analysis due to GCP violations.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Asthma Control Questionnaire (ACQ) Composite Scores to Week 12, measured over Baseline and Week 12.

The primary endpoint was change from baseline in the Asthma Control Questionnaire composite score through Week 12. The registry describes ACQ as an instrument used in clinical research and practice to evaluate asthma control or impairment. It combines responses to questions about nighttime awakenings, symptoms on awakening, activity limitation, shortness of breath, wheeze frequency, short-acting bronchodilator use, and forced expiratory volume in 1 second.

Each response ranges from zero, indicating no impairment or limitation, to six, indicating total impairment or limitation. The total score is the sum of the question values divided by the number of questions. The registry states that higher scores indicate worsening of condition and a negative change from baseline indicates improvement. Thus, the direction of change is important when reading the group means and comparisons: a more negative change represents a larger decrease in the score, not a worsening on this scale.

A composite outcome can represent several related aspects of a condition in one measure. That can make it useful for an overall assessment of asthma control, while also meaning that the composite does not show which individual component accounts for any observed change. The registry record does not report a threshold for interpreting a change as clinically meaningful, so the estimates should be read as score differences rather than classified as clinically important or unimportant based on a threshold not stated in the record.

7. Results Posted in the Registry

• AMG 827 140 mg: mean -0.52 versus -0.43 on control; difference -0.09 (95% CI -0.35 to 0.16), standardized effect d = -0.12; z-test p = 0.468.
• AMG 827 210 mg: mean -0.51 versus -0.43 on control; difference -0.09 (95% CI -0.33 to 0.16), standardized effect d = -0.11; z-test p = 0.490.
• AMG 827 280 mg: mean -0.56 versus -0.43 on control; difference -0.13 (95% CI -0.38 to 0.13), standardized effect d = -0.16; z-test p = 0.322.

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 the Frequency of Rescue Short Acting β-Agonist (SABA) Use to Week 12

• 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 deviations.
• Result:
  • AMG 827 140 mg: mean -0.19 versus -0.63 on control; difference 0.44 (95% CI -0.49 to 1.37), standardized effect d = 0.15; z-test p = 0.351.
  • AMG 827 210 mg: mean -0.73 versus -0.63 on control; difference -0.10 (95% CI -1.02 to 0.82), standardized effect d = -0.03; z-test p = 0.836.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AMG 827 140 mg 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 AMG 827 210 mg and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in Daily Asthma Symptom Score to Week 12

• 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 deviations.
• Result:
  • AMG 827 140 mg: mean -1.82 versus -1.53 on control; difference -0.29 (95% CI -1.39 to 0.82), standardized effect d = -0.08; z-test p = 0.615.
  • AMG 827 210 mg: mean -1.63 versus -1.53 on control; difference -0.10 (95% CI -1.11 to 0.92), standardized effect d = -0.03; z-test p = 0.850.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AMG 827 140 mg 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 AMG 827 210 mg and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in Asthma Quality of Life Questionnaire (AQLQ) Score at Week 12

• 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 deviations.
• Result:
  • AMG 827 140 mg: mean 0.60 versus 0.63 on control; difference -0.03 (95% CI -0.35 to 0.29), standardized effect d = -0.03; z-test p = 0.855.
  • AMG 827 210 mg: mean 0.54 versus 0.63 on control; difference -0.09 (95% CI -0.36 to 0.19), standardized effect d = -0.11; z-test p = 0.525.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AMG 827 140 mg 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 AMG 827 210 mg and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists secondary outcomes spanning lung function, peak expiratory flow, rescue medication use, symptom scores, quality of life, symptom-free days, and pharmacokinetic measures. The posted numeric summaries include change in rescue short-acting β-agonist use, daily asthma symptom score, and Asthma Quality of Life Questionnaire score. Across those reported summaries, the group means vary in direction and size, and the available sponsor comparisons do not present a consistent pattern of clear separation from placebo. The secondary results therefore do not provide a persuasive, coherent counterpoint to the primary outcome comparisons.

These outcomes relate to the primary ACQ result because they assess other aspects of asthma symptoms, function, quality of life, or drug exposure. Agreement across related outcomes can sometimes add context, while differences may reflect that each measure captures a distinct aspect of health or has different variability. The registry record does not report numerical results for every listed secondary outcome, so no conclusion about the full set of listed outcomes can be drawn from the numeric summaries available.

Secondary outcomes generally carry less weight than a prespecified primary outcome when many measures and group comparisons are examined. In general, testing many outcomes creates more opportunities for apparently notable results to occur by chance. Adjustment for multiplicity can address some of that issue, but the registry record does not report a multiplicity procedure for these secondary analyses. Their p-values should therefore be treated as exploratory rather than as a collection of independent confirmatory findings. The registry does not report a hierarchy or testing sequence for the secondary outcomes.

The secondary outcomes are best considered as supporting context, not as a replacement for the primary endpoint. Their interpretation also depends on endpoint-specific missingness and analysis details, which are not fully reported in the registry record. No secondary result should be singled out as definitive without considering the number of outcomes examined, the uncertainty around its estimate, and the analysis plan that is not reported.

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

• Change From Baseline in Pre- and Post-Bronchodilator Forced Expiratory Volume in 1 Second (FEV1)
• Change From Baseline in AM and PM Peak Expiratory Flow Rate (PEFR) to Week 12
• Change From Baseline in the Frequency of Rescue Short Acting β-Agonist (SABA) Use to Week 12
• Change From Baseline in Daily Asthma Symptom Score to Week 12
• Change From Baseline in Asthma Quality of Life Questionnaire (AQLQ) Score at Week 12
• Proportion of Asthma Symptom-free Days
• Time to Maximum Observed Concentration (Tmax) of AMG 827 at Week 8 to 10
• Maximum Observed Concentration (Cmax) of AMG 827 at Week 8 to 10

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a score measured at baseline and Week 12, summarized as mean change from baseline with a standard deviation. A mean describes the group's average change, while a standard deviation describes the spread of individual observations around that mean. The two quantities answer different questions: the mean indicates the average direction and size of change, and the standard deviation indicates how variable the changes were within the group.

What does the sponsor's ANCOVA estimate represent?

The registry reports an analysis of covariance, or ANCOVA, and gives least-squares mean differences for each dose group compared with placebo. In general, ANCOVA compares group outcomes while accounting for covariates included in the model. Least-squares means are model-based group estimates, and their difference is not necessarily identical to subtracting the two raw group means. The registry record names ANCOVA but does not report the covariates, model specification, handling of missing values within that model, or other implementation details. The sponsor-reported estimates and p-values therefore describe the posted analysis, but the record does not provide enough detail to reproduce its adjustment.

The registry gives 95% confidence intervals around the sponsor-reported differences. In general, a confidence interval expresses the uncertainty around an estimate under the assumptions of the method used. A narrower interval indicates greater precision, while a wider interval indicates more uncertainty. For these primary comparisons, all three intervals include zero. That means the posted interval estimates are compatible with no difference as well as with differences in either direction within the interval. It does not mean that zero is proven, or that every value in the interval is equally plausible.

What are the unadjusted recalculations?

The recalculated comparisons use the posted group means, standard deviations, and analysis sample sizes to estimate a simple difference between each dose-group mean and the placebo mean. These are unadjusted comparisons, not ANCOVA results. They can help readers connect the group summaries to a comparison estimate, but they do not replace the sponsor's model-based analysis. The standard deviations and group sample sizes also contribute to the uncertainty of these recalculations; a mean difference should not be read as if it were exact.

The recalculated p-values are unadjusted and are not a substitute for the registry's reported p-values. The registry's ANCOVA p-values reflect the sponsor's analysis method, while the recalculated values arise from a different, simpler comparison. The record does not describe whether the primary comparisons were adjusted for making comparisons across multiple dose groups. As a result, the p-values should be interpreted in the context of that reporting gap and the confidence intervals, rather than as stand-alone evidence.

Why can over-dispersion matter?

Over-dispersion is a general issue for some outcome models when the observed variability is greater than the model assumes. If a model understates variability, its standard errors may be too small and its intervals too narrow; if it accounts for variability appropriately, uncertainty estimates can be more realistic. The primary endpoint is analyzed as a continuous score with means and standard deviations, and the registry record does not identify an over-dispersion issue or describe a model that requires a particular over-dispersion adjustment. It would therefore be inappropriate to claim that over-dispersion affected these results.

How should the analysis population be understood?

The registry defines the primary outcome population as randomized participants who received at least 1 dose of study treatment, were not excluded due to GCP violations, and had non-missing baseline and post-baseline data. This is the stated population for the endpoint, rather than an assumption that every enrolled participant contributed to the analysis. The record does not report the number excluded for each reason, the handling of specific patterns of missing data, or whether additional analysis populations were used. Such details matter because the participants contributing observed endpoint data may differ from those without observed data.

More generally, methods for missing data depend on assumptions about why values are absent and what information is used to address them. The registry record's population definition does not establish which assumptions or imputation methods were applied. It is not possible to infer those methods from the endpoint summaries alone.

11. Confidence Intervals and Precision

The sponsor-reported 95% confidence intervals are the most direct guide to the precision of the primary comparisons. For AMG 827 140 mg versus placebo, the least-squares mean difference was -0.068, with a 95% confidence interval from -0.311 to 0.175. For AMG 827 210 mg versus placebo, it was -0.075, with a 95% confidence interval from -0.316 to 0.166. For AMG 827 280 mg versus placebo, it was -0.113, with a 95% confidence interval from -0.355 to 0.129.

All three intervals span zero and include estimates on both sides of zero. On this score, negative change from baseline indicates improvement, but a comparison interval crossing zero does not settle whether one group had a larger average improvement than placebo. The intervals show that the estimates are not precise enough to isolate a direction of difference. They do not prove that the groups are identical, and they do not identify which possible value is the true effect.

Confidence intervals are a general way to express sampling uncertainty, conditional on the method and its assumptions. A 95% confidence interval is not a statement that there is a 95% probability that the true effect lies within this particular interval. Rather, the confidence level describes the long-run behavior of the interval procedure under repeated use in comparable circumstances.

The registry also reports standard deviations for the group changes, which describe variation among participant outcomes rather than precision of the comparison itself. Precision depends on the comparison estimate and its uncertainty, as reflected in the interval. The record does not report a clinically important difference threshold for the ACQ score. Consequently, statistical uncertainty can be described, but the interval cannot be divided into clinically meaningful categories using a threshold absent from the record.

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 illustrations are planning calculations based on the observed effect, not the original trial design. For these continuous outcomes, the values are participants per arm, not total events. The displayed 80% power column estimates how many participants per arm would be needed, under the listed assumptions, for an 80% chance of detecting the stated effect if that effect were the true one. The 90% power column uses a higher target chance of detection under the same general planning framework.

The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is a stated fraction of the observed effect. The registry record does not report the original trial's sample-size assumptions or justification, so these illustrations should not be mistaken for a reconstruction of the study's original planning.

The rows for planning at 100% of the observed effect use the effect as observed in the posted comparison; the rows for planning at 75% and 50% use smaller fractions of that observed effect. Smaller effects are harder to distinguish from background variation, so planning for a smaller effect requires more participants per arm under the same assumptions. The curve provides a visual way to consider how the planning requirement changes with the effect size; it is not evidence that the study actually had a particular level of power.

These figures are planning illustrations, not observed power. Observed power is not a useful substitute for reading the estimate and its confidence interval after a study. The observed data are better understood through the reported effects and their uncertainty, while the planning figures show how assumptions about a future target effect can inform sample-size choices.

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 = -0.094; pooled standard deviation σ = 0.790.
2. n = 2 × (1.96 + 0.842)² × 0.790² / -0.094² = 1,108.5, rounded up to 1,109 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 <- -0.094                      # observed difference in means
sigma <- 0.789900329543066                     # 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(1199289)
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 = 1,109 (80% power), 1,484 (90% power)
 75% of observed effect: n per arm = 1,971 (80% power), 2,639 (90% power)
 50% of observed effect: n per arm = 4,434 (80% power), 5,936 (90% power)
power.t.test: 1109.4 
Simulated power at n = 1,109 per arm: 0.798 (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 reports sponsor analyses using ANCOVA, with least-squares mean differences and 95% confidence intervals. The additional recalculations are simple unadjusted comparisons reconstructed from the posted group means and standard deviations. They are included as a transparent summary-statistic comparison, not as a correction to or replacement for the sponsor's model.

The recalculated differences are -0.094 for AMG 827 140 mg versus placebo, -0.085 for AMG 827 210 mg versus placebo, and -0.129 for AMG 827 280 mg versus placebo. The sponsor's corresponding ANCOVA estimates are -0.068, -0.075, and -0.113. Differences between these values are expected because the ANCOVA estimates are model-based and the recalculations do not adjust for covariates. The registry record does not give enough information about the ANCOVA specification to reproduce its estimates from the group summaries.

Both sets of posted comparisons have confidence intervals that include zero, but the sponsor's reported analysis is the analysis to emphasize when describing the trial's own findings. Its p-values are 0.584, 0.539, and 0.358 for the three dose comparisons, respectively. These results do not establish a difference from placebo. The simple recalculations are useful as a check on the direction and approximate scale of the summary-data contrasts, but they should not be presented as if they were the sponsor's adjusted estimates.

15. Multiplicity, Missing Data and Analysis Populations

The primary outcome compares three AMG 827 dose groups with placebo. When several comparisons are made, a general statistical concern is multiplicity: the chance of finding at least one apparently unusual result can increase as more tests are considered. A prespecified adjustment or testing strategy can control the interpretation of such multiple comparisons. The registry record reports an ANCOVA estimate, confidence interval, and p-value for each dose-versus-placebo comparison, but does not report whether or how multiplicity was handled for the primary comparisons.

The registry also lists numerous secondary outcomes. Considering many outcomes and comparisons increases the importance of knowing whether analyses were prespecified and whether p-values were adjusted or arranged in a testing hierarchy. Those details are not reported in the registry record. Accordingly, secondary p-values should be regarded cautiously and should not be interpreted as if each were the sole planned test.

The endpoint population is defined as randomized participants who received at least 1 dose, were not excluded because of GCP violations, and had non-missing baseline and post-baseline data. That definition gives useful information about who contributed to the primary endpoint, but it does not explain how missing values were handled or why values were missing. The registry record does not report an imputation method, assumptions about missingness, sensitivity analyses, or a detailed accounting of exclusions from this population.

The record identifies ANCOVA as the sponsor's method but does not report its covariates or full model specification. It also does not report whether interim analyses were conducted, what stopping rules might have applied, or the original sample-size justification. These omissions limit independent assessment of the analysis plan and the degree to which the posted results reflect all planned analyses. They are reporting limitations; they do not establish that a particular method was or was not used.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: herpes zoster (1.4%), meningitis viral (1.4%), alanine aminotransferase increased (1.4%), degeneration of uterine leiomyoma (1.4%), otitis externa (1.3%).

The registry reports serious adverse events by group as the number of participants affected and the number at risk. The figures are: placebo, 1 of 76; AMG 827 140 mg, 3 of 74; AMG 827 210 mg, 1 of 76; and AMG 827 280 mg, 2 of 76. No deaths are reported in any group. These counts describe the serious adverse event information posted in the registry; they do not, by themselves, establish a causal relationship to study treatment.

The registry also gives selected serious adverse event terms with percentages: Herpes zoster, 1.4%; Meningitis viral, 1.4%; Alanine aminotransferase increased, 1.4%; Degeneration of uterine leiomyoma, 1.4%; Otitis externa, 1.3%; and Hypoglycaemia unawareness, 1.3%. The record does not provide enough detail in these summaries to infer individual risk or compare specific event types across all groups. Serious events are distinct from the separately reported count of participants with other adverse events; the latter should not be treated as a serious-event count.

These safety figures are descriptive. They do not establish that a treatment is safe or unsafe, and the small event counts do not support strong comparative conclusions. Interpretation would also depend on event definitions, follow-up, attribution, and completeness of reporting, details that are not reported in the registry record.

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

The posted results answer a focused question: how did the average ACQ score change through Week 12 in each dose group compared with placebo? The group means were negative, which the registry identifies as improvement from baseline. The sponsor's adjusted comparisons were also negative, but each 95% confidence interval included zero. The estimates therefore do not establish that any AMG 827 dose differed from placebo on this primary outcome.

That conclusion is not the same as proving that the groups had identical outcomes or that a difference is impossible. The confidence intervals cover a range of values in both directions, and the registry does not specify a threshold for a clinically important ACQ difference. Thus the results can be described in terms of observed estimates and uncertainty, but not classified against an unreported clinical threshold.

The findings also do not establish what would happen in populations or circumstances beyond this trial. The record concerns a phase 2 randomized, double-blind, placebo-controlled study in participants with inadequately controlled asthma, using the stated injection schedule and Week 12 endpoint. The registry information does not support generalizing the results to other groups, treatment schedules, follow-up periods, or outcomes.

Statistical evidence and medical decisions are not interchangeable. A p-value does not quantify the probability that a treatment works, and a non-significant comparison is not proof of no effect. Likewise, an observed numerical difference is not proof of a treatment effect. The appropriate reading is that the sponsor-reported analyses did not demonstrate a difference from placebo on the primary endpoint, with uncertainty remaining around the estimates.

The secondary summaries and safety counts add context but do not change the primary result into a definitive conclusion. Their interpretation is constrained by multiple outcomes, incomplete reporting of analysis details, and the absence of a reported multiplicity strategy. The record should be read as a report of estimates and their limits, rather than as a universal judgment about treatment effect.

18. Why This Trial Matters Statistically

This trial illustrates why a randomized comparison is read through both the point estimate and its uncertainty. The group means show how the ACQ score changed on average, while the sponsor's ANCOVA contrasts directly compare each dose with placebo. Confidence intervals make clear that the estimates remain compatible with more than one interpretation, including no difference.

It also illustrates why reported analysis methods matter. The unadjusted recalculations based on summary statistics are not the same as the sponsor's ANCOVA estimates. Both may be informative for different purposes, but they answer slightly different statistical questions. Giving priority to the sponsor's reported analysis avoids presenting a simple reconstruction as though it were the trial's planned model.

The outcome is a composite score with a stated direction: negative change represents improvement, and higher scores indicate worsening. Correctly identifying that direction prevents a common interpretive error. At the same time, a directionally favorable estimate does not establish a between-group effect when uncertainty is substantial.

Finally, this record highlights the importance of reporting the analysis population, missing-data methods, multiplicity handling, and model specification. The registry identifies the endpoint population and names ANCOVA, but leaves important implementation details unreported. Those details help readers judge reproducibility and the strength of conclusions. The statistical lesson is not that one particular result should be preferred because of its direction; it is that estimates, uncertainty, design, and reporting completeness need to be considered together.

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 specific to this trial arise from the information reported in the registry. The primary endpoint population required treatment exposure and non-missing baseline and post-baseline data, as well as no exclusion due to GCP violations. The record does not give a detailed account of participants excluded for each reason, so it is not possible to assess how those exclusions may have affected the analyzed population.

The sponsor's method is identified as ANCOVA, but the registry does not report the covariates, model specification, or missing-data approach. This limits independent reproduction of the least-squares mean differences and assessment of the assumptions behind the model. The simple recalculations are not a substitute for those missing details because they use unadjusted group summaries.

The record does not report the original sample-size justification, the planned effect size, or whether interim analyses were conducted. The planning illustrations based on observed effects cannot establish what the study was designed to detect. The registry also does not report how multiplicity was addressed across dose comparisons or secondary outcomes, which limits confirmatory interpretation of multiple p-values.

The primary outcome is a composite ACQ score, and the registry does not give a threshold for a clinically meaningful change. It is therefore possible to describe the score differences and their statistical uncertainty, but not to judge them against a trial-specific clinical threshold from the record. The Week 12 timeframe also limits conclusions to the endpoint window reported; longer-term outcomes are not established by these data.

Finally, only the posted registry summaries can be evaluated. The registry's listed secondary outcomes extend beyond the outcomes for which numerical summaries are available in the record, and the safety counts are descriptive. These reporting limits should not be mistaken for proof that unreported analyses or information do not exist elsewhere; they mean that such information is not reported in the registry record.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Evaluate the Dosing of AMG 827 for Subjects With Inadequately Controlled Asthma — NCT01199289.
• Registry record last updated 2021-11-26; status last verified 2021-10.
• Results first posted 2021-11-26.
• 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

NCT01199289 is a completed, randomized, double-blind, placebo-controlled phase 2 parallel-group study sponsored by Amgen. The registry reports enrollment of 315 participants and compares placebo with AMG 827 at 140 mg, 210 mg, and 280 mg.

The primary endpoint was change from baseline in ACQ composite score through Week 12. Negative change indicates improvement on the registry's described scale. The sponsor-reported ANCOVA comparisons with placebo had 95% confidence intervals that included zero, and the posted analyses did not establish a difference on the primary endpoint. The record also includes secondary outcome summaries and descriptive adverse event counts, with important analysis and reporting details not reported in the registry record.