A Two Week Study to Assess the Tolerability of AZD9668 in Chronic Obstructive Pulmonary Disease (COPD) Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Two Week Study to Assess the Tolerability of AZD9668 in Chronic Obstructive Pulmonary Disease (COPD) Patients (NCT00703391), a randomized phase 2 study. The primary outcome is Alanine Aminotransferase (ALT).

2. Clinical Question

The registry describes a two-week study of AZD9668 in patients with mild to moderate chronic obstructive pulmonary disease (COPD). Its stated purpose was to assess tolerability and pharmacokinetics. The primary outcome reported was the number of participants whose alanine aminotransferase (ALT) level was greater than 3 times the upper limit of normal during the study’s specified measurement period.

The posted outcome is a count of participants meeting the ALT threshold, not a continuous summary of ALT values. The record reports no participants meeting that threshold in either arm. That count addresses this particular laboratory safety outcome; it does not, by itself, describe all aspects of tolerability or pharmacokinetics. The registry lists sputum neutrophil counts, AZD9668 sputum concentrations and quantitative sputum bacteriology among the secondary outcomes, but the primary question remains the occurrence of the specified ALT elevation.

3. Trial Design

This was a randomized, phase 2, parallel-group study comparing AZD9668 with placebo. The official title describes a two-week, double-blind, placebo-controlled study of orally administered AZD9668. The registry classifies masking as quadruple. It does not identify which roles were masked or provide further operational details about how masking was implemented.

The registry lists an enrollment of 18. The AZD9668 arm has 12 participants listed for the primary outcome, and the placebo arm has 6. These are the arm-level counts associated with that posted outcome; the record does not explain whether any participants were excluded from its analysis or how missing measurements were handled. The study status is completed. The recorded start is 2008-06, and primary completion is recorded as 2008-09.

The study’s stated purpose was to assess tolerability and pharmacokinetics in patients with mild to moderate COPD. The primary outcome presented is an ALT threshold event. The listed secondary outcomes concern sputum absolute neutrophil count, sputum differential neutrophil count, AZD9668 sputum concentrations and quantitative sputum bacteriology. The registry records 20 outcome measures, while four secondary outcomes are named in the record.

Randomization and a concurrent placebo group provide a design for comparing assigned groups, while parallel-group allocation means the arms are followed as separate groups rather than as a crossover. Those general design features do not establish that the groups were comparable in every respect or that the posted outcome analysis adjusted for any baseline characteristics. Such details, along with the randomization procedure and any allocation-concealment procedures, are not reported in the registry record.

4. Participant Flow and Accounting

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

Overall Study

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 Alanine Aminotransferase (ALT), measured over Throughout the duration of the study (pre-dose, cmax, steady state, end of dosing and post dose).

The primary outcome counts participants whose alanine aminotransferase (ALT) level was greater than 3 times the upper limit of normal. The registry describes the measurement period as throughout the study, including pre-dose, cmax, steady state, end of dosing and post dose. The reported unit is participants, so the result is a count of people who crossed the stated threshold, rather than a mean ALT concentration or a change in ALT.

ALT is a laboratory measure used in monitoring liver-related safety. A threshold outcome can make a notable laboratory elevation easier to identify and count than a summary such as an average value, which could conceal individual high readings. That is a general reason for using threshold-based safety outcomes. The registry does not state why this particular threshold was selected, how many ALT measurements were available per participant, or how values between scheduled measurements were treated.

The posted count is zero in the AZD9668 arm and zero in the placebo arm. This means that no participants in either listed arm were recorded as meeting the specified threshold in the posted primary outcome. It does not mean that all ALT values were identical, that no participant had any ALT increase, or that other laboratory or clinical safety outcomes were absent. Those conclusions would require information beyond this endpoint count.

The registry record does not report an analysis population definition, a comparison estimate with uncertainty, or a statistical test for this endpoint. Consequently, the outcome can be described as a pair of observed counts, but the record does not support a quantified estimate of the difference between assigned groups.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a randomized, parallel-group comparison of AZD9668 and placebo. For the primary outcome, each participant is classified according to whether ALT exceeded the stated threshold during the recorded period. Such an outcome is binary: a participant either meets the event definition or does not. The posted values are counts of participants with the event, alongside arm-level counts for that endpoint.

What does a binary outcome show?

A binary count focuses on whether a defined event occurred, rather than the size or trajectory of a measurement. That can be useful when the clinical or safety question concerns crossing a prespecified threshold. The trade-off is that different underlying patterns can produce the same count. For example, a count alone does not reveal how close non-event values came to the threshold, the distribution of all ALT measurements, or the timing and duration of any elevations. Those details are not reported in the registry record.

For this endpoint, the registry reports zero events in each arm. When event counts are zero in both groups, the observed data do not distinguish the groups on this event count. That is not the same as demonstrating equivalence or establishing that the true event probabilities are identical. In general, estimates based on uncommon outcomes can be imprecise, particularly when the amount of information is limited. The registry does not provide a confidence interval or another inferential estimate that would quantify the range of plausible comparative effects.

Why is no two-arm comparison reported?

The record indicates that values are missing or not estimable for a two-arm comparison of this binary endpoint. This is consistent with the limitation of trying to estimate a relative or absolute group contrast when neither arm has a recorded event. A model or test may require a method designed for sparse or zero-event data; the registry does not report an alternative method, a model specification, or a statistical test for this endpoint. No unadjusted re-calculation is reported, and there is no reported discrepancy requiring the sponsor’s analysis to take precedence.

In general, an effect estimate summarizes the contrast between groups, while an uncertainty interval indicates how much that estimate could vary given the data and method. A p-value, when used, addresses compatibility with a specified null hypothesis under assumptions of the test. Neither a p-value nor an interval replaces the observed counts, and neither can be reconstructed as a reported result when the registry provides no such analysis. No p-value or confidence interval for the primary outcome is reported in the registry record.

What do the planning assumptions mean?

The record describes planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation between arms and a normal approximation. The effect to detect is expressed as a stated fraction of the observed effect. These are assumptions for prospective planning illustrations, not evidence about what the study was originally designed to detect. The registry does not report the trial’s own sample-size justification, a target effect used in its original design, or its planned power.

In general, a planning calculation asks how much information might be needed to detect a chosen effect under specified assumptions. It depends on the assumed effect, the outcome’s variability or event frequency, the allocation scheme and the chosen error criterion. If an observed effect is used as the starting point, that value can be unstable, especially when the observed event counts are zero. A planning illustration based on such an effect should not be read as observed power, nor as evidence that the trial was adequately or inadequately powered. The registry does not report numerical planning targets in the record summarized for this analysis.

For outcomes that are counts or rates, over-dispersion is a general statistical concern: it occurs when observations vary more than a simple model assumes. Ignoring it can make standard errors too small and uncertainty appear narrower than warranted. The posted primary outcome is a participant count for a binary event, not a reported count-rate model, and the registry does not report an over-dispersion analysis for it. This general concept should not be mistaken for a feature of the analysis actually used in this trial.

How should missing information affect interpretation?

A registered outcome count does not reveal by itself how missing laboratory measurements were treated. If a participant has no usable measurement during a relevant time, an analysis needs a rule for deciding how that participant contributes. Different rules can affect the count and the comparison. The registry does not report the amount or handling of missing primary-outcome data, the analysis population, or a sensitivity analysis. Therefore, the posted counts can be described, but the completeness and robustness of the endpoint analysis cannot be assessed from this record.

The registry lists 20 outcome measures and 4 named secondary outcomes. Multiple outcomes can raise a multiplicity issue because considering many statistical tests may increase the chance of at least one apparently unusual finding. That is a general concern. The record does not report a multiplicity adjustment, a hierarchy among outcomes, or a prespecified strategy for interpreting multiple analyses. It also lists zero posted analyses, so no detailed analysis plan or model outputs can be assessed from the registry record.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("AZD9668", "Placebo")
n      <- c(12, 6)
events <- c(0, 0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  AZD9668  12  0  0.0  0.0  26.5  0.0  24.2  0.2  24.7
ARM  Placebo  6  0  0.0  0.0  45.9  0.0  39.0  0.4  41.0

11. Multiplicity, Missing Data and Analysis Populations

The record lists 20 outcome measures and names 4 secondary outcomes: sputum absolute neutrophil count, sputum differential neutrophil count, AZD9668 sputum concentrations and quantitative sputum bacteriology. Considering multiple endpoints can create a multiplicity problem in general: when many hypotheses are tested, the chance of at least one apparently notable result can increase. The registry record does not report whether any adjustment was planned or applied, how outcomes were prioritized, or whether a formal testing hierarchy was used.

The registry lists zero analyses. It does not provide a detailed statistical analysis plan, identify the analysis population for the primary endpoint, or report how missing outcome data were handled. The primary endpoint counts are posted, but the record does not say whether all randomized participants had complete ALT measurements throughout the stated period, how partial follow-up was treated, or whether exclusions occurred. These omissions limit assessment of how the counts were derived and whether the analysis followed a prespecified plan.

In general, missing data can affect both the number of participants contributing to an analysis and the interpretation of a group comparison. The effect depends on why data are missing and how the analysis handles them. No missingness assumptions or sensitivity analyses are reported in the registry record, so their consequences cannot be evaluated. The available counts should therefore be read as the posted outcome information, without assuming a particular missing-data method or analysis population.

12. 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 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events for AZD9668 as 0 among 12 participants at risk and for placebo as 0 among 6 participants at risk. These are event counts with the corresponding at-risk counts stated for serious adverse events. They indicate that no serious adverse events were recorded in either group in the posted figures. Zero observed events do not establish that the underlying risk is zero; they describe what was recorded in this study.

Other adverse event counts are listed as 5 for AZD9668 and 4 for placebo. The registry does not report at-risk counts for these figures, so rates cannot be determined from the record. The counts alone do not establish whether events were mild or severe, related to treatment, or reported by unique participants rather than as multiple events per participant. Those details are not reported in the registry record.

Death counts are not reported in the registry record. No top serious adverse event terms are listed. The serious-event counts should be interpreted separately from the primary ALT outcome: they are different safety summaries with different event definitions. None of these figures alone captures the full tolerability or pharmacokinetic profile.

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

The posted primary result is straightforward as a count: no participant in either arm was reported to have ALT greater than 3 times the upper limit of normal during the stated period. The study also reports no serious adverse events in either group. These findings describe the recorded outcomes in this two-week study; they are not proof that AZD9668 cannot be associated with ALT elevations or serious adverse events.

There is no reported inferential comparison for the primary binary outcome. With no events recorded in either group, the posted results do not quantify a comparative effect or its precision. A lack of an observed difference is not the same as evidence that groups are equivalent. Establishing equivalence or non-inferiority would require an appropriate design, a defined margin and an analysis with uncertainty; these details are not reported in the registry record.

The counts also do not answer every question about tolerability. The registry lists other adverse event counts, but does not report their at-risk denominators or enough detail to interpret them as rates or treatment-related events. It does not report death counts. Pharmacokinetics is part of the study’s stated purpose, and AZD9668 sputum concentrations are listed as a secondary outcome, but the outcome values and analyses are not reported in the registry record for this interpretation.

Likewise, the findings should not be generalized beyond the recorded study population, intervention, comparator, duration and outcomes. The registry describes patients with mild to moderate COPD and a two-week study of oral AZD9668 versus placebo. It does not establish longer-term effects, effects in other populations, or outcomes not posted in the record. The most defensible conclusion is limited: the registry’s primary ALT threshold count was zero in each arm, while the comparative uncertainty and broader safety profile remain unquantified in the posted record.

14. Why This Trial Matters Statistically

This study illustrates why a safety count and a treatment comparison are not interchangeable. The primary endpoint is clinically interpretable as a defined ALT threshold event, and the registry reports zero such events in both arms. Yet the same zero counts that are easy to describe do not yield an informative estimate of the difference between groups through a conventional comparison. The absence of a reported interval or test also means the reader cannot use those statistics to judge precision.

It is a useful example of the limits of small, short-duration studies for uncommon binary outcomes. In general, when events are rare, a study may observe no events even when the underlying probability is not zero. A zero count is therefore an observation, not proof of absence. The record’s arm-level outcome counts and serious adverse event figures can be stated exactly, while broader claims about comparative safety cannot be derived from them.

The record also highlights the importance of analysis transparency. The registry lists multiple outcome measures and several secondary outcomes, but it does not report a detailed statistical analysis plan, a missing-data approach, or a multiplicity strategy. Without those details, readers cannot assess all analytical choices that may shape results across endpoints. The registry’s indication that a two-arm binary comparison is not estimable is itself important: it signals that a numerical contrast should not be improvised from the posted outcome.

Finally, planning illustrations must be separated from evidence observed in the trial. The stated planning assumptions describe a calculation framework, not observed power or the trial’s original justification. Keeping those distinctions clear helps prevent a null or unestimable comparison from being mistaken for proof of no effect. The value of the record is its transparent description of a narrow outcome and its limits, rather than a definitive safety conclusion.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The primary outcome is a threshold count, and the registry does not provide the underlying ALT measurements, their distribution, changes from baseline or timing of individual elevations. A count of zero in each group cannot show whether values differed below the threshold. It also cannot show how consistently measurements were obtained across the stated pre-dose, cmax, steady state, end-of-dosing and post-dose time points.

The record does not report an inferential estimate, confidence interval or p-value for the primary comparison, and it states that the two-arm binary comparison is not estimable from the posted values. Therefore, comparative precision cannot be evaluated from this endpoint. The registry does not explain how the primary-outcome counts relate to any missing measurements, nor does it identify the analysis population or describe exclusions.

The study is described as lasting two weeks and enrolling patients with mild to moderate COPD. Its findings therefore cannot, from this record alone, address longer-term outcomes or other patient groups. The registry classifies the study as randomized and quadruple masked, but does not report the randomization procedure, the identities of masked roles, or checks on whether masking was maintained.

Safety reporting is also incomplete for broader interpretation. The serious adverse event counts have at-risk denominators, but the other adverse event counts do not. The record does not report the other events’ terms, severity, relatedness or whether counts represent participants or event occurrences. Death counts are not reported. Although the study’s purpose includes pharmacokinetics and several secondary outcomes are named, detailed outcome results and analyses for those measures are not included in the record.

Finally, the registry lists multiple outcome measures but does not report a multiplicity plan, a detailed analysis plan, missing-data methods or the original sample-size justification. These omissions prevent a full assessment of prespecification and analytic robustness.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Two Week Study to Assess the Tolerability of AZD9668 in Chronic Obstructive Pulmonary Disease (COPD) Patients — NCT00703391.
• Registry record last updated 2012-01-26; status last verified 2012-01.
• Results first posted 2010-12-29.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry record describes a completed, randomized, phase 2, quadruple-masked, parallel-group, placebo-controlled study of orally administered AZD9668 in mild to moderate COPD. The study’s stated duration was two weeks, its enrollment was 18, and its recorded start and primary-completion dates are 2008-06 and 2008-09.

The primary outcome was the number of participants with ALT greater than 3 times the upper limit of normal during the specified study period. The registry lists 0 among 12 for AZD9668 and 0 among 6 for placebo. It reports no estimable two-arm binary comparison. Serious adverse event counts are 0 among 12 at risk for AZD9668 and 0 among 6 at risk for placebo. Other adverse event counts are 5 and 4, but their at-risk counts are not reported; death counts are not reported in the registry record.

The registry names 4 secondary outcomes and lists 20 outcome measures. It does not report a detailed analysis plan, missing-data approach, analysis population, multiplicity strategy, or inferential comparison for the primary outcome.