The Effect of Inhibition of B7-mediated Costimulation on Allergic Airway Inflammation in Mild Atopic Asthmatics: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Effect of Inhibition of B7-mediated Costimulation on Allergic Airway Inflammation in Mild Atopic Asthmatics (NCT00784459), a randomized phase 2 study. The primary outcome is Baseline Percentage of Eosinophils Recovered in the BAL Prior to Segmental Allergen Challenge.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group phase 2 study comparing intravenous abatacept with intravenous placebo in people with mild atopic asthma. The registry describes triple masking. Enrollment was 24, with 13 participants in the abatacept arm and 11 in the placebo arm for the posted baseline outcome. The study is listed as completed.

The registry describes a sequence beginning with eligibility assessment, a titrated skin prick test, and baseline evaluation. Participants then underwent fiberoptic bronchoscopy with segmental allergen challenge and were randomized to abatacept or placebo. After 12 weeks of study drug, the study description says participants would undergo repeat segmental allergen challenge. The treatment schedules were the same in structure: intravenous doses every 2 weeks for 3 doses, followed by doses every 4 weeks for 2 doses. Abatacept and placebo were each described as administered at 10 mg/kg.

The study description identifies a planned primary response based on whether the percentage of eosinophils recovered in bronchoalveolar lavage fluid after segmental allergen challenge decreased by at least 50% compared with placebo. It also lists secondary areas of interest: airway obstruction and hyperreactivity, airway inflammation, symptoms, and safety. The registry record does not post analyses for those secondary outcomes. It posts two outcome measures, but the available numerical results concern baseline eosinophils before randomization. The registry record does not report the study’s original sample-size justification, detailed statistical analysis plan, or further allocation procedures.

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

Reasons for not completing this period:

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 Baseline Percentage of Eosinophils Recovered in the BAL Prior to Segmental Allergen Challenge, measured over baseline.

The posted outcome is the baseline percentage of eosinophils recovered in bronchoalveolar lavage fluid before segmental allergen challenge. It is recorded as a mean with a standard deviation and is described as collected before randomization to abatacept or placebo. Eosinophils are a cell type, and this measure expresses their share of the recovered cells in the lavage fluid.

The study description gives a different planned primary question: whether the percentage of eosinophils recovered after segmental allergen challenge would decrease by at least 50% with abatacept compared with placebo. Measuring cells in lavage fluid after a controlled allergen challenge was intended to assess allergic airway inflammation. The registry’s posted baseline outcome does not answer that treatment-response question: it describes values measured before randomization, rather than the response after treatment and repeat challenge.

Baseline measurements can characterize the groups and their starting values. They do not establish what happened after treatment. For that reason, the posted baseline means should not be interpreted as estimates of abatacept’s effect on the planned post-challenge endpoint. The registry record does not post numerical results for that planned comparison.

7. Results Posted in the Registry

• Treatment With Abatacept: mean 19.00 versus 42.00 on control; difference -23.00 (95% CI -42.51 to -3.49), standardized effect d = -1.00; z-test p = 0.021.

8. Statistical Interpretation

9. Statistical Methods Explained

The posted outcome is continuous: it is a percentage of cells, summarized by a mean and standard deviation in each group. A mean describes the group’s arithmetic average. A standard deviation describes how spread out individual observations are around that average. These summaries are useful for describing a continuous measure, but they do not show the full distribution of participant values or whether the measure followed a particular statistical model.

What does the mean difference describe?

The mean difference compares the two group means by subtracting the placebo mean from the abatacept-assigned group mean. For the baseline outcome, the reported difference is -23.0 percentage points. A negative sign indicates that the mean in the abatacept-assigned group was lower than the mean in the placebo group. Because the outcome was collected before randomization, this difference describes the observed starting values rather than a response to treatment.

In general, a group difference is a summary of separation between group averages. It does not tell whether every participant had a value in the same direction, and it does not describe all features of the distributions. A mean difference is most informative when read alongside the group sizes, the standard deviations, the timing of measurement, and the study question.

What does the standard deviation add?

The standard deviation gives a measure of participant-to-participant variability within each group. The placebo standard deviation was 30.0, and the abatacept-group standard deviation was 15.0. These are substantial relative to the corresponding means, particularly for the placebo group. A standard deviation is not the same as the uncertainty around a mean: it describes variation among observations, whereas a confidence interval for a mean or difference describes uncertainty in an estimated quantity.

General statistical note: continuous outcomes can have distributions that are skewed or contain influential values. Means and standard deviations alone cannot establish whether those features are present. The registry record does not report distributional checks, transformations, robust methods, or a detailed analysis plan for this outcome. Those methods should not be inferred from the summary statistics.

How should the p-value be read?

The unadjusted p-value is 0.021. In general, a p-value measures how incompatible the observed data are with a specified null model, assuming that model and the analysis assumptions hold. It is not the probability that the null hypothesis is true, nor is it the probability that the observed difference arose by chance. It also does not measure the size or importance of a difference.

The comparison is explicitly unadjusted. The registry record does not describe adjustments for covariates, stratification factors, or multiple testing in the posted result. The p-value therefore should be read as an unadjusted test of the baseline group comparison, not as a complete analysis of the planned post-treatment endpoint.

What does the effect estimate represent?

The reported mean difference is the estimate in the original percentage-point scale, which is generally easier to interpret than a standardized effect measure. A standardized effect expresses a difference relative to a measure of variability. Such effects can assist comparisons across scales, but their interpretation depends on the population and variability. The registry record does not give a clinical threshold for the baseline difference, so no conclusion about clinical importance follows from its size alone.

General statistical note: random allocation is intended to make treatment groups comparable on average, but it does not guarantee identical baseline characteristics in a particular study. Chance differences may occur, especially in a study with a small enrollment. In this record, the baseline measurement was collected before randomization, so group labels refer to the assignment participants later received; they cannot explain the earlier measurements.

What is the role of recalculation?

The reported comparison is an unadjusted recalculation from the posted group summaries. Such a calculation can provide a transparent comparison when individual participant data and a sponsor analysis are not posted. It does not reproduce an analysis that would require individual-level data, account for features not contained in summary statistics, or establish what the trial’s prespecified model was. The registry record does not report the trial’s detailed statistical plan or the original sample-size justification.

General statistical note: over-dispersion occurs when variability is greater than a chosen model expects. If an outcome has over-dispersion, standard errors and tests from a model that assumes less variability may be misleading unless the model accounts for it. The posted endpoint is a percentage, but the registry record does not report a model-based analysis or an assessment of over-dispersion. It is therefore not possible to say whether over-dispersion was assessed or how it would affect a post-treatment analysis.

Why does timing govern interpretation?

Statistical summaries must be matched to the question and measurement time. The stated primary question concerns a response after treatment and repeat allergen challenge. The posted numerical outcome is baseline, collected before randomization. Even a carefully calculated difference and confidence interval cannot turn a baseline comparison into an estimate of change, a post-treatment contrast, or the probability of reaching the planned response threshold. The registry record does not report numerical results for those analyses.

10. Confidence Intervals and Precision

The 95% confidence interval for the unadjusted baseline mean difference is -42.51 to -3.486 percentage points. In general, a confidence interval represents a range of values compatible with the observed data under the calculation’s assumptions. A narrower interval indicates greater precision; a wider interval indicates that the estimate is compatible with a broader range of values. The interval covers a wide span, so the exact size of the baseline difference is uncertain.

The interval does not include zero, in line with the unadjusted p-value of 0.021. This indicates evidence of a baseline difference under the unadjusted comparison’s assumptions. It does not indicate a treatment effect, because the measure was collected before randomization. The interval also does not establish whether any difference is important in clinical terms; the registry record does not report a clinical importance threshold for this baseline outcome.

Precision depends on the amount of information in the data and on variability. The posted outcome includes 11 participants in the placebo group and 13 in the abatacept group, with standard deviations of 30.0 and 15.0, respectively. These group sizes are limited, and the observed variability is not negligible. The confidence interval should therefore be considered alongside the small enrollment and the timing of measurement.

This interval concerns only the unadjusted baseline mean difference. It is not a confidence interval for the planned post-treatment primary response, and it cannot establish whether the 50% response criterion was reached. The registry record does not post a confidence interval for that planned comparison.

11. Planning a Similar Trial: Sample Size

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

The sample-size figures are planning illustrations based on the observed baseline effect, not the trial’s original design calculations. They report participants per arm for 80% power and participants per arm for 90% power. They are not total event counts: the event-count convention applies only to time-to-event calculations, and this outcome is continuous.

The listed planning 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 specified as a fraction of the observed effect. The registry record does not report the assumptions used in the trial’s original sample-size justification, so these illustrative calculations should not be taken as a reconstruction of the study’s planned design.

In general, power is the probability that a planned statistical test will detect an effect of a specified size if that effect is present and the assumptions hold. The 80% power column gives the number of participants per arm used in the illustration to target an 80% chance of detecting the stated effect. The 90% power column gives the number per arm for a 90% target. These are planning targets, not guarantees about the result of a particular trial.

The rows consider planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. Smaller effects need more participants to detect under otherwise unchanged assumptions. The calculations use the observed baseline comparison, which is not the planned post-treatment response and is not necessarily a realistic treatment effect. This is not observed power; it is an illustration of sample sizes under stated assumptions.

12. 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 = -23.000; pooled standard deviation σ = 23.061.
2. n = 2 × (1.96 + 0.842)² × 23.061² / -23.000² = 15.8, rounded up to 16 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 <- -23                      # observed difference in means
sigma <- 23.0611834435742                     # 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(784459)
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 = 16 (80% power), 22 (90% power)
 75% of observed effect: n per arm = 29 (80% power), 38 (90% power)
 50% of observed effect: n per arm = 64 (80% power), 85 (90% power)
power.t.test: 16.8 
Simulated power at n = 16 per arm: 0.780 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry record lists two outcome measures and no posted analyses. It identifies a planned primary question in the study description and describes secondary areas involving airway obstruction and hyperreactivity, airway inflammation, symptoms, and safety. However, numerical results for the planned post-challenge primary response and for secondary outcomes are not posted in the available record. The numerical outcome reported for interpretation is the baseline percentage of eosinophils before randomization.

Multiplicity arises when several hypotheses or outcomes are tested. In general, conducting multiple tests can increase the chance of at least one apparently unusual result if each test is interpreted without accounting for the collection of tests. Possible approaches include defining a primary outcome in advance, specifying a testing order, or adjusting significance levels. The registry record does not report whether multiplicity was addressed in the trial’s statistical plan or whether the posted p-value was part of a wider testing strategy. The reported p-value is unadjusted and should not be treated as accounting for unreported analyses.

The record also does not report how missing outcome observations were handled, whether all randomized participants were included in each analysis, or whether analyses followed an intention-to-treat or another population definition. The posted group sizes for the baseline outcome are available, but they do not by themselves establish the analysis population for post-treatment outcomes. The number of participants with missing post-challenge measurements, if any, is not reported in the registry record.

These omissions limit assessment of the planned treatment comparison. Missing-data methods can affect an estimate when outcome availability is related to participant characteristics or outcomes. Analysis populations can also differ in whether they include all randomized participants or only participants with particular follow-up data. Without the relevant registry entries or posted analyses, neither the direction nor the size of any such effect can be determined.

14. 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.

The posted adverse-event figures list 11 participants at risk in the placebo group and 13 in the abatacept group. The registry reports zero participants affected by serious adverse events in each group and zero affected by other adverse events in each group. These counts describe the posted adverse-event categories and their recorded participant counts.

Zero affected participants in the registry record should not be expanded into a claim that no safety concerns occurred in every sense. The record does not post serious adverse-event terms, and death counts are not reported in the registry record. It also does not provide event timing, severity grades, withdrawals due to adverse events, or a detailed account of how adverse events were collected. The posted counts therefore offer limited information about safety and should be interpreted within those reporting limits.

These figures are not a comparison of risks adjusted for follow-up time, and the record does not provide a safety analysis plan. With the posted group sizes, the counts alone cannot characterize the full range of possible adverse outcomes. They also do not resolve the study’s planned efficacy question, which concerns post-treatment eosinophil response after allergen challenge.

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

The numerical result shows that the baseline mean percentage of eosinophils in bronchoalveolar lavage fluid was lower in the group later assigned to abatacept than in the placebo group. The reported unadjusted mean difference was -23.0 percentage points, with a 95% confidence interval from -42.51 to -3.486 and an unadjusted p-value of 0.021. This is evidence of a difference in the recorded baseline values under the unadjusted comparison.

It does not show that abatacept reduced eosinophils. The baseline outcome was collected before randomization, and therefore before either assigned treatment could have caused a change. The study description’s planned primary question was about a post-treatment reduction after repeat segmental allergen challenge. The registry record does not post numerical results for that treatment-response question. It also does not post analyses of the listed secondary outcomes.

The result does not establish that the observed baseline difference is clinically important. The registry record gives no clinical interpretation threshold for the baseline percentage. Nor does the confidence interval resolve the treatment question: it quantifies uncertainty about the baseline group contrast, not about a post-treatment effect. The reported adverse-event counts are limited to the categories posted, and death counts are not reported in the registry record.

In general, randomized comparisons are designed to estimate the effects of assigned interventions by comparing outcomes measured after assignment, using an appropriate analysis. Baseline measurements can help describe the groups and may be relevant to an analysis plan, but a baseline contrast alone is not evidence of benefit or harm. Here, the most defensible reading is descriptive: the posted baseline values differed between the two groups. Conclusions about abatacept’s effect on allergic airway inflammation, symptoms, airway function, or safety cannot be drawn from these baseline results alone.

16. Why This Trial Matters Statistically

This trial illustrates why measurement timing is central to statistical interpretation. A registry record may describe a planned treatment question while posting numerical results for a different time point or outcome. Here, the study description identifies a post-treatment eosinophil response after repeat allergen challenge as the planned primary question, while the numerical means concern baseline values collected before randomization. Keeping those estimands separate prevents a baseline group difference from being mistaken for a treatment effect.

The example also shows why statistical significance and scientific relevance are distinct. The unadjusted baseline comparison has a p-value of 0.021 and a confidence interval that does not include zero. Those features concern evidence for a baseline difference under the unadjusted calculation. They do not change the outcome’s timing or establish clinical importance. A statistically unusual baseline imbalance is not an intervention effect.

The registry record’s small enrollment and limited posted analyses make transparency about uncertainty especially important. The confidence interval spans a broad range of baseline differences, and the record does not post the planned primary analysis or secondary outcome analyses. Readers need to know not only what an estimate and p-value say, but also which participants, measurement time, and outcome they refer to.

For clinical researchers, the record is a useful reminder to align registry outcomes, analysis populations, missing-data handling, multiplicity procedures, and sample-size planning with the scientific question. Several of those details are not reported in the registry record. Their absence does not establish that they were not considered; it means the record does not allow those methods to be evaluated. Clear reporting of the post-treatment endpoint and its uncertainty would be needed to assess the planned treatment comparison.

17. 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 key limitation for interpreting the numerical result is that the posted outcome is a baseline measure collected before randomization. It does not measure change after treatment, and it is not the post-challenge response described as the planned primary question. The registry record does not post numerical results for that planned endpoint.

The enrollment was 24, and the posted baseline outcome includes 11 participants in the placebo group and 13 in the abatacept group. The confidence interval is broad, so the exact size of the baseline difference is uncertain. The registry record does not report individual participant data or distributional summaries beyond the means and standard deviations. It is therefore not possible to assess the shape of the outcome distribution, identify influential observations, or examine participant-level changes.

The registry record does not report the trial’s original sample-size justification, detailed statistical analysis plan, interim analyses, missing-data methods, or analysis-population definitions. It also does not report whether multiplicity was addressed. These are reporting gaps; they do not establish what the investigators did or did not do. No numerical secondary outcome analyses are posted, and the record does not provide the planned post-treatment comparison needed to assess the stated primary question.

Safety interpretation is also limited. The posted serious and other adverse-event counts are zero in both groups, but serious-event terms are not listed and death counts are not reported in the registry record. The record does not provide further details on event ascertainment, timing, or withdrawals. These limitations mean the posted counts should not be treated as a complete characterization of safety.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: The Effect of Inhibition of B7-mediated Costimulation on Allergic Airway Inflammation in Mild Atopic Asthmatics — NCT00784459.
• Registry record last updated 2014-02-05; status last verified 2013-12.
• Results first posted 2014-02-05.
• 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.

21. Record Summary

The registry record describes a completed, randomized, triple-masked, parallel-group phase 2 study of intravenous abatacept versus intravenous placebo in mild atopic asthma. Enrollment was 24. The study description identifies a planned primary question about a post-treatment reduction in eosinophils recovered from bronchoalveolar lavage fluid after segmental allergen challenge.

The posted numerical result is a baseline measure collected before randomization. Mean percentages were 42.0 in the placebo group and 19.0 in the abatacept group; the unadjusted mean difference was -23.0, with a 95% confidence interval from -42.51 to -3.486 and an unadjusted p-value of 0.021. These values describe baseline group differences, not a treatment effect. The registry record does not post numerical results for the planned primary response or secondary outcomes.