Phase 2a, AMP Challenge, Dose Escalation Study to Assess the Dose Response for Topical Efficacy and Systemic Activity in Asthmatic Subjects: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase 2a, AMP Challenge, Dose Escalation Study to Assess the Dose Response for Topical Efficacy and Systemic Activity in Asthmatic Subjects (NCT02991859), a randomized phase 2 study. The primary outcome is Provocative Concentration (PC) of Adenosine 5' Monophosphate (AMP) Causing a 20 Percent (%) Reduction in Forced Expiratory Volume in 1 Second (FEV1) (AMP PC20)- Dose Response Analysis.

2. Clinical Question

The registry record describes a randomized, quadruple-masked, incomplete-block, two-period crossover study. Its objective was to examine topical efficacy in the airways alongside systemic activity, rather than to answer a broad question about long-term asthma control or clinical outcomes. The primary endpoint uses an AMP challenge to assess airway responsiveness, while the study description also identifies 24-hour plasma cortisol suppression as a measure of systemic activity. The results available in the registry record do not include a cortisol outcome, so the combined topical-efficacy and systemic-activity question cannot be evaluated from these results alone.

3. Trial Design

This completed Phase 2a study enrolled 56 participants with asthma. It was randomized, placebo-controlled, quadruple-masked and used an incomplete-block crossover design. Participants were assigned to one of 5 or 12 treatment sequences, with one or two treatment periods. The registry describes each period as five consecutive 7-day phases of escalating doses of FF, FP, BUD or placebo. A washout period of 25 to 42 days separated periods. The planned study duration for an individual participant was approximately 13 or 24 weeks, including follow-up.

The crossover structure means that a participant could receive different treatments in separate periods, rather than being assigned only one treatment for the entire study. In general, within-participant comparisons can reduce the influence of stable differences between participants. That advantage depends on the conduct and analysis of the crossover design, including the handling of period effects and incomplete treatment sequences; detailed results for those features are not reported in the registry record. The registry identifies the design as an incomplete block, so not every participant necessarily received every treatment.

The primary analysis used participants who were randomized, received at least one dose and had at least one post-dose pharmacodynamic measurement. Analysis at a specified time point was limited to participants with available data at that time. The primary measurement was assessed 12 hours after the last dose on Day 7. This is a defined, short-term pharmacodynamic assessment during dose escalation, not a measure of sustained outcomes over months or years.

The registry describes treatment sequences and escalating doses, but the available arm descriptions are incomplete. A full account of the planned sequence allocation, treatment-period analysis, sequence effects and sample-size justification is not reported in the registry record. These omissions limit how fully the crossover analysis can be reconstructed from the registered information.

4. Participant Flow and Accounting

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

Before assignment: A total of 165 participants were screened of which 109 failed screening and 56 participants were enrolled in study. Of the 56 participants enrolled,only 54 were randomized and received at least one dose of study medication. Only 7 participants were withdrawn from the study, 2 participants who were not withdrawn, entered Period 2 esc Phase 1.

Only the first 6 of 17 groups are shown.

Period 1,Escalation Phase 1,Upto Day 7

Reasons for not completing this period:

Period 1,Escalation Phase2,Upto Day 14

Period1, Escalation Phase3, Upto Day21

Reasons for not completing this period:

Period1, Escalation Phase4, Upto Day28

Analysis population for the primary outcome, as defined in the registry: Pharmacodymanic (PD) population. Participants who were randomized and received at least one dose of trial medication and also have at least one post dose PD measurement. Only those participants with data available at the specified time points were analyzed.

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.

Only the first 6 of 18 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Provocative Concentration (PC) of Adenosine 5' Monophosphate (AMP) Causing a 20 Percent (%) Reduction in Forced Expiratory Volume in 1 Second (FEV1) (AMP PC20)- Dose Response Analysis, measured over 12 hours post last dose on Day 7.

The primary endpoint is AMP PC20: the concentration of adenosine 5′-monophosphate that produces a 20% reduction in forced expiratory volume in 1 second (FEV1). It is expressed in milligrams per milliliter. The registry reports the endpoint 12 hours after the last dose on Day 7 and presents treatment-specific predicted means with 95% confidence intervals.

In general, PC20 is used as a measure of airway responsiveness to a provocation challenge. A higher concentration needed to produce the specified FEV1 reduction corresponds to a less responsive airway under that test condition. This is a pharmacodynamic measure; it is not itself a measure of symptoms, exacerbations, everyday lung function or long-term disease control.

The registry describes how the percentage fall in FEV1 was calculated from the highest measurements after saline and after AMP. It then specifies a three-parameter Emax model using log2-transformed AMP PC20 as the outcome, with a common Emax across FF, FP and BUD and an unstructured variance-covariance matrix. The reported treatment means and confidence intervals are predicted estimates from that model. The model-based presentation allows the dose-response patterns to be considered, but the registry does not report a single direct estimate comparing the medicines at a common dose or a formal test of differences between their dose-response curves.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary analysis is a model-based dose-response analysis rather than a collection of independent pairwise comparisons. The registry specifies a three-parameter Emax model with log2-transformed AMP PC20 as the outcome. Emax models are generally used to describe a response that changes with dose and approaches a plateau. The three-parameter form describes the response curve through a limited set of parameters; the registry record does not explain the clinical interpretation assigned to each fitted parameter in the additional analysis rows.

What does the Emax model contribute?

In general, a dose-response model uses the observations across dose levels to estimate a curve, rather than treating every dose as an unrelated treatment group. The predicted means and their confidence intervals represent model-based estimates at the listed treatments and doses. Such predictions depend on whether the chosen curve adequately describes the observed relationship. The registry does not report model diagnostics, sensitivity analyses or other checks of fit, so those aspects cannot be assessed from the record.

The model assumes a common Emax across FF, FP and BUD. In general, this assumption constrains the modeled maximum response to be shared among the medicines, even while other aspects of their dose-response patterns may differ. An unstructured variance-covariance matrix allows the model to represent covariance among measurements without imposing a simpler prespecified pattern. These are modeling choices; the registry does not report results from alternative assumptions, so their influence on the estimates is not known.

How should the confidence intervals be read?

The primary endpoint is reported as a predicted mean with a 95% confidence interval. In general, a confidence interval describes the precision of an estimate under the model and its assumptions. A narrower interval indicates greater precision than a wider interval, but neither interval is a guarantee about the range of individual responses. The intervals in the registry are the uncertainty estimates for the modeled means, not standard deviations among participants.

The endpoint is log2-transformed for model fitting, while the reported unit is milligrams per milliliter and the displayed means are predicted estimates. A transformation changes the scale on which the model is fitted. General interpretation therefore needs to distinguish the model’s analysis scale from the reported clinical measurement scale. The record does not give a detailed account of the back-transformation or the specific parameterization used to generate every displayed estimate.

What is the crossover implication?

In general, a crossover design allows treatment responses from different periods to be compared within participants, which can account for stable individual characteristics. It also requires attention to the ordering of treatments, period effects and possible carryover from an earlier treatment. This study used an incomplete-block, two-period design and included a washout period between treatment periods. The registry does not report detailed model terms for sequence, period or carryover, nor does it report an assessment of these effects. The design label and washout duration alone do not establish how these issues were handled analytically.

The primary population is the pharmacodynamic population: randomized participants who received at least one dose and had at least one post-dose pharmacodynamic measurement. For the specified analysis time, only participants with data available at that time were included. This is not the same as analyzing every randomized participant regardless of whether they received treatment or contributed a measurement. The registry does not report an imputation method or a sensitivity analysis for missing outcome data.

Why are some comparisons not calculated?

The posted primary outcome is a multi-dose modeled response, not a simple two-arm comparison with a single treatment and placebo. A recalculation based on two groups’ summary means would not reproduce the sponsor’s dose-response model or account for the crossover design, common-Emax assumption and covariance structure. The reported treatment estimates and model analyses are therefore the relevant primary results. Recalculated pairwise comparisons are available for selected secondary baseline measurements, but they answer a different and narrower question.

For those secondary comparisons, the recalculated mean differences are unadjusted and based on posted summary statistics. In general, an unadjusted test does not account for examining multiple comparisons or outcomes. The planning illustrations associated with these recalculations are hypothetical sample-size calculations based on the observed effect and stated assumptions. They are not observed power, and they are not the trial’s original sample-size justification. The registry record does not report the study’s sample-size justification or a multiplicity plan.

10. Analyses Reported by the Sponsor

The primary endpoint is reported using the sponsor’s model-based dose-response analysis: a three-parameter Emax model, log2-transformed AMP PC20, a common Emax across FF, FP and BUD, and an unstructured variance-covariance matrix. The treatment-specific means are described as predicted estimates, with 95% confidence intervals. The registry also lists estimates and confidence intervals from analyses labeled as three-parameter Emax models, but does not clearly identify the interpretation of each listed parameter or provide p-values for those rows.

No simple recalculation from two treatment-group summaries would replicate this primary model or its crossover analysis. The recalculated results elsewhere concern selected secondary measurements taken before dosing on Day 1 in Period 1; they are not recalculations of AMP PC20. The registry record does not flag a discrepancy between a sponsor-reported estimate and a recalculated primary comparison. A direct recalculated comparison is not available for the primary endpoint because it is reported as a dose-response model rather than a two-arm contrast.

Accordingly, the sponsor’s model-based results are the appropriate results to describe for the primary outcome. Their interpretation remains bounded by the registered analysis details: the model assumptions are stated, but diagnostics, alternative model results and a direct formal comparison of the medicines’ dose-response curves are not reported in the registry record.

11. Multiplicity, Missing Data and Analysis Populations

The registry reports 50 outcome measures and 21 analyses, while results are posted for 45 secondary outcome measures. These counts indicate that the registered study included multiple endpoints and analyses, but do not by themselves establish which outcomes were prespecified as confirmatory, how the testing hierarchy was organized or whether every planned analysis was completed. A multiplicity adjustment or testing strategy is not reported in the registry record.

Multiplicity matters because repeated statistical testing can increase the likelihood of finding at least one apparently notable result when there is no true effect. In general, a prespecified strategy can control how results are interpreted across several comparisons. Since the registry does not describe such a strategy, the selected secondary recalculations should be read as unadjusted and exploratory. The Emax model is reported for the primary endpoint, but details about how the model-based analyses fit into an overall hypothesis-testing plan are not reported.

The primary population includes randomized participants who received at least one dose and had at least one post-dose pharmacodynamic measurement. The record further states that only participants with data available at a specified time point were analyzed. This clarifies eligibility for the analysis but does not explain the reasons for missing measurements or whether missingness differed by treatment or response. An imputation method, missing-data sensitivity analysis and detailed participant disposition by analysis time are not reported in the registry record.

The registry does not report the trial’s sample-size justification, interim analyses or a complete statistical analysis plan. Those omissions mean the registry record does not establish the intended power for the primary dose-response analysis, the treatment of all missing values, or how multiplicity was meant to be addressed. Planning illustrations for selected recalculated secondary comparisons are not substitutes for these protocol details and should not be interpreted as observed power.

12. 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 3% reporting threshold.

Most frequent non-serious adverse event terms

The safety results list zero serious adverse events and zero deaths in each reported treatment group. For serious adverse events, the registry gives the number at risk and the number affected; the listed affected counts are zero. This is a description of the observed groups and study periods, not proof that serious events cannot occur or that treatments have identical safety profiles.

Other adverse events were reported in the placebo and active-treatment groups. Those counts are descriptive, and the group sizes are modest. The record does not provide a comparative statistical analysis of adverse events, nor do these figures establish whether any particular event was caused by treatment. The study involved escalating doses over defined treatment periods, so the results do not answer questions about longer-term exposure or rare events. Adverse-event counts should therefore be read as the safety observations recorded for these participants, not as a definitive comparison among medicines.

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

The posted primary results describe modeled AMP PC20 values after escalating doses of FF, FP and BUD. The predicted means rise across the listed doses for each medicine, with 95% confidence intervals indicating uncertainty around those estimates. This is evidence about a short-term pharmacodynamic response under the study’s challenge-test conditions; it is not by itself evidence about how participants felt, how often asthma symptoms occurred, or whether exacerbations or other clinical outcomes changed.

The results do not establish that one medicine is superior to another. The registry reports dose-response estimates but does not provide a formal direct comparison of the three curves, and the listed dose levels are different across medicines. Nor do the results establish a relative therapeutic index: the study description links airway response to systemic activity via 24-hour plasma cortisol suppression, but cortisol results are not reported in the registry record.

The analyses also do not demonstrate equivalence. An absence of a reported comparison, or overlapping confidence intervals, is not a test that treatments have the same effect. The primary estimates are model-dependent, and the record does not report diagnostics or sensitivity analyses that would show how robust they are to alternative modeling choices. Further, the pharmacodynamic analysis includes participants with the required post-dose measurement, so it does not automatically represent every enrolled participant.

The safety record lists no serious adverse events or deaths in the reported groups, but those observations are limited to the participants and treatment periods recorded. They do not rule out uncommon or delayed harms. Overall, the registry results are best understood as short-term dose-response and safety observations in a crossover study, with uncertainty and reporting gaps that constrain broader conclusions.

14. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates a dose-response analysis across three medicines within an incomplete-block crossover design. Rather than relying only on treatment-versus-placebo contrasts, the primary analysis fits a three-parameter Emax model to AMP PC20 across escalating dose levels. That structure can describe how response changes as dose increases, while also making conclusions dependent on model assumptions such as a common Emax across the medicines.

The crossover framework has a distinct analytical purpose: participants may contribute measurements under more than one treatment, which can help account for stable differences between people. At the same time, crossover studies require attention to period and sequence effects and potential carryover. The registry identifies the design and washout period, but does not report enough analytical detail to assess those issues fully. This makes the design a useful example of why knowing the study label is not a substitute for a complete statistical analysis plan.

The study also links two kinds of pharmacology: a measure of airway responsiveness and a measure of systemic activity. That pairing motivates a therapeutic-index question, but the posted primary results cover only AMP PC20. The absence of cortisol results in the registry record leaves the combined question unresolved. Statistically, it underscores that a measured endpoint can illuminate one part of a research objective without answering the entire objective.

Finally, the record distinguishes sponsor-model results from simpler recalculations of secondary baseline summaries. These approaches answer different questions and should not be treated as interchangeable. Clear interpretation depends on identifying the estimand, analysis population, time point and assumptions behind each result.

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 analysis is reported as a model-based dose-response analysis, but the registry does not report model diagnostics, goodness-of-fit information, sensitivity analyses or results under alternative assumptions. The common-Emax assumption may matter to comparisons among the medicines, and the record does not show how estimates would change without that constraint. The specific interpretation of the additional parameter estimates and confidence intervals is also not fully described.

The crossover design is incomplete-block and includes one or two periods per participant. Although a washout period is described, detailed information on sequence allocation, period effects, carryover assessment and the handling of incomplete treatment sequences is not reported in the registry record. These gaps make it difficult to independently assess whether the model fully reflects the design.

Only participants with available measurements at a specified time point were included in that time-point analysis. The registry does not report the amount or reasons for missing primary endpoint data by treatment, an imputation method or sensitivity analyses for missingness. The analyzed number varies across the listed doses, so the set of contributing participants is not constant throughout the dose-response results.

The study’s description emphasizes comparison of topical efficacy and systemic activity, but the registry results available for the primary endpoint concern AMP PC20. Plasma cortisol suppression and the resulting relative therapeutic index are not reported. The secondary results include baseline FEV1 and FVC comparisons, not treatment-response results for these measures. The registry also does not report a multiplicity plan or the original sample-size justification. These limits narrow both the conclusions available from the posted results and the ability to assess the original statistical design.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Phase 2a, AMP Challenge, Dose Escalation Study to Assess the Dose Response for Topical Efficacy and Systemic Activity in Asthmatic Subjects — NCT02991859.
• Registry record last updated 2020-07-31; status last verified 2020-07.
• Results first posted 2020-03-30.
• 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

NCT02991859 is a completed randomized, quadruple-masked Phase 2a crossover study in participants with asthma, with 56 enrolled. It examined escalating doses of fluticasone furoate, fluticasone propionate and budesonide, with placebo included in treatment sequences. The primary endpoint was AMP PC20, analyzed using a three-parameter Emax model and reported as predicted means with 95% confidence intervals at 12 hours after the last dose on Day 7.

The posted estimates rise across the listed doses for each medicine, but the registry does not report a formal direct comparison of the medicines’ dose-response curves or plasma cortisol results for the broader therapeutic-index question. The record lists zero serious adverse events and zero deaths in the reported groups. Missing-data methods, multiplicity strategy, sample-size justification and several details needed to evaluate the crossover model are not reported in the registry record.