This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Effect of Gefapixant (AF-219/MK-7264) on Cough Reflex Sensitivity (MK-7264-015) (NCT02397460), a randomized phase 2 study. The primary outcome is Cough Reflex Sensitivity to Capsaicin Measured by Maximal Cough Response (Emax).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02397460 |
| Phase | Phase 2 |
| Condition | Refractory Chronic Cough |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | Quadruple |
| Enrollment | 50 |
| Arms in the primary outcome | 9 |
| Lead sponsor | Afferent Pharmaceuticals, Inc., a subsidiary of Merck & Co., Inc. (Rahway, New Jersey USA) |
| Results posted | Yes |
2. Clinical Question
Population
The study concerned healthy participants and participants with refractory chronic cough. The primary endpoint population was defined as randomized participants who received at least one dose and had at least one post-dose assessment of capsaicin-evoked Emax. The listed primary results distinguish healthy males and chronic-cough males and females.
Intervention
Participants received single doses of gefapixant: 50 mg or 300 mg. The 50 mg dose was administered as one tablet and the 300 mg dose as six tablets. The study assessed cough reflex sensitivity after dosing.
Comparator
The comparator was a single dose of placebo-matching tablets. The crossover design included placebo and gefapixant treatment conditions; the registry record does not report the treatment sequence or washout details.
Primary question
At 2 hours post-dose, how did single doses of gefapixant relate to capsaicin-evoked maximal cough response, or Emax, in healthy participants and participants with chronic cough? Emax was expressed as explosive coughs in 15 seconds and estimated using a model.
The primary question is about a laboratory cough challenge, not a direct measure of long-term symptoms or everyday functioning. Capsaicin was inhaled at doubling concentrations, and cough sounds were counted after each inhalation. The registry also describes assessment of cough reflex sensitivity to ATP and lists other outcomes, including cough thresholds and urge-to-cough. These are related measurements, but they address different aspects of the response to a challenge.
The registry record lists 50 enrolled participants and describes the study as randomized, quadruple-masked, and crossover. Those design features frame interpretation, but they do not establish the size or precision of any treatment contrast. For the primary result, the record gives model-based estimates across participant and treatment groupings and a mixed-model analysis. It does not provide a separate estimated effect with a confidence interval for each dose-versus-placebo comparison. The primary question therefore needs to be read alongside the limits of the reported estimates and analysis.
3. Trial Design
This was a phase 2 randomized, quadruple-masked crossover study of gefapixant and placebo. The registry record lists an enrollment of 50 and a completed status. The recorded study start was 2015-04-29, and primary completion was 2016-04-22. The study description identifies healthy participants and participants with chronic cough; the condition listed is refractory chronic cough.
The intervention arms were single doses of gefapixant 50 mg, gefapixant 300 mg, and placebo. The study description states that the primary objective was to assess the effect of the two gefapixant doses on capsaicin cough reflex sensitivity in healthy participants and participants with chronic cough. It also describes assessment of ATP cough reflex sensitivity. The primary challenge assessment was conducted 2 hours post-dose in Periods 1 and 2.
In a crossover study, participants receive more than one treatment condition at different times. As a general statistical principle, this can allow comparisons within the same participant and reduce the influence of stable differences between participants. It also means that period, sequence, and possible carryover considerations may matter to the analysis. The registry record does not report the treatment sequences, washout interval, or how period and carryover effects were handled. It reports a mixed-model analysis for the primary outcome, but the detailed model specification is not reported in the registry record.
The primary outcome population was defined as randomized participants who received at least one dose and had at least one post-dose primary endpoint assessment of capsaicin Emax. This is a post-randomization assessment-based definition, rather than a statement that every randomized participant contributed to the primary analysis. The number excluded under that definition and the handling of incomplete assessments are not reported in the registry record.
The record says the study was completed, but completion status alone does not describe adherence, missing observations, or whether the study followed its original protocol in every respect. The sample-size justification, allocation ratio, planned analysis details beyond those stated for the endpoint, and any interim analyses are not reported in the registry record. These absences limit reconstruction of the design and its intended statistical properties.
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.
Only the first 6 of 8 groups are shown.
Treatment Period 1
| Stage | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/Healthy | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/Healthy | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/CC | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/CC | Cohort 2: PBO→Gefapixant 50 mg→PBO→Gefapixant 50 mg/Healthy | Cohort 2: Gefapixant 50 mg→PBO→Gefapixant 50 mg→PBO/Healthy |
|---|---|---|---|---|---|---|
| STARTED | 7 | 7 | 6 | 6 | 6 | 6 |
| COMPLETED | 7 | 7 | 6 | 6 | 6 | 6 |
| NOT COMPLETED | 0 | 0 | 0 | 0 | 0 | 0 |
Treatment Period 2
| Stage | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/Healthy | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/Healthy | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/CC | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/CC | Cohort 2: PBO→Gefapixant 50 mg→PBO→Gefapixant 50 mg/Healthy | Cohort 2: Gefapixant 50 mg→PBO→Gefapixant 50 mg→PBO/Healthy |
|---|---|---|---|---|---|---|
| STARTED | 7 | 7 | 6 | 6 | 6 | 6 |
| COMPLETED | 7 | 7 | 6 | 6 | 6 | 6 |
| NOT COMPLETED | 0 | 0 | 0 | 0 | 0 | 0 |
Reasons for not completing this period:
| Reason | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/Healthy | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/Healthy | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/CC | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/CC | Cohort 2: PBO→Gefapixant 50 mg→PBO→Gefapixant 50 mg/Healthy | Cohort 2: Gefapixant 50 mg→PBO→Gefapixant 50 mg→PBO/Healthy |
|---|---|---|---|---|---|---|
| Adverse Event | 0 | 0 | 0 | 0 | 0 | 0 |
Treatment Period 3
| Stage | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/Healthy | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/Healthy | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/CC | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/CC | Cohort 2: PBO→Gefapixant 50 mg→PBO→Gefapixant 50 mg/Healthy | Cohort 2: Gefapixant 50 mg→PBO→Gefapixant 50 mg→PBO/Healthy |
|---|---|---|---|---|---|---|
| STARTED | 7 | 7 | 6 | 6 | 6 | 6 |
| COMPLETED | 7 | 7 | 6 | 6 | 6 | 6 |
| NOT COMPLETED | 0 | 0 | 0 | 0 | 0 | 0 |
Treatment Period 4
| Stage | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/Healthy | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/Healthy | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/CC | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/CC | Cohort 2: PBO→Gefapixant 50 mg→PBO→Gefapixant 50 mg/Healthy | Cohort 2: Gefapixant 50 mg→PBO→Gefapixant 50 mg→PBO/Healthy |
|---|---|---|---|---|---|---|
| STARTED | 7 | 7 | 6 | 6 | 6 | 6 |
| COMPLETED | 7 | 6 | 6 | 6 | 5 | 6 |
| NOT COMPLETED | 0 | 1 | 0 | 0 | 1 | 0 |
Reasons for not completing this period:
| Reason | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/Healthy | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/Healthy | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/CC | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/CC | Cohort 2: PBO→Gefapixant 50 mg→PBO→Gefapixant 50 mg/Healthy | Cohort 2: Gefapixant 50 mg→PBO→Gefapixant 50 mg→PBO/Healthy |
|---|---|---|---|---|---|---|
| Adverse Event | 0 | 1 | 0 | 0 | 0 | 0 |
| Physician Decision | 0 | 0 | 0 | 0 | 1 | 0 |
Analysis population for the primary outcome, as defined in the registry: All randomized participants who received at least 1 dose of study medication and had at least 1 post-dose primary endpoint assessment of Emax in response to capsaicin challenge
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 9 groups are shown.
| Characteristic | Unit | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/Healthy | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/Healthy | Cohort 1: PBO→Gefapixant 300 mg→PBO→Gefapixant 300 mg/CC | Cohort 1: Gefapixant 300 mg→PBO→Gefapixant 300 mg→PBO/CC | Cohort 2: PBO→Gefapixant 50 mg→PBO→Gefapixant 50 mg/Healthy | Cohort 2: Gefapixant 50 mg→PBO→Gefapixant 50 mg→PBO/Healthy |
|---|---|---|---|---|---|---|---|
| Participants analysed | 7 | 7 | 6 | 6 | 6 | 6 | |
| Age, Continuous | Years | 34.1 (SD 11.87) | 40.9 (SD 6.20) | 61.0 (SD 9.25) | 59.5 (SD 8.38) | 35.0 (SD 8.17) | 34.7 (SD 7.42) |
| Sex: Female, Male: Female | Participants | 0 | 0 | 5 | 5 | 0 | 0 |
| Sex: Female, Male: Male | Participants | 7 | 7 | 1 | 1 | 6 | 6 |
6. Primary Endpoint
The registry lists the primary outcome as Cough Reflex Sensitivity to Capsaicin Measured by Maximal Cough Response (Emax), measured over 2 hours post-dose.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | Emax (Explosive coughs/15 sec) |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was cough reflex sensitivity to capsaicin, measured by maximal cough response (Emax). In this study, Emax represents the modeled maximum number of explosive cough sounds in the first 15 seconds after inhalation. Capsaicin challenge solutions were prepared at doubling concentrations from 0.49 μM to 1000 μM and administered by inhalation. The challenge was performed 2 hours post-dose in Periods 1 and 2.
As a general explanation, a concentration-response challenge can characterize how cough response changes as the challenge concentration increases. Emax describes the upper response estimated by that relationship; it is not simply a count from one inhalation. The registry explicitly states that the presented primary values are model-based and that nonlinear mixed-effects modeling was used to estimate Emax. It also describes population pharmacodynamic modeling in NONMEM 7.3, with data exploration, goodness-of-fit plots, statistical analyses, and simulations performed in Matlab R2015a.
This endpoint is useful for quantifying a physiological response under a controlled challenge. It is different from a measure of cough frequency over everyday waking hours, and it does not by itself describe how participants felt or functioned outside the challenge. The registry lists cough frequency and urge-to-cough among other outcomes, but those outcomes are not interchangeable with modeled maximal response. The primary endpoint therefore speaks specifically to capsaicin-evoked cough reflex sensitivity at the stated assessment time. The model-based values should be interpreted as estimates from the fitted response model, not as raw average cough counts.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry’s primary results are organized by treatment, participant group, and sex where listed. For healthy males, the model-based Emax is 4.14 with placebo and 3.66 with either gefapixant dose. For chronic-cough males, the values are 4.14 with placebo and 3.37 with either dose. For chronic-cough females, the values are 7.57 with placebo and 6.17 with either dose. These are model-based estimates; the record does not give standard errors or confidence intervals for these individual values.
Descriptively, the posted estimates are lower for each gefapixant dose than placebo within the listed strata. The same values are reported for the two doses within each listed stratum. This pattern does not establish that the doses have equal effects, nor does it quantify the uncertainty in any specific dose-versus-placebo contrast. No separate primary contrast estimate and confidence interval for either dose against placebo is reported in the registry record.
The record reports a mixed-model superiority analysis with p < 0.001 across the listed groups. That result is not accompanied by a contrast estimate or confidence interval in the registry record. It should therefore be read as the reported result of the overall model analysis, rather than as a standalone estimate of the size of a particular dose’s effect versus placebo. The particular null hypothesis, model terms, and contrast underlying that reported test are not described in enough detail to assign it a more specific interpretation.
A p-value describes how compatible the observed data are with a specified statistical model and null hypothesis. It does not measure effect size, establish clinical importance, or give the probability that a treatment effect is true. Here, the missing contrast estimates and intervals are especially relevant: the reported p-value does not supply the magnitude or precision of the dose-specific differences.
The secondary cough-frequency recalculations are separate from the primary model-based endpoint. Their directions are not uniformly backed by equally precise evidence: one unadjusted comparison has an interval that excludes no difference, while the other reported intervals include no difference. Those recalculations should be treated as exploratory in the absence of a stated multiplicity strategy. They do not replace the primary analysis or resolve the lack of a dose-specific primary estimate with uncertainty bounds.
Overall, the registry record supports a description of modeled capsaicin response values and reports an overall mixed-model test. It does not report enough detail to quantify the uncertainty of individual primary dose-versus-placebo comparisons, or to infer a broader effect beyond the measured challenge response. The outcome is a physiological measure obtained under study conditions, not a direct statement about longer-term cough symptoms.
9. Statistical Methods Explained
How should the crossover design be understood?
As a general statistical concept, crossover studies compare treatment conditions that participants receive at different times. A within-participant comparison can reduce the contribution of characteristics that remain stable within a person. This advantage depends on the design and analysis appropriately accounting for the order and timing of treatment periods. In this study, the registry states that the primary capsaicin challenge was performed in Periods 1 and 2 and reports a mixed-model analysis. The sequence of treatments, washout interval, and handling of period or carryover effects are not reported in the registry record. It is therefore not possible to reconstruct those parts of the analysis from the record.
What does model-based Emax mean?
The primary outcome was not a single raw cough count. Cough sounds were recorded after inhalation across a range of capsaicin concentrations, and a nonlinear mixed-effects model was used to estimate the maximal response. In general, nonlinear models describe relationships that do not have to follow a straight line, while mixed-effects models can represent both population-level patterns and participant-level variation. The record identifies population pharmacodynamic modeling and names software used, but does not report the model equation, parameter uncertainty, model diagnostics, or the detailed assumptions used for the final estimates. The values should consequently be understood as fitted estimates, not direct averages of observed counts.
How should the overall test be read?
The sponsor-reported primary analysis is described as a mixed-model analysis and reports superiority with p < 0.001. In general, a p-value is calculated relative to a null hypothesis under a statistical model. A small p-value does not quantify the size of an effect, its clinical importance, or its precision. Without a clearly stated contrast and a corresponding estimate and confidence interval, the reported test cannot be translated into a dose-specific effect size. The registry record does not give the exact test specification or identify which model contrast the result represents.
A confidence interval, as a general concept, gives a range of values compatible with an estimation procedure and its assumptions. Narrower intervals usually indicate greater precision than wider ones. No confidence intervals are reported for the primary model-based group estimates, so their precision cannot be assessed from the posted primary values alone. The absence of an interval does not mean that the estimate has no uncertainty; it means that the registry record does not quantify it in that form.
What do the secondary recalculations add?
For the posted daytime cough-frequency outcomes, recalculated comparisons use unadjusted differences between group means and confidence intervals based on the posted summary statistics. These are simple recalculations, not the sponsor’s primary model. The estimates are presented as mean differences, with negative values indicating a lower reported mean in the gefapixant group than in the placebo group. A confidence interval that includes no difference is compatible with a range of effects that includes no mean difference under the calculation assumptions. These recalculations do not incorporate any unreported adjustment or crossover-specific modeling.
General statistical caution is warranted for count outcomes such as cough frequency. Counts can vary more than a simple model assumes; over-dispersion is the general term for variability greater than expected under a basic count model. Methods that account for over-dispersion can affect standard errors and intervals. The recalculations use the reported means and standard deviations and do not establish whether a count model or an over-dispersion adjustment would be appropriate. The registry does not report sponsor analyses for these secondary comparisons.
How should planning illustrations be interpreted?
The sample-size calculations are planning illustrations based on observed effects, not a reconstruction of the trial’s original design. The stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The scenarios consider detecting the full observed effect and stated fractions of that effect. They are not observed power and do not show the probability that the completed study detected an effect. Because these calculations use observed results, they should not be treated as independent evidence for the findings or as the original sample-size justification. That justification is not reported in the registry record.
What statistical details remain unavailable?
The record describes the primary population, endpoint, modeling approach, and overall test, but does not report the detailed statistical analysis plan, missing-data method, imputation procedure, interim analyses, or multiplicity strategy. It also does not report whether all randomized participants contributed to every endpoint. These details matter because analysis choices can affect estimates, uncertainty, and the interpretation of multiple outcomes. In general, statistical methods should be interpreted in light of their assumptions and the data available for each analysis. Where those details are not reported in the registry record, the result should not be given a more specific interpretation than the record supports.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Mixed Models Analysis | not reported | < 0.001 |
The sponsor’s primary result is the reported mixed-model superiority analysis, with p < 0.001 across the listed primary-result groups. The registry gives that test but does not provide a dose-specific contrast estimate, confidence interval, or detailed test specification. The p-value is therefore best described as the result of the sponsor’s stated overall model analysis, not as a precise estimate of either dose’s difference from placebo.
A separate unadjusted recalculation is not available for the primary endpoint. The posted endpoint is a modeled count outcome, and the record does not provide the summary information needed for a simple two-arm recalculation with uncertainty. The group Emax values alone do not supply the variance or covariance information needed to reconstruct the crossover contrasts. Consequently, there is no recalculated primary estimate to set against the sponsor’s analysis.
The secondary cough-frequency comparisons are different: they are recalculations from posted means and standard deviations, not sponsor analyses. Their methods and limitations should not be transferred to the primary endpoint. The registry record does not report sponsor analyses for those secondary outcomes. The primary interpretation should remain anchored to the sponsor’s model-based analysis, while recognizing that its contrast-level estimates and precision are not reported in the registry record.
11. Multiplicity, Missing Data and Analysis Populations
The registry record reports a primary mixed-model superiority test and lists multiple secondary outcomes, including challenge sensitivity, cough thresholds, urge-to-cough, and daytime cough frequency. It does not report a multiplicity adjustment, testing hierarchy, or other strategy for controlling the interpretation of multiple statistical tests. This omission matters because multiple comparisons can make an isolated unadjusted result less persuasive than a result tested under a prespecified multiplicity plan. The secondary recalculations should therefore be read as exploratory unless another analysis plan is available.
The primary analysis population is described as randomized participants who received at least one dose and had at least one post-dose primary endpoint assessment. The registry does not report how many participants failed to meet that definition, how missing endpoint observations were handled, or whether incomplete challenge data were included in modeling. Imputation methods, sensitivity analyses for missing data, and reasons for missingness are not reported in the registry record. The same information is not reported for the secondary outcomes.
The record names the primary endpoint modeling approach and reports an overall mixed-model result, but does not give the full statistical analysis plan or enough detail to reconstruct the test. It also does not report interim analyses, the sample-size justification, or the treatment-sequence and period handling details relevant to a crossover study. These are distinct uncertainties: multiplicity concerns how a set of tests is interpreted, while missing-data and model details concern which observations and assumptions contribute to each estimate.
As a general principle, a prespecified plan helps distinguish primary confirmatory analyses from exploratory comparisons and clarifies how incomplete observations and multiple endpoints are handled. In the absence of those details in the registry record, interpretation should stay close to the reported endpoint definitions and results. Missing reporting does not establish that a method was or was not used; it means that the method is not reported in the registry record.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Cohort 1: Gefapixant 300 mg/Healthy | 0 | 14 | 0.0% |
| Cohort 1: Placebo/Healthy | 0 | 14 | 0.0% |
| Cohort 1: Gefapixant 300 mg/Chronic Cough | 0 | 12 | 0.0% |
| Cohort 1: Placebo/Chronic Cough | 0 | 12 | 0.0% |
| Cohort 2: Gefapixant 50 mg/Healthy | 0 | 12 | 0.0% |
| Cohort 2: Placebo/Healthy | 0 | 12 | 0.0% |
| Cohort 2: Gefapixant 50 mg/Chronic Cough | 0 | 12 | 0.0% |
| Cohort 2: Placebo/Chronic Cough | 0 | 11 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Cohort 1: Gefapixant 300 mg/Healthy | 0 of 14 |
| Cohort 1: Placebo/Healthy | 0 of 14 |
| Cohort 1: Gefapixant 300 mg/Chronic Cough | 0 of 12 |
| Cohort 1: Placebo/Chronic Cough | 0 of 12 |
| Cohort 2: Gefapixant 50 mg/Healthy | 0 of 12 |
| Cohort 2: Placebo/Healthy | 0 of 12 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Cohort 1: Gefapixant 300 mg/Healthy | 14 of 14 |
| Cohort 1: Placebo/Healthy | 5 of 14 |
| Cohort 1: Gefapixant 300 mg/Chronic Cough | 12 of 12 |
| Cohort 1: Placebo/Chronic Cough | 7 of 12 |
| Cohort 2: Gefapixant 50 mg/Healthy | 9 of 12 |
| Cohort 2: Placebo/Healthy | 4 of 12 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Cohort 1: Gefapixant 300 mg/Healthy | Cohort 1: Placebo/Healthy | Cohort 1: Gefapixant 300 mg/Chronic Cough | Cohort 1: Placebo/Chronic Cough | Cohort 2: Gefapixant 50 mg/Healthy | Cohort 2: Placebo/Healthy |
|---|---|---|---|---|---|---|---|
| Dysgeusia | Nervous system disorders | 13 of 14 | 1 of 14 | 9 of 12 | 1 of 12 | 4 of 12 | 0 of 12 |
| Headache | Nervous system disorders | 3 of 14 | 2 of 14 | 5 of 12 | 2 of 12 | 2 of 12 | 1 of 12 |
| Paraesthesia oral | Gastrointestinal disorders | 4 of 14 | 0 of 14 | 4 of 12 | 1 of 12 | 0 of 12 | 0 of 12 |
| Ageusia | Nervous system disorders | 0 of 14 | 0 of 14 | 3 of 12 | 0 of 12 | 4 of 12 | 0 of 12 |
| Hypogeusia | Nervous system disorders | 2 of 14 | 0 of 14 | 3 of 12 | 0 of 12 | 2 of 12 | 0 of 12 |
| Cough | Respiratory, thoracic and mediastinal disorders | 0 of 14 | 1 of 14 | 0 of 12 | 0 of 12 | 2 of 12 | 3 of 12 |
| Pharyngeal hypoaesthesia | Respiratory, thoracic and mediastinal disorders | 2 of 14 | 0 of 14 | 1 of 12 | 0 of 12 | 1 of 12 | 0 of 12 |
| Dry mouth | Gastrointestinal disorders | 1 of 14 | 0 of 14 | 0 of 12 | 1 of 12 | 0 of 12 | 0 of 12 |
| Dyspepsia | Gastrointestinal disorders | 1 of 14 | 0 of 14 | 1 of 12 | 0 of 12 | 0 of 12 | 0 of 12 |
| Hypoaesthesia oral | Gastrointestinal disorders | 1 of 14 | 0 of 14 | 1 of 12 | 0 of 12 | 0 of 12 | 0 of 12 |
The registry lists serious adverse event counts separately by cohort, treatment, and health-status group. It reports zero serious adverse events in every listed group and zero deaths in every listed group. These entries describe the safety data as recorded for the listed groups; they do not by themselves establish that serious harms cannot occur or provide a precise estimate of risk.
The record also gives counts of participants affected by other adverse events. For example, the listed counts are 14 of 14 for the healthy gefapixant 300 mg group in Cohort 1 and 5 of 14 for the corresponding placebo group. In the chronic-cough groups in that cohort, the counts are 12 of 12 with gefapixant 300 mg and 7 of 12 with placebo. Cohort 2 also reports other adverse events separately for healthy and chronic-cough groups receiving gefapixant 50 mg or placebo. These are descriptive counts and should be interpreted within the listed cohort and treatment group.
Serious adverse events and other adverse events are different categories; a count of other events is not a count of serious events. The registry does not provide event-term detail in the serious-event listing. Because the trial used a crossover design and the safety entries are grouped by cohort and treatment, the group counts should not be added to infer a total number of distinct participants. Safety reporting in the registry is not a substitute for a full assessment of event timing, severity, relatedness, or duration.
13. What the Result Does, and Does Not, Mean
The primary result describes modeled capsaicin-evoked maximal cough response at 2 hours post-dose. Across the listed strata, the model-based values are lower with gefapixant than with placebo, and the registry reports an overall mixed-model superiority result. These findings describe the posted analysis of a controlled cough challenge. They do not, on their own, quantify a specific dose-versus-placebo effect with a confidence interval because those contrast-level estimates are not reported.
As a general distinction, a physiological challenge endpoint measures a response under defined test conditions. It is not automatically equivalent to a change in daily symptoms, quality of life, disease course, or other outcomes that matter to participants. The record separately lists daytime cough frequency and urge-to-cough outcomes, but those endpoints have their own definitions and limitations. The primary Emax result should not be expanded to outcomes that were not measured or reported.
The reported p-value does not say how large or important an effect is. The model-based values give a descriptive scale for the endpoint, but the registry does not report uncertainty intervals for those values or detailed dose-specific primary contrasts. The secondary frequency recalculations add some information about other measured outcomes, but they are unadjusted and do not replace a complete prespecified analysis.
The record also does not report enough information to evaluate all design and analysis details, including treatment sequence, washout, missing-data handling, and multiplicity. Those gaps constrain how confidently the findings can be generalized beyond the posted analyses. The results are best understood as evidence about the measurements and participants represented in this completed phase 2 crossover study, not as a universal estimate for every population or clinical setting.
No medical decision follows from the statistical summaries alone. The results describe measurements, model outputs, and uncertainty as reported; they do not establish a treatment recommendation or a broader clinical conclusion.
14. Why This Trial Matters Statistically
This trial illustrates how endpoint choice shapes statistical interpretation. The primary endpoint was a modeled maximal response to a capsaicin challenge, whereas secondary outcomes included cough frequency and other challenge-related measures. These outcomes are connected by the broad topic of cough, but they are not interchangeable. A result on one measurement does not automatically establish a result on another.
It also demonstrates the difference between an overall model test and an interpretable treatment contrast. The registry reports a mixed-model p-value for the primary analysis, yet it does not provide dose-specific contrast estimates and confidence intervals. Without those quantities, readers can identify the reported statistical test and describe the group estimates, but cannot assess the precision or size of each primary comparison in the usual way.
The crossover design is another important statistical feature. In general, crossover studies can make within-participant comparisons useful, but interpretation depends on treatment order, periods, and the handling of carryover. The registry confirms the crossover structure and periods for the primary challenge assessment, while leaving important implementation and analysis details unreported. This shows why design labels alone are not enough to reconstruct an analysis.
Finally, the secondary results illustrate why multiplicity and uncertainty matter. Several distinct outcomes were recorded, while the registry does not describe a multiplicity strategy. An unadjusted result may be informative as a descriptive finding, but it deserves less weight than a prespecified analysis that accounts for the full testing context. Careful separation of endpoint, estimand, model, and uncertainty helps prevent a narrow statistical finding from being interpreted more broadly than the data warrant.
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 values are explicitly model-based, yet the registry does not report confidence intervals or standard errors for the individual group estimates. It also does not report dose-specific primary contrast estimates. This restricts assessment of the precision and magnitude of the primary treatment comparisons. The overall mixed-model p-value is not a substitute for those missing quantities.
The displayed primary groupings include healthy males and chronic-cough males and females. Although the endpoint description says the assessment was conducted in male and female healthy participants and participants with chronic cough, a separate healthy-female primary estimate is not listed. The registry record does not explain this difference in how the population is described and how the results are grouped.
The listed primary group counts are 26.0 for healthy males, 5.0 for chronic-cough males, and 18.0 for chronic-cough females in each treatment grouping. The chronic-cough male stratum is particularly small in the posted results, so its model-based value should not be assumed to have the same precision as a value from a larger group. No uncertainty intervals are posted for these stratum estimates. The record also does not explain how repeated observations or incomplete challenges contributed to the fitted model.
Important crossover details are absent, including treatment sequence, washout interval, and the handling of period or carryover effects. Missing-data methods, the full statistical plan, and multiplicity handling are also not reported in the registry record. These omissions prevent a complete appraisal of model assumptions and analysis decisions.
The secondary comparisons are simple unadjusted recalculations from summary statistics, not the sponsor’s stated primary model. They may not account for within-participant dependence or other features of the crossover design. Their planning illustrations are based on observed effects and stated assumptions, rather than the trial’s original design justification. These limitations mean the secondary estimates and planning scenarios should be treated as descriptive tools, not as a substitute for the full trial analysis.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Effect of Gefapixant (AF-219/MK-7264) on Cough Reflex Sensitivity (MK-7264-015) — NCT02397460.
- Registry record last updated 2021-02-12; status last verified 2021-01.
- Results first posted 2021-02-12.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
The registry record describes a completed phase 2 randomized, quadruple-masked crossover study of single-dose gefapixant 50 mg and 300 mg versus placebo in healthy participants and participants with chronic cough. Enrollment was 50. The primary endpoint was model-based capsaicin cough reflex sensitivity, measured as maximal cough response 2 hours post-dose.
The listed primary values are lower with either gefapixant dose than with placebo within the displayed participant strata. The registry reports an overall mixed-model superiority result of p < 0.001, but does not give dose-specific primary contrast estimates or confidence intervals. Secondary outcomes cover cough frequency and other cough challenge responses; the posted unadjusted frequency recalculations have differing degrees of uncertainty.
The record reports zero serious adverse events and zero deaths in each listed safety group. Treatment sequence, washout, missing-data methods, multiplicity strategy, and the full statistical analysis plan are not reported in the registry record. These results therefore describe the posted model-based and descriptive findings, with important limits on precision and reconstruction of the analysis.