Study of the Efficacy and Safety of MK-0476 in Japanese Pediatric Participants With Seasonal Allergic Rhinitis (MK-0476-519): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of the Efficacy and Safety of MK-0476 in Japanese Pediatric Participants With Seasonal Allergic Rhinitis (MK-0476-519) (NCT01857063), a randomized phase 3 study. The primary outcome is Change From Baseline in Total Nasal Symptom Score (TNSS) Averaged During 3 Hours of Exposure.

2. Clinical Question

The central question is whether nasal symptom scores during controlled pollen exposure differed between montelukast and placebo in Japanese pediatric participants with seasonal allergic rhinitis. The primary hypothesis in the registry is that montelukast is superior to placebo for treating nasal symptoms. The primary endpoint is change from baseline in the Total Nasal Symptom Score (TNSS), averaged during 3 hours of exposure on Day 7 of each treatment period.

Because this was a crossover trial, the treatment comparison concerns periods in which participants received different treatments, rather than two groups that each received only one treatment. The registry reports the primary result by study drug as taken. It does not report enough detail to establish how every design feature was handled in the analysis.

3. Trial Design

This completed Phase III study was randomized, triple-masked, placebo-controlled, and used a crossover model. The registry lists an enrollment of 220. Participants were assigned to one of two treatment sequences: montelukast followed by placebo, or placebo followed by montelukast. Each treatment was taken for 7 days, with a 7-day washout period between the treatment periods. The study began on 2013-06-10, and primary completion was recorded as 2013-09-01.

A crossover design lets each participant receive both treatments in sequence. In general, this can make a treatment comparison more precise by allowing within-participant information to contribute to the contrast. It also means the comparison can be affected by period effects, changes over time, and possible carryover from an earlier treatment. Whether and how these issues were assessed or modeled is not reported in the registry record.

The primary endpoint was assessed on Day 7 of each treatment period. Participants completed symptom questionnaires before entering a chamber room and during 3 hours of Japanes cedar pollen exposure. The post-baseline TNSS was averaged across assessments at 30, 60, 90, 120, 150 and 180 minutes. The registry says the analysis was done by study drug as taken.

The record lists the primary endpoint and secondary symptom outcomes, including overall and individual nasal symptom scores and measurements at specified times during exposure. Nine secondary outcomes are listed as posted. The record does not report the intended hierarchy among outcomes, a multiplicity adjustment, or a detailed account of the analysis population for each secondary outcome. The primary analysis population is described as the Full Analysis Set, with eligibility based on randomization, study-drug use, and availability of baseline and post-baseline assessments.

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.

Period 1 (7 Days)

Reasons for not completing this period:

Period 2 (7 Days)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Full Analysis Set (FAS) population consisted of all randomized participants who took at least one dose of study drug, and had a baseline assessment and at least one post-baseline assessment in a treatment 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 Change From Baseline in Total Nasal Symptom Score (TNSS) Averaged During 3 Hours of Exposure, measured over Baseline and after 3 hours of pollen exposure on Day 7 of each treatment period.

The primary endpoint is change from baseline in the Total Nasal Symptom Score (TNSS), averaged during 3 hours of pollen exposure on Day 7 of each treatment period. TNSS combines three symptoms: nasal congestion, nasal discharge, and sneezing. Each symptom score ranges from 0 to 4, making the total possible score range from 0 to 12. A higher score indicates more severe nasal symptoms.

Baseline was assessed before entering the chamber room on Day 7 of a treatment period. The post-baseline score was the average of assessments taken at 30, 60, 90, 120, 150 and 180 minutes during exposure. Averaging repeated measurements over the exposure period gives an overall symptom measure for that period, rather than relying on a single assessment time. In general, repeated measurements can describe symptom experience across a period, although the average may not show how symptoms changed at each individual time point.

Change from baseline places the post-baseline assessment in relation to the participant’s score before chamber exposure. The registry reports least-squares means for the treatment-specific results and a difference in least-squares means for the comparison. The direction used to calculate change, such as post-baseline minus baseline or the reverse, is not reported in the registry record. The posted score scale indicates that higher TNSS represents more severe symptoms, but interpreting the direction of a change estimate also requires knowing the change convention.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary result is reported as least-squares means and a difference in least-squares means. In general, a least-squares mean is a model-based estimate of an outcome mean, potentially adjusted for terms included in the statistical model. It is not necessarily the simple arithmetic average of participants’ observed scores. The registry names Longitudinal Data Analysis (LDA) as the method for the treatment comparison, but does not report the model specification, the terms included, or the assumptions used to obtain the estimates.

What does the reported treatment contrast represent?

The record reports the difference in least-squares means as -0.01, with a 95% confidence interval from -0.11 to 0.10. The listed comparison is montelukast and placebo. The confidence interval includes zero, which is the value corresponding to no difference on the contrast scale. Its width communicates uncertainty around the estimated contrast. In general, a narrower interval indicates greater precision than a wider interval, but precision should not be confused with clinical importance. No threshold for a clinically important TNSS difference is reported in the registry record.

The treatment-specific least-squares means are 1.17 for montelukast and 1.18 for placebo. Their confidence intervals are reported separately, but the uncertainty in the treatment contrast is not obtained by simply comparing whether those two intervals overlap. The contrast has its own estimated uncertainty, reported directly as the confidence interval from -0.11 to 0.10. The registry’s reported analysis is the primary basis for interpreting that comparison.

How should the p-value be read?

The reported p-value is 0.913. In general, a p-value measures how compatible the observed data are with a specified null hypothesis under the statistical model and test. A large p-value does not show that the null is true, and it is not the chance that the result arose “by chance.” It also does not express the size or practical importance of the treatment contrast. Those questions require looking at the estimated difference, its confidence interval, and any meaningful-effect threshold defined for the outcome. Such a threshold is not reported in the registry record.

Why does the crossover structure matter?

In general, crossover studies compare treatments across periods within participants, so observations from the same participant are related rather than independent. An analysis should account for that dependence and may also need to address treatment sequence and period. Possible carryover is another design consideration: an effect from treatment in an earlier period could persist into a later period. The registry identifies the study as a crossover and names LDA, but the model details, handling of within-participant correlation, period and sequence terms, and any carryover assessment are not reported in the registry record. These omissions limit independent evaluation of how the design was represented in the analysis.

Can the posted summaries be recalculated independently?

A simple unadjusted re-calculation of the two-treatment contrast from posted summary statistics is not available for this continuous endpoint. The registry record does provide treatment-specific least-squares means and confidence intervals, but it does not provide the variance information needed to reconstruct a valid contrast and its uncertainty from those summaries alone. The sponsor’s reported LDA analysis is therefore the available treatment comparison. The sample-size illustrations are separate planning calculations based on the observed effect and stated assumptions; they are not an independent verification of the trial analysis or observed power.

What details remain unknown?

The registry does not report the complete statistical analysis plan, model formula, covariance structure, handling of missing outcome assessments, or whether transformations or other model checks were used. It also does not state whether a prespecified analysis population or a sensitivity analysis differed from the population described for the primary endpoint. General statistical interpretation depends on those choices, especially in a crossover study with repeated measurements. The reported result can be described as posted, while the unavailable details should not be inferred.

10. Analyses Reported by the Sponsor

The sponsor’s reported primary comparison uses Longitudinal Data Analysis and gives a difference in least-squares means of -0.01, with a 95% confidence interval from -0.11 to 0.10 and p = 0.913. The registry also reports least-squares mean changes of 1.17 for montelukast and 1.18 for placebo. These are the sponsor’s posted results and are the available basis for reading the treatment comparison.

A separate simple, unadjusted re-calculation of the contrast is not available from the posted summary statistics for this continuous outcome. Treatment-specific estimates alone do not provide all the variance information needed to reproduce the contrast and its confidence interval. Consequently, there is no independent recalculated estimate to set against the sponsor’s analysis. The registry does not flag a discrepancy between an available re-calculation and the sponsor result.

The reported LDA result should therefore be interpreted as the sponsor’s model-based analysis, not as a result independently reproduced from the treatment summaries. The registry does not provide enough detail about the model specification to assess all analytic choices. This limits the extent to which the comparison can be independently checked from the posted record.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists the primary TNSS endpoint and nine posted secondary outcomes. These include weighted TNSS, individual symptom scores, and scores assessed at specified times during chamber exposure. The record also lists 11 outcome measures and 5 analyses. It does not report a testing hierarchy for these outcomes, whether the primary endpoint was the only confirmatory test, or whether any adjustment was applied for multiple comparisons.

Multiplicity matters when several outcomes or repeated tests are considered. In general, testing many hypotheses can increase the chance of obtaining at least one apparently notable result even if no corresponding effect exists. An adjustment or prespecified hierarchy can help control that risk, but the appropriate approach depends on the scientific questions and analysis plan. No particular method should be assumed for this trial because it is not reported in the registry record.

The primary analysis population is described as the Full Analysis Set: randomized participants who took at least one dose of study drug and had a baseline assessment and at least one post-baseline assessment in a treatment period. The record reports treatment-specific outcome counts of 213 for montelukast and 211 for placebo. It does not provide a participant flow account explaining all differences between those counts and other populations or denominators.

Missing-data handling is also not reported in the registry record. In general, analyses can be affected when outcome measurements are unavailable, and the implications depend on why data are missing and how the analysis addresses them. The registry does not describe imputation, a missing-data assumption, or sensitivity analyses for the primary endpoint. These gaps make it difficult to assess how robust the result is to missing outcome information. The available interpretation should therefore remain tied to the analysis and population as described in the record.

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 no serious adverse events in the montelukast group among 216 participants at risk and no serious adverse events in the placebo group among 218 participants at risk. No serious adverse event terms are listed. These figures describe the posted serious-event counts; they do not establish that no serious events could occur, or that the treatments have identical safety profiles.

For other adverse events, the record reports 10 among 216 participants at risk in the montelukast group and 17 among 218 at risk in the placebo group. These are counts, not percentages. The registry does not provide event-specific terms in the registry record reported for these figures, so the counts alone do not describe the type, severity, timing, or duration of those events. The death counts are not reported in the registry record.

The safety denominators differ from the primary outcome counts, and the record does not explain all population differences. In general, adverse-event counts should be interpreted in light of how events were defined, collected, and attributed, as well as the period of observation. Details about those procedures are not reported in the registry record. These figures are descriptive registry results, not a complete assessment of comparative safety.

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

The primary result is an estimated difference in least-squares mean change in TNSS of -0.01, with a 95% confidence interval from -0.11 to 0.10 and p = 0.913. The estimate is close to zero, and the interval includes zero. The posted analysis therefore does not provide clear evidence of a difference in the primary endpoint under the model used. The confidence interval should remain central to interpretation because it conveys uncertainty and includes values on either side of no difference.

This is not the same as demonstrating equivalence or proving that the treatments have identical effects. An equivalence or non-inferiority conclusion requires a prespecified margin and an analysis designed for that question. Such a margin and analysis are not reported in the registry record. A non-significant result also does not establish that no effect exists; it means the reported analysis did not distinguish the observed contrast from zero with the evidence summarized by its test.

The endpoint is a symptom score measured during controlled cedar-pollen exposure in a chamber, with the primary result averaged across assessments during 3 hours on Day 7 of each treatment period. The estimate directly concerns that outcome, time window, and analysis population. It does not automatically describe symptoms in other settings, at other times, or in different populations. The registry’s population is Japanese pediatric participants with seasonal allergic rhinitis.

The registry’s crossover design and LDA result are important context. The record does not report enough model detail to independently evaluate all assumptions or the handling of period and carryover effects. It also does not report missing-data methods or a multiplicity plan. Those omissions limit how far the result can be generalized beyond the posted primary analysis.

The safety figures are descriptive and should be kept distinct from the efficacy estimate. They report serious-event counts and other adverse-event counts in their respective at-risk groups, but do not provide a full account of event types or other safety-analysis details. Taken together, the posted information describes the primary score comparison and reported adverse-event counts; it does not settle every question about effect size, clinical importance, generalizability, or safety.

14. Why This Trial Matters Statistically

This trial illustrates why the design and analysis must be read together. In a crossover study, treatment comparisons use information from treatment periods within participants. That structure can be efficient, but it introduces dependence between measurements and makes period and possible carryover effects relevant. The registry identifies a crossover design and a longitudinal analysis method, yet does not provide the model specification needed to assess how those features were handled. This is a useful reminder that naming a statistical method is not the same as documenting its implementation.

It also demonstrates the distinction between a point estimate and its uncertainty. The reported contrast is -0.01, while its 95% confidence interval ranges from -0.11 to 0.10. The estimate alone could suggest a very small contrast; the interval adds information about the range of values compatible with the analysis. The p-value of 0.913 answers a related but different question and cannot replace the interval or a threshold for clinical importance.

The endpoint’s construction matters as well. TNSS combines congestion, discharge, and sneezing, and the primary value averages repeated symptom assessments during controlled pollen exposure. That creates a defined measurement target, but an average does not convey every feature of symptom change over time. The other posted outcomes include individual symptoms and measurements at multiple times, raising questions about how multiple analyses were prioritized. The registry does not report a multiplicity plan.

Finally, the record shows why registry results must be interpreted with attention to what is and is not documented. The treatment-specific summaries are not sufficient to reconstruct an unadjusted contrast and its uncertainty, so the sponsor’s LDA result is the available comparison. Missing-data handling and several model details are also not reported. Statistical interpretation can be precise about the posted estimate while remaining appropriately limited about the analytic choices and broader conclusions.

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 registry record does not include the full statistical model specification for the primary LDA analysis. In particular, it does not report the model terms or how within-participant correlation, treatment sequence, period, or possible carryover were addressed. These details are especially relevant to interpreting a crossover comparison. The reported treatment contrast can be quoted, but the record does not allow an independent assessment of every modeling assumption.

The record also does not describe missing-data methods, assumptions about unavailable symptom assessments, or sensitivity analyses. Although the Full Analysis Set is defined, the registry does not provide a detailed participant flow account or explain all differences among outcome and safety denominators. The population and estimates should therefore be understood as those posted, without inferring unreported reasons for the differences.

Multiplicity procedures are not reported despite the listing of secondary symptom outcomes and repeated assessment times. The record does not state whether those analyses were confirmatory, exploratory, or arranged in a prespecified hierarchy. The clinical importance threshold for a change in TNSS is also not reported, limiting interpretation of the magnitude of the estimated contrast beyond its statistical uncertainty.

The primary outcome was measured during controlled pollen exposure in a chamber, and the posted result is specific to that setting and time frame. The registry does not provide information that would establish how closely the chamber measurements represent symptoms in other settings. In addition, a simple unadjusted recalculation of the primary comparison is not possible from the posted summaries, so the sponsor’s model-based result cannot be independently reproduced from the reported treatment-specific estimates alone. These limitations do not change the numerical result, but they narrow the conclusions that can be drawn from the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study of the Efficacy and Safety of MK-0476 in Japanese Pediatric Participants With Seasonal Allergic Rhinitis (MK-0476-519) — NCT01857063.
• Registry record last updated 2024-06-17; status last verified 2022-02.
• Results first posted 2014-05-26.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry record describes a completed Phase III, randomized, triple-masked, placebo-controlled crossover study of montelukast in Japanese pediatric participants with seasonal allergic rhinitis. Participants received montelukast 5 mg chewable tablets and placebo in separate 7-day treatment periods, with a 7-day washout between periods. Enrollment was 220.

The primary endpoint was change from baseline in TNSS averaged during 3 hours of pollen exposure on Day 7 of each treatment period. The posted least-squares mean changes were 1.17 for montelukast and 1.18 for placebo. The reported difference was -0.01, with a 95% confidence interval from -0.11 to 0.10 and p = 0.913, using Longitudinal Data Analysis.

The registry also reports no serious adverse events in either at-risk group and lists other adverse events as 10 among 216 at risk for montelukast and 17 among 218 at risk for placebo. Details of the primary model, missing-data handling, and multiplicity procedures are not reported in the registry record.