A Study of the Efficacy and Safety of Desloratadine (MK-4117) in Japanese Participants With Perennial Allergic Rhinitis (MK-4117-200): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of the Efficacy and Safety of Desloratadine (MK-4117) in Japanese Participants With Perennial Allergic Rhinitis (MK-4117-200) (NCT01918033), a randomized phase 3 study. The primary outcome is Change From Baseline in Total Nasal Symptom Score (TNSS) Assessed by the Investigator at Week 2.

2. Clinical Question

This is a superiority question about the difference between each active-treatment group and placebo, not simply whether participants’ scores changed during the trial. A decrease in TNSS represents fewer or less severe nasal symptoms because higher scores indicate more frequent or severe symptoms. The comparison therefore focuses on whether the decrease differed between a desloratadine group and the placebo group.

The trial also recorded outcomes concerning symptom sub-scores, nasal findings, eye symptoms, global improvement, interference with daily activities, and participant diary assessments. Those outcomes provide other perspectives on symptoms and their effect on daily life, but they are secondary to the specified Week 2 investigator-assessed TNSS endpoint. The registry record does not report a threshold for deciding whether a difference in TNSS would be clinically important. The statistical comparison can describe the estimated difference and its precision, but the registry record does not establish how large a difference participants would consider meaningful.

3. Trial Design

The registry describes a Phase III, randomized, parallel-group, placebo-controlled, quadruple-masked trial. It records enrollment of 608 and a completed study. Participants were assigned to desloratadine 5 mg, desloratadine 10 mg, or placebo. The posted primary outcome uses the Full Analysis Set, defined as participants who received at least one dose and had a baseline or post-baseline TNSS assessment. The group counts reported for that outcome are 202 for desloratadine 5 mg, 203 for desloratadine 10 mg, and 201 for placebo.

The parallel-group structure means that each participant was assigned to a treatment group rather than crossing between treatments for the primary comparison. Randomization is intended, in general, to support fair comparisons by distributing participant characteristics across groups without relying on deliberate selection into treatment. Masking is intended, in general, to limit the influence of knowledge about treatment assignment on participants’ experiences, clinical assessments, and study conduct. The registry record labels the trial quadruple-masked but does not specify in the posted information which roles were masked or how masking was maintained.

The treatment descriptions specify oral tablets taken once daily for up to 2 weeks. Participants in the 5 mg group received one desloratadine 5 mg tablet and one placebo tablet; those in the 10 mg group received two desloratadine 5 mg tablets; and those in the placebo group received two placebo tablets. The registry record gives the primary endpoint timeframe as baseline and Week 2, with the primary hypothesis focused on treatment after 2 weeks.

The registry record reports an enrollment figure, but does not report the trial’s original sample-size justification, design effect assumption, or other planning assumptions. It also does not report detailed randomization procedures or the specific masking arrangements. The planning illustrations associated with the recalculations use stated assumptions and should not be mistaken for the trial’s original design rationale.

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:

Analysis population for the primary outcome, as defined in the registry: The Full Analysis Set (FAS) population consisted of all participants who took at least one dose of study drug and had a baseline or at least one post-baseline assessment for TNSS.

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) Assessed by the Investigator at Week 2, measured over Baseline and Week 2.

The primary endpoint is change from baseline in investigator-assessed Total Nasal Symptom Score at Week 2. An investigator interviewed and examined participants and rated sneezing, rhinorrhea, nasal congestion, and nasal itching. Each symptom contributes a sub-score, and the four sub-scores are summed to form TNSS. Scores range from 0 to 12, with higher scores indicating more frequent or severe nasal symptoms.

Change from baseline expresses the difference between a participant’s starting assessment and the assessment at the specified follow-up. For this endpoint, a negative change indicates a lower score at Week 2 than at baseline. The reported group results are least-squares means, a model-based way of expressing the average outcome under the analysis used; the registry record does not provide all model details needed to describe exactly how those means were estimated.

The endpoint is used because it combines four nasal symptoms into one measure, allowing the primary comparison to address overall nasal symptom burden rather than one symptom alone. That summary is also a trade-off: a combined score can conceal different patterns among its component symptoms. The registry lists secondary outcomes for symptom sub-scores and other assessments, but the primary conclusion should remain tied to the prespecified Week 2 TNSS comparison. The registry record does not report a minimum clinically important difference for TNSS, so the statistical estimates alone do not define what magnitude of change would be important to participants.

7. Results Posted in the Registry

• Desloratadine 5 mg: mean -1.96 versus -1.87 on control; difference -0.09 (95% CI -0.50 to 0.32), standardized effect d = -0.04; z-test p = 0.667.
• Desloratadine 10 mg: mean -1.94 versus -1.87 on control; difference -0.07 (95% CI -0.48 to 0.34), standardized effect d = -0.03; z-test p = 0.738.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary result is a continuous-outcome comparison. TNSS is a score formed by adding four symptom sub-scores, so the reported group summaries describe average changes on that scale. In general, a mean is useful for describing the center of a continuous measure, while also potentially concealing variation between participants. The posted group results include a least-squares mean and a 95% confidence interval for each group.

What does a least-squares mean represent?

A least-squares mean is an estimate produced by a statistical model, rather than simply the unadjusted arithmetic average of all observed changes. It is often used to express the mean outcome for a group while accounting for variables or repeated measurements included in the model. The registry names Constrained Longitudinal Data Analysis for the treatment comparisons and reports differences in least-squares means. It does not report enough model specifications to establish which covariates, covariance structures, or other details were used. Those features should not be inferred from the method name alone.

The posted arm summaries are -1.96 for desloratadine 5 mg, -1.94 for desloratadine 10 mg, and -1.87 for placebo. Their intervals describe uncertainty around the estimated group means. A treatment contrast instead subtracts the placebo estimate from the active-treatment estimate within the analysis. The registry reports those contrasts directly: -0.09 for 5 mg versus placebo and -0.08 for 10 mg versus placebo. Interpreting a contrast is more relevant to the superiority question than comparing whether either group changed from baseline on its own.

How should the confidence interval be read?

In general, a confidence interval gives a range of estimates compatible with the data and the method used, under the assumptions of that method. It communicates precision as well as direction. A narrower interval indicates more precision than a wider interval on the same scale. The reported 95% intervals for both treatment contrasts include zero, so they do not distinguish the observed contrast from no difference at that confidence level. The interval is not a guarantee that the true value lies within its endpoints, and it does not assign a probability to any one value after the data have been observed.

The interval is also not a range of individual participant responses. It concerns the estimated contrast in group means. Participants’ individual symptom changes can vary more widely than the uncertainty around the group estimate. The posted spread values for the arms describe variation in the outcome, while the confidence intervals around means describe uncertainty in estimated means; these quantities answer different questions.

What does the unadjusted recalculation add?

The recalculation compares the posted group means using an unadjusted mean difference and an approximate pooled standard deviation. It gives a quick, transparent check based on summary statistics, rather than reproducing the trial’s full longitudinal model. The recalculated estimates are -0.09 for the 5 mg comparison and -0.07 for the 10 mg comparison; the unadjusted p-values are 0.667 and 0.738. Because the posted summary statistics do not capture every feature of the trial analysis, these calculations are best treated as supplementary. The registry’s reported Constrained Longitudinal Data Analysis results remain the results from the trial’s own analysis.

For general context, unadjusted comparisons do not account for multiple testing or for adjustment variables unless those are built into the calculation. The registry record does not state whether the reported p-values were adjusted for multiplicity, and the registry’s model details are limited. It is therefore not possible to describe the recalculation as a replication of every element of the trial’s analysis.

Why can over-dispersion matter in general?

For count outcomes, over-dispersion means that observed variation is greater than a simple model expects; ignoring it can make standard errors too small and confidence intervals too narrow. The primary endpoint is a continuous summed score, not a count of events. That general issue does not establish a problem with this trial’s TNSS analysis. The registry record does not report a specific variance assumption or diagnostic assessment for the primary model, so no claim about model fit should be made.

The primary population is described as the Full Analysis Set, requiring at least one dose and a baseline or post-baseline TNSS assessment. This definition clarifies who could contribute to the primary outcome, but it does not by itself explain how incomplete follow-up was handled. The registry record does not report the missing-data method, the amount of missing TNSS data, or sensitivity analyses for missing outcomes. Those details affect how robustly the model represents participants with incomplete assessments and remain important uncertainties when interpreting the result.

10. Confidence Intervals and Precision

The sponsor-reported 95% confidence interval for desloratadine 5 mg versus placebo runs from -0.49 to 0.31 around an estimated difference of -0.09. For desloratadine 10 mg versus placebo, the interval runs from -0.48 to 0.32 around an estimated difference of -0.08. Both intervals include zero. The estimates are slightly negative, but the intervals include values on both sides of zero, so the direction of the underlying difference is not determined precisely by these results.

In general, confidence intervals communicate the range of values that remain compatible with the observed data and the analysis assumptions. Their width reflects uncertainty about the group contrast, not the range of scores experienced by individual participants. The posted arm-level confidence intervals likewise concern estimated group means, whereas the reported spread values describe participant-level variability. These are related but distinct descriptions of the data.

The intervals do not establish that the doses and placebo are equivalent. Equivalence requires an explicitly defined margin and an analysis designed to assess whether the difference is contained within that margin. The registry record does not report an equivalence margin. Nor does it provide a clinical-importance threshold for TNSS, so the endpoints of the intervals cannot be classified as clinically unimportant on the basis of the registry information alone.

The unadjusted recalculated intervals are based on approximate summary-statistic methods and are not identical to the intervals from the trial’s longitudinal analysis. The sponsor-reported intervals are the more relevant precision estimates for the trial’s own analysis. Overall, the intervals show that the estimated differences are small in magnitude but not precise enough to rule out differences in either direction at the stated confidence level.

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 planning illustrations give participants per arm for 80% power and participants per arm for 90% power. They are not total event counts: the outcome is continuous, and the entries refer to participants in each of the two groups in a comparison. The illustrations use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set as the stated fraction of the observed effect. These are planning assumptions for the illustrations; the registry record does not report the trial’s original sample-size assumptions.

The 80% power column describes the planned number of participants per arm under which the calculation would have an 80% chance of detecting the specified effect, if the assumptions hold and the effect is real. The 90% power column uses a higher probability of detecting that same specified effect, so it requires more participants per arm. Power is a planning property of a proposed design, not a measure of how likely the observed result is to be true.

For each comparison, the entries also show planning for 75% of the observed effect and 50% of the observed effect. Smaller effects are harder to distinguish from random variation, and therefore require more participants per arm under the stated assumptions. The displayed values grow as the assumed fraction of the observed effect becomes smaller. This pattern is a general feature of power planning, not evidence that those effects were observed in a new study.

The figures are planning illustrations computed from the observed effect, not the trial’s original design calculations. They are not observed power. The original design justification, including the effect size and assumptions used when planning the trial, is not reported in the registry record. These illustrations should therefore be read as hypothetical planning scenarios, not as an account of how the trial’s enrollment was chosen.

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 = -0.090; pooled standard deviation σ = 2.100.
2. n = 2 × (1.96 + 0.842)² × 2.100² / -0.090² = 8,549.3, rounded up to 8,550 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 <- -0.0899999999999999                      # observed difference in means
sigma <- 2.10033599772003                     # 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(1918033)
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 = 8,550 (80% power), 11,446 (90% power)
 75% of observed effect: n per arm = 15,199 (80% power), 20,347 (90% power)
 50% of observed effect: n per arm = 34,198 (80% power), 45,781 (90% power)
power.t.test: 8550.3 
Simulated power at n = 8,550 per arm: 0.802 (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. Analyses Reported by the Sponsor

The registry reports the trial’s own primary comparisons using Constrained Longitudinal Data Analysis. For desloratadine 5 mg versus placebo, the reported difference in least-squares means is -0.09, with a 95% confidence interval from -0.49 to 0.31 and p = 0.661. For desloratadine 10 mg versus placebo, the reported difference is -0.08, with a 95% confidence interval from -0.48 to 0.32 and p = 0.707.

The recalculations use posted group summaries to make unadjusted comparisons. They give estimates of -0.09 for 5 mg versus placebo and -0.07 for 10 mg versus placebo, with unadjusted p-values of 0.667 and 0.738. The close estimates are a useful check on the general scale and direction of the comparisons, but the recalculations use an approximate pooled standard deviation and do not reproduce the longitudinal analysis.

The reported results are not flagged as discrepant. The sponsor’s analysis should be interpreted as the trial’s own analysis; the recalculated estimates are supplementary and should not replace it. Small differences between results can arise because the methods differ. The registry record does not provide enough detail to reproduce every component of the sponsor’s model, and the recalculation should not be presented as a corrected version of that analysis.

14. Multiplicity, Missing Data and Analysis Populations

The registry record lists 11 outcome measures and 36 analyses. It identifies the Week 2 change in investigator-assessed TNSS as the primary outcome and lists secondary outcomes involving other symptom scores, nasal findings, global improvement, daily activity interference, and participant diaries. Eight secondary outcomes are posted. These counts make clear that the trial considered more than the primary comparison, but they do not establish how the full set of analyses was organized or interpreted.

Multiplicity arises when several hypotheses are tested: even when no true difference exists, a larger family of tests can increase the chance that at least one result appears statistically notable by chance. In general, investigators may address this with a prespecified testing sequence or an adjustment to significance levels or p-values. The registry record does not report whether a multiplicity procedure was used, which outcomes were grouped together for that purpose, or whether secondary outcomes were intended to support confirmatory conclusions. The primary comparisons should therefore be distinguished from findings among secondary analyses.

The registry record reports a Full Analysis Set definition: participants had to take at least one dose and have a baseline or at least one post-baseline TNSS assessment. This is useful information about eligibility for the primary analysis, but it does not fully explain the analysis population or the treatment of participants with missing follow-up. The registry record does not report the number of missing TNSS assessments, reasons for missingness, the method used to handle missing data, or sensitivity analyses assessing the effect of alternative missing-data assumptions.

These omissions limit assessment of robustness. General statistical practice recognizes that results can depend on whether missing observations are related to participants’ symptoms or treatment experiences. The missingness mechanism and the method used to address it are not reported in the registry record, so no particular assumption should be attributed to the trial. The record also does not provide enough detail to determine whether the analysis followed a complete prespecified statistical analysis plan. Such a plan, along with multiplicity and missing-data procedures, would help clarify how the reported results should be read alongside the other outcomes.

15. 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: epilepsy (0.5%).

The posted serious adverse event figures are 1 participant in the desloratadine 5 mg group, 0 in the desloratadine 10 mg group, and 0 in the placebo group. The group counts at risk are 202, 203, and 201, respectively. The registry also identifies epilepsy at 0.5% among the top serious adverse event terms. The other adverse event counts reported are 11 for desloratadine 5 mg, 17 for desloratadine 10 mg, and 9 for placebo.

These are descriptive counts from the registry record. They do not by themselves establish that a treatment caused or prevented an event, and a small number of events offers limited information about uncommon harms. Serious adverse events and other adverse events are separate summaries and should not be combined. The registry record does not report deaths, so death information is not reported in the registry record. Further detail on event definitions, timing, severity, and the method for attributing events to study treatment is not reported in the registry record.

Safety counts should be read with the study’s duration and participant groups in mind: treatment was described as once daily for up to 2 weeks. The counts do not describe longer-term safety, and the registry record does not provide enough detail to assess risks beyond the events listed. These figures are not a substitute for the full safety dataset.

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

The primary results show small estimated differences between each desloratadine dose and placebo in Week 2 change from baseline in investigator-assessed TNSS. The sponsor-reported estimates are -0.09 for desloratadine 5 mg and -0.08 for desloratadine 10 mg. The 95% confidence intervals include zero, and the reported p-values are 0.661 and 0.707. Taken together, the posted primary analysis does not provide clear statistical evidence of a difference from placebo on this endpoint.

This interpretation is not the same as concluding that the treatments and placebo are identical. The intervals include a range of possible differences in either direction, and the registry record does not give an equivalence margin. Nor does it give a threshold for clinical importance. The statistical results therefore cannot establish that any possible difference is too small to matter, or that a particular difference would matter to participants.

The group mean changes were negative in all three groups: -1.96 for 5 mg, -1.94 for 10 mg, and -1.87 for placebo. A decrease indicates a lower TNSS at follow-up than baseline. Within-group improvement is not itself evidence that the study treatment produced greater improvement than placebo; the randomized between-group contrast is the relevant primary test. The registry’s comparisons estimate only small differences between groups, with uncertainty that includes no difference.

The endpoint summarizes four nasal symptoms and does not cover every dimension of allergic rhinitis. Secondary outcomes include other assessments, but their results and multiplicity handling are not detailed in the primary comparison. The trial’s findings should be understood as applying to the specified population, treatment duration, outcome measure, and analysis—not to different populations, longer use, or outcomes not reported. These data describe statistical evidence; they do not constitute medical advice or a recommendation about treatment.

17. Why This Trial Matters Statistically

This trial illustrates why a change within a treatment group is not the same as a treatment effect. All three groups had negative average changes in TNSS, but the primary question is whether the active-treatment groups changed more than placebo. The sponsor-reported contrasts are close to zero relative to the uncertainty in their estimates. Reading the between-group estimates and their confidence intervals helps avoid attributing all observed improvement in an active-treatment group to the active treatment itself.

It also shows why statistical significance and clinical importance are separate questions. The registry provides confidence intervals and p-values, but does not report a clinically important difference threshold. A result may be statistically distinguishable without being important to participants, or an uncertain result may still leave open potentially meaningful differences. Here, the intervals include zero, and the record does not provide a threshold for judging clinical relevance.

The registry’s own longitudinal analysis and an unadjusted summary-statistic recalculation are close in broad direction, yet they are not identical methods. This offers a practical reminder that a recalculation can act as a check on scale and direction without replacing the analysis used in the trial. Model details, missing-data handling, and multiplicity procedures remain relevant when evaluating how much confidence to place in any posted result.

Finally, the planning illustrations demonstrate how strongly participant requirements depend on the assumed effect size. Smaller assumed effects require larger per-arm samples under the stated assumptions. Those illustrations do not reveal the original design rationale or provide observed power. Keeping planning calculations distinct from the completed trial’s results is important for an accurate statistical reading.

18. 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 report the original sample-size justification or the assumptions used to choose enrollment. The planning illustrations are based on the observed effects and should not be treated as the trial’s design rationale. The registry also does not report a clinical-importance threshold for TNSS, limiting interpretation of whether the estimated differences are meaningful to participants.

Important analysis details are not reported in the registry record, including the full model specification, missing-data handling, the extent and pattern of missing outcome assessments, and sensitivity analyses. Although the primary population is defined, that definition alone does not establish how incomplete assessments affected the estimates. The registry record also does not state whether or how multiplicity was addressed across the listed outcomes and analyses.

The primary endpoint is an investigator-assessed sum of four symptom scores. It offers a compact measure of nasal symptom burden, but the summary can obscure different changes in its component symptoms. The record lists sub-score and other secondary outcomes, but their findings are not included in the primary result. Details about assessment procedures beyond the endpoint description are not reported in the registry record.

The treatment period is up to 2 weeks, so the posted outcome does not establish what happens with longer treatment. The enrolled population is Japanese participants with perennial allergic rhinitis; generalization beyond that population is not established by this registry record. The safety summaries are counts, and the record does not provide enough information about event timing, attribution, or longer-term follow-up to characterize uncommon or delayed events. These limitations concern what can be concluded from the posted record, not necessarily what may have been assessed elsewhere.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Study of the Efficacy and Safety of Desloratadine (MK-4117) in Japanese Participants With Perennial Allergic Rhinitis (MK-4117-200) — NCT01918033.
• Registry record last updated 2024-06-18; status last verified 2022-02.
• Results first posted 2014-08-15.
• 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.

22. Record Summary

The registry describes a completed Phase III, randomized, parallel-group, placebo-controlled, quadruple-masked trial of desloratadine in Japanese participants with perennial allergic rhinitis. The primary outcome is change from baseline in investigator-assessed TNSS at Week 2. The Full Analysis Set includes participants who took at least one dose and had a baseline or post-baseline TNSS assessment.

The posted least-squares mean changes were -1.96 for desloratadine 5 mg, -1.94 for desloratadine 10 mg, and -1.87 for placebo. The registry’s treatment contrasts were -0.09 and -0.08, with 95% confidence intervals spanning zero. The primary results do not provide clear statistical evidence of a difference from placebo, while not establishing equivalence.

The registry record does not report the original sample-size rationale, detailed missing-data or multiplicity procedures, or a clinical-importance threshold for TNSS. Safety counts are descriptive and should be read in light of the short treatment period and the limits of the posted information.