Phase III Cat-PAD Follow-on Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase III Cat-PAD Follow-on Study (NCT02040844), a randomized phase 3 study. The primary outcome is Mean Combined Score (CS) Consisting of TRSS/8+Allergy Medication Score[AMS]).

2. Clinical Question

The central question is whether the mean combined symptom-and-medication score one year after completion of the original CP007 study differed among participants previously assigned to Cat-PAD treatment and those previously assigned to placebo. This was a follow-on study: participants received no further investigational product in CP007A, so the comparison concerns outcomes after the earlier study treatment rather than a new course of treatment during follow-up.

The primary outcome reflects both symptom severity and allergy rescue medication use. The registry describes symptom ratings from absent to severe and a medication score from no rescue medication use to use of at least one daily dose of systemic corticosteroid. Because the outcome combines these components, a group mean is not a symptom-only measure.

The comparison is descriptive of the recorded follow-up results. It does not by itself establish whether prior treatment caused any later difference. Interpretation also depends on who completed the assessment, how the original study groups were formed, and the analysis methods used. Further details about those methods are not reported in the registry record.

3. Trial Design

The registry classifies NCT02040844 as a phase III, randomized, quadruple-masked, parallel follow-on study for cat allergy. It is described as an optional prospective follow-up to the phase III, double-blind, placebo-controlled CP007 study. Its stated purpose was to evaluate continued efficacy and safety for up to five years after the start of treatment, without administering further investigational product in CP007A.

The recorded enrollment is 430. The primary outcome population, however, is defined as subjects who completed the End of Year 1 assessments. The posted primary outcome groups contain 88 participants in the one-course Cat-PAD group, 109 in the two-course group, and 97 in the placebo group. The enrollment figure and these outcome-specific group counts describe different aspects of the record and should not be treated as interchangeable.

The primary assessment was one year after completing CP007. The outcome description also characterizes this as two years after treatment began. The posted secondary outcomes use the same general one-year-after-completion timeframe. The registry lists additional symptom, symptom-component, medication, and quality-of-life outcomes, although a posted inferential analysis is not included for them.

The follow-on arms are defined by the treatment received in CP007: Cat-PAD Treatment 1, Cat-PAD Treatment 2, or placebo. The follow-on record says no further treatment was received. Thus, the reported follow-up comparison is between histories of earlier study treatment, not between active and placebo treatments newly administered in this study.

The original study's detailed allocation procedures, the follow-on study's own masking procedures in practice, and the protocol's sample-size justification are not reported in the registry record. The registry classification of randomized and quadruple-masked should therefore be read as the recorded design description, not as a substitute for those missing operational details. The record lists the study as completed, but it does not provide a detailed account of follow-up completeness or reasons for non-completion.

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 number of subjects included in the analysis is the total number of subjects that completed the End of Year 1 assessments (1 year after the completion of CP007 and 2 years after the start of treatment).

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 Mean Combined Score (CS) Consisting of TRSS/8+Allergy Medication Score[AMS]), measured over 1 year after completing CP007.

The primary endpoint is the mean Combined Score (CS), assessed one year after completion of CP007. It combines the Total Rhinoconjunctivitis Symptom Score (TRSS) and the Allergy Medication Score (AMS). The endpoint description specifies CS as TRSS divided by eight, plus AMS. Dividing TRSS by eight expresses the symptom component as an average across the eight symptoms rather than as their total.

The eight symptoms comprise four nasal symptoms and four ocular symptoms. Each is rated on a scale from 0, absent, to 3, severe. The average symptom component therefore ranges from 0 to 3. AMS also ranges from 0 to 3, with higher scores reflecting greater use of allergy rescue medication under the stated scoring rules. The overall combined score has a stated maximum of 6. AMS is not additive across medication use, so the medication component is capped at 3.

Combining symptoms with medication use can represent two aspects of allergy burden in one outcome. It also means the combined mean cannot show, on its own, whether any group difference arose from symptoms, medication use, or both. The separate secondary measures offer related symptom and medication summaries, but the registry does not post inferential analyses for them.

The registry reports least-squares means with standard errors as the primary endpoint summaries. Details of the model used to produce those least-squares means, including covariates and handling of missing observations, are not reported in the registry record. The measure is therefore best read as a group average on the defined combined scale at the specified follow-up, with the limitations of the posted analysis detail in mind.

7. Results Posted in the Registry

• Cat-PAD Treatment 1 (1 Course): mean 1.99 versus 1.93 on control; difference 0.06 (95% CI -0.37 to 0.49), standardized effect d = 0.04; z-test p = 0.784.
• Cat-PAD Treatment (2 Courses): mean 1.91 versus 1.93 on control; difference -0.02 (95% CI -0.44 to 0.40), standardized effect d = -0.01; z-test p = 0.925.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Mean TRSS

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Cat-PAD Treatment 1 (1 Course): mean 12.95 versus 12.92 on control; difference 0.03 (95% CI -1.86 to 1.92), standardized effect d = 0.00; z-test p = 0.975.
  • Cat-PAD Treatment (2 Courses): mean 13.15 versus 12.92 on control; difference 0.23 (95% CI -1.61 to 2.07), standardized effect d = 0.03; z-test p = 0.807.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cat-PAD Treatment 1 (1 Course) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cat-PAD Treatment (2 Courses) and the control. The interval gives the range of effects that remain compatible with the data.

Mean Component Scores of the TRSS (Nasal)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Cat-PAD Treatment 1 (1 Course): mean 7.06 versus 6.96 on control; difference 0.10 (95% CI -0.97 to 1.17), standardized effect d = 0.03; z-test p = 0.854.
  • Cat-PAD Treatment (2 Courses): mean 7.17 versus 6.96 on control; difference 0.21 (95% CI -0.83 to 1.25), standardized effect d = 0.06; z-test p = 0.692.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cat-PAD Treatment 1 (1 Course) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cat-PAD Treatment (2 Courses) and the control. The interval gives the range of effects that remain compatible with the data.

Mean Component Scores of the TRSS (Ocular)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Cat-PAD Treatment 1 (1 Course): mean 5.90 versus 6.00 on control; difference -0.10 (95% CI -1.06 to 0.86), standardized effect d = -0.03; z-test p = 0.838.
  • Cat-PAD Treatment (2 Courses): mean 5.97 versus 6.00 on control; difference -0.03 (95% CI -0.96 to 0.90), standardized effect d = -0.01; z-test p = 0.950.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cat-PAD Treatment 1 (1 Course) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cat-PAD Treatment (2 Courses) and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcomes cover total rhinoconjunctivitis symptoms, nasal and ocular symptom components, allergy medication use, and quality of life. Their group means generally show small separations between the prior Cat-PAD groups and placebo in the posted summaries. The registry does not provide inferential analyses for these secondary outcomes, so the reported means and standard errors offer descriptive context rather than a complete set of tested treatment comparisons.

This broad pattern is consistent with the primary outcome in the limited sense that the posted group averages do not show a marked separation at the follow-up assessment. The secondary measures cannot independently confirm or overturn the primary comparison. They assess related but distinct aspects of symptoms, medication, and quality of life, and their measurement scales are not interchangeable with the combined score.

Because the combined endpoint incorporates symptom and medication components, the secondary outcomes might help distinguish which elements contribute to the overall pattern if fully analyzed. However, the registry's lack of reported inferential analyses prevents conclusions about whether any component differed between groups beyond what the descriptive summaries suggest. The record also does not report a prespecified hierarchy among secondary outcomes.

Unadjusted secondary findings carry less weight when multiple outcomes and group comparisons are considered. Testing many outcomes increases the chance that at least one result appears notable by chance. Without a reported multiplicity approach or a clear prespecified testing sequence, isolated findings should be treated cautiously rather than as definitive evidence. This is a general statistical concern, not evidence that any particular result is spurious.

The secondary outcomes are most useful as a descriptive view of whether the broad pattern resembles the primary endpoint and which dimensions were measured. They do not resolve the primary outcome's uncertainty, address missing-data concerns, or establish a causal effect. Their contribution is limited by the absence of posted analysis details.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Mean TRSS
• Mean Component Scores of the TRSS (Nasal)
• Mean Component Scores of the TRSS (Ocular)
• Mean Allergy Medication Score (AMS)
• Mean RQLQ Score

9. Statistical Interpretation

10. Statistical Methods Explained

Continuous outcomes summarize measurements that can take a range of values. The combined score is a continuous measure on a bounded scale, so group summaries are expressed as means. A mean is useful for describing the average score in a group, but it does not show the full distribution: participants can have different scores around the same mean.

What does a least-squares mean represent?

A least-squares mean is generally a model-based group mean, often adjusted for variables included in the model. It is distinct from a simple arithmetic average when adjustment is used. The registry identifies the primary and secondary means as least-squares means, but does not report the model specification or the variables used to obtain them. The posted summaries therefore do not allow a reader to reconstruct the adjustment or assess whether it was appropriate.

What is a standard error?

A standard error describes the estimated uncertainty in a mean, rather than the spread of individual participants' scores. Smaller standard errors generally indicate more precise estimates of the group mean, all else equal. The standard errors posted for the primary outcome are 0.16, 0.15, and 0.15 across the one-course, two-course, and placebo groups. They should not be interpreted as the range containing most individual scores. The registry does not post standard deviations for these groups.

How are mean differences read?

A mean difference compares a treatment-group mean with the placebo mean. A positive estimate means the treatment-group mean is higher; a negative estimate means it is lower. On the primary outcome, the posted differences are 0.06 for one course and -0.02 for two courses. The outcome's scale runs from a lower to a higher combined score, but these estimates alone do not say whether a difference is clinically important. The registry does not provide a threshold for clinical importance.

What does a confidence interval add?

A confidence interval gives a range of values compatible with the data and the calculation method at the stated confidence level. It communicates precision as well as direction. When an interval crosses zero for a mean difference, the data are compatible with no difference and with some differences in either direction. That does not prove no difference; it indicates that the estimate is not precise enough to exclude zero under that interval procedure. The primary comparison intervals cross zero.

What does an unadjusted p-value mean?

A p-value evaluates how unusual the data would be under a specified null hypothesis and statistical procedure. It is not the probability that the null hypothesis is true. The posted primary p-values are unadjusted, meaning they are not presented with a correction for testing multiple outcomes or comparisons. The registry does not report whether another multiplicity procedure was used in an analysis outside the posted results. Because several endpoints and comparisons appear in the record, unadjusted values should not be treated as if each were the sole planned test.

Why can over-dispersion matter?

General statistical explanation: over-dispersion occurs when observed variability is greater than a model assumes. It can make standard errors too small if the model does not account for that extra variability, which can make confidence intervals too narrow and p-values too small. Over-dispersion is particularly discussed for count outcomes, but the broader lesson applies to model assumptions: uncertainty estimates depend on whether the statistical model represents the data adequately. The primary outcome is continuous, and the registry does not report a dispersion assessment or enough model detail to evaluate model assumptions.

What is the status of the posted comparisons?

The record lists no reported analyses for the primary endpoint or the posted secondary endpoints. The comparison estimates, intervals, and unadjusted p-values should be understood as summary-statistic comparisons based on the posted group means and standard errors. The pooled standard deviations used for planning illustrations are approximate, because the posted summaries give standard errors rather than participant-level variability. No participant-level data, model output, or full analysis protocol is included in the registry record.

General statistical explanation: estimates, uncertainty intervals, and tests answer related but different questions. An estimate gives the direction and size of a difference; an interval describes its uncertainty; and a p-value measures compatibility with a null model under a specified procedure. None of these alone establishes causation, clinical importance, equivalence, or the absence of an effect. The study's follow-on structure and limited analysis reporting make it especially important to distinguish the observed summaries from stronger claims.

11. Confidence Intervals and Precision

The means for the primary outcome are close: 1.99 in the one-course group, 1.91 in the two-course group, and 1.93 in placebo. Standard errors are 0.16, 0.15, and 0.15, respectively. These standard errors describe uncertainty around the group means, not participant-to-participant variability. They offer one indication of the precision of the mean estimates, but do not substitute for a full account of the outcome distributions.

The mean-difference estimates are 0.06 for one course versus placebo and -0.02 for two courses versus placebo. The posted confidence intervals for both comparisons cross zero, so the direction of the underlying difference is not pinned down by the intervals. The intervals also allow values farther from zero than the point estimates. Exact endpoints are not needed to see the main limitation: the posted estimates do not exclude no difference or all potentially relevant differences.

There is no registry-reported threshold for a clinically meaningful difference on the combined scale. Statistical precision and clinical importance are separate issues. A narrow interval could still cover differences that are unimportant, while a wide interval could include differences that matter; judging that requires a prespecified interpretive threshold that is not reported in the registry record.

Precision is also conditional on the analysis population and method. The primary population consists of people who completed the End of Year 1 assessment, and the registry does not report a missing-data analysis or the model details underlying the least-squares means. The posted interval estimates are based on summary statistics and approximate variability, rather than a fully described sponsor analysis. This leaves uncertainty about how alternative modeling choices or incomplete follow-up might affect the estimates.

Accordingly, the posted data support a cautious statement about the group averages at the specified assessment, not a precise claim of equivalence or a definitive estimate of a long-term treatment effect.

12. Planning a Similar Trial: Sample Size

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

The sample-size illustrations translate the observed differences into hypothetical planning quantities. For these continuous outcomes, the figures are participants per arm, not total events. The assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Each row plans for detecting the stated fraction of the observed effect. The trial's own original sample-size justification and design assumptions are not reported in the registry record.

The 80% power column gives the number of participants per arm that the planning calculation estimates would provide an 80% chance of detecting the specified effect under those assumptions, if that effect were the true effect and the assumptions held. The 90% power column expresses the analogous planning target with a higher chance of detection. Higher power generally requires more participants, all else equal.

Rows for 100%, 75%, and 50% of the observed effect explore progressively smaller effects. Smaller effects are harder to distinguish from random variation, so the corresponding planning illustrations require more participants per arm. This relationship is particularly pronounced when the observed difference is very small relative to the variability used in the calculation.

These are planning illustrations based on the observed effects and approximate standard deviations, not the sample-size basis documented for the original trial. They are not observed power. Observed power is not a useful substitute for reading the estimates and their uncertainty after a study; the relevant information is the observed effect, its interval, and the limitations of the data and analysis.

The calculations concern two-arm comparisons, while the follow-on study reports three groups. The planning quantities are therefore best understood as hypothetical pairwise illustrations, not as a reconstruction of the study's enrollment target or a recommendation for a future trial. They do not account for unreported design features, follow-up loss, multiplicity procedures, or a clinically meaningful effect threshold.

13. 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 = 0.060; pooled standard deviation σ = 1.489.
2. n = 2 × (1.96 + 0.842)² × 1.489² / 0.060² = 9,662.5, rounded up to 9,663 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 <- 0.0600000000000001                      # observed difference in means
sigma <- 1.48859709507825                     # 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(2040844)
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 = 9,663 (80% power), 12,936 (90% power)
 75% of observed effect: n per arm = 17,178 (80% power), 22,997 (90% power)
 50% of observed effect: n per arm = 38,650 (80% power), 51,742 (90% power)
power.t.test: 9663.5 
Simulated power at n = 9,663 per arm: 0.800 (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.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and several secondary outcomes, with comparisons between each prior Cat-PAD group and placebo. When many outcomes or comparisons are tested, the chance of at least one apparently unusual result can rise even if there are no real group differences. Statistical methods can address this multiplicity, for example by adjusting significance thresholds or defining a prespecified sequence of tests. The registry does not report a multiplicity adjustment or a testing hierarchy, so whether one was used in the study's analysis is not reported in the registry record.

The primary comparison p-values are identified as unadjusted. They therefore should not be interpreted as family-wise adjusted evidence across the outcome set. The secondary outcomes have no reported inferential analyses, making it inappropriate to treat their descriptive means as formal positive or negative tests. The absence of posted results does not establish that no analyses were performed; it means those analyses are not reported in the registry record.

The analysis population for the primary outcome is defined by completion of the End of Year 1 assessment. This is a clear description of who contributed to the endpoint summary, but the registry does not report the number eligible for follow-up, the number missing the assessment, or reasons for missingness. It also does not explain whether any imputation, weighting, or sensitivity analysis was used. If completion is related to participants' outcomes or group assignment, a completer analysis may not fully represent all enrolled participants. That is a general limitation to consider, not a claim that such bias occurred.

The record does not provide a detailed definition of the analysis population for each secondary outcome, nor a flow of participants across follow-up assessments. It also does not report whether outcome analyses followed an intention-to-treat principle, used a per-protocol population, or applied another approach. Those details are not reported in the registry record.

The absence of a described analysis plan limits assessment of prespecification, covariate adjustment, handling of repeated measures, and control of false-positive findings. It also makes it difficult to know whether the posted summary-statistic comparisons align with the primary analysis used by the sponsor. These omissions do not invalidate the descriptive results, but they constrain how strongly they can support inferential conclusions.

15. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: squamous cell carcinoma (0.7%), haemolytic uraemic syndrome (0.7%), colonic fistula (0.7%), colonic abscess (0.7%), gastronteritis escheria coli (0.7%).

The safety record reports serious adverse events in 3 participants at risk in the one-course Cat-PAD group, 4 in the two-course group, and 3 in placebo. It reports no deaths in any of the three groups. The recorded numbers at risk for serious adverse events are 138, 148, and 144, respectively. These safety populations differ from the primary outcome analysis groups, which are based on completion of the specified assessment.

The registry also reports other adverse events affecting 54, 62, and 56 participants in the respective groups, with the same group-specific numbers at risk listed for those events. These counts describe participants affected, not the total number of individual event episodes. The record lists several serious-event terms with a reported 0.7% beside each; the listing does not establish that those events were caused by study treatment.

Counts alone do not describe severity beyond the registry's serious-event classification, timing, duration, relatedness, or differences in exposure and follow-up. No deaths were recorded, but that fact by itself does not summarize all safety outcomes. Since CP007A involved no further investigational product, these figures belong to the follow-on record and should be interpreted in the context of the earlier study treatment groups, not as adverse events following newly administered treatment in this follow-on study.

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

The results describe a follow-up assessment of people who had received one of the prior CP007 treatment assignments and completed the relevant assessment. The mean combined scores were close across the three groups, and both posted treatment-versus-placebo intervals cross zero. The most direct interpretation is that the registry's summary comparisons do not show a clear separation in mean combined score at this assessment, while leaving uncertainty about the size and direction of any underlying difference.

This is not proof that the groups are equivalent. Equivalence or non-inferiority requires a defined margin and an analysis designed to test that question; neither is reported in the registry record. Likewise, a non-significant unadjusted p-value does not demonstrate that the treatment has no effect. The results are compatible with a range of possible differences, and the registry does not provide a threshold for deciding which differences would matter clinically.

The follow-on design is also important. Participants received no further investigational product in CP007A, so the results address outcomes after earlier treatment rather than the effect of a treatment newly given during follow-up. The record's stated purpose extends to evaluation over up to five years after treatment began, but the posted primary outcome concerns the assessment one year after CP007 ended. The posted primary comparison should not be taken as a complete account of outcomes across the entire intended follow-up period.

The outcome combines an average symptom rating with allergy medication use. Similar combined means do not tell whether the symptom component and medication component were individually similar, nor whether different combinations of those components produced similar total scores. The secondary summaries measure related dimensions, but their inferential analyses are not reported.

Finally, these results do not establish a causal conclusion without considering the original study allocation, follow-up completion, analysis population, and statistical methods. The registry reports randomized allocation and a prior placebo-controlled study, but detailed analysis procedures and missing-data handling are not reported. The data support a careful description of the posted averages and their uncertainty, not medical advice or a claim about treatment effectiveness.

17. Why This Trial Matters Statistically

This study illustrates a common statistical challenge in long-term follow-up: the question may concern persistence of outcomes after an earlier intervention, while the follow-on study itself administers no new treatment. Keeping that distinction clear prevents a follow-up comparison from being read as if it were a new active-treatment trial. The groups represent prior treatment histories, and the follow-up assessment represents a later point in time.

The study also makes clear why point estimates should be read together with intervals. The primary means and differences are small in the posted summaries, but the intervals crossing zero do not demonstrate equality. They indicate that the results do not precisely isolate a difference in one direction. A p-value alone cannot communicate this range of uncertainty or the possible relevance of values inside it.

A further statistical lesson is the distinction between an enrollment figure and the population contributing to a particular outcome. The record lists enrollment of 430, while the primary assessment summaries are based on participants completing the named follow-up assessment. Attrition and assessment-specific denominators matter for interpreting who the results describe. The registry's limited follow-up flow and missing-data reporting restrict evaluation of that issue.

The record includes multiple related outcomes and comparisons, but little inferential reporting beyond the primary summary comparisons. This makes multiplicity and prespecification important considerations. Without a stated testing hierarchy or correction, unadjusted secondary results would have limited confirmatory weight even if inferential estimates were available.

Finally, the example shows the limits of retrospective sample-size illustrations based on observed effects. Such calculations can describe hypothetical planning needs under stated assumptions, but they cannot reveal the original design rationale or replace the estimate-and-interval interpretation of completed data. The trial's own sample-size justification is not reported in the registry record.

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 provides limited detail on the statistical analysis. It identifies least-squares means and standard errors, but does not describe the model, covariates, missing-data method, or sensitivity analyses. The posted mean-difference comparisons are based on group summaries and approximate variability, rather than a full model output. This makes it difficult to evaluate whether those comparisons match the sponsor's planned primary analysis.

The primary outcome population is restricted to participants who completed the End of Year 1 assessment. The record does not provide a participant flow across follow-up, reasons for missing assessments, or the relationship between completion and outcomes. The enrollment figure should therefore not be read as the number contributing to each endpoint. The extent to which missing follow-up could affect the group comparisons cannot be assessed from the registry record.

The outcome is a composite of symptoms and medication use. A similar combined average could arise from different symptom and medication patterns, and the registry does not provide posted inferential analyses for those components. Nor does the record report a threshold for a clinically meaningful difference, limiting interpretation of the size of any observed difference.

Multiple outcomes and pairwise comparisons are listed, but a multiplicity strategy and hierarchy of testing are not reported. Secondary findings therefore have limited confirmatory interpretation. The study's original sample-size justification and detailed analysis plan are also not reported in the registry record.

Finally, this follow-on study assessed prior treatment groups without further investigational product. The primary result addresses a specific follow-up assessment, not the full stated period of evaluation up to five years. It cannot by itself establish the duration of any effect or represent every possible outcome during long-term follow-up.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Phase III Cat-PAD Follow-on Study — NCT02040844.
• Registry record last updated 2018-06-15; status last verified 2018-05.
• Results first posted 2018-05-14.
• 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

NCT02040844 is recorded as a completed phase III randomized, quadruple-masked, parallel follow-on study in cat allergy. It evaluated outcomes after prior CP007 assignments to Cat-PAD treatment or placebo; no further investigational product was administered in CP007A.

The primary endpoint was the mean combined symptom-and-medication score one year after CP007 completion. The posted group means were 1.99 for one course, 1.91 for two courses, and 1.93 for placebo. The mean differences versus placebo were 0.06 and -0.02, with unadjusted p-values of 0.784 and 0.925. Both intervals cross zero.

The registry also lists secondary symptom, medication, and quality-of-life outcomes, but does not report inferential analyses for them. Detailed analysis methods, missing-data handling, multiplicity procedures, and the original sample-size justification are not reported in the registry record. The results are best read as follow-up summaries with limited analysis detail, rather than as proof of equivalence or a definitive long-term causal effect.