A Safety and Activity Study of NS2 in Subjects With Allergic Conjunctivitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Safety and Activity Study of NS2 in Subjects With Allergic Conjunctivitis (NCT02578914), a randomized phase 2 study. The primary outcome is Ocular Itching Score at Day 14, 60 Minutes Into the Allergen Challenge.

2. Clinical Question

The study evaluates activity under a controlled allergen-provocation setting rather than describing itching across everyday exposure. The endpoint focuses on a specified time during the conjunctival allergen provocation test (CAPT), with baseline recorded at Day -7 and the efficacy assessment at Day 14. Ocular itching was rated on a scale from 0, meaning none, to 4, meaning severe.

The reported treatment contrast is the mean difference in change between the NS2 and vehicle groups. The registry’s arm estimates are least squares means adjusted for baseline through an analysis of covariance (ANCOVA). The primary question is therefore about the adjusted average change in this particular study population and setting. It does not by itself establish how results would generalize to other populations, exposure conditions, dosing schedules or follow-up periods. Those details are not reported in the registry record.

3. Trial Design

The registry identifies this as a randomized, parallel-group Phase II study at a single center, with vehicle control. The official title describes it as double-masked; the brief summary also calls it double-masked. The structured masking field, however, says quadruple. Which roles were masked and how masking was implemented are not reported in the registry record.

The study compared NS2 ophthalmic drops (0.5%) with vehicle ophthalmic drops (0.0%). The summary says participants were randomized 1:1 and received multiple doses. The number of doses, schedule, treatment duration and details of allocation generation or concealment are not reported in the registry record. The registry lists 100 participants enrolled, with 50 participants in each arm for the primary outcome.

The listed condition is allergic conjunctivitis, with seasonal allergies also recorded. The brief summary describes seasonal allergic conjunctivitis induced by grass, tree or ragweed pollen. The study began in 2015-08, had primary completion in 2015-12, and is marked completed. The registry does not report the exact enrollment and completion dates or the duration of follow-up beyond the stated primary assessment.

The primary outcome was mean change from baseline in ocular itching score at Day 14, 60 minutes into the allergen challenge. Baseline was Day -7. The posted result is for the intent-to-treat population and is presented as a least squares mean with a standard error for each treatment group. The analysis description specifies ANCOVA with baseline as a covariate and treatment group as a factor.

The registry record does not report the original sample-size justification, planned primary analysis details beyond the endpoint description, or whether any interim analyses were planned or conducted. It lists no posted secondary outcomes and no analysis entries, although the primary outcome result is present. These missing details limit what can be reconstructed about the full prespecified design and analysis plan.

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

Analysis population for the primary outcome, as defined in the registry: Intent-to-treat population

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 Ocular Itching Score at Day 14, 60 Minutes Into the Allergen Challenge, measured over The efficacy assessment period was assessed at Day 14; baseline was Day -7..

The primary endpoint is change from baseline in ocular itching score on Day 14, measured 60 minutes into the allergen challenge. The score ranges from 0, none, to 4, severe. A change score describes how the participant’s recorded score at the assessment differs from baseline; a more negative change indicates a larger decrease in the score relative to baseline.

The allergen challenge provides a defined testing context and timing for assessing itching. That makes the endpoint a measure of response during a specified provocation assessment, rather than a summary of symptoms over an unspecified period. The registry’s assessment description identifies baseline as Day -7 and the efficacy assessment as Day 14.

The posted group results are least squares means, not simple unadjusted averages. The registry says they were estimated using ANCOVA, including baseline score as a covariate and treatment group as a factor. Accounting for baseline in this way can make the group comparison reflect differences in outcome after adjustment for starting scores. The registry does not report the model’s other details, including assumptions checks or handling of missing measurements.

This endpoint is useful for asking whether average itching-score changes differed between the randomized groups at the specified assessment. It does not capture every possible symptom, outcome or experience, and the record does not report additional primary endpoints. Its interpretation should remain tied to the scale, challenge conditions, assessment time and population stated in the registry.

7. Results Posted in the Registry

• NS2 Ophthalmic Drops (0.5%): mean -1.55 versus -1.01 on control; difference -0.54 (95% CI -1.00 to -0.08), standardized effect d = -0.46; z-test p = 0.022.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is continuous: it is a score whose numerical values represent ordered levels of ocular itching, and the registry reports a mean change. For a continuous outcome, group summaries describe average values, while a treatment contrast describes how those averages differ. The score’s stated scale runs from 0, none, to 4, severe; a negative change from baseline means a decrease in score.

What does the least squares mean represent?

A least squares mean is a model-based group mean, adjusted for covariates included in the analysis. In this trial, the registry specifies ANCOVA with baseline as a covariate and treatment group as a factor. This means the reported arm estimates are not simply the raw arithmetic averages of observed changes; they are estimates from a model that accounts for baseline score. The registry does not report additional covariates or the full model specification.

Adjustment for a baseline measurement is commonly used in randomized trials of continuous outcomes. Baseline scores may be associated with follow-up scores, and including baseline can improve the precision of the treatment comparison when the model is appropriate. This is a general statistical explanation, not a claim about diagnostics or performance in this specific study. Model checks and diagnostic results are not reported in the registry record.

How should the mean difference be read?

The comparison is NS2 minus vehicle: -0.537 units. The negative sign indicates that the NS2 group’s estimated change was lower than the vehicle group’s estimated change. Since the endpoint is change in a scale where 0 is none and 4 is severe, lower scores and more negative changes represent a larger decrease in the score. The contrast is expressed in the scale’s units and is not a ratio or a percentage.

A treatment difference is an estimate, not an exact description of every participant’s experience. Individual scores can vary around a group mean, and the group contrast does not imply that every participant had the same change. The registry reports standard errors for the two arm means but does not provide individual-level outcome distributions.

What do a p-value and confidence interval add?

The unadjusted p-value is 0.022. A p-value measures how incompatible the observed data are with a specified null model, under the assumptions of the analysis; it is not the probability that the null hypothesis is true, nor does it measure the clinical importance of an effect. The registry does not report a multiplicity adjustment, so the p-value should be read as unadjusted.

A confidence interval gives a range of values compatible with the data and the statistical procedure at the stated confidence level. For a treatment difference, it conveys uncertainty about both direction and magnitude. Here, the reported interval lies below zero. Its breadth is also relevant: values across the interval represent a range of plausible mean differences under the analysis. The confidence level itself is not stated in the registry record, and the exact interval is not reproduced in this explanation.

Why does over-dispersion matter?

Over-dispersion occurs when outcome variability is greater than a statistical model expects. In general, ignoring over-dispersion can make standard errors too small and uncertainty appear lower than it is. This concept is particularly relevant to count outcomes or outcomes modeled with a restricted variance structure. The primary endpoint is reported as a continuous score analyzed with ANCOVA; the registry does not report an over-dispersion assessment or a separate variance-modeling method. There is therefore no basis in the record to say that over-dispersion was detected or that a correction was used.

The arm estimates include standard errors of 0.168 for NS2 and 0.163 for vehicle. A standard error describes the estimated precision of a mean under the analysis; it is not the spread of individual participant scores. The pooled standard deviation used for an illustrative standardized effect is approximate, and no standardized effect estimate is needed to interpret the endpoint in its original units.

What is and is not documented?

The registry provides the endpoint definition, its analysis population, the ANCOVA description, arm-level least squares means and standard errors, and the posted comparison. It lists no analysis entries, despite the primary result being posted, and no secondary outcomes. The full statistical analysis plan, model diagnostics, missing-data handling, interim-analysis plan and original sample-size calculation are not reported in the registry record. General statistical principles can explain the methods, but those missing details cannot be inferred from the posted result.

10. Confidence Intervals and Precision

Precision concerns how much uncertainty surrounds an estimate. The mean difference is -0.537 units, and the reported confidence interval lies entirely below zero. That indicates that the interval’s range of compatible values is negative under the stated calculation. The interval is not a guarantee that the true difference lies within its bounds; it is an uncertainty summary whose interpretation depends on the analysis assumptions.

The interval is important because it gives more information than the point estimate alone. The point estimate identifies the center of the reported comparison, while the interval communicates a range of values consistent with the data and method. Its width reflects uncertainty: a wider interval allows a broader range of possible differences, while a narrower one indicates greater precision. The registry record does not state the confidence level, and the exact endpoints are not repeated.

The comparison is based on 50 participants per arm. The reported standard errors for the adjusted arm means are 0.168 for NS2 and 0.163 for vehicle. Standard errors describe uncertainty in estimated means; they do not show how variable participants’ individual scores were. The registry does not provide individual-level data or a distribution of scores, so variability among participants cannot be assessed from these summaries alone.

The negative direction is consistent across the reported point estimate and interval, but uncertainty remains about the precise magnitude. The p-value of 0.022 is another summary of evidence under a null model; it does not replace the confidence interval or convey the range of compatible effect sizes. Further, multiplicity procedures and missing-data methods are not reported in the registry record, which constrains a fuller assessment of precision and inferential certainty.

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 sample-size figures are planning illustrations, not the trial’s original design calculation. They are expressed as participants per arm, not total events. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect used for each planning row is the stated fraction of the observed effect. The trial’s own original sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the number of participants per arm that the illustration estimates would be needed to detect the chosen effect under those assumptions with 80% power. The 90% power column gives the corresponding per-arm planning figure for 90% power. Power is a design-stage probability calculated under specified assumptions; these figures do not describe the chance that the observed result is true. This is not observed power.

For planning at 100% of the observed effect, the illustration gives 75 participants per arm for 80% power and 100 participants per arm for 90% power. For planning at 75% of the observed effect, it gives 133 and 178 participants per arm, respectively. For planning at 50% of the observed effect, it gives 299 and 400 participants per arm, respectively.

Smaller effects are harder to distinguish from random variation, so the illustration requires more participants per arm as the effect used for planning becomes smaller. These are assumptions-based examples using the observed comparison, not predictions of results in a future study. The registry record does not report whether its original design used these assumptions or any of these target effects.

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 = -0.537; pooled standard deviation σ = 1.170.
2. n = 2 × (1.96 + 0.842)² × 1.170² / -0.537² = 74.6, rounded up to 75 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 <- -0.537                      # observed difference in means
sigma <- 1.17039523238947                     # 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(2578914)
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 = 75 (80% power), 100 (90% power)
 75% of observed effect: n per arm = 133 (80% power), 178 (90% power)
 50% of observed effect: n per arm = 299 (80% power), 400 (90% power)
power.t.test: 75.5 
Simulated power at n = 75 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.

13. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, time points, comparisons or subgroup analyses are tested. Testing many hypotheses can increase the chance of at least one apparently notable result by chance. Statistical plans may address this through a prespecified hierarchy or adjustment procedure. The registry record does not report a multiplicity strategy for this trial, so it is not possible to determine whether or how multiple testing was controlled.

The registry lists one outcome measure and no posted secondary outcomes. The primary outcome is assessed at a specified time, Day 14, 60 minutes into the allergen challenge. The record does not establish whether additional outcomes or analyses were prespecified but not posted. The comparison’s p-value is explicitly unadjusted and should not be described as multiplicity-adjusted.

The primary analysis population is identified as intent-to-treat. The registry does not report the number included in that population for the analysis beyond the arm counts, nor does it describe exclusions, protocol deviations or how participants were classified if they did not complete the assessment. It also does not report the amount or pattern of missing outcome data.

Missing-data handling can matter because an analysis of participants with observed outcomes may differ from an analysis that makes assumptions about those without them. Common approaches rely on assumptions about why data are missing and how missing outcomes relate to observed information. These are general considerations; the registry record does not identify an approach used here, such as imputation, a likelihood-based method or a sensitivity analysis. No such method should be inferred from the ANCOVA description alone.

The record also does not report interim analyses, stopping rules, model diagnostics or a full statistical analysis plan. These gaps do not invalidate the posted summaries, but they limit independent assessment of prespecification and robustness. The available account supports interpretation of the stated endpoint and comparison only, with the unreported analysis decisions kept distinct from what is documented.

14. 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 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events for 50 participants at risk in each group. There were 0 serious adverse events in the NS2 group and 0 in the vehicle group. Deaths were 0 in each group. No serious adverse-event terms are listed.

These figures are counts within the registry’s stated groups. They do not show that no non-serious events occurred: the separate “other” adverse-event figures list 49 affected participants among 50 at risk in the NS2 group and 11 among 50 at risk in the vehicle group. The record does not provide further details in the listed serious-event terms, and the safety observation period is not reported in the registry record.

Zero serious events in both groups means that no serious events were recorded in these summaries; it does not establish that serious events are impossible or that the groups have identical risks. The trial enrolled 100 participants overall, so the safety counts should be interpreted within the size and reporting scope of this study. These descriptive data are not medical advice and do not establish safety in other populations, settings or durations.

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

The posted results show a lower adjusted mean change in ocular itching score for NS2 than for vehicle at Day 14, 60 minutes into the allergen challenge. The estimated difference is -0.537 units, with an unadjusted p-value of 0.022. The reported confidence interval lies below zero, while still representing uncertainty about the exact size of the mean difference.

This finding applies to the endpoint and conditions described in the registry: an ocular itching scale from 0 to 4, a baseline at Day -7, an assessment on Day 14 during a CAPT, and an intent-to-treat population. It is an average group comparison and does not describe every individual response. It also does not directly answer questions about symptoms outside the challenge setting or outcomes that were not posted.

The result does not establish how the estimate would translate to routine exposure, another treatment schedule, another population or longer follow-up. Those questions are not answered by the registry record. The record identifies a single-center Phase II study and provides no posted secondary outcomes, so the result should not be generalized beyond its stated scope without additional evidence.

The safety record reports no serious adverse events or deaths in either group, alongside separate counts for other adverse events. These are descriptive counts in the study, not proof that an event cannot occur or a comprehensive characterization of safety over time. The safety observation period and detailed event descriptions are not reported in the registry record.

Statistical evidence and clinical importance are distinct. A p-value does not measure how important a difference is to participants, and the record does not provide a threshold for a meaningful change in this score. No treatment recommendation follows from the posted statistics alone.

16. Why This Trial Matters Statistically

This trial illustrates why an endpoint’s measurement context matters. Ocular itching was measured during a defined allergen challenge at a specified time, rather than as an open-ended symptom report. The resulting estimate answers a focused statistical question about average change in that setting. Keeping that context attached to the result prevents the estimate from being treated as a general measure of symptoms in every circumstance.

The study also illustrates the role of baseline adjustment. The posted arm results are least squares means from ANCOVA, with baseline as a covariate and treatment group as a factor. Comparing adjusted means can account for starting scores in the model, while the mean difference remains expressed in the original scale units. The interpretation still depends on the model and its assumptions, details of which are only partly reported.

Precision adds a necessary counterpart to the point estimate. The mean difference has a negative direction, and its interval lies below zero, but the interval communicates uncertainty in magnitude. Reporting the interval alongside the p-value helps readers avoid treating a single threshold-crossing result as a complete account of the evidence.

The registry also makes clear the difference between a trial’s actual design and a later planning illustration. The sample-size figures use the observed effect under explicit assumptions; they do not reveal the original rationale for enrolling 100 participants. Such distinctions matter because planning calculations are sensitive to the effect size chosen and should not be presented as observed power.

Finally, gaps in the registry record—such as unreported multiplicity procedures and missing-data methods—shape how confidently the analysis can be reconstructed. Identifying what is documented and what is not is part of sound statistical interpretation, not a substitute for the results themselves.

17. 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 results come from a single-center study, and the registry record does not report details about the center or recruitment procedures. The listed population is seasonal allergic conjunctivitis induced by grass, tree or ragweed pollen. Generalizability to other forms of conjunctivitis, other exposure settings or populations is not established by this record.

The primary assessment was tied to a conjunctival allergen provocation test on Day 14, 60 minutes into the challenge. The result therefore describes the specified testing context and time point. Follow-up beyond this efficacy assessment, the dosing schedule and the number of doses are not reported in the registry record.

The masking description is not fully consistent across registry fields: the official title and brief summary say double-masked, while the structured masking field says quadruple. The roles masked and operational details are not reported, so the nature of masking cannot be clarified from the record alone.

The record reports an intent-to-treat population but does not detail analysis-population counts, missing outcome data, missing-data handling, protocol deviations, model diagnostics or sensitivity analyses. It also does not report multiplicity procedures, interim analyses or the full statistical analysis plan. These omissions limit evaluation of the assumptions and robustness of the posted estimate.

Only the primary outcome result is posted; no secondary outcomes are listed. The record does not provide a threshold for a clinically meaningful change in ocular itching score. Its safety summaries list serious events and other adverse events, but the safety observation period and additional event details are not reported in the registry record.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: A Safety and Activity Study of NS2 in Subjects With Allergic Conjunctivitis — NCT02578914.
• Registry record last updated 2023-12-05; status last verified 2016-05.
• Results first posted 2023-12-05.
• 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.

21. Record Summary

NCT02578914 is a completed randomized, parallel-group, vehicle-controlled Phase II study of NS2 ophthalmic drops (0.5%) versus vehicle (0.0%) for seasonal allergic conjunctivitis. The registry lists 100 participants enrolled, with 50 participants per arm in the primary outcome results. The study began in 2015-08 and reached primary completion in 2015-12.

The primary endpoint was change from baseline in ocular itching score at Day 14, 60 minutes into an allergen challenge. The adjusted mean change was -1.55 for NS2 and -1.013 for vehicle; the estimated mean difference was -0.537, with an unadjusted p-value of 0.022. The reported confidence interval lies below zero.

The registry reports 0 serious adverse events and 0 deaths in each group. It lists no secondary outcomes, and several analysis details—including missing-data methods, multiplicity procedures and the original sample-size justification—are not reported in the registry record.