Exploratory Study of OPC-12759 Ophthalmic Suspension: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Exploratory Study of OPC-12759 Ophthalmic Suspension (NCT01493180), a randomized phase 2 study. The primary outcome is Change in the Keratoconjunctival Staining Score From Baseline.

2. Clinical Question

The registry describes the study as an exploratory evaluation of efficacy and safety. Its primary comparison concerns a continuous score: the mean change from baseline in staining score. The reported group means were -3.4 for OPC-12759 and -3.1 for sodium hyaluronate. Since the score is a measure of epithelial staining and 0 is better, a negative change indicates movement toward a lower score from baseline; the registry does not provide baseline group means or individual participant trajectories.

The question is comparative, not simply whether either group changed. The posted results estimate the difference between average changes and provide an uncertainty interval for that contrast. The interval spans values on either side of no difference, so the results do not precisely distinguish the groups on this measure. That uncertainty should be considered alongside the study’s exploratory purpose and the information not reported in the registry record, including details about analysis populations and missing observations.

3. Trial Design

This was a completed phase 2, randomized, single-masked, parallel-group study sponsored by Otsuka Pharmaceutical Co., Ltd. The registry lists a start date of 2011-12 and primary completion in 2012-11. It reports total enrollment of 102 participants, divided evenly between the two posted groups: 51 received OPC-12759 ophthalmic suspension and 51 received sodium hyaluronate ophthalmic solution.

The parallel design means the groups received different interventions rather than switching between them as part of the reported comparison. The registry labels the study single-masked but does not state which people involved in the trial were masked. Further details of the randomization procedure, allocation concealment, treatment duration, dosing schedule, and participant flow are not reported in the registry record. The brief summary identifies evaluation of efficacy compared with sodium hyaluronate and assessment of safety as the study’s purposes.

One primary outcome measure and one analysis are posted. The primary outcome is change in keratoconjunctival staining score from baseline, with the timeframe described as baseline and 4 weeks. The registry says changes at the end of instillation were compared using a two-sided t-test and describes the outcome data as last observation carried forward. It does not report the original sample-size justification or the assumptions used when planning the trial. The separate sample-size figures are planning illustrations based on the observed contrast, not the trial’s original design plan.

Some details in the primary outcome description name the groups by concentration and rebamipide, while the intervention listings use the names OPC-12759 ophthalmic suspension and sodium hyaluronate ophthalmic solution. The registry identifies the comparison as the 2% rebamipide group versus the 0.1% sodium hyaluronate group. It does not provide further treatment-delivery details in the posted record.

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

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 in the Keratoconjunctival Staining Score From Baseline, measured over Basekine, 4 weeks.

The primary endpoint is change from baseline in the keratoconjunctival staining score. The outcome description explains that staining is used to indicate damage to the corneal and conjunctival epithelium. The cornea is divided into 3 fractions and the conjunctiva into 2 fractions; each fraction receives a score from 0 to 3, and the scores are combined into a total from 0 to 15. A score of 0 is better.

Change from baseline expresses how the score at the later assessment differs from the participant’s starting score. Comparing average changes between randomized groups focuses the analysis on the difference in mean change, rather than on either group’s change considered alone. The registry describes the timeframe as baseline and 4 weeks and says the comparison uses change at the end of instillation, with last observation carried forward for the endpoint analysis.

This kind of continuous measure can preserve variation across the score range rather than reducing outcomes to a simple event/no-event category. Its interpretation nevertheless depends on how the scale behaves and on the analysis population and missing-data handling. The registry describes the scale and the scoring direction, but does not report a threshold for a clinically important change or details about the measurement process beyond the scoring scheme. The posted results therefore describe average score changes and their estimated between-group contrast; they do not by themselves establish how a particular score change would be experienced by an individual participant.

7. Results Posted in the Registry

• Sodium Hyaluronate Ophthalmic Solution: mean -3.10 versus -3.40 on control; difference 0.30 (95% CI -0.69 to 1.29), standardized effect d = 0.12; z-test p = 0.553.

8. Statistical Interpretation

9. Statistical Methods Explained

A mean summarizes the average value of a continuous outcome across participants in a group. For this study, the group means are average changes in the staining score from baseline. The mean difference compares those two averages. The contrast is stated as sodium hyaluronate minus OPC-12759, so a positive estimate corresponds to a higher average change in the sodium hyaluronate group. Since lower staining scores are described as better, the direction of the contrast can be interpreted against the score’s stated direction, while keeping in mind that the estimate concerns group averages.

What does the standard deviation describe?

A standard deviation describes how spread out participant values are around the group mean. The reported standard deviations are 2.7 in the OPC-12759 group and 2.4 in the sodium hyaluronate group. They are not standard errors and are not confidence intervals. The standard error concerns the precision of an estimated quantity, such as a mean or mean difference; it generally reflects both observed variability and the amount of information available. The recalculation reports group standard errors of 0.378 for OPC-12759 and 0.336 for sodium hyaluronate. These values describe uncertainty in the group means rather than the spread of individual scores.

Why use a t-test?

The registry says the primary outcome comparison used a two-sided t-test. In general, a t-test assesses whether the observed difference in means is compatible with a null hypothesis of no mean difference, subject to the method’s assumptions. A two-sided test considers departures in either direction rather than only a prespecified direction. The registry does not report whether the test was paired, adjusted, or otherwise modified, nor does it describe diagnostics for assumptions. The posted group summaries allow an unadjusted recalculation of the mean difference and its uncertainty, but they do not reproduce any analysis that might depend on participant-level data or additional modeling.

What is a p-value?

In general, a p-value measures how unusual the observed result, or a result more extreme under the test statistic, would be if the null hypothesis and the test assumptions held. It is not the probability that the null hypothesis is true, nor is it a measure of the probability that the treatment has an effect. The sponsor-reported two-sided p-value is 0.576. The recalculated unadjusted p-value is 0.553. The values are close but not identical; the registry does not give a detailed explanation for the difference. The recalculation is based on posted summary statistics and is not a replacement for the sponsor’s analysis.

What does the confidence interval add?

In general, a confidence interval gives a range of estimates compatible with the data and the method used to construct it. Its width conveys precision: a wider interval indicates more uncertainty about the location of the estimate. The recalculated interval for the mean difference runs from -0.691 to 1.291. Because it spans zero, it includes no mean difference and also includes estimates in either direction. The interval should not be read as a list of equally likely values, and its confidence level is not reported in the registry record. The posted interval is a recalculation from the summary statistics, not a sponsor-reported interval.

Why can variability affect precision?

Greater participant-to-participant variability tends to make a group mean or between-group contrast less precise when the amount of data is held fixed. Conversely, more information can improve precision when other conditions are comparable. This is a general statistical principle, not a conclusion that changes the observed result. The two standard deviations describe variability in the endpoint, while the confidence interval describes uncertainty in the estimated contrast. They answer related but distinct questions.

What is not established by the methods description?

The registry names a two-sided t-test and last observation carried forward, but does not report the trial’s full statistical analysis plan. Details such as the analysis population, the handling of protocol deviations, baseline adjustment, model diagnostics, and any prespecified sensitivity analyses are not reported in the registry record. The number of planned interim analyses and the original sample-size justification are also not reported in the registry record. Without those details, the posted summaries support a straightforward comparison of group means but do not allow a complete audit of all analysis decisions.

Over-dispersion is a general concern for some count outcomes when observed variability exceeds what a simple model assumes. It is not the central issue for this continuous mean comparison, and the registry does not describe a count model for the primary endpoint. The methods should therefore be understood in terms of a continuous score, its mean change, and the uncertainty around the group contrast.

10. Confidence Intervals and Precision

The recalculated mean difference is 0.3, with a confidence interval from -0.691 to 1.291. The estimate is the center of the reported contrast, while the interval indicates that the data are compatible with a range of possible differences. Its negative values would correspond to a lower mean change in the sodium hyaluronate group relative to OPC-12759 under the stated contrast; positive values correspond to a higher mean change in that group. Because 0 is included, the interval includes no difference between group means.

The interval also extends into both directions by amounts that leave uncertainty about which group had the lower mean change in the underlying comparison. A confidence interval does not show that all values inside it are equally plausible, and it does not mean that the true effect has a particular probability of lying within this specific interval. In general, intervals describe uncertainty under the calculation method and its assumptions. The confidence level for this recalculated interval is not reported in the registry record.

The group standard deviations, 2.7 and 2.4, describe variation in observed changes among participants. The recalculated standard errors of the group means are 0.378 and 0.336. These quantities are not interchangeable: standard deviations describe spread among observations, while standard errors describe precision of estimates. The interval for the difference is the most direct summary of precision for the between-group question.

The sponsor’s posted analysis reports a two-sided p-value but no estimate or confidence interval. The recalculated interval therefore adds an uncertainty summary derived from posted group statistics. It should be interpreted as an unadjusted recalculation, not as an interval reported by the sponsor. The registry does not include individual participant data or enough detail to assess whether alternative analysis choices would materially change the interval.

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 based on fractions of the observed mean difference, not the original enrollment plan. They give participants per arm for 80% power and participants per arm for 90% power. They are not total numbers across both arms and are not total event counts. The assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The targeted effect is the stated fraction of the observed effect.

For planning for 100% of the observed effect, the figures are 1,139 participants per arm for 80% power and 1,524 participants per arm for 90% power. For planning for 75% of the observed effect, they are 2,024 participants per arm for 80% power and 2,709 participants per arm for 90% power. For planning for 50% of the observed effect, they are 4,553 participants per arm for 80% power and 6,095 participants per arm for 90% power.

In plain language, the 80% power column represents the number of participants per arm estimated under the stated assumptions to give an 80% chance of detecting the targeted effect if that effect were present. The 90% column uses a higher target chance of detection and consequently calls for more participants under these calculations. Smaller effects need more participants to distinguish them from random variation with the same planned power and significance level. These are planning calculations, not observed power, and they should not be treated as the number of participants the original trial needed or as evidence about the trial’s success.

The trial enrolled 51 participants per arm. That enrollment is the observed study size; the much larger planning figures answer a different question, using the observed contrast as the starting point. The original sample-size assumptions and justification are not reported in the registry record. The curve associated with these planning calculations should be read as illustrating the relationship between the assumed effect size and the required number per arm under the same assumptions, not as another observed trial outcome.

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 = 0.300; pooled standard deviation σ = 2.554.
2. n = 2 × (1.96 + 0.842)² × 2.554² / 0.300² = 1,138.1, rounded up to 1,139 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 <- 0.3                      # observed difference in means
sigma <- 2.55440795488896                     # 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(1493180)
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 = 1,139 (80% power), 1,524 (90% power)
 75% of observed effect: n per arm = 2,024 (80% power), 2,709 (90% power)
 50% of observed effect: n per arm = 4,553 (80% power), 6,095 (90% power)
power.t.test: 1139 
Simulated power at n = 1,139 per arm: 0.801 (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 sponsor’s primary comparison as a two-sided t-test with p = 0.576. It does not post the sponsor’s estimated mean difference or a sponsor-reported confidence interval. The recalculation from the posted group summaries gives a mean difference of 0.3, a confidence interval from -0.691 to 1.291, and an unadjusted p-value of 0.553.

The recalculation is a simple unadjusted analysis from the summary means, standard deviations, and group sizes. It is useful for showing an estimated contrast and its precision, but it may not capture analysis features that require participant-level data or details of the sponsor’s analysis. The registry does not explain why the p-values differ. Both analyses address the same comparison and both reported p-values are presented as two-sided tests.

The recalculated confidence interval should not be attributed to the sponsor, and the recalculated p-value should not replace the reported analysis. The registry’s sponsor-reported analysis is the trial’s posted inferential result; the recalculation is supplementary context based on the group summaries. No discrepancy flag is recorded for this comparison.

14. Multiplicity, Missing Data and Analysis Populations

The registry posts one primary outcome measure and one analysis, and it reports no secondary outcome measures. These registry entries limit the scope of multiplicity concerns that can be assessed from the posted results, but they do not establish that the study considered only one hypothesis or conducted only one analysis. Whether additional outcomes, time points, subgroups, or alternative analyses were planned or examined is not reported in the registry record. Any multiplicity adjustment or prespecified strategy for controlling false-positive findings is also not reported in the registry record.

The primary outcome description states that the end-of-instillation change was analyzed using last observation carried forward. This indicates that an earlier recorded observation was carried forward for participants with a later missing outcome, but the registry does not report how many observations were missing, when they were missing, or how many participants were affected by this procedure. It does not describe reasons for missingness or provide sensitivity analyses using other missing-data assumptions. Last observation carried forward is an analysis choice, not proof that missing outcomes would have remained unchanged.

The analysis population is not reported in the registry record. It is therefore unclear from the posted information whether the analysis included all randomized participants, only participants who completed treatment, or another defined group. The registry also does not report participant flow, exclusions after randomization, adherence, or protocol deviations in the information summarized for this endpoint. These omissions make it difficult to determine how closely the analyzed groups correspond to the randomized groups.

The registry names a two-sided t-test but does not report a full statistical analysis plan or the timing of any interim analyses. Those details should not be inferred from the endpoint description. Consequently, the posted estimates can be interpreted as a comparison of the reported group summaries, while the effects of unreported multiplicity, population definitions, and missing-data decisions cannot be assessed from the registry record.

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 1.0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: meningitis (2.0%).

Serious adverse events are reported for 1 of 51 participants in the OPC-12759 group and 0 of 51 participants in the sodium hyaluronate group. The registry also lists meningitis with a value of 2.0% among the serious-event terms. The record does not specify in the posted summary how that term relates to the group-level counts, so the term should not be assigned to a particular group based on this information alone.

Other adverse events are reported for 16 of 51 participants receiving OPC-12759 and 12 of 51 receiving sodium hyaluronate. Those counts are distinct from the serious-event counts. The registry does not provide further detail in this summary about event severity categories, timing, recurrence, or whether each event was considered related to treatment. Death counts are not reported in the registry record.

These are descriptive counts from a study with 51 participants in each group. A small number of events can provide limited information about how frequently uncommon harms occur, and a difference in counts alone does not establish a causal relationship. Safety findings should be read as the events recorded in the registry rather than as a complete characterization of risk.

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

The posted primary results show mean staining-score changes of -3.4 with OPC-12759 and -3.1 with sodium hyaluronate. The estimated difference is 0.3, with a recalculated interval extending from -0.691 to 1.291. This describes a small observed difference in group means alongside uncertainty that includes differences in either direction and no difference. The result does not precisely establish which group had the lower average change.

The findings are specific to the enrolled patients, interventions, endpoint, and analysis described in the registry. They concern average change in a staining score, not every participant’s experience, and they do not establish whether a change is important to patients. The registry does not report a threshold for clinically important change, individual-level results, or detailed baseline scores. Both group means moved in the direction of a lower score, but that within-group observation is not itself the randomized treatment comparison.

The sponsor-reported two-sided p-value is 0.576, while the unadjusted recalculated p-value is 0.553. These values should not be read as the probability that either intervention is effective or ineffective. The reported interval and the modest observed contrast provide more direct information about the range and precision of the estimated difference, although the recalculated interval is not a sponsor-reported result.

The trial is described as exploratory, and its original sample-size justification, analysis population, and detailed missing-data information are not reported in the registry record. Therefore, the results are not a definitive answer about all patients or settings. They also do not establish comparative safety from the limited event counts. The appropriate statistical reading is that the registry posts group-level changes and an uncertain estimated contrast, with important design and analysis details unavailable in the record.

17. Why This Trial Matters Statistically

This trial illustrates why a comparison of mean changes needs both an effect estimate and an account of uncertainty. The group means alone could invite attention to the fact that both changed, while the between-group mean difference directly addresses the randomized comparison. Its confidence interval makes clear that the observed estimate does not pin down a single precise comparative effect.

It also demonstrates that a p-value is not a substitute for effect size or precision. The registry reports the sponsor’s two-sided t-test p-value, while the summary-statistic recalculation adds an estimated difference and interval. Keeping the sponsor analysis distinct from the unadjusted recalculation helps readers understand which result was reported and what can be reconstructed from group summaries without overstating either.

The endpoint itself has a defined score range and direction, which helps interpretation: 0 is better. But a statistical contrast in a score does not automatically indicate the importance of that change to patients. The registry does not post a clinically important threshold, so statistical interpretation should stay focused on the score and its estimated difference rather than implying a clinical meaning the record does not establish.

Finally, the record highlights the value of transparent reporting. The endpoint description identifies a t-test and last observation carried forward, but the analysis population, original sample-size justification, and details of missingness are not reported. Those gaps limit reproducibility and make it harder to judge the sensitivity of the findings to analysis choices. The trial therefore offers a useful example of reading an effect estimate carefully while separating reported findings, recalculated context, and unanswered methodological questions.

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 primary comparison is based on 51 participants per group, and the recalculated interval is wide enough to include estimates in both directions. This limits precision for the comparative effect. The sample-size illustrations use the observed effect and do not document how the trial’s enrollment was originally chosen; the registry does not report the original sample-size rationale.

The study is identified as single-masked, but the registry does not say who was masked or describe how masking was maintained. It labels allocation randomized but does not report the randomization sequence or allocation concealment. Treatment dosing, adherence, and participant flow are also not described in the record information. These details matter for judging implementation and the credibility of a randomized comparison.

The endpoint is described as change at the end of instillation, with last observation carried forward, but missing-data counts, reasons, and sensitivity analyses are not reported in the registry record. The analysis population is not reported either. As a result, the relationship between randomized participants and those included in the reported means cannot be fully assessed. The registry also does not provide baseline group scores or individual participant outcome distributions.

The sponsor reports a p-value but not a sponsor estimate or confidence interval, and the registry does not explain the difference between that p-value and the recalculated value. The recalculation is unadjusted and relies on summary statistics, so it cannot address analysis choices that require participant-level data. Finally, the posted safety summary gives limited detail about event timing and treatment attribution, and death counts are not reported in the registry record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Exploratory Study of OPC-12759 Ophthalmic Suspension — NCT01493180.
• Registry record last updated 2015-03-10; status last verified 2015-02.
• Results first posted 2014-03-03.
• 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

NCT01493180 is a completed phase 2 randomized, single-masked, parallel-group study sponsored by Otsuka Pharmaceutical Co., Ltd. It enrolled 102 patients with keratoconjunctival epithelial disorder, with 51 assigned to OPC-12759 ophthalmic suspension and 51 to sodium hyaluronate ophthalmic solution. The registry lists the study start as 2011-12 and primary completion as 2012-11.

The primary outcome was change from baseline in keratoconjunctival staining score, for which 0 is better. Mean changes were -3.4 and -3.1, respectively. The sponsor-reported two-sided t-test p-value was 0.576; an unadjusted recalculation gives a mean difference of 0.3, a confidence interval from -0.691 to 1.291, and p = 0.553.

The registry reports serious adverse events in 1 participant in the OPC-12759 group and 0 in the sodium hyaluronate group. Other adverse events were reported in 16 and 12 participants, respectively. The original sample-size justification, analysis population, and detailed missing-data information are not reported in the registry record.