A Phase III Study Assessing the Safety and Efficacy of DE-117 Ophthalmic Solution Compared With Latanoprost Ophthalmic Solution in Subjects With OAG or OHT: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase III Study Assessing the Safety and Efficacy of DE-117 Ophthalmic Solution Compared With Latanoprost Ophthalmic Solution in Subjects With OAG or OHT (NCT02981446), a randomized phase 3 study. The primary outcome is Mean Diurnal IOP at Month 3.

2. Clinical Question

3. Trial Design

The registry describes a Phase III, randomized, observer-masked, active-controlled, parallel-group study. It was multinational and multicenter. The two arms received DE-117 ophthalmic solution or latanoprost ophthalmic solution 0.005%. The registry lists enrollment as 370 and status as completed. The sponsor was Santen Pharmaceutical Co., Ltd.

The study began in 2017-01, with primary completion in 2019-01. The registry classifies masking as single. It does not give further detail about which study personnel were masked or how masking was maintained. The title identifies the study as the PEONY Study.

The primary endpoint was continuous: mean diurnal intraocular pressure at Month 3, in mmHg. The posted arm values are least-squares means with standard errors. The reported sample counts are 184 for DE-117 and 185 for latanoprost. One subject in the DE-117 arm was excluded from the full analysis set because there was no post-baseline intraocular pressure measurement. The registry does not describe the full analysis set definition beyond that note.

The registry also lists secondary outcomes concerning intraocular pressure at nine timepoints and mean diurnal intraocular pressure at Week 1. Two secondary endpoint summaries are posted, including the Week 1 result. There are no reported analyses in the registry record. Details about the randomization sequence, allocation concealment, visit schedule beyond the endpoint timeframes, treatment adherence, and the trial's original sample-size justification are not reported in the registry 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

Analysis population for the primary outcome, as defined in the registry: One subject in the DE-117 arm was excluded from the FAS due to no data for at least 1 post-baseline IOP measurement.

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.

Population: All 370 subjects received at least 1 dose of their assigned study medication and were included in the Safety population. Data for all subjects except 1 were included in the Full Analysis Set (FAS); 1 subject in the DE-117 arm was excluded from the FAS due to no data for at least 1 post-baseline IOP measurement.

6. Primary Endpoint

The registry lists the primary outcome as Mean Diurnal IOP at Month 3, measured over Month 3 (average of IOP at 3 time points: 09:00, 13:00, 17:00).

The primary endpoint is mean diurnal intraocular pressure at Month 3, measured in mmHg. It averages the study-eye readings at 09:00, 13:00, and 17:00. Intraocular pressure was measured with a calibrated Goldmann applanation tonometer. The registry says the same investigator and authorized study staff were preferred for each subject during the study.

A mean across several daily measurement times represents the endpoint across that specified measurement window rather than a single reading. In general, using a defined set of timepoints can help characterize a measure that may vary during the day. The posted endpoint is therefore a continuous measurement, and the arm summaries express the estimated mean pressure at Month 3.

The posted least-squares mean was 17.5 mmHg for DE-117 and 16.8 mmHg for latanoprost, each reported with a standard error of 0.25. A least-squares mean is a model-based estimate; however, the registry record does not describe the model or its covariates. The reason for choosing this endpoint and the clinical threshold for an important difference are not reported in the registry record. Consequently, the endpoint can be interpreted as a quantitative comparison of pressure measurements, while its clinical importance requires context not stated in the record.

7. Results Posted in the Registry

• Latanoprost Ophthalmic Solution 0.005%: mean 16.80 versus 17.50 on control; difference -0.70 (95% CI -1.39 to -0.01), standardized effect d = -0.21; z-test p = 0.048.

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 Diurnal IOP at Week 1 (Second Key Secondary Endpoint)

• 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:
  • Latanoprost Ophthalmic Solution 0.005%: mean 18.50 versus 18.50 on control; difference 0.00 (95% CI -0.73 to 0.73), standardized effect d = 0.00; z-test p = 1.000.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Latanoprost Ophthalmic Solution 0.005% and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcomes cover pressure at nine timepoints and mean diurnal pressure at Week 1. The registry lists two secondary endpoint summaries, including a comparison for mean diurnal pressure at Week 1. The broad pattern across the posted summaries is that the Week 1 arm estimates are the same, while the primary Month 3 estimates differ. This describes the posted endpoints without implying that the measurements across timepoints follow a particular trajectory.

The Week 1 result offers an earlier-timepoint complement to the Month 3 primary result, but it does not replace the primary endpoint. The endpoint timing and summary differ, and the Week 1 estimate has its own uncertainty. The registry record does not provide a complete analysis of all nine timepoints, despite listing the timepoint outcome, and does not report analyses for the posted secondary outcomes.

Secondary results generally carry less confirmatory weight when the testing strategy and multiplicity handling are unclear. Testing several outcomes or timepoints can increase the chance that at least one result appears noteworthy by chance. In general, prespecified hierarchies or other multiplicity procedures can help manage that issue, but whether such a strategy was used is not reported in the registry record. The Week 1 comparison is unadjusted, and the record does not state how it relates to any formal testing sequence.

Accordingly, the secondary endpoint summaries provide descriptive context for the primary comparison, rather than independent confirmation of it. They should be interpreted alongside the primary estimate, their own uncertainty, and the absence of reported multiplicity methods. No unreported secondary results or overall pattern across individual timepoints can be inferred from the summaries that are posted.

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

• IOP at 9 Timepoints (First Key Secondary Efficacy Endpoint)
• Mean Diurnal IOP at Week 1 (Second Key Secondary Endpoint)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous, meaning it is measured on a scale where differences between values can be expressed in the measurement unit. Here the unit is mmHg. For a continuous outcome, an arm mean gives a concise description of the central value, while a measure of uncertainty helps show how precisely that mean has been estimated. The registry reports least-squares means and standard errors for the Month 3 endpoint.

What does a least-squares mean represent?

In general, a least-squares mean is an estimate from a statistical model rather than necessarily the raw arithmetic average of all observed values. Depending on the model, it may account for selected factors or repeated measurements. The registry record does not describe the model used for the posted least-squares means, its covariates, or whether additional factors were included. It is therefore appropriate to report the posted arm estimates as given, without attributing a particular adjustment to them.

The standard error measures the estimated sampling variability of an estimate. In general, a smaller standard error indicates a more precisely estimated mean, though interpretation also depends on the analysis and data. The standard errors for the Month 3 means are 0.25 in both arms. They are not standard deviations: a standard deviation describes variation among observations, whereas a standard error describes uncertainty in an estimated quantity.

How should the arm comparison be read?

The comparison is reported as a mean difference. The posted estimate is -0.7 mmHg for latanoprost compared with DE-117. Its negative sign indicates that the estimated value for the comparator is lower under the reported contrast. The interval from -1.393 to -0.007 describes uncertainty around that difference as reported in the registry. A difference estimate gives the magnitude and direction of the contrast in the original unit, while the interval indicates a range of values compatible with the estimate under the method used.

A p-value evaluates how unusual the observed data, or data more extreme, would be under a specified null model. It does not give the probability that the null hypothesis is true, and it is not a measure of clinical importance. The posted primary p-value is unadjusted. The registry does not provide the complete test specification or say whether other analyses were accounted for in the primary test.

What can an interval tell readers?

In general, an interval around an estimate conveys precision: a narrower interval indicates less sampling uncertainty under the analysis assumptions, and a wider interval indicates more. An interval that includes zero is compatible with no difference under its stated construction; an interval that does not include zero indicates the estimate's range is on one side of zero. The interval's confidence level and construction are not reported in the registry record, so no particular coverage interpretation should be assigned. The posted Month 3 interval reaches close to zero at its upper limit, which makes the exact size of the difference uncertain even though the reported values are negative throughout the interval.

What does the recalculation add?

The posted comparison includes a simple unadjusted recalculation based on summary statistics, alongside the registry's arm-level estimates. Such a recalculation can be useful as a descriptive cross-check, but it does not reproduce an adjusted model unless the needed model details are available. In this record, the discrepancy flag is false, so no conflict between the posted recalculation and sponsor analysis is indicated. The registry record does not report a sponsor analysis specification or adjusted effect estimate, and the recalculated comparison should not be treated as a substitute for unreported model details.

The standard deviation used for the recalculated comparison is approximate. This matters because the arm summaries are reported as standard errors rather than standard deviations. In general, converting or deriving a standard deviation from summaries depends on the information and assumptions available. An approximate pooled spread is useful for a standardized description, but it is not a directly reported individual-level variability estimate. The raw participant-level data and the exact sponsor analysis are not reported in the registry record.

Why do distributional assumptions matter?

Many common methods for comparing means rely on assumptions about the distribution of estimates or residuals. In general, a normal approximation can be useful for a mean comparison, especially when the analysis is designed around a continuous endpoint, but it does not remove the need to examine whether the model fits the data. The record does not report diagnostic checks, the distribution of individual pressure measurements, or the exact inferential procedure for the primary result. No conclusion about those assumptions can be drawn from the summary estimates alone.

Repeated measurement can also affect analysis. The Month 3 endpoint averages readings taken at three specified times. The record gives this endpoint definition but does not state whether the model analyzed each reading separately, the average only, or a larger repeated-measures structure. That distinction can affect the estimated uncertainty and model interpretation, so the reported endpoint and estimates should be treated as the available summary rather than a full account of the analysis.

11. Confidence Intervals and Precision

The primary estimate is a mean difference of -0.7 mmHg for latanoprost compared with DE-117. Its reported interval ranges from -1.393 to -0.007. The interval is relatively close to zero at its upper end, so the posted result allows for a small difference as well as larger negative values within that range. The point estimate alone would not convey this uncertainty.

In general, an interval is a way to express the precision of an estimate under the method used to calculate it. A narrow interval indicates that the estimate is more tightly bounded; a wide interval leaves more possible values compatible with the data and assumptions. An interval is not a guarantee that the true value lies within it, and its interpretation depends on the construction method and confidence level. The confidence level and interval method are not reported in the registry record.

The arm-level standard errors for the Month 3 least-squares means are 0.25 for DE-117 and 0.25 for latanoprost. They describe uncertainty around the estimated means, not the spread of individual measurements. The reported comparison interval is more directly informative about the uncertainty in the contrast between arms.

The registry also reports a Week 1 interval from -0.735 to 0.735 around a mean difference of 0.0 mmHg. It spans negative and positive values, so the summary is compatible with a range of directions for the difference. Both endpoint intervals should be interpreted with the limitation that analysis details and multiplicity handling are not reported in the registry record. Precision in the statistical sense does not by itself establish clinical importance.

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 planning illustrations are expressed as participants per arm, not total events. They use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set as a stated fraction of the observed effect. These are planning assumptions for the illustration; the trial's own original 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 be needed to have an 80% chance of detecting the specified effect under those assumptions, if the effect and assumptions were correct. The 90% power column gives the corresponding planning estimate for a higher chance of detection. Higher planned power requires more participants for a fixed effect and assumptions.

The illustration uses planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. For 100% of the observed effect, the estimates are 370 participants per arm for 80% power and 495 participants per arm for 90% power. For 75% of the observed effect, they are 657 and 880 participants per arm, respectively. For 50% of the observed effect, they are 1,478 and 1,979 participants per arm, respectively. Smaller effects need more participants because they are harder to distinguish from random variation under the same assumptions.

These figures are prospective planning illustrations based on the observed effect, not observed power. They do not tell whether the completed trial had adequate power under its original design, because the original design assumptions and sample-size justification are not reported in the registry record.

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 = -0.700; pooled standard deviation σ = 3.396.
2. n = 2 × (1.96 + 0.842)² × 3.396² / -0.700² = 369.4, rounded up to 370 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 <- -0.699999999999999                      # observed difference in means
sigma <- 3.39578196441759                     # 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(2981446)
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 = 370 (80% power), 495 (90% power)
 75% of observed effect: n per arm = 657 (80% power), 880 (90% power)
 50% of observed effect: n per arm = 1,478 (80% power), 1,979 (90% power)
power.t.test: 370.4 
Simulated power at n = 370 per arm: 0.796 (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 record does not report a multiplicity adjustment, a testing hierarchy, or a prespecified order for evaluating the primary and secondary outcomes. This matters because the study lists a primary endpoint as well as secondary outcomes at multiple timepoints. In general, examining multiple endpoints can increase the chance of finding an apparently unusual result even when no true difference exists. Statistical procedures can control that risk, but the method used in this study is not reported in the registry record.

The primary comparison is labelled unadjusted. Its p-value should therefore be read as an unadjusted result, not as evidence that any family-wise error or false discovery rate was controlled. The record does not say whether the primary analysis was part of a broader prespecified testing strategy or whether any adjustment was made outside the posted summary.

There are no reported analyses in the registry record, despite posted summary results for the primary endpoint and two secondary outcomes. The registry also lists an outcome concerning intraocular pressure at nine timepoints, but the full set of analyses for those timepoints is not reported. This limits the ability to judge whether the posted results represent all planned comparisons.

The record notes that one subject in the DE-117 arm was excluded from the full analysis set because no post-baseline intraocular pressure measurement was available. The handling of other missing data, the analysis population definition, and any sensitivity analyses are not reported in the registry record. In general, missing outcome data can affect both the estimate and its uncertainty, especially if missingness is related to the outcome or treatment. The available note is not enough to determine whether missingness introduced bias or how robust the estimate is to alternative assumptions.

Details of the statistical analysis plan, model specification, interim analyses, and sample-size justification are not reported in the registry record. Their absence prevents a full assessment of whether the posted analyses followed the original plan, but it does not alter the descriptive values stated in the registry.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: tinnitus (0.5%), urinary retention (0.5%), intraductal proliferative breast lesion (0.5%), food poisoning (0.5%).

The posted serious adverse event counts are 2 among 185 at risk in the DE-117 arm and 2 among 185 at risk in the latanoprost arm. The registry reports zero deaths in each arm. These figures are counts over the safety population denominators stated in the record; they should not be confused with the efficacy analysis counts or interpreted as a complete description of all safety outcomes.

The record lists tinnitus, urinary retention, intraductal proliferative breast lesion, and food poisoning among the top serious adverse event terms, each reported as 0.5%. The registry excerpt does not specify the arm attribution for these terms, so they should not be assigned to one arm from this information alone.

Other adverse events are reported for 74 of 185 in the DE-117 arm and 55 of 185 in the latanoprost arm. The registry does not provide further detail about event types, severity, duration, or relationship to study treatment. Differences in these counts are descriptive and do not establish a treatment-related cause. Safety summaries require context about collection, coding, follow-up, and event definitions, which are not fully reported in the registry record.

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

The posted Month 3 estimates show a lower mean diurnal intraocular pressure for the latanoprost arm than for the DE-117 arm, with a reported mean difference of -0.7 mmHg. The interval from -1.393 to -0.007 and the unadjusted p-value of 0.048 describe the statistical uncertainty and test result for this comparison. They do not establish whether the observed difference is important to patients or clinicians.

The result applies to the endpoint definition used in the trial: pressure in the study eye at Month 3, averaged across measurements at 09:00, 13:00, and 17:00. It is not a result for every possible timepoint, every eye, or every person with open-angle glaucoma or ocular hypertension. The registry's population description and the absence of detailed eligibility criteria limit broader interpretation.

Statistical evidence and clinical meaning are distinct. An estimate can be statistically suggestive while its practical importance remains uncertain. The registry record does not state a threshold for a clinically important difference, patient-centered outcomes related to the pressure measure, or how the difference should be weighed against other outcomes. The posted result should not be treated as a recommendation about treatment.

The study also reports selected adverse event counts, but these summaries do not provide a full account of benefits and harms. The registry does not report sufficient detail about event attribution, event duration, or all safety analyses to support a comprehensive comparison of safety. The endpoint summaries and safety counts are descriptive evidence from the completed study, rather than a complete clinical decision framework.

Finally, the record does not describe the analysis model, missing-data methods, multiplicity approach, or the trial's original sample-size justification. These omissions leave uncertainty about some aspects of the statistical analysis. They do not change the reported estimates, but they constrain how confidently the results can be generalized or interpreted as confirmatory evidence.

17. Why This Trial Matters Statistically

This trial is statistically interesting because it compares two active ophthalmic treatments using a continuous pressure endpoint and reports both arm-specific estimates and a direct mean contrast. The endpoint averages readings at three specified times, so the primary result is tied to a defined diurnal summary rather than a single measurement. This makes the estimand understandable in its measurement context, while leaving the model details important for interpreting how the estimates were derived.

The Month 3 contrast is close to zero at one end of its reported interval, and the unadjusted p-value is 0.048. Together, these features illustrate why a point estimate should not be read in isolation. The direction of the estimate, its precision, the testing strategy, and clinical relevance are distinct considerations. The posted result invites attention to each rather than reliance on a threshold alone.

The record also demonstrates the difference between primary and secondary evidence. A Week 1 summary is available, and other timepoint outcomes are listed, but the complete analyses and multiplicity strategy are not reported. When several timepoints or endpoints may be examined, readers need to know how the confirmatory question was defined and how repeated testing was handled. Without that information, secondary results have a more descriptive role.

Finally, the planning illustration makes clear that sample-size requirements depend strongly on the effect size targeted. Smaller effects require more participants under the stated assumptions. Since these calculations use the observed effect rather than the trial's original design rationale, they are useful for understanding planning sensitivity but cannot reconstruct the study's design decision.

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.

Important details for evaluating the trial's statistical results are not reported in the registry record. These include the primary analysis model, covariates, handling of repeated measurements, interval construction and confidence level, and any diagnostic checks. Without them, the posted least-squares means and comparison can be described but not fully reconstructed.

The registry notes one DE-117 subject excluded from the full analysis set for having no post-baseline intraocular pressure measurement. It does not provide a broader accounting of missing data, reasons for missingness, imputation methods, or sensitivity analyses. The extent to which missingness could affect the estimates therefore cannot be assessed from the record.

The listed nine-timepoint secondary outcome is not accompanied by a complete set of posted analyses. The multiplicity strategy, if any, is not reported, and the posted secondary summaries are not described as adjusted. This makes it difficult to determine the confirmatory status of findings beyond the primary endpoint.

The clinical threshold for an important pressure difference and the trial's original sample-size assumptions are also not reported. The planning illustration is based on the observed effect and should not be taken as evidence about the original design. Details about eligibility criteria, adherence, treatment administration, and follow-up completeness are not reported in the registry record, limiting assessment of generalizability and implementation.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Phase III Study Assessing the Safety and Efficacy of DE-117 Ophthalmic Solution Compared With Latanoprost Ophthalmic Solution in Subjects With OAG or OHT — NCT02981446.
• Registry record last updated 2024-03-15; status last verified 2019-06.
• Results first posted 2024-03-15.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

The completed Phase III PEONY Study was a randomized, observer-masked, active-controlled, parallel-group study of DE-117 ophthalmic solution 0.002% and latanoprost ophthalmic solution 0.005% in people with open-angle glaucoma or ocular hypertension. The registry lists enrollment as 370 and identifies Santen Pharmaceutical Co., Ltd. as sponsor.

The primary endpoint was mean diurnal intraocular pressure at Month 3. The posted means were 17.5 mmHg for DE-117 and 16.8 mmHg for latanoprost, with a reported mean difference of -0.7 mmHg, an interval from -1.393 to -0.007, and an unadjusted p-value of 0.048. Secondary summaries and selected safety counts are also reported. Analysis details, multiplicity methods, and the original sample-size justification are not reported in the registry record.