Phase 2 Dose-Response Study Evaluating the Safety and Efficacy of NCX 470 vs Latanoprost in Subjects With Open-Angle Glaucoma or Ocular Hypertension: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase 2 Dose-Response Study Evaluating the Safety and Efficacy of NCX 470 vs Latanoprost in Subjects With Open-Angle Glaucoma or Ocular Hypertension (NCT03657797), a randomized phase 2 study. The primary outcome is Change From Baseline in Study Eye Mean Diurnal IOP at the Week 4.

2. Clinical Question

3. Trial Design

This was a randomized, multicenter, masked, parallel-group Phase 2 study of NCX 470 doses versus latanoprost 0.005%. The registry record lists quadruple masking and an enrollment of 656. The study is marked completed. Its recorded start date is 2018-08-01 and its primary completion date is 2019-08-23.

Four groups received once-daily ophthalmic solution for 4 weeks: NCX 470 0.021%, NCX 470 0.042%, NCX 470 0.065%, or latanoprost 0.005%. Participants used medication in both eyes. At baseline, one eye was designated the study eye: the eye with the highest mean diurnal intraocular pressure, with the right eye selected if both eyes had the same value.

The primary outcome was change from baseline in study-eye mean diurnal intraocular pressure at Week 4. The primary endpoint population is defined as randomized participants with diurnal pressure values at Week 4. This is not necessarily the same as every randomized participant; the registry record does not report a complete accounting of missing Week 4 endpoint values by group in the registry record for this analysis.

The listed secondary outcomes were change from baseline in study-eye mean diurnal pressure at Week 1, Week 2, and the exit visit, and the percentage of subjects with treatment-emergent ocular adverse events. The registry record identifies three outcome measures and six analyses. Further information about allocation procedures, masking implementation, protocol deviations, and the trial’s original sample-size justification is 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.

Before assignment: Potential participants were screened for eligibility and those on intraocular pressure (IOP) lowering medication were required to undergo a wash out of 5 to 42 days (dependent on the class of medication). After the appropriate washout period, or after a minimum of 5 days for participants who were not on IOP lowering medication at screening, participants underwent 2 eligibility visits held 3 to 7 days apart. Those who qualified were randomized to one of the 4 treatment groups.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All participants who were randomized and who had diurnal IOP values at week 4

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: Participants contribute a study eye (fellow eye also treated)

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Study Eye Mean Diurnal IOP at the Week 4, measured over Baseline, Week 4.

The primary endpoint measures the average change in pressure in the designated study eye between baseline and Week 4. Mean diurnal intraocular pressure is the average of measurements taken at 8AM, 10AM, and 4PM at Week 4. The result is expressed in mmHg, and the group summaries are means with standard deviations.

A change score places the focus on how each group’s pressure values shifted from its own baseline, rather than only comparing endpoint values. Negative mean changes in the posted results indicate reductions from baseline. A between-group mean difference then describes the contrast between those average changes; its direction depends on the order of the groups in the comparison.

Using a set of measurements across the day can describe pressure at more than a single measurement time. The study-eye definition also establishes which eye contributes to the primary endpoint when medication is used in both eyes. These endpoint choices define the quantity being compared, but they do not by themselves establish that any observed contrast is clinically important. The registry record does not report a threshold for clinical importance or a rationale for selecting this endpoint.

7. Results Posted in the Registry

• NCX 470 0.042%: mean -8.28 versus -7.79 on control; difference -0.49 (95% CI -1.17 to 0.19), standardized effect d = -0.19; z-test p = 0.160.
• NCX 470 0.065%: mean -8.70 versus -7.79 on control; difference -0.91 (95% CI -1.69 to -0.13), standardized effect d = -0.32; z-test p = 0.022.
• Latanoprost 0.005%: mean -7.41 versus -7.79 on control; difference 0.38 (95% CI -0.32 to 1.08), standardized effect d = 0.14; z-test p = 0.288.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a continuous measure, and the registry record reports each group’s mean change and standard deviation. A mean describes the center of the observed group values. A standard deviation describes how dispersed individual values are around that mean; it is not the same quantity as the uncertainty in the estimated group mean or in a between-group contrast.

The recalculated comparisons are unadjusted comparisons based on the posted means, standard deviations, and group sizes. A mean difference expresses the contrast between two group means on the original mmHg scale. For example, a negative estimate in a comparison of an NCX 470 dose with 0.021% represents a more negative mean change in the dose group, given that order. A positive estimate for latanoprost versus 0.021% reflects the comparison in the stated order. These recalculations do not adjust for baseline values or other covariates.

A confidence interval gives a range of values compatible with the estimate under the model and assumptions used. If an interval includes zero, it is compatible with no difference as well as with values on either side of zero. If it excludes zero, the interval supports a contrast in one direction under that analysis. Interval width also matters: a wide interval leaves greater uncertainty about the size of the contrast. These interpretations do not establish clinical importance, and they do not substitute for the sponsor’s prespecified analysis.

The sponsor reports ANCOVA analyses of mean differences in final values for each NCX 470 dose compared with latanoprost. ANCOVA, or analysis of covariance, estimates group contrasts while accounting for covariates included in the model. Such adjustment can yield estimates different from simple contrasts of posted group summaries. The registry record does not report the full ANCOVA specification, including the covariates used, so the sponsor’s results cannot be independently reconstructed from the available details.

The sponsor analyses are labeled non-inferiority. A non-inferiority analysis evaluates whether an intervention is not worse than a comparator by more than a defined margin. The margin and its justification are not reported in the registry record. Therefore, the reported estimates and p-values can be described, but the non-inferiority criterion cannot be fully evaluated from these details alone. The record’s discrepancy flag also matters: where sponsor results differ from unadjusted recalculations, the sponsor’s own analysis takes precedence.

There are multiple dose-versus-comparator contrasts and more than one outcome measure. Testing multiple hypotheses can increase the chance of at least one apparently noteworthy result if no adjustment or prespecified hierarchy is used. The record does not report whether multiplicity adjustment, a hierarchical testing procedure, or another method was used. The p-values from simple recalculations are explicitly unadjusted. The sponsor’s multiplicity approach is not reported in the registry record, so the contrasts should not be treated as independent confirmatory findings without further information.

The primary analysis population is described as participants who were randomized and had Week 4 diurnal pressure values. This definition identifies who contributed to the reported primary endpoint but does not explain how missing measurements were handled, whether any values were imputed, or whether other analysis populations were used. Those details are not reported in the registry record. Missing-data methods matter because conclusions may depend on which participants contribute measurements and on assumptions about absent values.

General statistical context: variability among participants can make a group contrast less precise even when group means differ. Standard deviations describe that variability, while confidence intervals describe uncertainty in the estimated contrast. In settings where variability is not well represented by the model, the interval and p-value may also be affected. The posted primary outcome is a continuous measure reported with standard deviations; no additional distributional diagnostics are reported in the registry record.

General statistical context: power is the probability that a specified analysis will meet its statistical criterion under an assumed effect and model. A sample-size calculation uses assumptions about the effect to detect, variability, significance level, allocation, and analysis. It is not a measure of whether a completed trial’s observed result is true or important. The planning illustrations associated with this study use the observed effect as a starting point, with specified fractions of that effect; they are not the original design justification.

10. Confidence Intervals and Precision

The sponsor’s ANCOVA estimate for NCX 470 0.021% versus latanoprost is -0.40, with a 95% confidence interval from -1.11 to 0.31. The interval includes zero, and the reported p-value is 0.267. For 0.042% versus latanoprost, the estimate is -0.81, with a 95% confidence interval from -1.52 to -0.09 and p = 0.028. For 0.065% versus latanoprost, the estimate is -1.23, with a 95% confidence interval from -1.96 to -0.51 and p < 0.001.

The intervals communicate uncertainty around each sponsor estimate. The 0.021% interval includes both negative and positive values, so the reported data are compatible with a range of contrasts on either side of zero under that analysis. The intervals for 0.042% and 0.065% are below zero, indicating estimates in the negative direction within their reported ranges. These interval positions do not establish that a difference is clinically important, and the record does not report the non-inferiority margin needed to evaluate that criterion fully.

The simple unadjusted recalculations have their own confidence intervals and should be distinguished from these sponsor intervals. Because the registry record flags a discrepancy, the sponsor’s analyses take precedence. The registry entries do not report confidence intervals for the individual group means or additional precision measures for the endpoint. The reported estimates are therefore best read as analysis-specific summaries, with uncertainty interpreted in light of the model and the information not reported in the registry record.

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 the observed effect, not the trial’s original design assumptions. For the continuous-outcome comparisons, figures are participants per arm. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is the stated fraction of the observed effect. Time-to-event displays would instead give total events under the stated Schoenfeld approach, but the comparisons for this continuous endpoint are expressed as participants per arm.

The 80% power column gives the planned number of participants per arm under which the calculation would have an 80% probability of meeting its statistical criterion if the assumed effect and other planning assumptions held. The 90% power column represents a higher target probability and therefore generally calls for more participants per arm. These are planning quantities, not counts observed in the study.

For example, the displayed planning values vary by comparison and by the fraction of the observed effect used. Planning for 75% of the observed effect requires more participants per arm than planning for 100% of the observed effect, and planning for 50% requires more still. This reflects a general principle: smaller effects are harder to distinguish from random variation, so detecting them at the same power generally requires more participants. The figures are illustrative rather than a reconstruction of the trial’s original enrollment rationale. This is not observed power.

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 = -0.490; pooled standard deviation σ = 2.560.
2. n = 2 × (1.96 + 0.842)² × 2.560² / -0.490² = 428.5, rounded up to 429 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 <- -0.489999999999999                      # observed difference in means
sigma <- 2.56005548282012                     # 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(3657797)
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 = 429 (80% power), 574 (90% power)
 75% of observed effect: n per arm = 762 (80% power), 1,020 (90% power)
 50% of observed effect: n per arm = 1,714 (80% power), 2,295 (90% power)
power.t.test: 429.5 
Simulated power at n = 429 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. Analyses Reported by the Sponsor

The sponsor’s reported analyses use ANCOVA and present mean differences in final values for each NCX 470 dose compared with latanoprost. The reported estimates are -0.40 for 0.021%, -0.81 for 0.042%, and -1.23 for 0.065%; each is accompanied by a 95% confidence interval and p-value. They are labeled non-inferiority analyses.

The recalculated contrasts use posted group summary statistics and are simple, unadjusted comparisons. They are not a reproduction of ANCOVA and can differ because the analytical methods and endpoint forms differ. The sponsor’s reported parameter is described as a mean difference in final values, whereas the main endpoint is change from baseline. The registry record does not give enough model detail to reconcile the estimates fully.

The discrepancy flag is true, so the sponsor’s own analysis takes precedence. The recalculations are useful as a transparent check on the posted summaries, but they should not replace the sponsor’s reported findings. The non-inferiority margin and full analysis specification are not reported in the registry record, limiting independent assessment of the sponsor’s inferential claim.

14. Multiplicity, Missing Data and Analysis Populations

The study reports a primary endpoint, two listed secondary outcomes, and six analyses. The reported sponsor comparisons include each of the three NCX 470 concentrations versus latanoprost. When several doses, outcomes, or analysis times are evaluated, the overall chance of a false-positive finding can depend on how the analyses are organized and whether statistical adjustment is used.

The simple recalculated p-values are unadjusted. The registry record does not report whether the sponsor adjusted for multiple comparisons, used a testing hierarchy, designated a primary comparison among the doses, or controlled a family-wise error rate through another approach. The sponsor’s p-values should therefore be interpreted as reported, without assuming a multiplicity procedure that is not described. The registry record also does not give enough information to determine whether all planned analyses are posted.

The primary endpoint population includes randomized participants who had Week 4 diurnal pressure values. The registry entries do not state how many randomized participants lacked those measurements by group, the reasons for missingness, or how missing endpoint values were handled. They also do not report whether imputation or sensitivity analyses were used. These details are relevant because conclusions can depend on which participants contribute to the analysis and on assumptions about missing values.

Although the registry describes the primary analysis population, it does not report further analysis populations such as a per-protocol population or a safety population definition. The full statistical analysis plan, covariate specification, non-inferiority margin, and missing-data strategy are not reported in the registry record. Without those details, the reported results can be described but some design and analysis choices cannot be independently assessed.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: b-cell lymphoma (0.9%), endophthalmitis (0.9%), hypertension (0.9%), transient ischaemic attack (0.9%).

The registry record reports serious adverse events affecting 1 participant in the NCX 470 0.021% group, 2 in the 0.042% group, 1 in the 0.065% group, and 0 in the latanoprost 0.005% group. Deaths are recorded as 0 in every group. These are the reported serious-event counts; the figures do not establish whether an event was caused by treatment.

The serious-event terms listed are B-cell lymphoma, endophthalmitis, hypertension, and transient ischaemic attack, each reported at 0.9%. The registry entries do not provide further detail about the timing, outcomes, or treatment relationship of these terms. They should not be interpreted as evidence of a causal association with any study group.

The record also reports other adverse events affecting 24 participants in the 0.021% group, 42 in the 0.042% group, 39 in the 0.065% group, and 20 in the latanoprost group. Those figures are distinct from the serious-event counts. The registry does not provide enough information to characterize severity beyond the serious-event category, duration, or specific ocular-event types. Safety counts are descriptive and should be read alongside the study’s group sizes at risk and the reporting definitions.

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

The posted results describe short-term changes in study-eye mean diurnal intraocular pressure during 4 weeks of once-daily treatment. All four group means show a reduction from baseline. The sponsor’s ANCOVA analyses report dose-versus-latanoprost contrasts with estimates and confidence intervals, and label the analyses as non-inferiority.

These findings do not by themselves establish whether any contrast is clinically meaningful. The registry record does not report a threshold for clinical importance or the non-inferiority margin. The sponsor’s reported analyses have precedence over simple unadjusted recalculations because the record flags a discrepancy, but the available information does not include enough detail to reconstruct the sponsor model or assess every analysis choice.

The results also do not establish how pressure changes would continue beyond the treatment period, whether they would persist, or how they relate to outcomes other than the measured pressure endpoint. The secondary outcomes include earlier pressure assessments and treatment-emergent ocular adverse events, but the registry entries summarized for interpretation do not include their detailed outcome results.

Safety counts describe events recorded during the trial and should not be treated as proof that treatment caused or prevented an event. The serious-event numbers are small descriptive counts, and the registry entries do not establish relatedness. The trial’s findings are limited to its enrolled population, measured endpoint, analysis population, and reported period. Broader conclusions require information beyond the posted results.

17. Why This Trial Matters Statistically

This study offers a useful statistical example of a multi-dose randomized comparison with a continuous endpoint. The primary outcome is expressed as change in mmHg, making the group means and between-group estimates interpretable on the endpoint’s original scale. The study-eye rule and averaging of measurements at multiple times define the target measurement clearly.

It also illustrates why group summaries and adjusted analyses should not be treated as interchangeable. Unadjusted recalculations from means and standard deviations offer one transparent view of the posted results, while the sponsor’s ANCOVA analyses use a different method and report final-value contrasts. The discrepancy flag makes that distinction especially important: the sponsor’s own analysis takes precedence, and recalculation should not be presented as a replacement.

Finally, the registry record demonstrates the importance of analysis details. Interpretation of the non-inferiority results depends on the margin and model specification, and assessment of the set of dose comparisons depends on the multiplicity approach. Missing-data handling also matters for an endpoint population defined by availability of Week 4 values. Because these details are not reported in the registry record, readers can understand the estimates and their stated uncertainty while recognizing limits on independent evaluation.

18. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record does not report the non-inferiority margin or its justification, although the sponsor analyses are labeled non-inferiority. It also does not provide the full ANCOVA specification, including covariates, or enough information to reproduce the sponsor’s estimates from the posted summaries. The record flags a discrepancy between the sponsor analysis and simple recalculations, requiring the sponsor’s own analysis to take precedence.

The multiplicity strategy is not reported, despite multiple dose comparisons and outcome measures. The registry record also does not describe how missing Week 4 pressure measurements were handled, give reasons for missing values, or report sensitivity analyses. The stated primary analysis population includes randomized participants with Week 4 values, so the handling of those without such values is relevant.

The endpoint covers a 4-week treatment period and a defined pressure measure. The registry entries do not establish persistence beyond that period or clinical importance thresholds. Detailed results for the listed secondary outcomes are not included in the outcome details used for these comparisons. The serious adverse-event counts and terms are descriptive, with no treatment relationship information.

The original sample-size justification and protocol assumptions are not reported in the registry record. The displayed planning illustrations use observed effects and stated planning assumptions, and cannot be treated as evidence of the original trial rationale. These limits constrain independent assessment of design choices, analysis robustness, and the broader meaning of the results.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Phase 2 Dose-Response Study Evaluating the Safety and Efficacy of NCX 470 vs Latanoprost in Subjects With Open-Angle Glaucoma or Ocular Hypertension — NCT03657797.
• Registry record last updated 2023-06-18; status last verified 2023-06.
• Results first posted 2022-12-08.
• 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

NCT03657797 is a completed, randomized, multicenter, masked, parallel-group Phase 2 study of NCX 470 versus latanoprost 0.005% in subjects with open-angle glaucoma or ocular hypertension. It enrolled 656 participants and compared three NCX 470 concentrations with once-daily latanoprost over 4 weeks.

The primary endpoint was change from baseline in study-eye mean diurnal intraocular pressure at Week 4. The sponsor reported ANCOVA non-inferiority analyses for each NCX 470 dose versus latanoprost. Simple unadjusted recalculations differ, and the sponsor’s analysis takes precedence. The non-inferiority margin, multiplicity approach, full model details, and missing-data methods are not reported in the registry record.