A Phase 2, Dose Finding Study of PF-03187207 in Patients With Primary Open Angle Glaucoma or Ocular Hypertension.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase 2, Dose Finding Study of PF-03187207 in Patients With Primary Open Angle Glaucoma or Ocular Hypertension. (NCT00595101), a randomized phase 2 study. The primary outcome is Change From Baseline Mean Diurnal Intraocular Pressure (IOP) in the Study Eye on Day 28.

2. Clinical Question

3. Trial Design

This was a Phase 2, randomized, parallel-group dose-finding study. The registry records quadruple masking, while the official title describes the study as double-masked. These are both descriptions in the record; it does not explain the difference between them. The trial compared PF-03187207 with 0.005% latanoprost in patients with primary open-angle glaucoma or ocular hypertension.

The official title describes a 28-day study, and the arm descriptions specify once-daily drops in the study eye for 28 days. The PF-03187207 regimens were low, medium, and high dose, each paired with Latanoprost Vehicle. The comparator arm received latanoprost with PF-03187207 Vehicle. The registry identifies PF-03187207, Latanoprost 0.005%, and both vehicles as drug interventions.

The registry lists enrollment of 117 and a completed status. The recorded start is 2007-12 and primary completion is 2008-06. The posted primary outcome is a continuous measure: change from baseline in mean diurnal intraocular pressure on Day 28, reported in mmHg. The registry describes the Per Protocol population as the primary analysis population for efficacy, excluding participants with major protocol violations before study completion from the intention-to-treat population. More detail about the protocol’s analysis procedures 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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: The Per Protocol (PP) population consisted of all subjects in the intention to treat (ITT) population excluding those who made major protocol violations before completing the study. The PP population was considered as the primary analysis population of efficacy (primary and secondary endpoints) of the 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 From Baseline Mean Diurnal Intraocular Pressure (IOP) in the Study Eye on Day 28, measured over Baseline, 28 days.

The primary endpoint is the change from baseline in mean diurnal intraocular pressure in the study eye on Day 28. Intraocular pressure was measured with a Goldmann applanation tonometer. Diurnal pressure was calculated as the mean of measurements at 8 AM, 10 AM, 1 PM, and 4 PM for each visit. The registry describes repeated readings at each time point and how readings were summarized when the first two measurements differed.

A change score expresses the Day 28 measurement relative to baseline. The reported negative group means therefore indicate a decrease from baseline in the measured pressure. The endpoint is continuous, so the reported group summaries use means and standard deviations rather than counts of participants meeting a threshold. A mean summarizes the average change, while the standard deviation describes the spread of individual values around that mean.

Using the mean diurnal value combines measurements across the specified times rather than focusing on a single daily measurement. It provides a common continuous outcome for comparing the study groups at the stated follow-up. The registry also lists change at specific time points on Day 14 and Day 28 and target intraocular pressure as secondary outcomes, but their results are not reported in the registry record.

7. Results Posted in the Registry

• PF-0318 0.024%: mean -5.94 versus -5.12 on control; difference -0.82 (95% CI -2.32 to 0.68), standardized effect d = -0.29; z-test p = 0.284.
• PF-0318 0.04%: mean -5.02 versus -5.12 on control; difference 0.10 (95% CI -1.12 to 1.32), standardized effect d = 0.04; z-test p = 0.872.
• 0.005% Latanoprost: mean -5.25 versus -5.12 on control; difference -0.13 (95% CI -1.44 to 1.18), standardized effect d = -0.05; z-test p = 0.846.

8. Statistical Interpretation

9. Statistical Methods Explained

The outcome is continuous, and the registry reports a mean and standard deviation for each group. A group mean describes the average change from baseline. The standard deviation describes how dispersed the participants’ changes were around that average; it is not the uncertainty in the mean itself. For the posted primary outcome, the group means are negative, reflecting average decreases from baseline, while the standard deviations show meaningful variation within each group.

How should the group contrast be read?

The pairwise estimates compare each listed arm with the PF-0318 0.006% reference group. The mean difference is the comparison arm’s mean change minus the reference group’s mean change. A negative estimate means the comparison arm’s mean change is more negative; a positive estimate means it is less negative. For example, the estimate for PF-0318 0.024% is -0.82 mmHg. This is a contrast between group averages, not the change observed by an individual participant.

The estimate for PF-0318 0.04% is 0.1 mmHg, and the estimate for 0.005% latanoprost is -0.13 mmHg. These estimates are near zero, but the associated confidence intervals are needed to understand their precision. The magnitude of a point estimate should not be interpreted apart from the range of values compatible with the data.

What does a confidence interval communicate?

A confidence interval gives a range of estimates consistent with the observed data under the statistical method used. It reflects sampling uncertainty: repeated studies could produce different estimates because different participants are observed. An interval that includes zero does not establish that the groups are the same. Rather, it indicates that the data do not rule out a zero difference at the confidence level used, while also allowing nonzero differences within the interval.

Each posted interval crosses zero. The interval for PF-0318 0.024% ranges from -2.32 to 0.680; that for PF-0318 0.04% ranges from -1.116 to 1.316; and that for 0.005% latanoprost ranges from -1.438 to 1.178. The displayed unadjusted p-values are 0.284, 0.872, and 0.846, respectively. These values summarize evidence against a zero contrast under the unadjusted comparisons, but they do not measure the probability that a treatment works or that a particular hypothesis is true.

Why does variation matter?

More variability within groups generally makes a difference between their means harder to estimate precisely, particularly when group counts are limited. The pooled standard deviations used in the recalculations are 2.828 for the PF-0318 0.024% comparison, 2.31 for PF-0318 0.04%, and 2.54 for 0.005% latanoprost. The calculations use the posted means, standard deviations, and group counts. They are unadjusted comparisons and do not account for covariates or any possible features of the original analysis that are not described in the registry record.

Over-dispersion is a concern in settings where observed variation exceeds what a statistical model assumes. In general, a model that understates variation can make uncertainty appear smaller than it is. These outcomes are reported as continuous measurements with standard deviations; the registry does not report a separate over-dispersion assessment or a model-based adjustment for it. No claim about over-dispersion is needed to interpret the posted group summaries.

What is the role of the p-value?

A p-value assesses how unusual data at least as incompatible with a specified null value would be if that null model were true and the analysis assumptions held. Here the null value for a mean difference is zero. The unadjusted p-values are not measures of effect size or clinical importance. They also do not resolve the uncertainty evident in the confidence intervals. As a general statistical principle, interpretation should consider the estimated contrast and its interval, not a p-value alone.

The registry lists the Per Protocol population as the primary efficacy analysis population. It describes that population as participants from the intention-to-treat population after excluding those with major protocol violations before completing the study. The precise handling of missing measurements, the exact analysis model, confidence level, and any adjustments are not reported in the registry record. Accordingly, the recalculated comparisons should be treated as simple unadjusted contrasts based on posted summary statistics, not as a reconstruction of every step in the sponsor’s analysis.

10. Confidence Intervals and Precision

Precision concerns how narrowly the data estimate a group contrast. The confidence intervals for all three posted comparisons include zero: PF-0318 0.024% versus the reference has an interval from -2.32 to 0.680; PF-0318 0.04% has an interval from -1.116 to 1.316; and 0.005% latanoprost has an interval from -1.438 to 1.178. Each interval permits a range of contrasts in either direction, rather than pinning down a single value.

The point estimates are -0.82 mmHg, 0.1 mmHg, and -0.13 mmHg, respectively. These estimates are small in relation to the ranges covered by their confidence intervals. The posted data therefore do not estimate the pairwise differences with high precision. In particular, an interval crossing zero is not proof of equivalence or of no effect; it means the data are compatible with zero and with other values inside the interval.

Confidence intervals capture statistical uncertainty under the calculation method, not every source of uncertainty. The registry record does not report the confidence level or all details of the sponsor’s analysis, and the recalculated contrasts are unadjusted. Interpretation should therefore keep both the range of estimates and the limitations of the calculation in view. The intervals describe uncertainty around these group contrasts, not uncertainty about an individual participant’s outcome.

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 planning figures are participants per arm for each two-arm comparison, not total events. They are illustrations based on the observed effect, not the trial’s original sample-size design. The stated 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 set to the stated fraction of the observed effect. The trial’s own original sample-size assumptions are not reported in the registry record.

The 80% power column gives the planned number of participants per arm for a design with an 80% chance of detecting the specified effect under those assumptions, if that effect is truly present. The 90% power column gives the corresponding number per arm for a higher chance of detection. These are planning targets, not guarantees that a future study will achieve a particular result.

Smaller effects are harder to distinguish from random variation and therefore require more participants under the same assumptions. For example, in the PF-0318 0.024% comparison, the listed 80% power illustration rises from 187 participants per arm for 100% of the observed effect to 747 participants per arm for 50% of that effect. These figures are not observed power. Observed power is not a useful substitute for interpreting the observed estimate and its confidence interval; the planning figures instead show how sample-size requirements change when the assumed effect changes.

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 = -0.820; pooled standard deviation σ = 2.828.
2. n = 2 × (1.96 + 0.842)² × 2.828² / -0.820² = 186.7, rounded up to 187 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 <- -0.82                      # observed difference in means
sigma <- 2.82782034870115                     # 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(595101)
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 = 187 (80% power), 250 (90% power)
 75% of observed effect: n per arm = 332 (80% power), 445 (90% power)
 50% of observed effect: n per arm = 747 (80% power), 1,000 (90% power)
power.t.test: 187.7 
Simulated power at n = 187 per arm: 0.798 (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

The registry lists two secondary outcomes: change from baseline in mean intraocular pressure at specific time points on Day 14 and Day 28, and target intraocular pressure. Results for those secondary outcomes are not reported in the registry record. The registry lists three outcome measures and no analysis entries. The posted pairwise comparisons are unadjusted, and the record does not report whether or how the trial’s original analyses accounted for multiple comparisons.

Multiplicity arises when several outcomes, time points, dose groups, or contrasts are tested. In general, running more tests increases the chance that at least one result appears unusual by chance. Methods for addressing multiplicity can change the interpretation of p-values and confidence intervals. Because the record does not describe an adjustment plan, the posted unadjusted p-values should not be treated as multiplicity-adjusted evidence.

The primary efficacy population is described as Per Protocol: participants in the intention-to-treat population were excluded from it if they made major protocol violations before completing the study. The posted summary does not establish how many participants were excluded for each reason, how missing endpoint data were handled, or whether sensitivity analyses were conducted. These details are not reported in the registry record.

The registry also does not report interim analyses, a complete statistical analysis plan, or the trial’s original sample-size justification. Those omissions limit the ability to assess whether the posted summaries match a prespecified analysis or how alternative analysis choices might affect the results. The recalculated contrasts are therefore best understood as unadjusted comparisons of the posted summary statistics, alongside the registry’s description of the intended Per Protocol efficacy population.

14. 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 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry lists zero participants with serious adverse events in each of the four safety groups. The corresponding numbers at risk are 29 for PF-0318 0.006%, 29 for PF-0318 0.024%, 29 for PF-0318 0.04%, and 30 for 0.005% latanoprost. These are the group-specific figures reported for serious adverse events; they are not a comparison adjusted for differences between groups.

The registry’s serious adverse event term list is empty. Death counts are not reported in the registry record. The figures describe what the record reports during the study and should not be read as proof that serious events cannot occur, or as a complete account of safety beyond the recorded follow-up. The record also lists other adverse events separately, so the serious-event counts should not be interpreted as counts of all adverse events.

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

The posted results describe average changes in diurnal intraocular pressure over the 28-day study and estimate contrasts between each comparison arm and the PF-0318 0.006% group. All four group means are negative. The three pairwise estimates are -0.82 mmHg, 0.1 mmHg, and -0.13 mmHg, and each confidence interval includes zero. The estimates are therefore uncertain, with intervals allowing differences in either direction.

These results do not establish that the groups are equivalent. Equivalence requires a prespecified margin and an analysis designed to test whether differences lie within that margin; such a margin and analysis are not reported in the registry record. Nor do the results establish that one regimen is superior, or that a particular dose has a dose-response relationship. The posted contrasts are unadjusted, and the registry record does not report results for the listed secondary outcomes.

The primary endpoint measures an average change from baseline, not an individual participant’s response. A group mean can conceal variation among participants, and the reported standard deviations indicate that individual changes were spread around the group averages. The confidence intervals quantify uncertainty in group contrasts under the calculation method, but do not cover every limitation, including unreported missing-data procedures and analysis details.

The safety information is also limited to what the registry records: zero serious adverse events are listed for each group, while death counts are not reported. These observations should not be taken to establish an absence of risk. Overall, the posted data are a record of the measured group summaries and their uncertainty, not a basis for medical advice or a definitive conclusion about treatment benefit.

16. Why This Trial Matters Statistically

This trial is statistically informative as a dose-finding comparison because it reports a continuous pressure-change outcome across three PF-03187207 dose groups and a latanoprost comparator. The registry records random allocation and a parallel-group design, and the endpoint is measured at a specified follow-up. These features make group-level contrasts a natural way to describe the posted efficacy summaries.

The results also illustrate why estimates and precision need to be considered together. The group means are all negative, but the pairwise estimates are near zero and the confidence intervals are broad enough to include zero and values on either side. A small point estimate does not settle whether a meaningful underlying difference exists, especially when the interval remains wide.

The dose-finding context makes the lack of reported multiplicity procedures relevant. Comparing several doses and outcomes can complicate interpretation, while the record does not report whether any adjustment was planned or applied. The registry also lists two secondary outcomes without reported results. This limits the conclusions that can be drawn about patterns across doses, measurement times, or outcome definitions.

The record’s description of the Per Protocol population and its missing analysis details highlight another statistical issue: the analysis population and handling of incomplete data affect how well results represent randomized participants. Those procedures cannot be fully assessed from the registry record. The study thus offers posted estimates for a short-term continuous endpoint, while also showing why careful reporting of analysis populations, multiplicity, and uncertainty is important in clinical trials.

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 available outcome summaries focus on the primary endpoint at Day 28. Although the registry lists change at specific time points on Day 14 and Day 28 and target intraocular pressure as secondary outcomes, those results are not reported in the registry record. This prevents assessment of how the groups compared on those additional measures or at the earlier visit.

The record describes the Per Protocol population as the primary efficacy population and defines it by exclusion of major protocol violations before study completion. It does not report the number excluded for each reason, the handling of missing primary outcome measurements, or an intention-to-treat analysis of the endpoint. The posted group counts therefore should not be assumed to describe every randomized participant.

The pairwise estimates are unadjusted recalculations from summary statistics. The registry record does not report the full analysis model, any covariate adjustment, the confidence level, a multiplicity strategy, interim analyses, or the original sample-size justification. The recalculations consequently cannot establish whether they reproduce the sponsor’s planned analysis.

The follow-up described for the primary endpoint is 28 days. The posted efficacy summaries and serious-event figures do not address longer-term outcomes. Serious adverse events are listed as zero in each group, but death counts are not reported, and the record’s safety summaries cannot establish that all possible adverse effects were captured. These limits call for a restrained reading of the available record rather than broader claims about longer-term efficacy or safety.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: A Phase 2, Dose Finding Study of PF-03187207 in Patients With Primary Open Angle Glaucoma or Ocular Hypertension. — NCT00595101.
• Registry record last updated 2020-09-14; status last verified 2020-09.
• Results first posted 2020-08-27.
• 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

NCT00595101 is a completed Phase 2 randomized parallel-group dose-finding study of PF-03187207 and 0.005% latanoprost in patients with primary open-angle glaucoma or ocular hypertension. The registry lists enrollment of 117, quadruple masking, and a 28-day dosing period. The official title describes the trial as double-masked.

The primary endpoint was change from baseline in mean diurnal intraocular pressure in the study eye on Day 28. The posted group means were -5.12, -5.94, -5.02, and -5.25 mmHg across the PF-0318 0.006%, PF-0318 0.024%, PF-0318 0.04%, and 0.005% latanoprost groups. All three unadjusted pairwise confidence intervals include zero. The registry describes the Per Protocol population as the primary efficacy population; details on missing data, multiplicity, and the original statistical plan are not reported.