A Study of AGN-195263 for the Treatment of Meibomian Gland Dysfunction: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of AGN-195263 for the Treatment of Meibomian Gland Dysfunction (NCT01633788), a randomized phase 2 study. The primary outcome is Percentage of Meibum Quality Responders in the Study Eye.

2. Clinical Question

The responder definition required a reduction of at least 50% in the number of meibomian glands in the study eye with a Meibum Quality Score of 2 or 3. The score grades expressed meibum from clear or nearly clear secretions to thick secretions that retain their shape or do not express. This endpoint therefore concerns a specified change in gland secretion quality, rather than a general measure of all symptoms or all aspects of meibomian gland dysfunction.

The registry record describes a safety and efficacy evaluation. It does not report a formal statistical analysis plan or the rationale for selecting this responder threshold. The primary comparison is between each of the three AGN-195263 concentrations and vehicle; no dose-to-dose primary comparison is reported.

3. Trial Design

This was a randomized, triple-masked, parallel-group Phase 2 study sponsored by Allergan. The registry record lists an enrollment of 232 and a completed status. The recorded start date is 2012-08-22, and the primary completion date is 2015-02-05. The study concerned meibomian gland dysfunction in the eyelids.

Four groups received AGN-195263 0.1%, AGN-195263 0.03%, AGN-195263 0.01%, or AGN-195263 vehicle. Each regimen was 1 drop in each eye twice daily. In a parallel-group design, participants are assigned to one treatment group and outcomes are compared across groups. Randomization is intended, in general, to reduce systematic differences between groups; the registry record does not describe the randomization procedure or how allocation was concealed.

Triple masking indicates that the study is recorded as masked at three levels, but the identities of those masked are not reported in the registry record. The record also does not report the sites, geographic locations, detailed eligibility criteria, treatment duration, adherence, or protocol deviations. The primary endpoint was measured at Month 6. The primary analysis population was modified intent-to-treat, defined as randomized and treated patients with a baseline meibum quality score, at least one postrandomization visit, and data at the noted time point. This is a narrower population definition than all randomized participants, and the record does not explain how patients without the required outcome data were handled.

Arm counts for the primary outcome were 54 for AGN-195263 0.1%, 56 for AGN-195263 0.03%, 53 for AGN-195263 0.01%, and 49 for vehicle. These endpoint denominators need not match overall enrollment: the record specifies outcome-specific population requirements, but does not provide reasons for the differences. The registry record lists 0 reported analyses.

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: Patients from one site were excluded from the modified intent-to-treat analysis population due to compliance issues. Patients from this site who received study treatment were included in the safety population and are reflected in the participant flow.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified intent-to-treat: all randomized and treated patients who have values for meibum quality score at randomization, at least one postrandomization visit, and data at the noted time point

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 Percentage of Meibum Quality Responders in the Study Eye, measured over Month 6.

The primary endpoint is the percentage of patients who were meibum quality responders in the study eye at Month 6. A responder was defined as a patient with a reduction of at least 50% in the number of glands scoring 2 or 3 on the Meibum Quality Score. Scores of 2 and 3 describe opaque, more viscous secretions, including secretions that retain their shape or do not express but are visible at the gland opening. Scores of 0 and 1 describe clearer or opaque secretions with normal viscosity.

This is a binary responder outcome: each patient either meets the defined response criterion or does not. Reporting the percentage gives the proportion meeting that criterion in each treatment group. In general, a predefined responder endpoint can make a result easier to interpret in terms of the proportion of participants crossing a specified threshold. It also depends on the chosen threshold and does not describe the full range of score changes among patients who do not meet it.

The study-eye qualification means the endpoint is based on the designated eye, rather than treating both eyes as independent observations. The registry record does not report how the study eye was selected. The endpoint is distinct from the secondary outcome based on maximum meibum quality score and from the secondary ocular discomfort responder outcome. It should be interpreted as evidence about this specific Month 6 response definition, not as a complete account of every aspect of the condition.

7. Results Posted in the Registry

• AGN-195263 0.1%: 46.3% (25/54) versus 42.9% (21/49) on control; risk difference 3.4 percentage points (95% CI -15.8 to 22.6), risk ratio 1.08 (0.70–1.66); unadjusted two-proportion z-test p = 0.726.
• AGN-195263 0.03%: 33.9% (19/56) versus 42.9% (21/49) on control; risk difference -8.9 percentage points (95% CI -27.5 to 9.7), risk ratio 0.79 (0.49–1.29); unadjusted two-proportion z-test p = 0.347.
• AGN-195263 0.01%: 32.1% (17/53) versus 42.9% (21/49) on control; risk difference -10.8 percentage points (95% CI -29.5 to 7.9), risk ratio 0.75 (0.45–1.24); unadjusted two-proportion z-test p = 0.260.

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.

Percentage of Maximum Meibum Quality Score (MMQS) Responders in the Study Eye

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • AGN-195263 0.1%: 20.4% (11/54) versus 22.4% (11/49) on control; risk difference -2.1 percentage points (95% CI -17.9 to 13.8), risk ratio 0.91 (0.43–1.90); unadjusted two-proportion z-test p = 0.797.
  • AGN-195263 0.03%: 23.2% (13/56) versus 22.4% (11/49) on control; risk difference 0.8 percentage points (95% CI -15.3 to 16.9), risk ratio 1.03 (0.51–2.09); unadjusted two-proportion z-test p = 0.926.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AGN-195263 0.1% and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AGN-195263 0.03% and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Complete Overall Ocular Discomfort Responders

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • AGN-195263 0.1%: 20.0% (11/55) versus 9.8% (5/51) on control; risk difference 10.2 percentage points (95% CI -3.2 to 23.6), risk ratio 2.04 (0.76–5.47); unadjusted two-proportion z-test p = 0.143.
  • AGN-195263 0.03%: 19.6% (11/56) versus 9.8% (5/51) on control; risk difference 9.8 percentage points (95% CI -3.4 to 23.1), risk ratio 2.00 (0.75–5.37); unadjusted two-proportion z-test p = 0.154.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AGN-195263 0.1% and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AGN-195263 0.03% and the control. The interval gives the range of effects that remain compatible with the data.

The two secondary outcomes address different dimensions from the primary endpoint: maximum meibum quality score response and complete overall ocular discomfort response. Their observed group patterns do not provide a uniform corroborating pattern across endpoints and concentrations. The registry record lists outcome values and unadjusted recalculations, but lists no reported secondary analyses.

These secondary measures can add context to the primary finding because one concerns a different summary of meibum quality and the other concerns ocular discomfort. Agreement across distinct outcomes can sometimes strengthen the coherence of a result, while disagreement can indicate that outcomes capture different aspects of a condition. In this record, however, the secondary findings should not be treated as independent confirmation of a treatment effect. Their interpretation depends on the outcome definitions, analysis populations, and the number of comparisons considered.

When many outcomes or treatment comparisons are examined, the chance of at least one apparently notable result arising by chance can increase. Adjusting for multiplicity is one general way to address that issue, but the registry record does not report a multiplicity procedure or identify a hierarchy for secondary testing. The secondary comparisons are unadjusted, and the record lists no reported analyses. They therefore carry less evidential weight than a prespecified, formally analyzed primary outcome and should be considered exploratory or descriptive in this record, without assuming that the original protocol classified them that way.

The secondary endpoint pattern also cannot resolve the imprecision in the primary comparisons. Each endpoint has its own definition and denominator, and the registry record does not report whether missing data or analysis methods were consistent across outcomes. Taken together, these results describe several measured outcomes at Month 6, but do not establish a single, comprehensive conclusion about all dimensions of meibomian gland dysfunction.

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

• Percentage of Maximum Meibum Quality Score (MMQS) Responders in the Study Eye
• Percentage of Complete Overall Ocular Discomfort Responders

9. Statistical Interpretation

10. Statistical Methods Explained

What does a responder percentage represent?

For a binary outcome, each participant is classified according to whether the stated event occurred. Here, the event is meeting the specified meibum quality response definition at Month 6. The reported percentage describes the share of the analyzed group classified as responders. The count of responders and the denominator both matter: the percentage is an estimate based on the participants included for that endpoint, not a fixed property known without uncertainty.

The risk difference compares the responder proportions in two groups by subtracting one proportion from the other. A positive estimate indicates a higher observed proportion in the AGN-195263 group than in vehicle; a negative estimate indicates a lower observed proportion. The reported risk differences are expressed as proportions, not percentage points. The registry record does not report an adjusted model for these comparisons, so they are read as unadjusted comparisons of the posted counts.

How should confidence intervals be read?

A confidence interval describes the range of values reasonably compatible with the observed data under the method used to calculate it. It conveys uncertainty in an estimate, rather than a guarantee that the true effect lies inside the particular interval. For a risk difference, zero represents no difference between group proportions. An interval that crosses zero includes both positive and negative differences among the values compatible with the data. General statistical interpretation also considers the interval's width: a wider interval indicates less precision about the effect's size and direction.

These intervals are recalculated, unadjusted comparisons based on the posted summary counts. The registry record does not specify the interval method or report sponsor analyses. The p-values are also unadjusted. A p-value measures how unusual data at least as incompatible with a specified null model would be if that model were true; it is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect.

Why distinguish observed results from planning illustrations?

Observed outcome estimates describe the participants who contributed data. A sample-size planning calculation asks a different question: how many participants might be needed under stated assumptions to detect a chosen effect with a chosen probability. The posted planning illustrations use a two-sided significance level of 0.05, equal allocation between the two groups, and a normal approximation. They use a stated fraction of the observed effect as the effect to detect. The trial's own original design assumptions and sample-size justification are not reported in the registry record.

The sample-size illustrations are not retrospective measures of the trial's ability to find an effect. They are conditional calculations based on assumptions, and changing the assumed effect changes the resulting planning requirement. Their role is to show how planning quantities respond to assumptions, not to establish what the original investigators expected or intended.

What does over-dispersion mean in general?

Over-dispersion arises when the observed variability in data is greater than a particular statistical model expects. It is a concern in some count or clustered data settings and can make model-based uncertainty too small if ignored. General explanations of over-dispersion do not imply that it applies to this binary responder endpoint. The registry record does not report a model or dispersion estimate for the primary outcome.

For this endpoint, the relevant general distinction is between the event proportion and the uncertainty around it. A percentage alone does not convey precision; counts, denominators, and an interval help describe the amount of information in the comparison. The outcome counts are modest relative to the full enrollment, and the uncertainty intervals reflect that the observed proportions do not pin down a narrow range of possible group differences.

What analysis details are unavailable?

The registry record does not report the original statistical analysis plan, the method used by the sponsor for the primary comparison, adjustment covariates, handling of missing outcome data, or an interim-analysis plan. It also does not provide a complete account of protocol deviations or outcome ascertainment. These details matter because analysis choices can affect the estimate, its uncertainty, and how evidence across multiple outcomes is interpreted. Their absence means the recalculated comparisons should be treated as a transparent descriptive analysis of posted counts, not as a reconstruction of the trial's full prespecified analysis.

11. Confidence Intervals and Precision

The precision of the primary comparisons is limited by the breadth of their uncertainty intervals. For AGN-195263 0.1% versus vehicle, the risk-difference interval runs from -0.158 to 0.226. For 0.03% versus vehicle, it runs from -0.275 to 0.097. For 0.01% versus vehicle, it runs from -0.295 to 0.079. All three intervals include zero and include effects in both directions. This means the posted counts do not sharply distinguish a positive difference from a negative one.

The endpoint denominators were 54, 56, and 53 for the three AGN-195263 groups and 49 for vehicle. The corresponding responder counts were 25, 19, 17, and 21. Confidence intervals are useful because the observed percentages are estimates from these groups; the interval gives information about uncertainty that the percentages alone cannot show. The intervals are not evidence that every value within them is equally likely, nor do they determine the clinical importance of any particular difference.

The registry record does not report arm-level confidence intervals for the percentages or the interval calculation method. The comparison intervals are unadjusted recalculations from the posted counts. No sponsor-reported inferential analyses are listed, and the registry record does not state how missing outcomes were handled. Those limitations matter when interpreting precision: statistical uncertainty in the calculated comparison is only one part of the uncertainty about the full trial result.

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 figures are participants per arm for 80% power and participants per arm for 90% power; they are not total events. The calculations use the assumptions stated for the planning illustrations: a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to the stated fraction of the observed effect. The original trial's sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the calculated number of participants in each group under which the assumed effect would be detected in 80% of repeated studies, if the planning assumptions and effect were correct. The 90% column gives the corresponding per-arm planning quantity for 90% power. Power is conditional on the chosen effect, significance level, allocation, and calculation method; it does not describe the probability that the study's result is true or that a treatment effect exists.

The planning rows use 100%, 75%, and 50% of the observed effect. As the assumed effect becomes smaller, more participants are needed to distinguish it from random variation under otherwise unchanged assumptions. The values are illustrations based on the observed estimate, not the sample size that the investigators necessarily planned for this trial. They are not observed power.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.429; treatment proportion p₁ = 0.463; average p̄ = 0.446.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7029 = 1.3777.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.7025 = 0.5913.
4. n = (1.3777 + 0.5913)² / (0.0344)² = 3,277.8, rounded up to 3,278 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
p0 <- 0.428571428571429                          # control proportion
p1 <- 0.462962962962963                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(1633788)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 3,278 (80% power), 4,388 (90% power)
 75% of observed effect: n per arm = 5,817 (80% power), 7,787 (90% power)
 50% of observed effect: n per arm = 13,062 (80% power), 17,485 (90% power)
power.prop.test: 3277.8 
Simulated power at n = 3,278 per arm: 0.802 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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 lists 0 reported analyses and does not report a statistical analysis plan, a testing hierarchy, or a procedure for controlling multiplicity. The primary endpoint is identified, but the record does not explain whether the three dose-versus-vehicle comparisons were treated as a family of tests or whether any adjustment was planned. It also does not state how the secondary outcomes were positioned relative to the primary endpoint for inferential testing.

In general, multiplicity matters when several hypotheses are tested: the more comparisons considered, the greater the opportunity for a chance finding. An adjustment or prespecified hierarchy can help clarify how results should be interpreted across a family of outcomes. Without a reported approach, the unadjusted p-values for separate comparisons should not be read as if they controlled the overall error rate across all doses and outcomes. This does not establish that no such procedure was used; it means the procedure is not reported in the registry record.

The primary endpoint population is defined as randomized and treated patients with a score at randomization, at least one postrandomization visit, and outcome data at the noted time point. This definition identifies who could enter the primary analysis, but it does not explain how missing data were addressed, whether any imputation was used, or how many otherwise eligible participants lacked Month 6 data. Those details are not reported in the registry record.

The registry record also does not report the handling of protocol deviations, analysis covariates, interim analyses, or whether analyses followed an intention-to-treat principle beyond the stated modified intent-to-treat population. The endpoint denominators differ by outcome, but reasons for those differences are not reported. These omissions limit the ability to assess the robustness of the results or reproduce the trial's original analysis. The recalculated comparisons are therefore best viewed as unadjusted summaries of the posted counts rather than a full audit of the prespecified statistical approach.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: gait disturbance (1.8%), atrial fibrillation (1.7%), muscle spasms (1.7%), bone cancer (1.7%), lung carcinoma cell type unspecified stage iv (1.7%).

The registry record reports serious adverse event counts among participants at risk in each group. The serious-event figures are 1 of 57 for AGN-195263 0.1%, 4 of 58 for AGN-195263 0.03%, 3 of 59 for AGN-195263 0.01%, and 0 of 58 for vehicle. These are participant counts, not estimates of a comparative treatment effect. The record does not provide a statistical comparison of these counts or report deaths; death information is not reported in the registry record.

Serious adverse events are not interchangeable with all adverse events. The record separately lists participants affected by other adverse events, but those counts do not change the interpretation of the serious-event figures. The listed serious-event terms include gait disturbance, atrial fibrillation, muscle spasms, bone cancer, lung carcinoma cell type unspecified stage IV, and migraine, with the recorded term percentages ranging from 1.7% to 1.8%. The record does not link these terms to specific treatment groups in the displayed listing.

These counts describe reported events during the study; they do not establish that an event was caused by a study intervention. Differences in counts should be interpreted cautiously, particularly without further detail on event definitions, exposure time, attribution, or statistical comparisons. Such details are not reported in the registry record.

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

The posted results describe Month 6 responder proportions for a specific study-eye meibum quality definition, together with unadjusted comparisons of each AGN-195263 concentration against vehicle. The arm percentages range from 32.1% to 46.3% across the active groups and are 42.9% for vehicle. The comparison estimates have confidence intervals that include zero and extend in both directions. This means the registry results do not provide a precise estimate of the direction or size of a difference for any of the three dose-versus-vehicle comparisons.

That interpretation is not equivalent to concluding that the groups are the same. A broad interval can remain compatible with a range of possible differences, including some in either direction. Conversely, a difference in observed percentages is not, by itself, evidence that a true treatment-group difference exists. The unadjusted p-values and recalculated intervals should be considered alongside the design, endpoint definition, and analysis limitations.

The result is specific to the modified intent-to-treat population defined in the registry record, the study eye, and the Month 6 visit. It does not establish what would occur at other time points, in groups not represented by the study, or on outcomes not measured. The two secondary endpoints provide additional outcome information, but the record lists no reported analyses and does not describe multiplicity control. Safety counts likewise describe reported events, not causal effects.

The registry record does not report the original sample-size rationale, the sponsor's inferential analysis, missing-data procedures, or detailed statistical plan. The planning illustrations are conditional calculations and do not reveal the trial's original design assumptions. Taken together, the record supports a cautious statistical description of the observed outcomes and their uncertainty; it does not, on its own, establish a definitive conclusion about treatment efficacy or safety.

17. Why This Trial Matters Statistically

This trial illustrates why an observed percentage and a treatment comparison answer different questions. The arm-level values show how many participants met the response definition in each group, while the risk difference directly contrasts each dose group with vehicle. Confidence intervals add a necessary account of uncertainty around those comparisons. In this study, the intervals include zero and allow differences in either direction, so the point estimates alone would give an incomplete impression.

The endpoint also illustrates the importance of operational definitions. A responder was determined by a threshold change in glands with specified meibum quality scores in the study eye. That makes the outcome interpretable as a proportion meeting a defined criterion, while also limiting what it can say about changes that do not cross the threshold or about other dimensions of the condition.

Finally, the record highlights the value of transparent reporting. The sample-size illustrations state their assumptions, but the original design justification is not reported. The analysis population is defined, yet handling of missing data and multiplicity procedures are not reported. Those gaps make it important to distinguish transparent recalculation from a sponsor's complete trial analysis. These are broadly useful lessons for interpreting binary clinical endpoints: clarify the event, compare groups directly, examine uncertainty, and identify which analysis details remain unavailable.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

Several limitations are specific to what the registry record reports for this study. The primary analysis uses a modified intent-to-treat population with outcome-availability requirements, but the record does not state how many randomized participants were excluded from that endpoint population or why the endpoint denominators vary. It does not report a missing-data strategy, imputation approach, or sensitivity analysis. As a result, the effect of incomplete outcome data on the reported proportions cannot be assessed from this record.

The registry record does not report the sponsor's inferential analyses, the statistical model, confidence-interval method, adjustment covariates, or multiplicity procedure. The comparisons discussed are unadjusted recalculations from posted summary counts. They may therefore not match a prespecified sponsor analysis, if one was conducted. The record also does not report the original sample-size calculation, intended detectable effect, interim-analysis plan, randomization details, allocation concealment, or identities of the groups included in triple masking.

For the clinical interpretation, the primary endpoint is limited to a Month 6 response threshold in the study eye. The registry record does not report how the study eye was selected or give detailed information on baseline group characteristics, adherence, protocol deviations, or outcome assessment procedures. The response measure does not capture every possible degree of change in meibum quality. Secondary outcomes add context but have no reported analyses, and the record does not describe how they relate to a testing hierarchy.

Safety information is presented as event counts and selected terms, without a reported causal assessment or statistical comparison. The registry record does not report deaths. These limitations do not establish that unreported procedures or information were absent; they restrict what can be concluded from the record itself.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Study of AGN-195263 for the Treatment of Meibomian Gland Dysfunction — NCT01633788.
• Registry record last updated 2017-09-08; status last verified 2017-08.
• Results first posted 2017-09-08.
• 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

NCT01633788 is a completed, randomized, triple-masked, parallel-group Phase 2 study sponsored by Allergan. The registry record lists 232 enrolled participants, a start date of 2012-08-22, and a primary completion date of 2015-02-05. The condition was meibomian gland dysfunction.

Participants received AGN-195263 0.1%, 0.03%, or 0.01%, or vehicle, as 1 drop in each eye twice daily. The primary outcome was the percentage of Month 6 meibum quality responders in the study eye, using a reduction of at least 50% in glands scoring 2 or 3. The two secondary outcomes concerned maximum meibum quality score responders and complete overall ocular discomfort responders.

The registry record lists 0 reported analyses. It does not report the original sample-size justification, a statistical analysis plan, missing-data procedures, or multiplicity handling. The outcome comparisons described are unadjusted recalculations from the posted counts.