Dexamethasone Intravitreal Implant for Treatment of Macular Edema After Plaque Radiotherapy of Uveal Melanoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Dexamethasone Intravitreal Implant for Treatment of Macular Edema After Plaque Radiotherapy of Uveal Melanoma (NCT01471054), a randomized phase 2 study. The primary outcome is Number of Participants for Whom Study Eye Showed >=2 Lines of Improvement in Best-corrected Visual Acuity.

2. Clinical Question

The study addressed macular edema after plaque radiotherapy of uveal melanoma. Its stated aim was to evaluate the safety and efficacy of dexamethasone intravitreal implant, Ozurdex, and compare it with intravitreal bevacizumab. The primary outcome was the number of participants whose study eye showed at least 2 lines of improvement in best-corrected visual acuity at 12 months.

The primary outcome is a count of participants meeting a defined improvement threshold, rather than a comparison of average visual-acuity change. The registry record also lists safety outcomes and change in central subfield retinal thickness. It does not report a reason for termination or a fuller account of participant eligibility.

3. Trial Design

This was a phase 2 study with randomized allocation, no masking, and a single-group model recorded in the registry. Those design labels do not fit together straightforwardly: randomized allocation and two named treatment arms suggest an intended comparison, while the model is listed as single group. The registry record does not explain this inconsistency, so the design cannot be characterized more precisely than the recorded details allow.

The record reports enrollment of 6 participants. The posted primary outcome counts are grouped by treatment: 5 participants in the Ozurdex arm and 1 in the bevacizumab arm. This is a markedly uneven distribution of participants across the arms, and the record does not explain why it occurred or provide details of the allocation process.

The study began in 2014-04 and had a primary completion date of 2015-07. Its status is terminated. The reason for termination is not reported in the registry record, and the listed dates alone do not establish how long each participant was followed.

Follow-up descriptions outline repeated visits after treatment. For Ozurdex, visits are described at 1 week and at 1, 2, 3, 4, 5, and 6 months, followed by visits every 2 months. For bevacizumab, the description specifies visits at 1 week and at 1, 2, 3, 4, 5, and 6 months, with subsequent examinations every 4-8 weeks depending on macular-edema status. The primary outcome timeframe is 12 months. The registry record does not report masking, a detailed analysis plan, or a reason for the different later visit intervals.

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

Reasons for not completing this period:

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 Number of Participants for Whom Study Eye Showed >=2 Lines of Improvement in Best-corrected Visual Acuity, measured over At 12 months.

The primary endpoint counts participants whose study eye had at least 2 lines of improvement in best-corrected visual acuity at 12 months. Visual acuity was measured with a Snellen eye chart placed 10 feet away. The outcome is binary at the participant level: each participant either met the stated improvement threshold or did not. The registry record reports 2 participants meeting the threshold in the Ozurdex group of 5, and 0 in the bevacizumab group of 1.

A threshold-based endpoint can be useful when the research question concerns how many participants reach a defined degree of change, rather than the average amount of change across all participants. It can make the outcome directly interpretable as a count of participants meeting the criterion. However, it also discards information about the size of changes that fall short of or exceed the threshold. Two participants can both meet the threshold despite having different amounts of improvement, while a small change on either side of the cutoff can place participants in different categories.

As a general statistical consideration, the endpoint’s interpretation depends on consistent measurement and a clear definition of baseline and follow-up assessments. The registry record describes the measurement chart, distance, threshold, and 12-month timeframe. Further details about assessment procedures, handling of incomplete measurements, and the analysis population are not reported in the registry record. The counts therefore describe the posted endpoint results but do not, by themselves, establish a treatment effect or its precision.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a binary outcome: participants are counted according to whether their study eye met the criterion of at least 2 lines of visual-acuity improvement at 12 months. The posted results give the number meeting the criterion and the number of participants in each arm. This form of reporting is straightforward to read, but a count alone does not show how uncertain the underlying outcome rate or between-group contrast may be.

What does a confidence interval describe?

As a general statistical concept, a confidence interval is a range of values compatible with the observed data under a specified method and its assumptions. It communicates precision as well as the direction and scale of an estimate. Narrow intervals indicate more precise estimates; wide intervals indicate less precision. For a binary endpoint, uncertainty is especially important when the number of participants is small or outcomes are sparse. The registry record does not report a confidence interval for the primary outcome comparison.

A p-value, when reported, addresses how compatible the observed data are with a specified null model under the analysis assumptions. It does not measure the probability that a treatment is effective, and it does not describe the magnitude or clinical importance of an effect. The registry record does not report a p-value or a formal two-arm analysis for this endpoint. The absence of an inferential result should not be replaced by an informal conclusion drawn from the raw counts.

Why does the group size matter?

The registry reports 5 participants in the Ozurdex arm and 1 in the bevacizumab arm. General statistical reasoning indicates that estimates from small groups are highly sensitive to individual outcomes. The single participant in the bevacizumab arm cannot characterize the range of outcomes that might occur among a larger group. The unequal group sizes also mean that the posted counts do not offer a balanced basis for comparing the two arms.

Randomization can, in general, help balance measured and unmeasured participant characteristics across treatment groups. Its value depends on how allocation was carried out and on the study’s conduct. Although the registry lists randomized allocation, it also lists a single-group model, and the record does not explain this combination. It does not report the allocation sequence, concealment procedures, or a detailed account of enrollment by arm. The design labels should therefore be described as recorded rather than reconciled by assumption.

What does a count-based endpoint leave out?

A threshold outcome simplifies interpretation by classifying participants according to whether they crossed a specified cutoff. The trade-off, as a general statistical matter, is loss of detail about the amount of change. The posted endpoint does not show the distribution of visual-acuity changes or how far participants were from the threshold. A continuous measure of change could retain more information, but the registry record does not provide continuous primary-outcome results.

Measurement consistency also matters. The outcome description identifies a Snellen eye chart at 10 feet and specifies the improvement threshold and 12-month timeframe. Other details that could inform interpretation, such as the handling of participants without a 12-month measurement, are not reported in the registry record. Without those details, the counts cannot establish whether the results represent all enrolled participants or only those with recorded assessments.

How should planning calculations be read?

General sample-size planning asks how many participants might be needed to detect an effect under a set of assumptions. Such calculations depend on the target effect, the significance level, allocation, and the statistical approximation. The planning approach described for this study uses a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation, with the target effect expressed as a fraction of the observed effect. These are planning assumptions, not evidence that a study of a particular size would achieve a specified result in practice.

Sample-size calculations based on an observed effect are illustrative. An effect observed in a very small study can be unstable, and calculations based on it inherit that uncertainty. These calculations are not observed power, and they do not establish that the original trial had an adequate sample size. The registry record does not report the trial’s own sample-size justification. Because the binary comparison is not estimable in the posted results, no precise comparative effect can be used to support a planning conclusion from these counts.

What can safety counts establish?

Safety summaries report participants at risk and the number with events. Zero events in a group means none were recorded among the participants counted as at risk for that outcome. As a general principle, observing no events in a small group does not show that the underlying risk is zero. A larger study or more complete information about event definitions and observation periods would be needed to characterize uncommon events more precisely. The registry record gives event counts but does not report detailed event definitions or inferential safety estimates.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Ozurdex", "Bevacizumab")
n      <- c(5, 1)
events <- c(2, 0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Ozurdex  5  2  40.0  5.3  85.3  11.8  76.9  11.8  77.7
ARM  Bevacizumab  1  0  0.0  0.0  97.5  0.0  79.3  1.3  84.2

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 5 secondary outcomes: change in central subfield retinal thickness, development of glaucoma, cataract, retinal detachment, and vitreous hemorrhage. It also reports 6 outcome measures in total. The numerical results and analyses for the secondary outcomes are not reported in the registry record. The record does not describe a prespecified hierarchy among outcomes or a method for addressing multiple statistical comparisons.

As a general statistical issue, testing several outcomes can increase the chance of at least one apparently notable result arising by chance. A prespecified primary outcome can help focus interpretation, while a stated multiplicity strategy can clarify how secondary findings are to be evaluated. The registry record identifies the primary endpoint but does not report a multiplicity procedure or a detailed statistical analysis plan. No conclusion should be drawn about whether multiplicity was addressed outside the information recorded.

Missing-data methods and analysis populations are not reported in the registry record. The posted primary outcome includes 5 participants in the Ozurdex group and 1 in the bevacizumab group, but those counts do not establish whether every enrolled participant had a 12-month assessment or how incomplete assessments were handled. The record does not state whether the analysis followed assignment regardless of treatment received, included only participants who completed follow-up, or used another population definition.

In general, missing outcomes can affect estimates if the likelihood of having an observed result is related to treatment or outcome. The consequences depend on how much information is missing and why. Those details are not reported in the registry record, so the potential effect of missing data cannot be assessed. The study’s terminated status does not, by itself, establish that data were missing or explain how termination affected the analysis.

12. 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.

The registry reports no serious adverse events among the 5 participants at risk in the Ozurdex group and no serious adverse events among the 1 participant at risk in the bevacizumab group. It also reports no other adverse events in either group. Deaths are reported as 0 in both groups. These are observed counts, not proof that the treatments have no safety risks.

The group sizes are very small, particularly for bevacizumab. As a general statistical principle, a study with few participants may fail to observe events that could occur in a larger population. Zero reported events therefore cannot establish that an event is impossible or that the underlying event rate is zero. The counts provide a description of what was recorded among the participants counted as at risk.

The registry record does not report detailed definitions of serious and other adverse events, event observation windows, or how adverse events were assessed. It lists no serious adverse-event terms. The absence of listed terms is consistent with the reported zero serious events, but the record does not provide further event-level detail. These safety summaries should be interpreted alongside the small enrollment and the limited information about follow-up and ascertainment.

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

The posted primary result is a count of participants who crossed a visual-acuity improvement threshold at 12 months: 2 of 5 in the Ozurdex group and 0 of 1 in the bevacizumab group. These figures describe the recorded outcomes for the participants in each group. They are not, on their own, a precise estimate of the comparative effect of treatment.

The comparison is constrained by the small total enrollment, the unequal group sizes, and the single participant in the bevacizumab group. The registry record does not report an estimable two-arm comparison, a confidence interval, or a p-value. Consequently, the posted counts cannot establish whether the apparent difference reflects a treatment contrast, chance variation, or other factors. They also cannot support a conclusion that either treatment works or does not work.

The design record adds uncertainty: allocation is described as randomized, while the model is listed as single group. The registry record does not explain the discrepancy. It also does not report the reason the study was terminated, the analysis population, or the handling of missing 12-month assessments. These omissions limit assessment of how the outcome counts relate to all enrolled participants and to the intended design.

The safety counts record no serious adverse events, other adverse events, or deaths in either group. With the number of participants at risk so small, those observations cannot exclude events that were not seen. Secondary outcomes are listed, but their numerical results are not reported in the registry record.

In general, a treatment comparison is most informative when its design, endpoint assessment, analysis population, and uncertainty estimates are clear. Here, the registry record supports a limited descriptive account: it reports the primary outcome counts and safety counts, but not enough information for a precise or definitive comparison. These findings should be understood as the results recorded for this small terminated study, not as a general conclusion about treatment effects.

14. Why This Trial Matters Statistically

This study raises several statistical issues that are important when reading small clinical trials. Its total enrollment was 6, and the posted primary outcome groups contain 5 participants receiving Ozurdex and 1 receiving bevacizumab. With such sparse data, each participant’s outcome has substantial influence on the descriptive comparison. The single-participant comparator group is particularly limited as a basis for estimating outcomes in a broader population.

The record also illustrates why design labels need to be interpreted carefully. It records randomized allocation but a single-group model, while reporting two named treatment arms. Without an explanation, readers cannot determine how the registry’s model designation relates to the intended comparison. Clear descriptions of allocation, analysis populations, and treatment-group construction are important for judging whether an observed contrast can be interpreted as a randomized comparison.

A threshold-based primary outcome makes the endpoint easy to state as a count of participants meeting a criterion. It also leaves out the magnitude and distribution of visual-acuity changes. This trade-off is a general feature of dichotomizing a continuous measure. The record gives the endpoint definition and assessment timeframe, but not a continuous primary-outcome analysis or an uncertainty estimate for the binary comparison.

Finally, the terminated status, absence of a reported reason for termination, and lack of reported inferential analyses limit interpretation beyond the raw counts. These features make the study a useful example of why outcome counts, group size, design documentation, and uncertainty need to be considered together. Statistical interpretation depends not only on what the observed counts are, but also on how much information supports them and what the registry record does not explain.

15. Limitations

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

The primary outcome is reported as a count at a single timeframe, 12 months. The registry record does not provide a confidence interval, p-value, or other inferential estimate for the comparison. It also does not report continuous visual-acuity results, so the amount of change among participants who did and did not meet the threshold cannot be assessed from the posted primary outcome.

The enrollment was 6, with 5 participants in the Ozurdex group and 1 in the bevacizumab group. This small and uneven distribution sharply limits the information available for a two-arm comparison. The record’s randomized-allocation label and single-group model label are not reconciled, and details of the allocation process are not reported. The reason for termination is also not reported in the registry record.

Although the primary outcome is defined by a Snellen chart at 10 feet and a threshold of at least 2 lines of improvement, the registry record does not report how missing assessments were handled, which analysis population was used, or whether all enrolled participants had results at 12 months. The listed secondary outcomes are not accompanied by numerical results or analyses in the registry record. No multiplicity method or detailed statistical analysis plan is reported.

Safety summaries give zero serious adverse events, zero other adverse events, and zero deaths in both groups, but event definitions and detailed ascertainment methods are not reported. With so few participants, the absence of observed events does not characterize uncommon risks precisely. These limitations constrain both the comparative interpretation of the primary outcome and the scope of conclusions that can be drawn from the safety counts.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Dexamethasone Intravitreal Implant for Treatment of Macular Edema After Plaque Radiotherapy of Uveal Melanoma — NCT01471054.
• Registry record last updated 2019-09-03; status last verified 2019-08.
• Results first posted 2019-01-10.
• 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.

19. Record Summary

The registry record describes a phase 2 study of dexamethasone intravitreal implant and intravitreal bevacizumab for macular edema after plaque radiotherapy of uveal melanoma. It lists randomized allocation, no masking, and a single-group model. The study enrolled 6 participants and is marked terminated. The reason for termination is not reported in the registry record.

The primary endpoint was the number of participants with at least 2 lines of study-eye best-corrected visual-acuity improvement at 12 months, measured with a Snellen chart at 10 feet. The posted counts are 2 of 5 for Ozurdex and 0 of 1 for bevacizumab. The registry record does not report an estimable two-arm comparison or inferential analysis for this endpoint.

Reported safety counts include no serious adverse events, no other adverse events, and no deaths in either group. Five secondary outcomes are listed, but their numerical results are not reported in the registry record. A detailed analysis plan, missing-data methods, and analysis populations are also not reported.