This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for The Prophylactic Use of Topical Cyclosporine A 0.05% to Prevent Graft Versus Host Disease Related Dry Eye (NCT00553735), a randomized phase 4 study. The primary outcome is Corneal Staining Score.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00553735 |
| Phase | Phase 4 |
| Condition | Dry Eye |
| Allocation | Randomized |
| Design model | Single Group |
| Masking | Triple |
| Enrollment | 8 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Massachusetts Eye and Ear Infirmary |
| Results posted | Yes |
2. Clinical Question
The study asked whether topical cyclosporine A 0.05% could prevent the onset or progression of dry eye associated with graft-versus-host disease in people who had recently received a bone marrow transplant and were considered at risk. Its primary endpoint was corneal staining score at 18 months. The registry record reports that no data were collected for this endpoint because participants withdrew, did not attend appointments, or were lost to follow-up. Consequently, the record does not show whether the treatment changed corneal staining or prevented dry eye.
Population
People described as having recently received a bone marrow transplant and being at risk for graft-versus-host disease. The enrollment was 8. Further eligibility details are not reported in the registry record.
Intervention
Topical cyclosporine A 0.05%, identified in the record as Restasis. The eye-level description says each participant’s eyes were randomized to the study therapies.
Comparator
Artificial tear, described as the placebo. The plan assigned this treatment to the other eye of the same participant.
Primary question
Would cyclosporine A 0.05% prevent the onset or progression of graft-versus-host disease-related dry eye, as assessed by corneal staining score at 18 months?
The intended comparison is therefore within each participant: one eye receives cyclosporine and the other receives artificial tear. This paired arrangement is different from assigning each participant to a separate treatment group. The record does not report an outcome comparison, an effect estimate, or an uncertainty interval for the primary question.
3. Trial Design
The registry describes a phase 4, randomized, triple-masked study with a single-group model. Enrollment was 8. The recorded start was 2007-08, the primary completion was 2010-12, and the status is terminated. The sponsor was Massachusetts Eye and Ear Infirmary. The reason for termination is not reported in the registry record.
The intervention descriptions state that both eyes of each eligible participant were randomized: one eye was assigned cyclosporine A 0.05%, and the other was assigned artificial tear as placebo. That is an eye-level, paired allocation plan. It differs from the registry’s single-group model label, which does not describe the eye-by-eye comparison clearly. The record does not report the randomization procedure, sequence generation, allocation concealment, or how treatment assignment was implemented.
Triple masking indicates that the registry classifies the study as masked at three levels, but it does not identify which people or roles were masked. The masking method, whether masking was maintained, and whether treatment assignment could be recognized are not reported in the registry record. The record also does not provide a protocol or a detailed schedule of visits and assessments.
The primary endpoint was corneal staining score, with a timeframe of 18 months. Listed secondary outcomes include the Symptom Assessment iN Dry Eye (SANDE) Patient Questionnaire, tear break-up time, and Schirmer testing with and without anesthesia. Although the record identifies 7 outcome measures, it reports no analyses. It states that no primary outcome data were collected because participants withdrew themselves, failed to show for appointments, or were lost to follow-up. The record does not clarify whether any secondary outcome measurements were collected or analyzed.
Because both eyes of a participant were to receive different treatments, measurements from the eyes would be paired rather than independent. This design can account for characteristics shared by the participant, but it also requires an analysis that respects the relationship between the eyes. The statistical plan for such an analysis 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
| Stage | All Study Participants |
|---|---|
| STARTED | 8 |
| COMPLETED | 1 |
| NOT COMPLETED | 7 |
Analysis population for the primary outcome, as defined in the registry: No data were collected because participants withdrew themselves, failed to show for their appointments, or were lost to follow-up.
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.
| Characteristic | Unit | Arm I |
|---|---|---|
| Participants analysed | 8 | |
| Age, Categorical: <=18 years | Participants | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 7 |
| Age, Categorical: >=65 years | Participants | 1 |
| Age, Continuous | years | 43 (SD 17) |
| Sex: Female, Male: Female | Participants | 7 |
| Sex: Female, Male: Male | Participants | 1 |
| Region of Enrollment: United States | participants | 8 |
6. Primary Endpoint
The registry lists the primary outcome as Corneal Staining Score, measured over 18 months.
| Element | Registry entry |
|---|---|
| Result type | N/A |
| Measure of spread | not applicable |
| Unit | not stated |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was corneal staining score at 18 months. Corneal staining is an examination-based measure used to assess surface changes in the eye. In a dry-eye study, such an endpoint can provide an objective sign to complement symptom reports and tests of tear production or tear-film stability. It is relevant to the study’s aim because the question concerned preventing or limiting eye-surface effects associated with graft-versus-host disease.
The registry record does not describe the staining scale, its possible values, the grading procedure, or the direction and size of a change that would be considered important. It also does not state whether the score was intended to be analyzed as a continuous measure, an ordered category, or another form of outcome. Those details matter because the measurement scale and analysis method determine how differences should be interpreted.
Most importantly, the record explicitly says that no data were collected for the primary endpoint. There is therefore no reported corneal staining result at 18 months, no comparison between cyclosporine-treated and artificial-tear-treated eyes, and no estimate of precision. The endpoint is specified as the primary measure, but its planned role cannot be converted into a finding when the outcome data are absent.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The central statistical fact is that the registry record contains no primary endpoint data. It therefore cannot establish whether corneal staining differed between eyes assigned cyclosporine A 0.05% and eyes assigned artificial tear. No treatment effect, confidence interval, or primary-outcome significance test is reported. The record’s status is terminated, but termination by itself does not establish why the study ended or what its results would have been.
The registry reports enrollment of 8 and gives reasons for the missing primary outcome data: participants withdrew themselves, did not attend appointments, or were lost to follow-up. These reasons indicate that the planned endpoint was not available for analysis as described. The record does not report how many participants experienced each reason, whether any measurements were collected before follow-up ended, or whether the reasons differed by assigned eye treatment. Those distinctions could affect how missingness is understood, but they cannot be resolved from the record.
There are no analyses reported, and the primary endpoint is not reduced to a comparison. Thus there is no observed contrast to interpret, whether as a difference in scores, a relative measure, or another effect scale. No inference about prevention or progression can be drawn from the absence of a result. Missing data do not show that treatments were equivalent, and they do not show that either treatment was better or worse.
The eye-level design could, in principle, support a paired comparison. A paired analysis would compare the cyclosporine-treated and artificial-tear-treated eyes within the same participant, preserving the randomization structure and accounting for shared participant characteristics. But such an analysis requires observed measurements and a method suitable for the score’s scale. Neither the measurements nor an analysis method is reported. The single-group model label also does not, by itself, convey the paired eye-level allocation described in the intervention text.
The record lists secondary measures, including symptoms, tear break-up time, and Schirmer testing. Listing an outcome does not establish that it was measured or analyzed. The record reports zero analyses, and whether any secondary results exist is not reported in the registry record. Consequently, the listed measures cannot be treated as evidence about benefit or harm.
The adverse-event entry reports no serious events among the 8 participants at risk and 2 other events among the 8 at risk. These are counts as recorded, not a comparison between the two eye treatments. The record does not give event descriptions or establish which treatment, if either, was associated with an event. Deaths are not reported in the registry record. The safety figures should therefore be read as limited descriptive reporting, not as a comprehensive assessment of safety.
Any planning illustrations based on an observed effect should also be interpreted cautiously in this setting. The registry record contains no primary outcome data from which a primary-endpoint effect can be interpreted. Planning calculations are not observed power and cannot substitute for missing results. Overall, the record documents an intended randomized comparison but does not provide an outcome-based answer to the study question.
9. Statistical Methods Explained
The study’s design and the information needed to interpret a statistical result are related but distinct. The registry identifies randomization and triple masking, and the intervention descriptions specify that different treatments were assigned to the two eyes of each participant. Those details describe the intended comparison. They do not supply the observed endpoint values or explain how those values would have been analyzed. Since no primary data were collected, the record contains no estimate of the treatment contrast to evaluate.
What does randomization mean in this design?
Randomization is intended to make treatment assignment independent of characteristics that might otherwise influence outcomes. In this study, the described assignment was between eyes within a participant, rather than between separate groups of participants. Each participant therefore serves as their own comparison for the eye-level treatments. This can reduce the influence of characteristics shared by both eyes, but it does not remove the need to account for the relationship between measurements from the same person.
The registry does not describe how the eye assignments were generated or concealed. It also does not explain whether the left or right eye had a fixed assignment rule, how any assignment was recorded, or how adherence was assessed. These methods are not reported in the registry record, so the quality and implementation of allocation cannot be assessed from the entry alone.
Why does the paired structure matter?
Measurements from two eyes belonging to the same participant are generally related because the eyes share features of the person and may share environmental or clinical influences. Treating those measurements as if they came from unrelated people can misstate the amount of independent information and can lead to inappropriate uncertainty estimates. A paired analysis instead uses the within-participant relationship in the treatment comparison.
The exact paired method depends on the outcome scale and distribution. For example, an analysis designed for a roughly continuous score may not be appropriate if the score is an ordered category or has a strong concentration at particular values. The registry does not report the staining scale, the planned statistical test, or a model. It therefore does not establish which method was intended or appropriate.
What would an effect estimate and confidence interval contribute?
An effect estimate expresses the direction and magnitude of the observed difference on a defined scale. A confidence interval describes the uncertainty around that estimate under the assumptions of the analysis. A narrow interval can indicate greater precision, while a wide interval can indicate substantial uncertainty. Neither an estimate nor an interval is reported for corneal staining, because the primary outcome data were not collected.
A significance test, if one had been planned and performed, would address compatibility of the observed data with a specified statistical null hypothesis under the test’s assumptions. It would not by itself measure clinical importance or establish that a treatment caused an outcome. No primary significance test or p-value is reported. The record does not report a statistical analysis plan, so the null hypothesis, test, model, and any adjustment strategy are also not reported in the registry record.
How should the planning assumptions be read?
The registry describes planning illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. It also describes the effect to detect as a stated fraction of the observed effect. These are assumptions for planning calculations, not observations from the trial and not evidence that a particular effect was present. Such calculations are not observed power.
For this study, the intended allocation was between eyes within each participant, while the planning description refers to equal allocation between arms. The record does not explain how that planning setup corresponds to the paired eye-level design. It also does not report a primary endpoint effect estimate, and it states that primary data were not collected. A planning illustration tied to an observed effect therefore cannot be interpreted as a result establishing an effect for the primary endpoint in this record.
More generally, a sample-size calculation depends on assumptions about the outcome, the effect worth detecting, variability, allocation, and the analysis. If assumptions do not match the study’s actual design or data, the resulting illustration may not characterize the study’s ability to answer its question. The registry does not report the original design justification or a complete statistical plan, so the planning assumptions should not be taken as a substitute for those details.
Why can over-dispersion matter?
As a general statistical point, over-dispersion occurs when observed variability exceeds what a chosen model assumes. If ignored, it can make uncertainty estimates too small and tests too optimistic. This issue can arise in some count or categorical outcome models, but the registry does not report a fitted model, outcome distribution, or analysis for which over-dispersion was assessed. It is therefore a general consideration, not a reported problem in this study.
Similarly, model choice should follow the measurement scale and design. For a paired eye comparison, the analysis should preserve participant pairing. If repeated measurements over time were collected, the method might also need to account for within-eye and within-participant correlation. The registry identifies an 18-month timeframe for the primary endpoint but does not describe repeated primary measurements, a longitudinal model, or any planned handling of correlation.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 7 outcome measures, including corneal staining score as the primary outcome and symptom assessment, tear break-up time, and Schirmer testing among the secondary measures. When multiple outcomes or analyses are examined, the chance of at least one apparently unusual finding can increase. Statistical plans may address this through a hierarchy of outcomes, adjusted testing, or a clearly designated primary comparison. The registry record does not report a multiplicity strategy, and it reports no analyses. Because no primary outcome data were collected, there is no primary result against which to evaluate secondary findings.
The record states that participants withdrew themselves, failed to show for appointments, or were lost to follow-up. It does not report the number of missing measurements by reason, the timing of loss, whether missingness differed between the eye assignments, or whether any outcome data were available before withdrawal. It also does not report an approach for handling missing data, such as an imputation method or sensitivity analysis. These details are important because conclusions from an analysis of incomplete data depend partly on why values are missing and on the assumptions used to address them.
The analysis population is not reported in the registry record. In particular, the record does not say whether the intended analysis would include all randomized participants, only participants with measurements, or another defined group. It does not report how deviations from assigned eye treatment would have been handled. With no primary observations and zero reported analyses, neither an intention-to-treat analysis nor a per-protocol analysis can be identified from the entry.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Arm I | 0 | 8 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Arm I | 2 of 8 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Arm I |
|---|---|---|
| Uveitis (bilateral) | Eye disorders | 1 of 8 |
| Bacterial Infection (bilateral) | Eye disorders | 1 of 8 |
The adverse-event record lists Arm I with 8 participants at risk for serious adverse events and 0 serious adverse events. It lists 8 participants at risk for other adverse events and 2 other adverse events. These are reported counts. They are not a comparison of cyclosporine-treated eyes with artificial-tear-treated eyes, and the record does not link the events to a specific eye treatment.
The registry does not provide the terms describing the 2 other events; the serious-event terms list is empty. It also does not report event timing, severity details for the other events, whether events were considered related to treatment, or how events were collected. Deaths are not reported in the registry record. The adverse-event information therefore has limited detail and should not be treated as a complete characterization of safety.
A zero count for reported serious events means that none were recorded in this entry for the stated at-risk group. It does not establish that a treatment has no risk, particularly when the study enrollment was 8 and the record does not describe the completeness or duration of safety follow-up. The other-event count is likewise descriptive and cannot establish causation or comparative risk.
12. What the Result Does, and Does Not, Mean
The record documents a randomized study intended to compare topical cyclosporine A 0.05% with artificial tear for preventing or limiting graft-versus-host disease-related dry eye. The intervention description assigns the treatments to different eyes of the same participant, and the primary endpoint is corneal staining score at 18 months. These details define the research question and planned comparison; they do not constitute an answer to that question.
The primary outcome data were not collected, according to the registry, because participants withdrew, did not attend appointments, or were lost to follow-up. There is no reported difference in corneal staining, no estimate of how large any difference might be, and no confidence interval describing its precision. There is also no reported primary analysis. Accordingly, the record cannot show whether the treatments differed, whether they were similar, or whether either changed the course of dry eye.
Absence of an outcome result is not evidence of no effect. It is also not evidence of an effect. A non-significant result, had one been reported, would itself require interpretation in light of its uncertainty; in this record even that result is unavailable. Likewise, the listed secondary outcomes cannot be used to infer an overall treatment effect because the record reports no analyses and does not state whether secondary measurements were collected.
The safety entry is limited to the adverse-event counts recorded for Arm I. It does not compare assigned eye treatments or establish that events were caused by a particular treatment. The absence of reported serious events should not be interpreted as proof of absence of risk. The record’s terminated status and its missing outcome data leave important questions unresolved, including why the trial ended and whether any unreported measurements exist.
The appropriate conclusion is therefore narrow: the registry describes an intended paired, randomized comparison, but reports no primary outcome result with which to assess its question. Neither the planned design nor the adverse-event counts justify a claim about treatment efficacy or ineffectiveness.
13. Why This Trial Matters Statistically
This trial is statistically notable because the intended comparison is within participants, at the level of the eye. Paired randomized designs can be efficient: each participant contributes information under both treatment assignments, allowing shared characteristics to be controlled through the comparison. That advantage depends on preserving the pairing in the analysis. A statistical method that treated the eyes as unrelated would fail to reflect the design.
The registry labels the model single group, while the intervention descriptions specify that one eye receives cyclosporine and the other artificial tear. This makes clear why registry design labels should be read alongside the allocation details. It also underscores the importance of reporting the unit of randomization, the unit of analysis, and how paired observations are handled.
The record also illustrates the distinction between a planned endpoint and an analyzable result. Corneal staining score is identified as the primary endpoint, but no data were collected for it. Without observed measurements, an analysis cannot quantify an effect or its precision. A clearly stated endpoint and a randomized design cannot compensate for an absent outcome dataset.
Finally, the entry emphasizes why missing-data reporting matters. Withdrawals, missed appointments, and loss to follow-up can leave the intended comparison unanswered. Knowing how many observations are missing, when they became unavailable, and whether missingness differs by treatment can help readers assess potential bias. Those details, along with the reason for termination and the analysis plan, are not reported in the registry record. The statistical importance of the study lies in these design and reporting issues, not in a treatment result that the record does not contain.
14. 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 limitation is that no primary endpoint data were collected, so there is no result for corneal staining score at 18 months. The registry attributes this to participant withdrawals, missed appointments, or loss to follow-up, but does not give counts by reason or indicate when measurements became unavailable. It is also not reported whether any outcome data were collected before follow-up ended.
The design description is incomplete and partly difficult to reconcile. The model is labeled single group, whereas the intervention descriptions specify randomization of the two eyes to different treatments. The record does not explain the randomization process, the masking roles, the analysis method for paired eyes, or how treatment adherence and deviations would be handled. It does not define the corneal staining scale or report a statistical analysis plan.
The record identifies secondary measures but reports zero analyses. It does not say whether those outcomes were measured, how multiple outcomes would be handled, or which analysis population was intended. The reason for termination is not reported in the registry record. The adverse-event entry provides counts but lacks event descriptions, treatment attribution, and detailed ascertainment information; deaths are not reported.
These limitations prevent assessment of the primary treatment contrast, its precision, the completeness of safety follow-up, and the potential impact of missingness. They also prevent readers from determining whether the planned paired design was implemented and analyzed as intended.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: The Prophylactic Use of Topical Cyclosporine A 0.05% to Prevent Graft Versus Host Disease Related Dry Eye — NCT00553735.
- Registry record last updated 2017-03-28; status last verified 2013-04.
- Results first posted 2017-03-28.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
This phase 4 randomized, triple-masked study was designed to evaluate topical cyclosporine A 0.05% for preventing the onset or progression of graft-versus-host disease-related dry eye. The intervention descriptions specify eye-level assignment: one eye received cyclosporine and the other artificial tear as placebo. The registry model label is single group. Enrollment was 8; the study started 2007-08, reached primary completion in 2010-12, and is listed as terminated.
The primary endpoint was corneal staining score at 18 months. The registry reports that no primary data were collected because participants withdrew, missed appointments, or were lost to follow-up. It reports zero analyses. The adverse-event entry lists 0 serious events and 2 other events among 8 at risk; event descriptions and treatment attribution are not reported. The reason for termination and the statistical analysis plan are not reported in the registry record. No primary treatment-effect estimate is available.