Perfluorohexyloctane (NOV03) for the Treatment of Signs and Symptoms of Dry Eye Disease (DED): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Perfluorohexyloctane (NOV03) for the Treatment of Signs and Symptoms of Dry Eye Disease (DED) (NCT03333057), a randomized phase 2 study. The primary outcome is Corneal Fluorescein Staining (CFS) Total (National Eye Institute (NEI) Grading).

2. Clinical Question

3. Trial Design

This was a Phase 2, randomized, double-masked, parallel-group study of perfluorohexyloctane (NOV03) for Dry Eye Disease (DED). The registry record lists 336 participants enrolled. Three groups appear in the posted primary outcome results: NOV03 2 times daily (BID), NOV03 4 times daily (QID), and Placebo BID/QID. The intervention descriptions identify the placebo solution as saline (0.9% sodium chloride solution), with separate BID and QID schedules.

Random allocation is intended to distribute participant characteristics across assigned groups without systematic selection by investigators. Double masking means the registry characterizes the study as masked for two parties, but the identities of those parties and the practical masking procedures are not reported in the registry record. A parallel-group design assigns participants to a group for the study comparison rather than describing a crossover between regimens.

The record identifies Novaliq GmbH as sponsor. The study status is completed; the listed start date is 2018-01-03 and primary completion date is 2018-05-18. Those dates describe the registry timeline and do not, by themselves, establish when each participant was assessed or how long each participant was followed.

The primary outcome concerns total corneal fluorescein staining using the National Eye Institute grading system, with assessment described at week 8 and a posted timeframe of baseline and 2 months. The record lists one outcome measure and no posted secondary outcomes. It does not report a detailed recruitment flow, the number randomized to each group separately from the outcome counts, or a detailed analysis population. The protocol, sample-size justification, and statistical analysis plan are 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

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 Corneal Fluorescein Staining (CFS) Total (National Eye Institute (NEI) Grading), measured over baseline and 2 months.

Corneal fluorescein staining (CFS) is a clinical grading measure of staining on the corneal surface. The registry specifies the National Eye Institute (NEI) grading scale and says that each of five areas on each cornea is graded from 0 to 3. A grade of 0 indicates no staining, and the maximum total score for each eye is 15. Higher values describe greater staining and corneal damage.

The primary outcome is titled Corneal Fluorescein Staining (CFS) Total (NEI Grading). The registry describes the primary efficacy variable as comparison of both absolute values and change from baseline between treatments at week 8. The posted timeframe is baseline and 2 months. The posted entries are mean estimates with 95% confidence intervals. Because the registry wording includes both absolute and change-from-baseline comparisons, interpretation of the listed estimates depends on the precise analysis definition, which is not fully detailed in the registry record.

This endpoint is used to quantify an observable sign of dry eye disease rather than relying only on reported symptoms. A standardized grading scheme gives the outcome a defined scale and direction: higher scores indicate more staining. It is one component of the clinical picture, not a complete account of symptoms, daily function, or all aspects of eye health. The record does not report a threshold defining a clinically important change.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean describe?

General statistical explanation: a mean is an average that summarizes the observed values in a group. It is useful for describing a continuous measure such as a staining score, but it compresses the distribution into a single estimate. It does not show whether participants' scores were clustered closely around that average or varied widely. The registry identifies the outcome parameter as a mean and gives confidence intervals as the dispersion information; it does not provide a standard deviation or a detailed distribution of individual scores.

In this trial, the posted mean estimates are -1.78 for NOV03 BID, -2.11 for NOV03 QID, and -0.93 for Placebo BID/QID. These are group-level estimates, not individual participant outcomes. The registry's description says the primary efficacy variable included absolute values and change from baseline. The precise correspondence between that description and each posted mean is not fully explained, so the estimates should be interpreted with that limitation in mind.

What does a 95% confidence interval communicate?

General statistical explanation: a confidence interval describes uncertainty around an estimate under the assumptions of the analysis. A narrower interval indicates greater precision for that estimate; a wider interval indicates less precision. It is not a range containing every participant's result, and it is not the probability that a particular treatment effect lies within the interval.

The registry reports 95% confidence intervals for each group mean: -2.27 to -1.3 for NOV03 BID, -2.59 to -1.63 for NOV03 QID, and -1.41 to -0.45 for Placebo BID/QID. These intervals describe uncertainty for the listed group estimates. They are not intervals for the difference between groups. A direct comparison requires an estimated contrast and its uncertainty, neither of which appears among the posted outcome results.

Why does the comparison matter?

General statistical explanation: a trial question about whether groups differ is answered by estimating a contrast, such as a difference between a treatment group and its comparator, with an uncertainty interval and an appropriate statistical test. Separate group means can help describe the data, but they do not replace that contrast. Comparing the means informally or subtracting one from another does not account for the analysis model, participant-level variation, or any adjustment specified by the study's plan.

The posted record has no analysis entries and identifies no direct comparison result for the continuous outcome. It also does not report a p-value for the primary comparison. The statistical analysis plan, including any model or adjustment used for the primary question, is not reported in the registry record. Consequently, the posted arm estimates support description but do not establish the inferential result of the planned treatment comparison.

What does randomization contribute?

General statistical explanation: random allocation is designed to make assigned groups comparable, on average, with respect to measured and unmeasured participant characteristics. It supports causal interpretation of a properly conducted comparison, but does not guarantee identical groups in a particular trial. Randomization also does not remove the need to report who was analyzed, how missing data were handled, or how the outcomes were compared.

The registry identifies allocation as randomized and the study model as parallel. It does not describe sequence generation, concealment procedures, or the analysis population. Those operational details are not reported in the registry record, so the design labels should not be expanded into assumptions about implementation.

How should the sample-size illustrations be read?

General statistical explanation: sample-size planning asks how many participants would be needed under specified assumptions to detect a target effect with a chosen significance level and power. Such figures are planning illustrations, not evidence that a trial had a particular probability of detecting its observed result. A calculation based on an effect observed in the same trial is especially dependent on that estimate and its uncertainty.

The record describes planning assumptions based on a fraction of the observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation. It distinguishes participant counts per arm for ordinary outcomes from total event counts for time-to-event outcomes. These assumptions describe an illustrative calculation, not the original trial design. The original sample-size justification is not reported in the registry record. No planning count should be read as observed power or as proof that the enrolled sample was adequate for the primary comparison.

Why can variation beyond a simple model matter?

General statistical explanation: statistical models rely on assumptions about the outcome and its variation. If observed variation is greater than a model expects, uncertainty may be understated unless the analysis accounts for that extra variation. This issue is often described as over-dispersion, although the registry record does not report that over-dispersion was present or that a model requiring such an adjustment was used. It is a general consideration, not a finding about this trial.

For this continuous outcome, the record reports mean estimates and confidence intervals but does not describe the distribution of scores, the model used, or diagnostic checks. It is therefore not possible to determine from the registry record whether a particular variance assumption was evaluated or whether an alternative analysis was considered.

Overall, the statistical information posted is limited to group-level means and confidence intervals for the primary outcome, accompanied by group counts. The record does not contain the analysis details needed to reproduce or fully evaluate the primary between-group inference. General principles about means, confidence intervals, and randomized comparisons clarify what the entries represent, but they cannot fill in the missing analysis specifications.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists one outcome measure and no posted secondary outcomes. It lists no analysis entries. This means the posted record does not provide a set of secondary outcome results or an account of the statistical tests performed. It does not establish that no other outcomes or analyses were considered during the study.

Multiplicity arises when several hypotheses or comparisons are tested. General statistical explanation: as the number of tests increases, the chance of at least one apparently positive result can increase unless the analysis plan addresses the issue. The record does not report whether there were multiple primary comparisons, whether any adjustment was planned, or how the two NOV03 regimens were to be compared with saline. The multiplicity strategy is not reported in the registry record.

Missing-data methods are also not described. The posted outcome counts are 111 for NOV03 BID, 114 for NOV03 QID, and 111 for Placebo BID/QID, but the record does not explain whether these represent all randomized participants, participants with observed endpoint data, or another analysis population. It does not report the number or reasons for missing measurements, imputation procedures, or sensitivity analyses. Therefore, the effect of missing outcome information on the estimates cannot be assessed from the registry record.

The record does not identify an intention-to-treat, per-protocol, or safety population for the primary efficacy analysis. It also does not report a prespecified hierarchy of tests, interim analyses, or a detailed statistical analysis plan. These items are not reported in the registry record and should not be inferred from the study's randomized, double-masked design.

11. 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 1% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: chest pain (0.9%), diverticulum (0.9%), endometrial adenocarcinoma (0.9%), abortion spontaneous (0.9%).

The registry reports serious adverse event counts among participants at risk: 1 for NOV03 BID, 3 for NOV03 QID, and 0 for Placebo BID/QID. The respective numbers at risk are 111, 114, and 111. The record lists zero deaths in each group. These are counts, not comparative estimates adjusted for follow-up or participant characteristics. The registry does not provide a statistical comparison of serious adverse events between groups.

The record also lists other adverse event counts of 1 for NOV03 BID, 4 for NOV03 QID, and 3 for Placebo BID/QID, with the same group sizes at risk. Serious and other adverse event counts should be read as separate reported categories; the registry does not explain whether any participant could be included in both categories or how events were classified and collected.

The listed serious adverse event terms are Chest pain, Diverticulum, Endometrial adenocarcinoma, and Abortion spontaneous, each reported as 0.9%. The registry excerpt does not assign those terms to a specific group or provide further event-level detail. The percentages should therefore not be used to infer group-specific risks. Sparse counts and the absence of comparative analyses limit what can be concluded from these entries; they do not establish equivalence between groups or prove that an event is related to a study intervention.

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

The results establish what the registry record reports: a randomized, double-masked Phase 2 study with three listed primary-outcome groups, mean estimates for corneal fluorescein staining, and 95% confidence intervals for those estimates. Higher scores on the stated NEI scale indicate greater staining and corneal damage. The outcome entries are negative, but the registry description does not fully clarify how the posted values correspond to the stated absolute and change-from-baseline analyses. That ambiguity matters when interpreting the direction and meaning of the estimates.

The results do not provide the direct difference between either NOV03 regimen and saline, a confidence interval for such a difference, or a p-value for the primary comparison. The group-level confidence intervals cannot substitute for a treatment contrast. Without the comparison estimate and analysis details, the precision of any treatment-versus-placebo difference cannot be determined from the posted results.

The record also does not report a threshold for a clinically important change, detailed participant characteristics, an analysis population, missing-data methods, or the statistical plan. Consequently, the mean estimates cannot show whether a change would be important to participants, how representative the analyzed participants were, or how robust the results are to missing outcome data and analysis choices.

Safety entries list serious adverse event and other adverse event counts, and zero deaths in each group. They are descriptive counts rather than evidence of a statistically established safety difference. The registry does not report enough detail to attribute the listed events to a treatment or to evaluate differences over a defined period.

These limitations do not negate the information that is posted; they define its scope. The record supports a description of the trial design, endpoint, group-level outcome estimates, and selected safety counts. It does not support a complete assessment of the primary comparative effect, the clinical importance of any change, or a broad conclusion about efficacy or safety.

13. Why This Trial Matters Statistically

This study illustrates why a randomized design and a reported treatment contrast are distinct parts of statistical interpretation. Randomization is a design feature intended to support fair comparisons. To understand the result, readers also need the estimated difference between groups, its uncertainty, and a description of the analysis used. The registry record identifies randomization and provides group-level outcome estimates, but it does not post the direct primary comparison.

The endpoint also illustrates the value of stating the scale's direction. The NEI grading description says that higher values represent greater staining and corneal damage. Without that direction, the sign of a mean estimate would be difficult to interpret. At the same time, the record's description of both absolute values and change from baseline shows why an outcome's analysis definition should be explicit: the meaning of a negative estimate depends on what quantity was summarized.

Confidence intervals help communicate precision, but their interpretation depends on which quantity they surround. Intervals around each group's mean do not answer the same question as an interval around a between-group difference. This distinction is important when registry results are used to assess comparative evidence.

The record further demonstrates how incomplete reporting can limit statistical assessment even when a trial design and outcome are identified. Details about missing data, analysis populations, multiplicity, and the original sample-size justification are not reported in the registry record. Their absence prevents readers from evaluating those aspects of the analysis, rather than proving that a particular method was or was not used.

Finally, the safety counts are useful as descriptive information but do not supply comparative estimates or establish causation. Taken together, the record is most informative when design labels, endpoint definitions, estimates, uncertainty, and reporting gaps are considered separately. That approach avoids treating an arm-level summary as a complete answer to the trial's comparative question.

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 outcome entry does not fully resolve whether the posted mean estimates represent absolute scores or change from baseline. The description refers to both types of comparison, while the timeframe is listed as baseline and 2 months and the results show negative means. This limits interpretation of the exact quantity represented by each estimate.

Only group-level means and confidence intervals are posted for the primary outcome. The registry record does not report the direct contrast between the NOV03 regimens and saline, a confidence interval for that contrast, or a p-value. It also does not describe the statistical model or adjustments used. The estimates therefore do not permit a full evaluation of the primary treatment comparison.

The record does not report participant characteristics, eligibility details, the number screened or randomized, or a participant flow. Although outcome counts are listed for each group, the analysis population and the relation of those counts to the randomized population are not specified. Missing measurements, their reasons, and methods for addressing them are not reported in the registry record.

The record lists no secondary outcomes and no analysis entries. The protocol, statistical analysis plan, multiplicity approach, interim analyses, and original sample-size justification are not reported in the registry record. It is not possible to determine from the posted information whether these matters were addressed elsewhere or what methods were used.

Safety reporting is limited to counts, deaths, and a short list of serious adverse event terms with percentages. Group attribution for those terms, event timing, exposure duration, and comparative analyses are not reported. These reporting limits mean that the posted safety information should be treated as descriptive rather than as a complete assessment of relative safety.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Perfluorohexyloctane (NOV03) for the Treatment of Signs and Symptoms of Dry Eye Disease (DED) — NCT03333057.
• Registry record last updated 2022-01-04; status last verified 2019-10.
• Results first posted 2021-05-24.
• 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.

18. Record Summary

The registry record describes NCT03333057, a completed Phase 2 randomized, double-masked, parallel-group study of NOV03 for Dry Eye Disease (DED). It lists Novaliq GmbH as sponsor and 336 enrolled participants. The listed study start date is 2018-01-03, and primary completion is listed as 2018-05-18.

The primary outcome is total Corneal Fluorescein Staining using the National Eye Institute grading scale, with assessment described at week 8 and a posted timeframe of baseline and 2 months. Group mean estimates and 95% confidence intervals are reported for NOV03 BID, NOV03 QID, and Placebo BID/QID. The record lists serious adverse event counts, other adverse event counts, and zero deaths in each group.

The posted results do not include a direct treatment-versus-placebo estimate or a detailed statistical analysis plan. Analysis populations, missing-data methods, and the multiplicity approach are not reported in the registry record.