Comparison Study of ISV-303 to Durasite Vehicle in Cataract Surgery Subjects: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparison Study of ISV-303 to Durasite Vehicle in Cataract Surgery Subjects (NCT01808547), a randomized phase 3 study. The primary outcome is Ocular Inflammation.

2. Clinical Question

3. Trial Design

The registry describes a completed Phase 3, randomized, parallel-group study comparing ISV-303 with DuraSite Vehicle in cataract surgery subjects. The sponsor was Sun Pharmaceutical Industries Limited. The study began in 2013-05, had primary completion in 2013-11, and recorded enrollment of 268 participants.

The record identifies masking as quadruple. It does not specify which parties were masked, how masking was maintained, or whether any procedures differed between groups. The allocation ratio and the detailed randomization procedure are not reported in the registry record. Arm descriptions beyond the intervention names are also not reported.

The study summary describes the purpose as evaluating ocular safety, tolerability and efficacy of topical ISV-303 compared with DuraSite Vehicle. The posted primary outcome is an ocular inflammation measure at 15 days. It is a binary participant count: participants either met the stated grade-0 and no-rescue criteria or did not. The registry reports outcome counts for the modified intent-to-treat population using LOCF, while the safety summaries use their own at-risk counts.

Although the overall enrollment is 268, the primary outcome counts are 168 in the ISV-303 group and 85 in the vehicle group. The reasons those counts differ from overall enrollment are not reported in the registry record. The record does not give the original sample-size justification, planned recruitment by arm, or the trial’s full statistical analysis plan.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Modified Intent-to-Treat (ITT) Population using Last Observation Carried Forward (LOCF) method.

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.

Population: Modified Intent-to-Treat: subjects who were randomized into the trial, underwent cataract surgery, and received at least one dose of study drug

6. Primary Endpoint

The registry lists the primary outcome as Ocular Inflammation, measured over 15 days.

The primary endpoint counts participants who had anterior chamber cell grade 0 at Day 15 and had not used rescue medications. The grade uses a 0 to 4 scale: grade 0 means 0 cells; grade 1 means 1-10 cells; grade 2 means 11-20 cells; grade 3 means 21-50 cells; and grade 4 means > 50 cells. Thus, the endpoint treats grade 0 without rescue medication as the event of interest.

This kind of binary endpoint gives a direct count of participants who satisfy a clinically defined response criterion, rather than describing each participant’s cell count in detail. Requiring no rescue medication makes rescue use part of the endpoint definition, rather than a separate outcome. The registry does not explain why this criterion was selected or give a separate outcome for cell grade and rescue medication use.

The analysis population is described as Modified Intent-to-Treat (ITT) using LOCF. LOCF carries a participant’s last observed outcome forward when a later measurement is unavailable. The specific missingness rules and timing of any carried-forward observation are not reported in the registry record.

7. Results Posted in the Registry

• ISV-303: 38.1% (64/168) versus 22.4% (19/85) on control; risk difference 15.7 percentage points (95% CI 4.2 to 27.2), risk ratio 1.70 (1.10–2.65); unadjusted two-proportion z-test p = 0.012.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a binary endpoint represent?

In general, a binary endpoint records whether each participant meets a defined condition or does not. Here, the condition is anterior chamber cell grade 0 at Day 15 together with no rescue medication. The reported numbers are participants with that event and the group counts used for the outcome. A binary summary is straightforward to interpret as a proportion, but it does not retain all the detail of the original cell-grade scale. For example, participants who do not meet the endpoint are not separated in the posted primary result according to their grade or rescue-medication status.

What does the risk difference say?

In general, a risk difference expresses the difference between two event proportions on an absolute scale. The posted unadjusted estimate is 0.157, with an interval from 0.042 to 0.273. The group proportions are 0.381 for ISV-303 and 0.224 for vehicle. The risk-difference scale is useful because it expresses the group contrast in proportion units. It does not, on its own, establish why the groups differed or whether the difference is important in a clinical context.

The registry also reports relative measures. In general, a risk ratio compares the event proportion in one group with the event proportion in the other. The reported estimate is 1.704, with an interval from 1.097 to 2.648. An odds ratio compares odds rather than proportions; its reported estimate is 2.138, with an interval from 1.176 to 3.886. Risk ratios and odds ratios can differ in size, so they should be named explicitly and interpreted on their own scales. All of these estimates summarize the same binary outcome, not separate endpoints.

How should the p-value be read?

In general, a p-value summarizes how unusual data at least as inconsistent with a specified null model would be if that model and the test assumptions held. The registry lists an unadjusted p-value of 0.012. It should not be interpreted as the probability that the observed difference happened by chance, or as the probability that there is no difference. The registry record does not report the test used, the prespecified decision rule, or whether the p-value was adjusted for any other analyses. A p-value also does not quantify the size or precision of the effect; those are addressed by the estimate and interval.

Why does the analysis population matter?

The primary result is attributed to a Modified Intent-to-Treat (ITT) Population using Last Observation Carried Forward. In general, an analysis population defines which participants contribute to an estimate, and its definition can affect the meaning and generalizability of the result. LOCF is a missing-data procedure that substitutes a participant’s last recorded observation for a later missing outcome. This approach can make assumptions about how outcomes behave after the last observation. Its appropriateness depends on the missingness pattern and clinical context; the registry does not report enough information to assess those features. The detailed definition of the modified population and the number of outcomes carried forward are not reported in the registry record.

What does an interval add?

In general, an interval around an estimate represents uncertainty under a specified statistical procedure. It communicates a range of values compatible with the observed data and method, rather than a guarantee that the true effect lies within that range. The registry gives interval limits for the risk difference, risk ratio, and odds ratio, but does not identify the confidence level or interval-construction method. Comparisons across estimates should therefore retain that reporting limitation.

What is not established by the posted analysis?

The registry record reports one primary outcome measure and no posted secondary outcome measures or analysis records. That does not establish whether additional analyses were planned or conducted outside the record. The record also does not describe covariate adjustment, interim analyses, or a multiplicity strategy. In general, multiplicity concerns the increased chance of false-positive findings when several hypotheses are tested; methods such as adjustment or a prespecified testing hierarchy can address that issue. Whether such methods were used in this study is not reported in the registry record.

The posted comparison is an unadjusted recalculation from summary counts. It summarizes the observed group data without accounting for covariates or other design information that is not available in the reported counts. The discrepancy indicator is false, so no conflict is flagged between the posted summary and the recalculation. The registry does not provide additional sponsor analysis details that would support a broader claim about the analysis plan.

10. Confidence Intervals and Precision

The risk-difference estimate is 0.157, with interval limits of 0.042 and 0.273. The interval communicates uncertainty around the estimate: values closer to either limit remain compatible with the data under the procedure used to calculate it. The range does not mean that every value in it is equally likely, and it is not a guarantee about the true effect.

The registry does not identify the confidence level for this interval or describe its construction. That missing information limits how fully the interval can be interpreted. The relative estimates have their own intervals: 1.097 to 2.648 for the risk ratio and 1.176 to 3.886 for the odds ratio. They are on different scales and should not be compared as though the numerical distance between their limits had the same meaning as the risk-difference interval.

The outcome counts also matter for precision. The primary analysis includes 168 participants in the ISV-303 group and 85 in the vehicle group, with 64 and 19 meeting the endpoint, respectively. The registry does not state why the outcome counts differ from overall enrollment or provide a breakdown of missing observations. Since the primary population uses LOCF, readers cannot assess from this record how many observations were carried forward or how alternative missing-data approaches might affect the estimate.

Precision is not the same as importance. An interval describes statistical uncertainty under a method; it does not decide whether a difference is clinically meaningful. The registry record does not state a threshold for a clinically important difference.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning illustration gives participants per arm for this binary outcome, not total events. Its assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set as a stated fraction of the observed effect. These figures are planning illustrations based on the observed effect; the trial’s own original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column describes the number of participants per arm indicated by the illustration for an 80% chance of detecting the selected effect under its assumptions, if that effect and those assumptions were correct. The 90% power column describes the corresponding planning number for a 90% chance. These are conditional design calculations, not guarantees about what a future study will observe. The row for 100% of the observed effect gives 133 participants per arm for 80% power and 177 for 90% power. Planning for 75% of the observed effect gives 228 and 304 participants per arm; planning for 50% gives 490 and 656.

Smaller effects require more participants in this illustration because they are harder to distinguish from sampling variation. The curve should be read as a visual representation of this planning relationship, not as a record of the trial’s enrollment process. This is not observed power: it does not describe the probability that the completed study detected an effect, and it does not replace interpretation of the estimate and its interval.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.224; treatment proportion p₁ = 0.381; average p̄ = 0.302.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6494 = 1.2729.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6398 = 0.5385.
4. n = (1.2729 + 0.5385)² / (0.1574)² = 132.4, rounded up to 133 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 <- 0.223529411764706                          # control proportion
p1 <- 0.380952380952381                          # 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(1808547)
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 = 133 (80% power), 177 (90% power)
 75% of observed effect: n per arm = 228 (80% power), 304 (90% power)
 50% of observed effect: n per arm = 490 (80% power), 656 (90% power)
power.prop.test: 132.4 
Simulated power at n = 133 per arm: 0.811 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome measure and no posted secondary outcome measures. It lists no analysis records. These entries describe what is recorded in the registry; they do not establish that no additional outcomes or analyses were considered elsewhere. The statistical plan, including any hierarchy of hypotheses or multiplicity adjustment, is not reported in the registry record.

In general, multiplicity arises when more than one statistical hypothesis is tested. Without an appropriate strategy, repeated testing can raise the chance of at least one false-positive result. A prespecified hierarchy, adjustment, or other strategy may address this, but the record does not say whether any was used. The unadjusted p-value of 0.012 should therefore be understood as an unadjusted result, not as evidence that multiplicity was addressed.

The primary population is described as Modified Intent-to-Treat (ITT) with LOCF. That states a general approach to missing outcome data, but does not report the number of missing Day 15 observations, the number carried forward, or the timing and source of last observations. The modified population’s detailed inclusion and exclusion rules are also not reported. The registry does not describe sensitivity analyses for missing data or explain how rescue medication was handled when outcome information was incomplete.

Interim analyses, stopping rules, and the full analysis plan are not reported in the registry record. It is therefore not possible to assess whether the posted result reflects a prespecified endpoint and analysis, or whether other analysis decisions were made during the study. These gaps limit assessment of the robustness of the inference, without changing the descriptive interpretation of the posted counts.

14. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: colitis (1.2%), chest pain (0.6%).

The safety record reports serious adverse events affecting 1 participant among 170 at risk in the ISV-303 group and 1 among 85 at risk in the vehicle group. It also reports other adverse events affecting 17 among 170 at risk in the ISV-303 group and 5 among 85 at risk in the vehicle group. These are counts of participants affected, not event rates recalculated for comparison. The registry does not provide a comparative statistical analysis of these safety counts.

The listed serious adverse event terms are Colitis at 1.2% and Chest pain at 0.6%. The record does not specify further detail about these cases in the data available for this summary. Deaths are not reported in the registry record. Safety at-risk counts differ from the primary outcome counts, and the reason for the difference is not reported. These summaries are descriptive and do not establish whether an event was caused by a study intervention.

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

The posted primary outcome indicates that more participants in the ISV-303 group than in the vehicle group met the registry’s combined Day 15 criterion of anterior chamber cell grade 0 and no rescue medication. The unadjusted risk-difference estimate is 0.157, and its interval runs from 0.042 to 0.273. The registry also reports a p-value of 0.012 without adjustment. Together, these statistics describe the observed group contrast and its uncertainty under the reported calculation, but they do not provide a complete account of the trial’s analysis plan.

The result does not isolate the contribution of cell grade from rescue-medication use, because the endpoint requires both. Nor does the primary outcome count show the distribution of grades among participants who failed the criterion. The registry does not report secondary outcomes, further time points, or a fuller analysis of the individual endpoint components.

The result should not be read as a determination of clinical importance. The registry does not state a clinically meaningful difference threshold, and the confidence level and calculation method for the reported intervals are not identified. The p-value is not a measure of effect size and does not establish the probability that a treatment effect is real.

Interpretation is also bounded by the available population and missing-data information. The primary analysis used a modified intent-to-treat population with LOCF, but the record does not explain its full definition or how many observations were carried forward. The outcome counts are smaller than overall enrollment, and the reasons are not reported. These limitations make the posted comparison useful as a description of the recorded endpoint results, while leaving important questions about analysis choices and generalizability unanswered.

16. Why This Trial Matters Statistically

This trial is statistically notable for reporting a binary endpoint with both absolute and relative comparisons. The risk difference expresses the group contrast on an absolute scale, while the risk ratio and odds ratio express relative contrasts. Reading the estimates together helps distinguish the scale of the effect from the question of its precision. In this record, the reported risk-difference interval extends from 0.042 to 0.273, and the relative estimates have their own intervals.

The endpoint also illustrates how an operational definition shapes interpretation. Grade 0 alone is not the full outcome: absence of rescue medication is required as well. That combined definition gives the analysis a clear event criterion but means the group-level result cannot reveal which part of the criterion explains the observed contrast. This is a useful reminder that a statistical result always refers to the endpoint as defined, not to a broader concept by implication.

The sample-size illustration provides a separate lesson. It shows how planning needs change when the target effect is reduced, under its stated assumptions. It does not reconstruct the trial’s design, because the original justification is not reported in the registry record. Keeping that distinction clear prevents a planning exercise based on the observed estimate from being mistaken for evidence about the study’s original power.

Finally, the record’s incomplete detail on multiplicity, missing data, and analysis methods illustrates why a numerical result and an analysis plan are not interchangeable. The posted counts and unadjusted estimate can be described, while questions about prespecification, robustness, and the handling of incomplete outcomes remain unresolved.

17. Limitations

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

Several trial-specific reporting gaps limit interpretation. The record does not provide the original sample-size rationale or the assumptions used to plan enrollment. The planning illustration is based on the observed effect and should not be treated as a description of the trial’s original design.

The registry does not report the full randomization method, allocation ratio, masking procedures, or which parties were masked. It gives no further arm descriptions and does not detail dosing. The group sizes used for the primary outcome differ from total enrollment, but the reasons are not reported. The record also does not explain the difference between primary-outcome and safety at-risk counts.

The primary endpoint’s analysis population is identified as modified intent-to-treat with LOCF, but its exact membership and missing-data details are absent. No counts of missing observations, LOCF substitutions, or sensitivity analyses are reported. The interval confidence level and construction method, the inferential test, and a multiplicity strategy are not reported. There are no posted secondary outcomes or analysis records, and interim analyses or stopping rules are not described.

These limitations do not invalidate the posted counts, but they constrain what can be inferred from them. In particular, they prevent a full appraisal of the prespecified analysis, missing-data robustness, and how the result relates to other potential outcomes.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Comparison Study of ISV-303 to Durasite Vehicle in Cataract Surgery Subjects — NCT01808547.
• Registry record last updated 2021-11-19; status last verified 2021-11.
• Results first posted 2016-07-13.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

NCT01808547 is a completed Phase 3 randomized, parallel-group study sponsored by Sun Pharmaceutical Industries Limited. It compared ISV-303 with DuraSite Vehicle in cataract surgery subjects, with enrollment of 268 and quadruple masking recorded.

The primary outcome was the number of participants with anterior chamber cell grade 0 at Day 15 and no rescue medication, analyzed in a Modified Intent-to-Treat population using LOCF. The posted counts were 64 of 168 for ISV-303 and 19 of 85 for vehicle. The unadjusted risk-difference estimate was 0.157, with interval limits of 0.042 and 0.273; the unadjusted p-value was 0.012.

The registry does not report the trial’s original sample-size justification, full statistical plan, detailed missing-data information, or a multiplicity strategy. Safety summaries are descriptive and report serious and other adverse events by group.