Study of the Effect of Intravenous AVE0005 (VEGF Trap) in Advanced Ovarian Cancer Patients With Recurrent Symptomatic Malignant Ascites: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of the Effect of Intravenous AVE0005 (VEGF Trap) in Advanced Ovarian Cancer Patients With Recurrent Symptomatic Malignant Ascites (NCT00327444), a randomized phase 2, phase 3 study. The primary outcome is Time to Repeat Paracentesis (TRP).

2. Clinical Question

The clinical question is therefore about the timing of repeat fluid drainage during the study period, not simply whether a participant ever underwent paracentesis. The registry also identifies safety, tolerability, paracentesis-related parameters, and participant-reported outcomes as secondary objectives. It lists the Area Under the Curve for the Participant Assessed Ascites Impact Measure, 60-day frequency of paracentesis, and plasma levels of free and VEGF-bound aflibercept among the secondary outcomes.

The study was randomized, double-blind, and placebo-controlled, so the primary contrast is between the assigned groups during the double-blind period. The open-label option after that period is part of the design description, but the registry’s primary endpoint timeframe is limited to Day 1 up to 6 months from randomization. The record does not report further details about how the open-label period was handled in analyses of the secondary outcomes.

3. Trial Design

This was a randomized, double-blind, parallel-arm study of intravenous aflibercept versus placebo in participants with advanced ovarian cancer and recurrent symptomatic malignant ascites. The official title describes a multicenter study and administration every 2 weeks. The registry lists the study as Phase 2 and Phase 3, with enrollment of 58 and completed status.

The double-blind period compared the aflibercept and placebo groups. The arm descriptions state that, during an open-label period, participants assigned to aflibercept could continue aflibercept or withdraw, while participants assigned to placebo could choose aflibercept or withdraw. This later option matters when interpreting the overall study course: the main randomized comparison is clearest for the double-blind period, whereas treatment choices in the open-label period were not randomized in the same way.

The primary endpoint analysis was defined for the intent-to-treat population, consisting of all randomized participants. The posted arm-level primary outcome estimates are based on 26 participants in the placebo group and 29 in the aflibercept group. The registry’s overall enrollment is 58. The reasons why the endpoint analysis denominators differ from overall enrollment are not reported in the registry record.

The registry identifies a primary objective of comparing the effect of aflibercept with placebo on repeat paracentesis. Secondary objectives include safety and tolerability, paracentesis-related measures, and participant-reported outcomes. The posted outcomes include the primary time-to-repeat-paracentesis measure and three secondary measures. The original sample-size justification, details of any interim analyses, and the full statistical analysis plan are not reported in the registry record.

The study began in 2006-07, and primary completion is listed as 2009-10. Those dates describe the registry’s study timeline; they do not establish the timing of individual participants’ follow-up or the duration of every analysis. The primary outcome timeframe is stated separately as Day 1 up to 6 months from randomization.

4. Participant Flow and Accounting

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

Before assignment: 55 were randomized (started population). 3 participants were not randomized but they were treated, and permanently withdrawn from the study due to disease progression (1 participant), fatal disease progression (1 participant) and a treatment emergent adverse event (1 participant).

Double Blind Treatment (DB) Period

Reasons for not completing this period:

Open Label Treatment (OL) Period

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The intent-to-treat (ITT) population - all participants who were randomized in the 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 Time to Repeat Paracentesis (TRP), measured over From Day 1 up to 6 months from randomization.

Time to Repeat Paracentesis (TRP) measures the number of days from randomization to the first post-randomization paracentesis. The registry specifies that the primary outcome was assessed from Day 1 up to 6 months from randomization. For participants who did not undergo a post-randomization paracentesis on study, TRP was calculated from randomization to the end of the double-blind treatment period.

This endpoint is used to describe how long participants went before another paracentesis was recorded. It is a continuous measure expressed in days, and the posted arm results are least-squares means with standard errors. A larger mean corresponds to a longer recorded time to the defined endpoint, but the interpretation is bounded by the registry’s endpoint rule: participants without a post-randomization paracentesis were assigned a time ending with the double-blind period.

That rule means the endpoint is not simply a count of procedures, nor is it a conventional time-to-event result as presented in the registry. The posted primary analysis reports mean values and an ANOVA comparison. The registry record does not report additional details about how incomplete follow-up, withdrawal, or other missing endpoint information was handled.

TRP was chosen to address the stated primary objective of comparing the effect of aflibercept and placebo on repeat paracentesis in symptomatic malignant ascites. As a general statistical point, a time-based measure can express delay in a clinically relevant event using a single continuous summary, but its interpretation depends on the event definition and how observations without an event are treated. In this trial, the no-paracentesis rule is therefore important to understanding what the reported mean represents.

7. Results Posted in the Registry

• Aflibercept: mean 55.10 versus 23.30 on control; difference 31.80 (95% CI 11.07 to 52.53), standardized effect d = 0.81; z-test p = 0.003.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Area Under the Curve (AUC) for Participant Assessed Ascites Impact Measure (AIM)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Aflibercept: mean 405.30 versus 29.80 on control; difference 375.50 (95% CI -28.24 to 779.24), standardized effect d = 0.66; z-test p = 0.068.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Aflibercept and the control. The interval gives the range of effects that remain compatible with the data.

60-Day Frequency of Paracentesis (FOP)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Aflibercept: mean 2.70 versus 5.10 on control; difference -2.40 (95% CI -4.26 to -0.54), standardized effect d = -0.68; z-test p = 0.011.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Aflibercept than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Taken together, the posted secondary measures offer additional views of ascites impact and paracentesis-related outcomes, alongside a pharmacologic measure of free and VEGF-bound aflibercept. The secondary estimates are not all equally precise, and the registry’s outcome list does not provide a complete basis for treating them as a single confirmatory test.

The frequency-of-paracentesis result points in a direction broadly aligned with the primary endpoint: the posted group means are lower in the aflibercept arm for frequency, while the primary mean time to repeat paracentesis is longer in that arm. These outcomes capture related but distinct aspects of paracentesis. Agreement in direction can provide context, but it does not make the measures interchangeable or independently establish the primary result.

The ascites impact AUC is less conclusive as a comparison because its sponsor-reported confidence interval includes zero and is broad. Its analyzed group sizes are smaller than those for the primary outcome. The plasma-level outcome is listed, but numerical results for it are not reported in the registry record. The record therefore does not support a statistical interpretation of that measure’s direction, magnitude, or precision.

Unadjusted secondary results carry less weight because several outcomes are examined and the registry does not report a multiplicity correction or a prespecified testing hierarchy. General statistical explanation: when many hypotheses are tested, the chance of at least one apparently notable result can rise, even if no corresponding effect exists. Without knowing which outcomes were confirmatory, which were exploratory, or whether adjustments were planned, secondary p-values should be interpreted cautiously. They may help describe patterns and generate questions, but they should not be read as independent proof that the primary finding is replicated.

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

• Area Under the Curve (AUC) for Participant Assessed Ascites Impact Measure (AIM)
• 60-Day Frequency of Paracentesis (FOP)
• Plasma Levels of Free and VEGF-bound Aflibercept

9. Statistical Interpretation

10. Statistical Methods Explained

The registry reports continuous outcomes as least-squares means and standard errors. A least-squares mean is an estimated group mean from a statistical model, rather than necessarily the simple arithmetic average of individual observations. The registry identifies ANOVA as the method for the sponsor’s reported comparisons. The model details beyond that label, including any covariates, are not reported in the registry record.

What does a mean difference describe?

A mean difference compares the estimated mean in one group with the estimated mean in the other. For the primary endpoint, the reported difference is aflibercept minus placebo, so a positive value indicates a longer estimated time to repeat paracentesis for the aflibercept group. For frequency of paracentesis, the reported difference is negative, which reflects a lower estimated frequency in the aflibercept group. The sign must always be interpreted with the outcome’s scale and the comparison’s direction in mind.

Means compress the data into a group-level summary. They do not show the full distribution, how much participants varied, or whether a small number of observations strongly influenced the estimate. The registry reports standard errors alongside the arm-level least-squares means. A standard error describes the estimated uncertainty in a mean, not the spread of individual participant values. General statistical explanation: a smaller standard error usually indicates a more precisely estimated mean, while a larger standard error indicates more sampling uncertainty.

What does a confidence interval add?

A confidence interval gives a range of estimates compatible with the observed data and the analysis procedure at the stated confidence level. The primary sponsor-reported 95% confidence interval ranges from 10.56 to 53.05 days for the mean difference. It stays on the positive side of zero, indicating that the reported range is consistent with a longer mean TRP in the aflibercept group under the stated comparison. A confidence interval does not give the probability that the true value lies within that particular interval. General statistical explanation: intervals are most useful when read for both their location and width, rather than as a binary significance label.

The recalculated intervals are based on the posted summary statistics and are simple unadjusted checks. In the primary analysis, the recalculated estimate is 31.8 days with an interval from 11.07 to 52.53. It is close to, but not identical to, the sponsor’s reported interval. Such small differences can arise because the registry’s reported model-based analysis and a summary-statistic re-calculation do not necessarily use identical information or procedures. The registry record does not provide enough model detail to reconstruct the sponsor’s calculation.

Why use a re-calculation?

A re-calculation can check whether the reported group summaries and comparison are broadly coherent. For these continuous outcomes, the re-calculation is unadjusted and uses the posted arm means and standard errors. The pooled standard deviation used for planning illustrations is estimated approximately from those standard errors and sample sizes. It should therefore be understood as an approximation, not as a participant-level reanalysis or a reconstruction of the sponsor’s model.

That distinction matters particularly for outcomes with small analysis denominators or substantial uncertainty. The AUC comparison has arm denominators of 15 and 16, and its recalculated interval spans zero. The sponsor’s reported interval also spans zero. A re-calculation can help describe the scale of uncertainty, but it cannot resolve questions about data collection, missingness, model specification, or assumptions that are not reported in the registry record.

Why does over-dispersion matter?

General statistical explanation: over-dispersion occurs when observed variability is greater than a model’s assumptions allow. When it is not accounted for, standard errors can be too small and confidence intervals too narrow, making estimates look more precise than they are. The registry reports continuous outcomes and standard errors, but it does not report an over-dispersion assessment or a specific adjustment for it. No conclusion about whether over-dispersion was present can be drawn from the registry record alone.

For count outcomes such as frequency of paracentesis, the outcome is a number of procedures, but the registry’s posted analysis treats it as a continuous measure and reports least-squares means and an ANOVA comparison. A count can have features such as skewness or extra variability. General statistical explanation: those features can affect how well a continuous-model approximation describes the data. The registry record does not report distributional diagnostics or alternative count-model analyses.

What can and cannot be inferred from a p-value?

A p-value measures how incompatible the observed data are with a specified null hypothesis, assuming the analysis model and its assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect. The recalculated primary p-value is 0.003, while the sponsor’s reported analysis gives a p-value in the registry record. The effect estimate and confidence interval are more informative about direction and scale; the p-value alone cannot describe how precisely the difference was estimated.

The record identifies ANOVA for the sponsor analyses, but does not report the full analysis plan, assumptions checks, multiplicity procedures, or interim-analysis details. These omissions prevent a complete independent audit of the inferential framework. They do not erase the posted results, but they limit what can be said about how the analyses were selected and whether every uncertainty source was addressed.

11. Confidence Intervals and Precision

The primary sponsor-reported difference is 31.8 days, with a 95% confidence interval from 10.56 to 53.05. The interval is entirely positive and spans a substantial range of possible mean differences. It supports a direction favoring a longer recorded time to repeat paracentesis in the aflibercept group, while also showing that the exact size of the difference is uncertain. It should not be read as a guarantee that the true effect is within those endpoints.

The simple unadjusted re-calculation gives the same estimated difference, 31.8 days, and a 95% confidence interval from 11.07 to 52.53. The closeness of the sponsor and recalculated intervals suggests that the posted summary statistics give a broadly similar picture of the primary contrast. The sponsor analysis remains the trial’s reported analysis; the re-calculation is not an independent participant-level fit.

Precision differs across outcomes. For the ascites impact AUC, the sponsor-reported interval ranges from -46.48 to 797.42, a broad range that includes zero. This leaves substantial uncertainty about the direction and size of the difference. The frequency-of-paracentesis interval ranges from -4.33 to -0.54 and does not include zero, although that interval still describes uncertainty in the magnitude of the estimate.

General statistical explanation: confidence-interval width depends on the amount of information and variability in the data, as well as the analysis method. Smaller analysis groups and greater variability tend to yield less precise estimates. The registry reports fewer analyzed participants for the AUC outcome than for the primary outcome, but it does not report why those denominators differ. Confidence intervals convey sampling uncertainty under the analysis; they do not automatically account for unreported missing-data mechanisms, model choices, or multiplicity.

12. Planning a Similar Trial: Sample Size

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

The sample-size table and curve are planning illustrations based on the observed effects, not the trial’s own original design calculation. For these continuous outcomes, the entries are participants per arm. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each row uses the stated fraction of the observed effect as the effect to detect.

The 80% power column gives the number of participants per arm that the illustration estimates would be needed for an 80% chance of detecting the specified effect under those assumptions, if that effect were the true effect and the assumed analysis conditions held. The 90% power column gives the corresponding per-arm number for a 90% chance. These are planning quantities, not outcomes observed in this trial.

The rows consider planning for the full observed effect and for smaller fractions of that effect. As the target effect becomes smaller, the required number of participants increases. This is a general feature of statistical planning: a subtler difference is harder to distinguish from random variation and generally requires more information. The entries depend on the observed effect, which itself is uncertain, so they should not be treated as definitive design recommendations.

The registry record does not report the trial’s original sample-size justification or its original assumptions. The planning illustrations use a normal approximation and equal allocation; they do not establish that the original trial used those assumptions. This is not observed power: it does not calculate the probability that the completed trial detected an effect after the results were known. The figures are prospective-style illustrations based on the posted effect estimates and specified planning assumptions.

13. How the Sample Size Is Derived

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

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 31.800; pooled standard deviation σ = 39.146.
2. n = 2 × (1.96 + 0.842)² × 39.146² / 31.800² = 23.8, rounded up to 24 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
delta <- 31.8                      # observed difference in means
sigma <- 39.1463750460408                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(327444)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 24 (80% power), 32 (90% power)
 75% of observed effect: n per arm = 43 (80% power), 57 (90% power)
 50% of observed effect: n per arm = 96 (80% power), 128 (90% power)
power.t.test: 24.8 
Simulated power at n = 24 per arm: 0.794 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

For the primary outcome, the sponsor’s reported ANOVA gives a least-squares mean difference of 31.8 days, with a 95% confidence interval from 10.56 to 53.05. The unadjusted re-calculation from the posted arm summaries also gives a difference of 31.8 days, with a 95% confidence interval from 11.07 to 52.53 and p = 0.003. The estimate is the same, while the interval limits differ modestly.

The re-calculation uses summary statistics and approximations; it does not reproduce the sponsor’s model or use participant-level data. The registry record does not report enough detail about the ANOVA model to reconstruct it fully. The sponsor’s reported analysis is therefore the appropriate reference for what the registry identifies as the trial analysis. The record does not flag a discrepancy between the primary sponsor result and the re-calculation.

For secondary outcomes, sponsor-reported estimates and intervals are also posted. The AUC estimate has a reported confidence interval that includes zero, while its reported p-value is also listed. That pairing should be read carefully because the interval and p-value do not convey the same simple signal in this record. The registry does not explain the difference in those inferential summaries. The frequency outcome has a reported negative difference and a confidence interval below zero. Neither secondary analysis changes the primary analysis or supplies information about unreported multiplicity procedures.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary outcome and multiple secondary outcomes, but the multiplicity strategy is not reported in the registry record. It is not reported whether secondary outcomes were tested with an adjustment for multiple comparisons, whether there was a hierarchy for testing them, or whether some were intended as exploratory. This matters because interpreting several unadjusted tests as if each were the only test can overstate the certainty of isolated findings.

The registry identifies the primary population as the intent-to-treat population, defined as all randomized participants. This is a useful statement about which participants belong to the primary analysis. However, the record does not report a full account of missing outcome data, exclusions from specific analyses, imputation methods, or sensitivity analyses. The differing denominators across posted outcomes make those details relevant, but they do not by themselves explain why the analysis populations differ.

For TRP, the registry gives a specific rule for participants with no post-randomization paracentesis on study: the time was calculated from randomization to the end of the double-blind treatment period. That rule is reported, but broader handling of incomplete observation or withdrawal is not. The open-label period allowed treatment choices or withdrawal, and the record does not report how such events were incorporated into each secondary analysis.

The registry reports ANOVA for sponsor comparisons, but the complete statistical analysis plan and model specification are not reported. Interim analyses and stopping rules are also not reported in the registry record. These omissions mean that readers cannot determine from the registry alone whether the displayed analyses were the only planned analyses, how analysis decisions were made, or whether additional sensitivity checks were performed. These are limits on interpretability, not evidence that any particular procedure was or was not used.

16. 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: disease progression (33.3%), vomiting (20.0%), dyspnoea (20.0%), dehydration (16.0%), abdominal pain (10.0%).

The registry reports serious adverse events for 18 of 25 participants at risk in the placebo group and 27 of 30 participants at risk in the aflibercept group. It also reports other adverse events for 22 of 25 participants at risk in the placebo group and 28 of 30 participants at risk in the aflibercept group. These counts describe the posted safety groups; the record does not provide enough information to establish causal attribution for individual events or to make a comparative safety conclusion.

The most frequent serious-event terms listed include disease progression at 33.3%, vomiting and dyspnoea at 20.0% each, dehydration at 16.0%, and abdominal pain and small intestinal obstruction at 10.0% each. The registry lists these as event terms and percentages. They should not be added together or interpreted as mutually exclusive categories; the record does not say that each term occurred in a different participant.

Deaths are not reported in the registry record. Serious adverse event counts are not a substitute for a full assessment of severity, timing, duration, relatedness, or exposure. The available figures describe the posted events, but do not by themselves establish whether a treatment caused an event or provide a complete safety comparison.

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

The primary registry result is an estimated difference in time to repeat paracentesis during the specified study period. The reported mean difference is 31.8 days, with a 95% confidence interval from 10.56 to 53.05. The interval lies above zero, and the posted group means are 23.3 days for placebo and 55.1 days for aflibercept. These statistics describe the trial’s randomized comparison under the registry-defined endpoint and analysis.

The result does not mean that every participant experienced the same delay, or that the group mean describes any individual participant’s course. It also does not establish a result beyond the measured population, endpoint definition, follow-up period, or analysis. For participants without an on-study post-randomization paracentesis, the registry assigned TRP through the end of the double-blind period; that definition is part of what the mean represents.

The secondary findings are not uniformly precise. The frequency measure points in a direction consistent with the primary paracentesis-related result, while the ascites impact AUC estimate has a broad interval that includes zero. The record does not report numerical plasma-level findings. Secondary analyses have less confirmatory weight when multiplicity procedures are not reported, and the record does not give enough detail to resolve all questions about their analysis populations or missing data.

The findings also do not establish a complete account of harms. The registry reports serious and other adverse event counts, along with selected serious-event terms, but deaths are not reported and event causality is not established by these counts. The trial record is therefore best read as a statistical account of posted outcomes and their precision, not as medical advice or a universal conclusion about treatment.

General statistical explanation: an estimate is most informative when its outcome definition, analysis population, uncertainty, and limitations are considered together. A confidence interval communicates uncertainty around an estimated contrast; it does not remove uncertainty about study conduct, missingness, multiplicity, or generalizability when those details are absent.

18. Why This Trial Matters Statistically

This trial illustrates why the estimand—the precise outcome being compared—matters as much as the numerical result. TRP is defined as time from randomization to the first post-randomization paracentesis, with a specific end-of-double-blind-period rule for participants without a post-randomization procedure. That rule shapes the meaning of the group means and prevents a reader from treating the endpoint as an unspecified time-to-event measure.

It also shows the value of reading estimates with their intervals. The primary sponsor estimate and the simple re-calculation are close, and both indicate a positive mean difference. The confidence interval makes clear that the exact magnitude remains uncertain. In contrast, a secondary AUC interval is broad and crosses zero. These examples show why statistical interpretation should consider magnitude, direction, precision, and endpoint meaning rather than relying on a single test result.

The record also demonstrates the distinction between randomized comparisons and later treatment choices. The double-blind period has the randomized placebo comparison; in the open-label period, participants could elect aflibercept or withdraw. Those phases do not have the same interpretive structure. Keeping that distinction clear helps avoid treating the entire follow-up as one unchanged randomized contrast.

Finally, the posted analyses show why analysis details matter. The registry identifies ANOVA and the primary intent-to-treat population, but does not report the full model, multiplicity plan, missing-data methods, or original sample-size justification. A statistical reader can describe the results that are posted while also identifying which parts of the inferential framework remain unknown. That is a central skill in registry-based trial interpretation.

19. 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 registry record does not report the original sample-size justification, so the planning illustrations cannot be taken as evidence of the trial’s intended design assumptions. The original statistical analysis plan, model covariates, interim analyses, and stopping rules are also not reported. These omissions limit independent assessment of how the analysis was planned and whether all analytic decisions were prespecified.

Although the primary population is identified as intent-to-treat, the registry does not report full missing-data procedures or reasons for the differences in outcome denominators. The primary endpoint’s special rule for participants without an on-study paracentesis is reported, but other handling of withdrawal and incomplete follow-up is not. The open-label option to continue treatment, start aflibercept, or withdraw also makes it important to know how each later outcome was analyzed; those details are not reported.

The AUC outcome has smaller posted analysis denominators than the primary endpoint and a broad sponsor-reported interval. The plasma-level endpoint is listed without numerical results in the registry record. The record also does not report a multiplicity strategy for the secondary outcomes, limiting the strength of conclusions based on separate secondary tests.

The simple re-calculations use posted summary measures, not participant-level data, and rely on an approximate conversion from standard errors for planning purposes. They cannot verify model assumptions, distributions, or the handling of individual observations. Finally, safety summaries include selected event counts and terms, but deaths are not reported, and event causality cannot be established from the listed counts alone.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of the Effect of Intravenous AVE0005 (VEGF Trap) in Advanced Ovarian Cancer Patients With Recurrent Symptomatic Malignant Ascites — NCT00327444.
• Registry record last updated 2013-01-01; status last verified 2011-07.
• Results first posted 2013-01-01.
• 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.

23. Record Summary

NCT00327444 is a completed, randomized, double-blind, placebo-controlled, parallel-arm study of intravenous aflibercept in advanced ovarian cancer with recurrent symptomatic malignant ascites. The registry lists Phase 2 and Phase 3, enrollment of 58, and a primary objective of comparing aflibercept with placebo on repeat paracentesis.

The primary endpoint was time to repeat paracentesis, defined from randomization to the first post-randomization paracentesis, with a specified end-of-double-blind-period rule for participants without one. The sponsor-reported mean difference was 31.8 days, with a 95% confidence interval from 10.56 to 53.05. The unadjusted re-calculation is similar in estimate and direction.

Secondary outcomes add context but vary in precision; multiplicity procedures, detailed missing-data methods, and the original sample-size justification are not reported in the registry record. Safety results include serious and other adverse-event counts, while deaths are not reported.