Vorinostat in Treating Patients With Progressive or Recurrent Glioblastoma Multiforme: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Vorinostat in Treating Patients With Progressive or Recurrent Glioblastoma Multiforme (NCT00238303), a randomized phase 2 study. The primary outcome is Proportion of Successes (Patients Alive and Progression-free).

2. Clinical Question

3. Trial Design

This was a phase II trial of vorinostat in patients with progressive or recurrent glioblastoma. The registry lists the allocation as randomized, the masking as none, and the model as parallel. Enrollment was 103. The study began in 2005-09, reached primary completion in 2008-07, and is listed as completed. The sponsor was the National Cancer Institute.

The intervention descriptions distinguish patients undergoing surgery from those not undergoing surgery. Stratum 1 patients were described as not undergoing surgery and received oral SAHA twice daily for 2 weeks, with courses repeated every 3 weeks in the absence of progression or unacceptable toxicity. Stratum 2 patients were described as undergoing surgery: SAHA began before the operation, patients then underwent tumor-removal surgery, and postoperative SAHA was to begin within 1-4 weeks. The postoperative course description is truncated in the registry record. The results also include Stratum 3, described as not undergoing surgery; a complete treatment description for this stratum is not reported in the registry record.

The primary analysis population is specifically described as 66 Stratum 1 patients who started treatment. The record says 68 Stratum 1 patients were enrolled and 2 did not receive treatment. Results for Stratum 2 and Stratum 3 use 15 and 20 patients, respectively. Although the overall trial allocation is labelled randomized, the record does not explain the randomization procedure or how the reported strata relate to that allocation. The original statistical analysis plan, including its intended comparisons and any planned interim analyses, 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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 68 Stratum 1 patients were enrolled. 2 stratum 1 patients did not receive treatment. Therefore, the remaining 66 patients were analyzed.

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 Proportion of Successes (Patients Alive and Progression-free), measured over At 6 months.

The primary endpoint is the proportion of participants who were alive and progression-free at 6 months. It combines survival status with disease status at a specified time: a patient counts as a success only if both conditions are met. This kind of endpoint gives a time-specific summary of whether patients are alive without recorded progression, rather than describing tumor response alone.

The registry describes the estimate as the number of successes divided by the total number of evaluable patients. It specifies a binomial point estimator and an exact 95% confidence interval. Progression was defined using changes in measurable disease or, for evaluable disease that was not bidimensionally measurable, an unequivocal increase in enhancement or mass effect as agreed upon by the primary physician and quality control physicians, or the appearance of new lesions. For bidimensionally measurable disease, the definition includes a >25% increase in the product of perpendicular diameters or new lesions.

The endpoint is useful for conveying a clinically interpretable milestone, but it does not describe when progression occurred within the interval or how long patients remained progression-free after 6 months. The registry’s definition and estimate therefore need to be read as a specific time-point proportion, not as a complete account of survival or disease course.

7. Results Posted in the Registry

• Stratum 2 (Undergoing Surgery): 26.7% (4/15) versus 15.2% (10/66) on control; risk difference 11.5 percentage points (95% CI -12.5 to 35.5), risk ratio 1.76 (0.64–4.86); unadjusted two-proportion z-test p = 0.287.
• Stratum 3 (Not Undergoing Surgery): 10.0% (2/20) versus 15.2% (10/66) on control; risk difference -5.2 percentage points (95% CI -20.9 to 10.6), risk ratio 0.66 (0.16–2.77); unadjusted two-proportion z-test p = 0.560.

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.

Confirmed Tumor Response

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Stratum 3 (Not Undergoing Surgery): 0.0% (0/20) versus 3.0% (2/66) on control; risk difference -3.0 percentage points (95% CI -7.2 to 1.1); unadjusted two-proportion z-test p = 0.431.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Stratum 3 (Not Undergoing Surgery) and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists survival, confirmed tumor response, and time to progression as secondary outcomes. Numerical results are posted for confirmed tumor response, while detailed numerical results for survival and time to progression are not reported in the registry record. The confirmed-response results therefore provide only a partial view of the secondary outcome set and should not be treated as a complete account of the trial’s secondary findings.

The response outcome addresses whether a tumor response was confirmed, whereas the primary endpoint asks whether a patient was both alive and progression-free at 6 months. These outcomes are related through disease status, but they are not interchangeable: a patient may be progression-free without a confirmed response, and response status alone does not capture survival at the primary time point. The secondary results therefore add context but do not directly verify or negate the primary result.

Secondary outcome findings generally carry less weight when the analysis plan and multiplicity approach are not reported. Testing or examining several outcomes and groups can increase the chance of apparently notable findings arising by chance. The record does not report a multiplicity adjustment or a hierarchy that would identify which secondary outcome was intended to be confirmatory. The reported secondary comparisons should consequently be read as descriptive or exploratory rather than as independent confirmation of the primary result.

The absence of posted survival and time-to-progression estimates limits any assessment of consistency across outcomes. It is not appropriate to infer their results from the primary endpoint or from the response outcome. Likewise, a lack of a reported analysis is not evidence that no events occurred. The secondary record supports a cautious, incomplete interpretation: it describes the scope of outcomes assessed and offers some response data, but does not provide a full set of results capable of establishing a consistent pattern across endpoints.

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

• Survival
• Confirmed Tumor Response
• Time to Progression

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary: each evaluable patient either met the definition of alive and progression-free at 6 months or did not. For a binary outcome, the basic proportion is the number meeting the definition divided by the number evaluated. The registry explicitly describes this binomial point estimator. It reports 15.2% for the treated Stratum 1 population, whose analysis denominator is 66. The confidence interval expresses uncertainty around that estimated proportion rather than a range containing a fixed share of individual patient outcomes.

The registry specifies an exact binomial method for the 95% confidence interval. Exact methods are based on the binomial distribution and are commonly used for proportions, including when the number of patients or observed successes is limited. The interval can be asymmetric around the point estimate; that is not an error. It reflects the discrete nature of counts and the limits of plausible underlying proportions given the observed data. The resulting intervals are broad for the smaller reported strata, consistent with the limited information contributed by those group sizes.

Progression is not simply a judgment based on one universally stated measurement. The registry definition distinguishes bidimensionally measurable disease from evaluable disease that is not bidimensionally measurable. For measurable disease, it describes a >25% increase in the product of perpendicular diameters of contrast enhancement or mass, or the appearance of new lesions. For evaluable disease, it describes an unequivocal increase in contrast enhancement or mass effect, with agreement by the primary physician and quality control physicians, or new lesions. This operational definition matters because classification as progression determines whether a patient can count as progression-free at the endpoint.

The posted comparisons use risk differences. A risk difference subtracts the reference group’s event proportion from the comparison group’s event proportion; its sign indicates which reported proportion is larger. The intervals for these recalculations are on the risk-difference scale, and an interval containing zero does not rule out either a lower or higher underlying proportion in the comparison group. The registry also includes unadjusted p-values for these comparisons. They are recalculated from posted summary statistics, and the record does not identify them as sponsor-reported inferential analyses. No adjustment for other factors is described for these calculations.

The risk differences and p-values require particular caution because the record presents Stratum 1, Stratum 2, and Stratum 3, but does not explain how the strata map to randomized assignment. The registry’s overall design label is randomized; it does not, by itself, show that the surgery and non-surgery strata form exchangeable groups. Differences may reflect patient selection or other features not described in the record. A statistical comparison of proportions cannot resolve that design uncertainty.

Confidence intervals and p-values answer related but different questions. A confidence interval conveys the range of effect estimates compatible with the data under the method’s assumptions and indicates precision. A p-value measures how unusual the observed result, or a more extreme result, would be under a stated null hypothesis and model. A p-value is not the probability that the null hypothesis is true, and it does not measure clinical importance. Here, the broad intervals are important for interpretation even where a single p-value is available.

The registry lists secondary outcomes of survival, confirmed tumor response, and time to progression. Confirmed tumor response has posted results, including estimates and intervals for some groups, while response for the surgery group is not estimable from the posted result because its reported denominator is zero. The record gives no reported analyses for that secondary outcome. Detailed survival and time-to-progression estimates are not reported in the registry record. The presence of outcome names in the registry is not equivalent to having complete numerical results or a prespecified analysis for each outcome.

The statistical analysis plan, including model choices beyond the described binomial estimator, handling of censoring, treatment of missing outcome status, covariate adjustment, and interim analyses, is not reported in the registry record. These details should not be inferred from the trial’s phase or design labels. Similarly, the registry does not report the original sample-size justification or the assumptions used in the trial’s own design. The separate sample-size illustrations use stated planning assumptions and are not evidence of the trial’s original design rationale or its achieved power.

11. Confidence Intervals and Precision

The primary estimate for Stratum 1 is 15.2%, with a 95% confidence interval of 7.5% to 26.1%. This interval indicates substantial uncertainty around the estimated proportion: values across a broad range remain compatible with the observed data under the exact binomial method. It should not be interpreted as a prediction interval for individual patients or as a guarantee that the underlying proportion lies within those limits.

The intervals for the other reported strata are also broad. Stratum 2 has an estimate of 26.7%, with limits from 7.8% to 55.1%; Stratum 3 has an estimate of 10.0%, with limits from 1.2% to 31.7%. The relatively small group sizes contribute to limited precision, particularly for Stratum 2. The point estimates differ, but the uncertainty intervals overlap broadly. Overlap is not itself a formal test of a difference; the recalculated risk-difference intervals provide a direct estimate of uncertainty for the comparisons.

For Stratum 2 versus Stratum 1, the recalculated risk-difference interval extends from -0.125 to 0.355. For Stratum 3 versus Stratum 1, it extends from -0.209 to 0.106. Both intervals include zero and encompass effects in opposite directions. Thus, the comparisons are imprecise and do not distinguish a difference from no difference. The trial’s stratification and allocation details are also insufficiently described to support a strong causal interpretation of these contrasts.

Precision is not the same as validity or importance. A narrow interval would describe statistical precision under the analysis assumptions, while design limitations could still affect whether an estimate answers a causal question. Conversely, a wide interval does not show that there is no effect; it shows that the data leave considerable uncertainty about its size and direction.

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 illustrations use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set to a stated fraction of the observed effect. The entries give participants per arm for 80% power and participants per arm for 90% power. They are planning illustrations based on the observed effect, not the original trial’s design assumptions, which are not reported in the registry record.

The 80% power column gives the number of participants per arm estimated to be needed for an 80% chance of detecting the specified effect under those planning assumptions, if that effect is truly present. The 90% power column describes the corresponding planning target for a 90% chance. These are prospective design calculations, not claims about what the completed trial achieved. This is not observed power.

Rows use different fractions of the observed effect. Planning for 75% of the observed effect or 50% of the observed effect means targeting a smaller effect than planning for 100% of the observed effect. Smaller effects are harder to distinguish from random variation, so the illustrations require more participants per arm as the target effect gets smaller. The displayed requirements can be large, particularly for the comparison with a smaller observed difference.

The figures apply only to the stated comparison and assumptions. They do not account for a different allocation scheme, losses to follow-up, covariate adjustment, multiplicity, or a different statistical model unless such features are explicitly included; those features are not part of the stated illustration. For time-to-event outcomes, the record’s general note says sample-size entries would refer to total events using the Schoenfeld approach, while other entries refer to participants per arm. These particular illustrations are expressed as participants per arm.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.152; treatment proportion p₁ = 0.267; average p̄ = 0.209.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5751 = 1.1272.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5693 = 0.4791.
4. n = (1.1272 + 0.4791)² / (0.1152)² = 194.6, rounded up to 195 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 <- 0.151515151515152                          # control proportion
p1 <- 0.266666666666667                          # 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(238303)
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 = 195 (80% power), 261 (90% power)
 75% of observed effect: n per arm = 329 (80% power), 440 (90% power)
 50% of observed effect: n per arm = 699 (80% power), 935 (90% power)
power.prop.test: 194.6 
Simulated power at n = 195 per arm: 0.806 (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.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple outcomes: the primary six-month alive-and-progression-free proportion, survival, confirmed tumor response, and time to progression. Comparisons are also presented across the reported strata. When several outcomes or group comparisons are examined, the chance of at least one apparently unusual result can increase. Statistical methods can address that multiplicity, for example by adjusting thresholds or defining a hierarchy of hypotheses, but no such approach is reported in the registry record.

The record does not provide a full statistical analysis plan. It does not report whether secondary outcomes were intended as confirmatory or exploratory, whether subgroup comparisons were prespecified, or whether interim analyses were planned. The number of analyses reported in the registry is zero, despite the presence of posted outcome results and recalculated comparisons. The recalculated unadjusted p-values should therefore not be mistaken for a documented sponsor analysis or for multiplicity-adjusted evidence.

The primary population description is informative about treatment initiation: 68 Stratum 1 patients were enrolled, 2 did not receive treatment, and the remaining 66 were analyzed. However, the record does not describe how missing six-month status was handled, whether all patients had complete follow-up at that time, or whether any censoring or imputation method was used. The exact binomial method is specified for the interval, but that alone does not describe the handling of missing outcome data.

The registry’s allocation label is randomized, but the relationship between randomization and the three reported strata is not explained. Details of sequence generation, concealment, stratification factors, and analysis by assigned group are not reported in the registry record. The posted population is described as patients who started treatment for Stratum 1, which is not necessarily the same as an analysis of all randomized patients. The record does not report an intention-to-treat analysis or a per-protocol definition.

These omissions do not establish that particular methods were absent from the study; they mean that the registry record does not document them. They limit assessment of selection, missing-data bias, and the risk of chance findings across multiple outcomes. Interpretation should rest on the reported endpoint definitions and estimates while recognizing that important analysis details are unavailable.

15. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: headache (13.3%), platelet count decreased (10.6%), central nervous system necrosis (6.7%), hydrocephalus (6.7%), disease progression (5.0%).

The registry reports serious adverse events for each of the three groups, with 13 of 66 participants affected in Stratum 1, 3 of 15 in Stratum 2, and 4 of 20 in Stratum 3. It also reports other adverse events affecting 65 of 66, 14 of 15, and 19 of 20 participants, respectively. These are counts of participants affected, not counts of separate events, so a participant could potentially contribute to more than one reported event term.

The listed serious adverse-event terms include headache at 13.3%, platelet count decreased at 10.6%, central nervous system necrosis at 6.7%, hydrocephalus at 6.7%, disease progression at 5.0%, and muscle weakness left-sided at 5.0%. The registry does not give additional context in these figures about severity grading, timing, duration, attribution to treatment, or overlap among terms. Death counts are not reported in the posted safety data; a missing value should not be read as zero.

These descriptive counts do not establish comparative safety. The group sizes differ, the record does not explain the stratum assignment, and no adjusted safety comparison is reported. The counts should be interpreted as the registry’s recorded safety information, not as a basis for individual treatment decisions.

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

The result describes the proportion of the analyzed Stratum 1 patients who were alive and progression-free at the six-month assessment, together with uncertainty around that proportion. It is a time-specific outcome in a phase II trial, not a complete measure of long-term survival, tumor control, quality of life, or the experience of every patient. The endpoint depends on the registry’s progression definitions and the assessment of evaluable disease.

The estimates for the three reported strata are not, on their own, proof that one approach is superior to another. Although the registry labels the overall allocation randomized, it does not explain how that allocation relates to the surgery and non-surgery strata in the results. In particular, the posted contrasts cannot establish that a difference is caused by surgery, by the timing of vorinostat, or by vorinostat itself. The recalculated comparison intervals include zero and are broad, leaving considerable uncertainty about the size and direction of any contrast.

The registry record also does not report the original sample-size rationale, detailed handling of missing outcomes, complete statistical analysis plan, or numerical results for all listed secondary outcomes. Those gaps restrict the conclusions that can be drawn from the posted information. The secondary response outcome is a different measure from being alive and progression-free at 6 months, and the unavailable survival and time-to-progression results cannot be inferred.

Accordingly, the data support a descriptive account of the recorded endpoint estimates, not a definitive causal comparison or a claim of clinical benefit or lack of benefit. The uncertainty intervals, group sizes, population definitions, and incompletely documented design all matter. The statistical results do not replace clinical judgment and should not be used as medical advice.

17. Why This Trial Matters Statistically

This trial illustrates why a registry’s design label and its outcome presentation must be considered together. The study is labelled randomized, yet its results are organized into surgery and non-surgery strata without a reported explanation of how those strata relate to randomization. Recognizing that distinction prevents a descriptive difference between groups from being treated automatically as a causal treatment effect.

The primary endpoint also demonstrates how an intuitive proportion can still be uncertain. The 6-month alive-and-progression-free measure is straightforward to describe, but the confidence intervals are broad, especially in the smaller groups. Reporting an estimate alongside its interval makes the limits of the data visible and discourages overinterpretation of a single percentage.

The record provides enough information to understand the primary endpoint definition, the exact binomial interval method, and the analyzed Stratum 1 population. At the same time, it leaves key questions about the statistical plan, missing outcomes, multiplicity, and secondary analyses unresolved. This contrast is instructive: a numerical result can be reproducible from registry summaries while still lacking the documentation needed to assess all aspects of its inferential status.

Finally, the sample-size illustrations show how design targets depend on the effect one wishes to detect and on explicit planning assumptions. They are not evidence about the power achieved by the completed trial and are not the original trial’s stated design justification. Keeping those distinctions clear is central to reading clinical trial statistics responsibly.

18. 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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

Several limitations are specific to the registry information for this trial. The primary analysis uses 66 Stratum 1 patients who started treatment, after 2 of the 68 enrolled Stratum 1 patients did not receive treatment. The record does not explain whether the primary analysis was intended to include all randomized participants, how the two untreated patients were handled in the trial’s original analysis, or whether any patients lacked six-month outcome data.

The design description is incomplete for interpreting group comparisons. The overall trial is labelled randomized, but the record does not describe the allocation procedure or explain why the results distinguish Stratum 1 and Stratum 3 as not undergoing surgery. It also does not provide a complete treatment description for Stratum 3. Consequently, the estimates should not be treated as a clean randomized comparison of surgery and no surgery, or of different treatment schedules.

The surgery group is small, and its confidence interval is wide. The Stratum 3 estimate is also based on a relatively small group. The resulting uncertainty makes point estimates unstable as summaries of the underlying proportions. The recalculated pairwise comparisons are unadjusted and do not account for potential differences between strata.

Other limitations include the absence of a reported original sample-size justification, statistical analysis plan, multiplicity strategy, missing-data approach, and interim-analysis information. Although survival and time to progression are listed as secondary outcomes, their numerical results are not reported in the registry record. Safety details are limited to participant counts and selected serious adverse-event terms; death counts are not reported. These documentation gaps constrain interpretation without proving that the underlying study did not conduct additional analyses.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Vorinostat in Treating Patients With Progressive or Recurrent Glioblastoma Multiforme — NCT00238303.
• Registry record last updated 2014-05-23; status last verified 2011-10.
• Results first posted 2013-07-10.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

NCT00238303 is a completed phase II trial of vorinostat in progressive or recurrent glioblastoma, with enrollment of 103 and primary completion in 2008-07. The registry labels the design randomized, unmasked, and parallel. Treatment descriptions include vorinostat alone in a non-surgery stratum and vorinostat around surgery in a surgery stratum.

The primary outcome is the proportion alive and progression-free at 6 months. The analyzed Stratum 1 population comprised 66 patients who started treatment. The registry reports binomial estimates with exact 95% confidence intervals and also presents results for two other strata. Recalculated unadjusted comparisons are imprecise and do not establish differences. Secondary outcome reporting is incomplete, and the statistical plan, missing-data methods, and original sample-size rationale are not reported in the registry record.