Valproic Acid, Radiation, and Bevacizumab in Children With High Grade Gliomas or Diffuse Intrinsic Pontine Glioma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Valproic Acid, Radiation, and Bevacizumab in Children With High Grade Gliomas or Diffuse Intrinsic Pontine Glioma (NCT00879437), a na phase 2 study. The primary outcome is 1-year Event Free Survival (EFS).

2. Clinical Question

3. Trial Design

This was a phase 2, single-group study sponsored by Baylor College of Medicine. The registry lists no allocation and no masking. Enrollment was 38, and the status is completed. The study began on 2009-09-01, with primary completion on 2015-08-30. The registry record describes a treatment sequence for children with newly diagnosed high-grade gliomas or brainstem gliomas, including diffuse intrinsic pontine glioma.

The intervention combined valproic acid with radiation, then continued valproic acid and added bevacizumab during maintenance. The radiation phase is described as weeks 1–6, with daily valproic acid and radiation for approximately 6 weeks. In the post-radiation phase, described as weeks 7–10, valproic acid continued daily. Maintenance began in week 11 and included daily valproic acid and bevacizumab once every 2 weeks, continuing for a maximum duration of 2 years.

The primary outcome was one-year event-free survival, measured at 12 months and expressed as a percentage of participants. The stated objective was comparison with historical series, using ACNS0126 for high-grade gliomas and CCG-9941 for diffuse intrinsic pontine glioma. Because the design is single-group, it does not provide randomized treatment groups with which to separate treatment effects from differences in prognosis, patient selection, or care over time. The registry record does not report the historical estimates, the analysis method for that comparison, or a detailed statistical analysis plan.

The record lists 45 outcome measures and 4 analyses. Secondary outcomes include median event-free survival, median overall survival, partial response in diffuse intrinsic pontine glioma, partial response in high-grade gliomas, and complete response in high-grade gliomas. The registry record does not report numerical estimates for these outcomes in the registry record for this analysis.

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: A total of 38 eligible patients, 20 diffuse intrinsic pontine glioma (DIPG), and 18 high-grade glioma (HGG), were enrolled from five institutions.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 1 patient with DIPG not evaluable due to early death (did not complete valproic acid and radiation) related to a surgical complication (unrelated to protocol therapy) 1 patient with HGG not evaluable due to non-compliance with protocol therapy

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: 20 participants with diffuse intrinsic pontine gliomas; 18 patients with high-grade gliomas

6. Primary Endpoint

The registry lists the primary outcome as 1-year Event Free Survival (EFS), measured over 12 months.

The primary endpoint is one-year event-free survival, reported as a percentage of participants at 12 months. Event-free survival describes whether participants remain alive and free of a defined event through a specified time. It is used in cancer studies because it can capture disease progression or other events that occur before death, depending on the protocol’s event definition. The particular event definition is not reported in the registry record.

A percentage at a fixed time point gives a direct summary of the proportion meeting the event-free criterion by that time, while a confidence interval communicates uncertainty around the estimate. The registry identifies a 95% confidence interval as the measure of dispersion. It also states that the primary objective was comparison with historical series, with separate historical references for high-grade gliomas and diffuse intrinsic pontine glioma.

The endpoint description records two participants as not evaluable: one participant with diffuse intrinsic pontine glioma after an early death related to a surgical complication, and one participant with high-grade glioma because of non-compliance with protocol therapy. The registry record does not report the one-year event-free survival estimate, its confidence interval, the number of events contributing to the estimate, or how the not-evaluable participants were handled in the primary analysis. Consequently, the endpoint’s intended meaning is clear, but its observed result and precision are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a fixed-time event-free survival percentage represent?

In general, event-free survival describes the time from a defined starting point until a qualifying event or death, with the exact event definition depending on the protocol. A fixed-time percentage, such as the endpoint at 12 months in this study, summarizes the estimated share of participants who remain event-free through that time. It is not the same as a count of participants who experienced an event, because time-to-event methods can account for differing follow-up and censoring. The registry record does not report the event definition, the survival-estimation method, or the event and censoring details for this endpoint.

The primary endpoint is described as a percentage of participants, with a 95% confidence interval. In general, a confidence interval gives a range of estimates reasonably compatible with the observed data under the statistical method used. It conveys precision: a narrower interval indicates greater precision than a wider interval. The interval is not reported in the registry record, so the precision of the primary estimate cannot be assessed from the available record.

Why does the single-group design matter?

In general, random allocation can help balance measured and unmeasured prognostic factors between treatment groups. This study was single-group, with allocation listed as not applicable and masking listed as none. It therefore has no randomized concurrent control group. The registry describes a historical comparison, using ACNS0126 for high-grade gliomas and CCG-9941 for diffuse intrinsic pontine glioma. A historical comparison can provide context, but differences in eligibility, patient mix, outcome definitions, follow-up, or care practices can affect the contrast. The registry record does not report sufficient details to evaluate those potential differences or the statistical method used for the historical comparison.

How should a confidence interval be distinguished from a significance test?

In general, a confidence interval describes uncertainty in an estimate, whereas a significance test evaluates data against a specified null hypothesis. Neither can be interpreted without knowing the method and the comparison being made. The registry entries for this study identify a 95% confidence interval for the primary endpoint but do not report a primary estimate, interval, p-value, or complete test specification. No conclusion about statistical significance can therefore be drawn from the available primary-outcome information.

What does the planning illustration mean?

In general, sample-size calculations ask how many participants or events might be needed under assumed conditions to detect a specified effect with a chosen error rate and power. The registry record describes a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It defines the effect to detect as a stated fraction of the observed effect. For time-to-event outcomes, the illustration is described in terms of total events using the Schoenfeld method; other outcomes are described in participants per arm.

These are planning assumptions, not a description of this trial’s single-group enrollment plan. The study’s design is not a two-arm randomized comparison, so an equal-allocation illustration does not turn the actual study into one. Planning calculations are conditional on their assumptions and should not be treated as observed power or as evidence that a clinical effect was established. The registry record does not report the trial’s own sample-size justification or enough numerical inputs in the registry record to interpret a particular planning result.

Why can over-dispersion matter?

In general, over-dispersion occurs when variability in count data exceeds what a statistical model assumes. If ignored, standard errors and confidence intervals can be too small, making results appear more precise than they are. The registry entries do not identify a count-data model or report an over-dispersion assessment for this trial. This general issue should not be mistaken for a method documented in the registry record.

What should be known about multiple outcomes?

When a study examines many outcomes or analyses, the chance of at least one apparently unusual result can increase if each is assessed independently. Statistical plans may address this through prespecified primary outcomes, adjusted testing, or an explicit exploratory interpretation. The registry lists 45 outcome measures and 4 analyses, but a multiplicity approach is not reported in the registry record. The number of listed measures alone does not establish which analyses were confirmatory or how multiplicity was handled.

For this study, the essential methods information for interpreting the primary endpoint would include the endpoint event definition, analysis population, treatment of the two participants noted as not evaluable, censoring rules, estimation method, and details of the historical comparison. These are not reported in the registry record. Without the numerical primary estimate and these analytic details, readers can identify the intended method framework only in broad terms, not reproduce or fully assess the primary analysis.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Diffuse Intrinsic Pontine Glioma", "High-grade Gliomas")
n      <- c(19, 17)
events <- c(2, 4)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Diffuse Intrinsic Pontine Glioma  19  2  10.5  1.3  33.1  2.9  31.4  3.2  31.7
ARM  High-grade Gliomas  17  4  23.5  6.8  49.9  9.6  47.3  9.7  47.6

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 45 outcome measures and 4 analyses. The primary endpoint is one-year event-free survival, while secondary outcomes include median event-free survival, median overall survival, partial response in diffuse intrinsic pontine glioma, partial response in high-grade gliomas, and complete response in high-grade gliomas. A multiplicity strategy—such as adjustment across outcomes or a distinction between confirmatory and exploratory analyses—is not reported in the registry record. The number of outcomes alone does not show whether or how multiplicity was addressed.

The primary endpoint description identifies two participants as not evaluable. One participant with diffuse intrinsic pontine glioma had an early death related to a surgical complication, described as unrelated to protocol therapy, and did not complete valproic acid and radiation. One participant with high-grade glioma was not evaluable because of non-compliance with protocol therapy. The registry record does not report how either participant was handled in the primary analysis, whether the analysis used an intention-to-treat or other defined population, or whether there were additional missing endpoint observations.

Rules for censoring, missing data, and analysis populations are not reported in the registry record. This leaves uncertainty about which participants contributed to the primary estimate and how incomplete follow-up or non-evaluability may have affected it. The record also does not report sensitivity analyses for missing data. These omissions are important for interpreting an event-free survival estimate, but the estimate itself is also not reported in the registry record.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: radiation necrosis (2.6%), amylase and lipase elevation (2.6%), subacute bone infarction (2.6%), intratumoral hemorrhage (2.6%).

The adverse-event record lists a group titled “38 Eligible Patients,” with 38 at risk for serious adverse events and 38 at risk for other adverse events. It reports 4 affected for serious adverse events and 33 affected for other adverse events. The record also lists 37 deaths. These are registry-reported counts; the available information does not describe the timing, attribution, or relationship of each event to study treatment. The counts should not be interpreted as evidence that treatment caused the events.

The most frequently listed serious adverse event terms are radiation necrosis, amylase and lipase elevation, subacute bone infarction, and intratumoral hemorrhage, each reported as 2.6%. The registry record does not give further detail about event grades, duration, management, recurrence, or causal assessment for these terms. A listed percentage describes the reported frequency in the registry, not the certainty of a treatment-related effect.

Safety counts need their denominators and definitions. The record supplies the at-risk counts for the listed group, but it does not clarify in the registry entries how adverse events were collected or whether events were counted by participant or occurrence across all reporting. The 37 deaths are not, by themselves, evidence about treatment attribution or the primary endpoint. The primary endpoint description separately notes one early death related to a surgical complication and described as unrelated to protocol therapy. The registry record does not provide a complete narrative linking individual deaths and adverse events.

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

The registry record establishes that a phase 2, single-group study enrolled 38 children with newly diagnosed high-grade or brainstem gliomas and tested a sequence of valproic acid, radiation, and maintenance valproic acid with bevacizumab. It identifies one-year event-free survival at 12 months as the primary endpoint and names historical series for the intended comparison. It also describes two participants as not evaluable for specified reasons.

Those design details do not establish what the primary result was. The one-year event-free survival estimate, its 95% confidence interval, the event count, and the numerical historical comparison are not reported in the registry record. The direction and size of any difference from historical outcomes, and the precision of that difference, cannot be inferred. The record’s description of an intended comparison is not itself evidence of a favorable or unfavorable result.

The single-group design also means that any observed outcome would need to be interpreted in context rather than as a randomized estimate of treatment effect. A historical comparison may be affected by differences between study populations, follow-up, endpoint definitions, and care. The available record does not provide enough information to assess these factors or to determine whether the historical groups were comparable to the enrolled participants.

Similarly, the safety counts describe reported events, not causal attribution. The record lists serious and other adverse events among eligible participants and identifies several serious event terms. It does not report detailed event-level information or a complete attribution assessment. The death count cannot independently answer whether the regimen affected survival or safety.

The defensible conclusion is therefore limited: the registry describes the study question, treatment sequence, enrollment, and some safety reporting, but it does not report the numerical primary efficacy result in the registry record for this analysis. No claim that the regimen did or did not improve outcomes follows from these data alone.

14. Why This Trial Matters Statistically

This study raises several important statistical issues in a pediatric oncology setting. First, its primary question is framed around a fixed-time event-free survival outcome and a historical comparison. That makes clear definitions of the event, follow-up, censoring, and analysis population essential. The registry identifies the endpoint at 12 months and a 95% confidence interval, but those methods and the numerical estimate are not reported in the registry record.

Second, the single-group design distinguishes descriptive outcome reporting from a randomized treatment comparison. A historical benchmark can be informative, but its interpretation depends on whether the study participants and historical patients are sufficiently comparable and whether endpoints were defined and measured consistently. The record names the historical series but does not report the details needed to evaluate comparability or the numerical contrast.

Third, the study reports two participants as not evaluable for different reasons: an early death related to a surgical complication and non-compliance with protocol therapy. Such cases make transparent analysis-population and missing-data rules particularly relevant. Those rules are not reported in the registry record, so readers cannot determine how the exclusions influenced the primary estimate.

Finally, the registry lists 45 outcome measures and 4 analyses. This makes it important to know which outcomes were prespecified as primary, how secondary analyses were classified, and whether multiplicity was addressed. The primary endpoint is named, but the multiplicity strategy is not reported. These are not merely technical details: they affect how confidently different results can be interpreted and compared. The record’s missing numerical efficacy result prevents assessment of the primary outcome itself.

15. Limitations

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

The primary limitation is that the numerical one-year event-free survival result is not reported in the registry record. The 95% confidence interval, event count, and numerical comparison with the named historical series are also absent. This prevents assessment of the primary outcome’s magnitude, direction, and precision.

The study was single-group, with no randomized concurrent comparator. Its stated historical comparison is vulnerable to differences in participant characteristics, eligibility, endpoint definitions, follow-up, and clinical practice. The registry record does not report the details needed to judge whether the historical populations were comparable or how the comparison was conducted.

The record identifies two participants as not evaluable, but does not report the analysis population, handling of those participants, censoring rules, missing-data procedures, or sensitivity analyses. The event definition for event-free survival is also not reported in the registry record. These gaps constrain interpretation and reproducibility.

The registry lists 45 outcome measures and 4 analyses, but does not report a multiplicity strategy or identify the status of each analysis as confirmatory or exploratory. The record also does not report the trial’s own sample-size justification, interim analyses, or a complete statistical analysis plan. The planning illustration described in the registry is not a substitute for those study-specific details.

Safety reporting includes counts and several serious event terms, but the registry entries do not give event-level timing, grading, attribution, or collection methods. These limitations mean the registry description can characterize the design and selected reported safety information, but cannot fully support an efficacy or causal safety assessment.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Valproic Acid, Radiation, and Bevacizumab in Children With High Grade Gliomas or Diffuse Intrinsic Pontine Glioma — NCT00879437.
• Registry record last updated 2021-07-21; status last verified 2021-06.
• Results first posted 2021-07-21.
• 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.

19. Record Summary

This completed phase 2 study, sponsored by Baylor College of Medicine, enrolled 38 children with newly diagnosed high-grade or brainstem gliomas. It used a single-group design without masking. Treatment consisted of valproic acid with radiation, followed by continued valproic acid and maintenance bevacizumab. The primary endpoint was one-year event-free survival at 12 months, with a stated comparison to historical series for high-grade gliomas and diffuse intrinsic pontine glioma.

The registry record notes two participants as not evaluable for specified reasons and lists 45 outcome measures and 4 analyses. It reports adverse-event counts and several serious event terms. The numerical primary endpoint estimate, its confidence interval, details of the historical comparison, and key analysis procedures are not reported in the registry record. The study description therefore defines the clinical question and design but does not provide enough information to characterize the primary efficacy result.