Radiation Therapy in Treating Young Patients With Gliomas: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Radiation Therapy in Treating Young Patients With Gliomas (NCT00238264), a na phase 2 study. The primary outcome is Marginal-failure Rate.

2. Clinical Question

The central statistical question is therefore how the study defined and planned to monitor a specific type of recurrence, rather than whether radiotherapy performed better than a comparator. Because no numerical primary result is reported, the endpoint’s observed magnitude and precision cannot be assessed from the registry record. The listed secondary outcomes include progression-free, event-free, and overall survival probabilities, quality of life, and measures involving the MIB-1 labeling index. Numerical results for those outcomes are not reported in the registry record either.

3. Trial Design

This was a phase II, single-group study sponsored by the Children’s Oncology Group. The registry lists enrollment of 92 and a completed status. The recorded start is 2006-11, and primary completion is dated 2013-12-30. These are registry entries about the study’s timeline and enrollment, not numerical efficacy results.

Participants received reduced-field conformal radiation therapy. The treatment description specifies sessions 5 days a week for 6 weeks, subject to disease progression or unacceptable toxicity. Masking is listed as none. Allocation is listed as NA, consistent with the single-group model; there is no randomized allocation or control arm described in the record.

The primary analysis population is described separately from enrollment: all evaluable patients were included, with n = 84. The registry says an eligible patient who progressed during treatment was deemed not evaluable and excluded from the primary analysis per protocol. The record does not explain how the remaining difference between enrollment and the primary analysis population arose, so it should not be attributed to particular reasons.

The study concerns gliomas, and the official title specifies low-grade gliomas. The record lists Brain Tumor and Central Nervous System Tumor as conditions. It does not report more detailed eligibility criteria, participant demographics, sites, or treatment-delivery adherence. The design is useful for describing outcomes under a treatment regimen in the enrolled study population, but without a comparator it cannot isolate treatment effects from other explanations for observed outcomes. The registry’s stated endpoint monitoring approach is Bayesian, using a binomial likelihood and a Beta distribution for the parameter representing the proportion of failures that are marginal failures. Further details of the monitoring rule are not reported in the registry record.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All evaluable patients were included in this analysis (n = 84). One eligible patient was deemed to be not evaluable due to progression during treatment, and was excluded from the analysis per protocol.

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 Marginal-failure Rate, measured over Up to 5 years.

The primary endpoint is called the marginal-failure rate, with a timeframe of up to 5 years. Its description defines the quantity as the proportion of failures that are marginal failures. For this endpoint, patients without recurrence and patients with central or distant recurrence count as treatment successes. A marginal recurrence is treated as a failure. The registry does not give a numerical marginal-failure estimate, event count, or uncertainty interval.

This endpoint focuses on the location or pattern of failure, rather than treating every recurrence as equivalent. That can be relevant when a study is evaluating a radiation approach that changes the treated field: the endpoint distinguishes marginal recurrence from central or distant recurrence. This is a general explanation of the endpoint’s statistical role, not evidence about what occurred in this study.

The registry says the monitoring approach uses a binomial likelihood and a Beta distribution for the parameter p, the proportion of failures that are marginal failures. In general, a binomial model describes counts of an event among a defined set of opportunities, while a Beta distribution can represent uncertainty about an event probability. Here the parameter is conditional on failures, rather than a proportion of all enrolled participants. The registry record does not report the prior’s parameters, a decision threshold, or the numerical endpoint result, so the planned monitoring cannot be reconstructed in detail.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a Bayesian monitoring plan for the primary endpoint. It specifies a binomial likelihood and a Beta distribution on p, where p is the proportion of failures that are marginal failures. This is a model for the event pattern among failures; it is not a report of the endpoint’s observed value. The registry does not give the prior parameters, monitoring thresholds, or numerical posterior summaries.

What does the binomial likelihood represent?

In general, a binomial likelihood represents the probability of an observed event count under a specified event probability and a defined number of opportunities. For this study’s endpoint, the opportunities are failures, and the event of interest is that a failure is marginal. The likelihood therefore concerns the share of failures that fall into that category. It does not directly describe marginal failures as a share of all enrolled participants.

The endpoint definition classifies patients with no recurrence, central recurrence, or distant recurrence as successes, and marginal recurrence as a failure. That classification is specific to the primary endpoint. It should not be confused with a general definition of clinical success, or with the definitions of progression-free, event-free, or overall survival.

What does the Beta distribution contribute?

In general, a Beta distribution is a probability model for an unknown probability. A Bayesian analysis combines a prior distribution with the likelihood from observed data to produce a posterior distribution. The posterior can express uncertainty about p and can be used to evaluate a monitoring rule. The particular prior matters: different prior choices can affect the resulting posterior, especially when the observed data are limited. The registry says a Beta distribution was used but does not report its parameters, so the planned prior cannot be reproduced from the record.

The registry says monitoring will be based on a Bayesian rule, but does not specify the rule’s decision threshold or other operational details. It also does not report whether monitoring occurred at particular times, whether a boundary was reached, or what conclusion followed. Such details are not reported in the registry record and should not be filled in by assumption.

Why does the analysis denominator matter?

The primary endpoint is conditional on failures. Its denominator is therefore the number of failures, not automatically the full primary analysis population. A percentage based on failures answers a different question from a percentage based on all evaluable patients. The registry record does not provide the failure count, the marginal-failure count, or the resulting estimate, so neither the denominator nor the endpoint value can be checked from the recorded results.

The primary analysis population is identified as all evaluable patients, n = 84. The record says an eligible patient who progressed during treatment was deemed not evaluable and excluded per protocol. This population statement does not supply the endpoint’s failure denominator. The record also does not report the handling of other missing observations, loss to follow-up, or incomplete endpoint assessments.

How should precision be understood?

In general, a confidence interval describes a range of values compatible with data under a specified frequentist procedure, while a Bayesian credible interval describes a range containing a stated share of posterior probability under a specified model and prior. Either type of interval can communicate uncertainty more fully than a point estimate alone. The registry record does not report a point estimate or interval for the primary endpoint, so its precision cannot be judged. No interval type or interval level is stated for the endpoint.

For outcomes such as progression-free survival, event-free survival, and overall survival, time-to-event methods can account for differing follow-up and censoring. This is a general description of those methods, not a statement that a particular method was used in this study. The registry names these survival probabilities as secondary outcomes but gives no analysis methods or numerical results for them. Details such as censoring rules, estimation methods, and time points beyond the primary endpoint’s stated timeframe are not reported in the registry record.

What do the planning illustrations mean?

The registry’s planning assumptions describe illustrations based on an observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They say that time-to-event illustrations refer to total events and other illustrations to participants per arm, and that the effect to detect is a stated fraction of the observed effect. These are planning assumptions, not evidence that this single-group trial had two randomized arms. No numerical illustration or observed effect is reported in the registry record. Such planning quantities are not observed power, and they should not be treated as an outcome of this study.

More generally, planning calculations depend on their assumed effect, variability, allocation, and analysis method. They describe a hypothetical design under those assumptions, not the probability that a completed trial has demonstrated a treatment effect. Here, the lack of a reported observed primary effect also prevents a reader from assessing the basis of an effect-scaled illustration. The registry does not report the study’s original sample-size justification, so the planning assumptions should not be mistaken for that justification.

What can the single-group design estimate?

A single-group study can describe outcomes observed in its treatment group, subject to the quality and completeness of follow-up and measurement. It cannot provide a concurrent randomized comparison. General causal interpretation is limited because differences from an external expectation may reflect factors other than treatment. In this record, the absence of numerical endpoint results adds a separate limitation: even a descriptive estimate of the primary outcome cannot be assessed.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists the marginal-failure rate as the primary outcome and names secondary outcomes involving progression-free survival, event-free survival, overall survival, quality of life, and MIB-1 labeling and its correlation with progression-free survival. It reports 7 outcome measures and 0 analyses. The record does not describe a multiplicity adjustment, a hierarchy for testing secondary outcomes, or a strategy for interpreting multiple comparisons. No numerical outcome analyses are reported, so there is no set of reported p-values or estimates on which to assess multiplicity in practice.

For the primary endpoint, the registry describes Bayesian monitoring using a binomial likelihood and a Beta distribution for the proportion of failures that are marginal failures. The prior parameters and monitoring decision rule are not reported in the registry record. It is therefore not possible to determine how the planned rule accounted for repeated monitoring or what decision it would produce from particular data.

The analysis population is stated as all evaluable patients, n = 84. An eligible patient who progressed during treatment was deemed not evaluable and excluded from the primary analysis per protocol. The registry does not report the handling of other missing data, incomplete follow-up, or missing secondary-outcome measurements. It also does not provide a full participant accounting that explains the difference between enrollment of 92 and the primary analysis population.

Because the trial is single-group, the listed outcomes do not form randomized treatment comparisons. The record does not report an intention-to-treat analysis or other analysis-population definitions beyond the primary endpoint population statement. These omissions limit assessment of selection, missingness, and the extent to which reported analyses, if any, would represent all enrolled participants.

11. 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: hydrocephalus (2.4%), periorbital edema (1.2%), anemia (1.2%), hypernatremia (1.2%), hypokalemia (1.2%).

For reduced-field conformal radiation therapy, the registry lists 85 at risk and 5 affected for serious adverse events. It separately lists 85 at risk and 5 affected for other adverse events. These are category-specific entries. The record does not say whether the affected participants overlap across categories, so the counts should not be combined or treated as independent groups.

The serious-event terms listed are hydrocephalus at 2.4%, periorbital edema at 1.2%, anemia at 1.2%, hypernatremia at 1.2%, hypokalemia at 1.2%, and hyponatremia at 1.2%. The registry does not specify a denominator alongside these term-level percentages or explain how they relate to the affected counts. The entries are not a comparison with another treatment group, because the study has no control arm.

These figures describe recorded safety events; they do not by themselves establish causation, severity beyond the serious-event classification, timing, duration, or an individual participant’s risk. Details of event ascertainment and attribution are not reported in the registry record. The adverse-event entries should therefore be read as descriptive registry data rather than as proof that radiation caused, or did not cause, a particular event.

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

The study record identifies a phase II evaluation of reduced-field conformal radiation therapy in young patients with gliomas and defines a primary endpoint focused on marginal failure. It also describes a planned Bayesian monitoring framework. Those design details clarify what the investigators intended to measure and how they described monitoring, but they are not a substitute for observed outcome estimates.

The registry record does not report a numerical primary endpoint result. It does not state the number of marginal failures, the total number of failures, or the estimated proportion of failures that were marginal. It also does not report numerical results for the secondary outcomes. Without those results, the magnitude and precision of outcomes cannot be evaluated. The absence of a reported estimate is not evidence that the event rate was high, low, or absent.

The single-group design also sets a limit on interpretation. There is no randomized comparator from which to estimate a difference in outcomes between treatment strategies. An observed outcome in the treatment group, if reported, would be descriptive; by itself it would not establish that the intervention caused the outcome. The registry’s rationale about directing radiation toward a tumor is a statement of purpose, not a trial result.

The safety entries offer some descriptive information, including counts of affected participants and listed serious-event terms. They do not provide a comparative safety estimate, and they do not establish whether an individual event was caused by treatment. The relationship between the serious and other event categories is not reported.

Accordingly, the registry information can support a careful account of design, endpoint definition, analysis population, and recorded safety entries. It cannot support a conclusion about treatment efficacy, comparative benefit, or the size of any treatment-related risk. Such conclusions would require outcome data and analysis details that are not reported in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically notable for the way its primary endpoint is framed. Rather than counting every recurrence as the endpoint event, the registry distinguishes marginal recurrence from central or distant recurrence and from no recurrence. The target is the proportion of failures that are marginal failures. That makes the endpoint’s denominator and event classification central to interpretation.

The planned Bayesian monitoring approach is also relevant. The registry specifies a binomial likelihood and a Beta distribution for the endpoint parameter. These choices describe a probabilistic framework for monitoring an event proportion, but the record does not provide enough detail to reproduce the plan: the prior parameters and decision rule are not reported. Nor are numerical monitoring results reported. The distinction between a stated method and a reported analysis is important when interpreting trial registries.

The single-group design illustrates a separate issue. An outcome description in one treatment group does not provide a randomized contrast. Statistical precision around a single-group estimate, even if available, would not resolve the lack of a control group or establish a causal treatment effect. In this record, numerical outcome estimates are not reported, so even the descriptive endpoint result cannot be assessed.

The stated analysis population further shows why endpoint populations should be read carefully. Enrollment is listed as 92, while the primary analysis population is n = 84, with an eligible patient excluded after progression during treatment because the patient was deemed not evaluable. The record does not account for the remaining difference, and it does not describe handling of other missing data. Reporting a clear population definition is useful, but incomplete accounting restricts assessment of who contributed to the analysis.

Finally, the record lists multiple secondary outcomes and adverse-event summaries, but no numerical analyses for efficacy outcomes. This highlights the need to distinguish outcome names, planned methods, and actual results. They answer different questions and should not be treated as interchangeable evidence.

14. Limitations

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

The main limitation for interpretation is that the registry record does not report a numerical primary endpoint estimate or numerical analyses for the listed secondary outcomes. The endpoint is defined, but its observed event counts, estimate, and uncertainty are absent. The record therefore does not allow assessment of the primary outcome’s magnitude or precision.

The primary analysis population is reported as n = 84, and the record identifies an eligible patient who progressed during treatment as not evaluable and excluded per protocol. Enrollment is listed as 92, but the record does not explain the full difference between enrollment and the primary analysis population. Further details about missing assessments, follow-up completeness, and analysis populations are not reported in the registry record.

The study is single-group, with no randomized comparator. This limits the ability to distinguish treatment-associated outcomes from other influences and prevents a direct comparison with another treatment strategy within the study. Masking is listed as none. The record does not report details that would allow assessment of outcome-assessment procedures or other measures to reduce bias.

Although the primary outcome description names Bayesian monitoring, the prior parameters and decision rule are not reported. The record also does not report the original sample-size justification, interim-analysis details, or a full statistical analysis plan. The planning assumptions described in the registry are not a substitute for these study-specific details.

For safety, the registry lists affected counts and serious-event terms but does not report whether participants overlap between the serious and other event categories, the denominator associated with each term-level percentage, or event attribution. These gaps limit interpretation of the safety entries. The registry record does not report more detailed eligibility criteria, participant characteristics, treatment adherence, or site-level information, which further restricts assessment of how broadly the study population and results could apply.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Radiation Therapy in Treating Young Patients With Gliomas — NCT00238264.
• Registry record last updated 2019-08-07; status last verified 2016-11.
• Results first posted 2017-02-23.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

The registry describes a completed phase II, single-group study of reduced-field conformal radiation therapy for young patients with gliomas. Enrollment is listed as 92, and the primary analysis population is reported as n = 84. The primary endpoint is the proportion of failures that are marginal failures, with monitoring planned up to 5 years using a Bayesian approach based on a binomial likelihood and a Beta distribution.

The record reports 0 analyses and no numerical estimate for the primary endpoint or listed secondary outcomes. It states that an eligible patient who progressed during treatment was deemed not evaluable and excluded from the primary analysis per protocol. Safety entries list 85 at risk and 5 affected for serious events, and the same at-risk and affected counts for other events; term-level serious-event percentages are also listed. The record supports description of the design and endpoint definitions, but does not provide numerical efficacy results or a randomized treatment comparison.