Observation or Radiation Therapy and/or Chemotherapy and Second Surgery in Treating Children Who Have Undergone Surgery for Ependymoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Observation or Radiation Therapy and/or Chemotherapy and Second Surgery in Treating Children Who Have Undergone Surgery for Ependymoma (NCT00027846), a non-randomized phase 2 study. The primary outcome is Event-free Survival.

2. Clinical Question

The trial addressed how children with localized ependymoma might be managed after an initial operation, with treatment pathways reflecting resection status, tumor location, and histology. Its stated purpose was to assess conformal radiation therapy, either alone or after chemotherapy and second surgery. A separate pathway involved observation for a specified group with completely resected, differentiated, supratentorial disease.

These pathways answer a descriptive clinical research question: what outcomes were recorded for eligible children managed through these protocol-defined approaches? They do not establish the effect of one treatment compared with another. The study was nonrandomized, and the treatment groups were formed according to clinical characteristics rather than allocation by chance. The registry record does not report a randomized treatment contrast.

Event-free survival was the main outcome specified for evaluation. The record defines an event broadly to include progression, relapse, a second neoplasm, or death. Although the endpoint is named and its method and time horizon are described, its estimate is not reported in the registry record. As a result, the central numerical result cannot be assessed from the registry information available.

3. Trial Design

The registry classifies this as a phase II, nonrandomized study with no masking and a single-group model. It reports enrollment of 378 and lists Children's Oncology Group as the sponsor. The study start is recorded as 2003-08, primary completion as 2010-01-01, and status as completed. These descriptors provide the administrative and design context, but do not establish how many participants were analyzed in each treatment pathway.

The clinical plan distinguishes groups by resection, tumor site, and histology. For patients with completely resected, differentiated, supratentorial ependymoma, the arm description specifies observation. A radiation group includes specified supratentorial and infratentorial histology and resection categories; its description states that conformal radiation therapy is given once daily, 5 days a week, for 6-6½ weeks. A subtotal-resection group is described as receiving chemotherapy, with second surgery among the treatment questions. The arm description available for that group is truncated after the initial chemotherapy details, so its full regimen and subsequent management are not reported in the registry record.

The listed interventions include radiation therapy, therapeutic conventional surgery, filgrastim, carboplatin, cyclophosphamide, etoposide, vincristine sulfate, and Mesna. Their presence in the intervention list does not mean that every participant received every intervention. The record's pathway descriptions indicate that care differed across groups.

Because allocation was nonrandomized, the groups should not be treated as interchangeable treatment arms. Clinical factors that determined the pathway may also be associated with outcomes. The record does not report a statistical adjustment that would make these groups comparable, and it does not report a planned between-group causal comparison. The design is therefore best read as a phase II evaluation of distinct clinical approaches rather than as a randomized comparative trial.

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: The product-limit (Kaplan-Meier) estimate is for estimation of event-free survival probability at 5 years. All eligible patients in the study were included.

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 Event-free Survival, measured over Up to 5 years after completion of study treatment.

The primary endpoint is event-free survival, reported in the registry as the probability of EFS at 5 years. The record defines the interval as beginning at study enrollment and ending at disease progression, disease relapse, occurrence of a second neoplasm, or death from any cause. It states that a product-limit, or Kaplan-Meier, estimate was used to estimate the 5-year probability. The analysis population is described as all eligible patients in the study.

Event-free survival is used in clinical research when the question concerns remaining free of a set of clinically important events over time. It combines the time to the first qualifying event, rather than describing only whether a participant experienced one particular event. The definition matters: changing which outcomes count as events would change the endpoint's meaning. Here, progression, relapse, a second neoplasm, and death are all included in the registry definition.

A time-to-event estimate can use information from participants whose follow-up ends before an event is recorded, under the method's assumptions. The Kaplan-Meier method accounts for such censoring rather than treating every participant as having the same observation period. A confidence interval communicates uncertainty around an estimated survival probability; it does not guarantee that the true value lies within the interval. The registry specifies a 95% confidence interval as the measure of dispersion, but does not report the estimate or interval. Consequently, neither the magnitude nor the precision of the primary endpoint can be evaluated from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a Kaplan-Meier estimate describe?

Kaplan-Meier is a product-limit method for describing the probability of remaining event-free over time. It uses the times at which events occur and incorporates follow-up that ends without a recorded event. Such an observation is commonly called censored: the participant contributes information up to the last known follow-up, but the method does not treat the follow-up end as an event. The estimate relies on the assumption that censoring is not related to a participant's underlying event risk in a way that makes the observed follow-up systematically misleading.

For this trial, the registry defines the starting point as study enrollment and the events as progression, relapse, a second neoplasm, or death from any cause. The endpoint is the estimated event-free probability at 5 years after study treatment, while the event-time origin is described as enrollment. Those details should be kept together when interpreting the endpoint: the event definition and time origin determine what the probability represents.

What does a confidence interval add?

A confidence interval expresses statistical uncertainty around an estimated quantity. In general, a narrower interval indicates greater precision, while a wider interval indicates less precision. Its width depends on factors such as the amount of usable follow-up and the number and timing of events. A 95% confidence interval is a method for expressing uncertainty; it is not a guarantee about the value in any one study. Although the registry identifies a 95% confidence interval for the primary outcome, it does not report either endpoint of that interval or the underlying estimate.

Why is this not a treatment-effect comparison?

Random allocation is a design feature that can help balance prognostic factors across treatment groups. This study is recorded as nonrandomized, and its pathways differ according to tumor and surgical characteristics. Those characteristics may be related to prognosis as well as treatment choice. Therefore, a contrast between observed outcomes across pathways could reflect these differences, treatment, or both. The registry record does not report a randomized contrast or a statistical adjustment that would support a causal comparison. The record explicitly cautions that the arms are not a like-for-like control comparison.

The record also labels the model as single-group while describing distinct clinical pathways. These registry descriptors should be read as administrative design classifications, not as evidence that all participants received one identical intervention. The arm descriptions make clear that management differed across groups.

How should event counts and safety fields be read?

Counts of participants affected and counts of events are not necessarily the same thing. A participant may experience more than one event, and the registry fields do not state whether the reported affected counts represent unique participants, event episodes, or a particular reporting window. The at-risk values identify the denominators associated with the safety entries, but do not establish that all participants had equal duration of observation. Serious and other adverse-event categories should be read according to their separate registry fields; they should not be added together or treated as mutually exclusive without a definition supporting that interpretation.

The listed serious-event terms and percentages are descriptive. The record does not provide the group assignment for that term list, definitions of the reporting thresholds, or a detailed account of attribution to study treatment. It also does not report death counts in the safety entries. These omissions limit what can be inferred from the safety information.

What can planning calculations establish?

The record describes sample-size assumptions as a planning illustration based on an observed effect, using a two-sided significance level of 0.05, equal allocation, and a normal approximation. For time-to-event outcomes, the illustration is described in terms of a total number of events; other outcomes are described in terms of participants per arm. These are planning concepts, not observed trial results or observed power. The record does not give numerical planning outputs in the information available for this analysis, and the primary endpoint estimate itself is not reported. The illustration therefore cannot be used to recover the missing outcome result.

In general, a sample-size calculation explores how much information a future study might need under specified assumptions. Its result depends on the assumed effect, variability, outcome frequency, design, and analysis method. An illustration based on an observed effect is not evidence that the trial was designed around that effect, and it does not show the probability that this completed study detected an effect. The registry record does not report the original sample-size justification, its statistical plan, or interim analyses.

Why can over-dispersion matter for counts?

As a general statistical point, some count outcomes vary more than a basic count model expects. This is called over-dispersion. If a model assumes less variability than the data contain, standard errors may be too small and confidence intervals too narrow. Appropriate modeling may need to account for extra variability, but the registry record does not identify a count model or report analyses of this kind. This general consideration should not be mistaken for a description of the trial's statistical plan.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures, including the primary event-free survival outcome and several secondary outcomes. The secondary outcome list includes overall survival, rate of gross-total or near-total resection and second surgery after chemotherapy, event-free survival, and local control and patterns of failure. Event-free survival appears twice in that list. The record does not explain whether those entries refer to distinct definitions, time points, or duplicate listings.

When a study evaluates multiple outcomes or multiple analyses, the chance of at least one apparently notable result can increase. Statistical methods for handling multiplicity may be used to control that risk, but whether such methods were planned depends on the protocol and analysis plan. The registry record does not report a multiplicity strategy or the statistical analysis plan. It reports no analyses, so there are no posted analysis results from which to determine how the listed outcomes were handled.

The primary outcome description states that all eligible patients were included in the product-limit estimate. It does not report the number eligible, the number analyzed, exclusions, or the handling of missing outcome or follow-up data. Missing-data procedures and reasons for loss to follow-up are not reported in the registry record. These details matter because incomplete follow-up can affect the amount of information available and, depending on its pattern, the validity of a time-to-event estimate.

The registry record does not report interim analyses, stopping rules, or the trial's original sample-size justification. It describes a planning illustration based on an observed effect, but that is not the same as the study's original design rationale and is not observed power. Without a reported estimate, interval, analysis population counts, or analysis plan, the precision and robustness of the primary result cannot be independently assessed from this record.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: leukocytes (total wbc) (62.5%), neutrophils/granulocytes (anc/agc) (62.5%), hemoglobin (hgb) (43.8%), platelets (39.1%), lymphopenia (23.4%).

The safety record reports separate serious and other adverse-event fields for the radiation group and the subtotal-resection group. In the radiation group, the serious-event field lists 281 at risk and 38 affected; the other-event field lists 281 at risk and 52 affected. In the subtotal-resection group, the serious-event field lists 64 at risk and 55 affected; the other-event field lists 64 at risk and 27 affected. These are the counts recorded for those entries and should not be treated as a randomized comparison.

The record also lists serious-event terms and percentages: leukocytes (total WBC), 62.5%; neutrophils/granulocytes (ANC/AGC), 62.5%; hemoglobin (Hgb), 43.8%; platelets, 39.1%; lymphopenia, 23.4%; and transfusion: pRBCs, 21.9%. The record does not clearly identify the group to which this list applies, nor does it specify in these fields whether percentages refer to participants or event episodes. They should therefore be read as reported safety descriptors, not as comparable treatment-specific risks.

The safety entries do not report death counts. They also do not describe the observation period, event recurrence, attribution, or full event definitions. The counts cannot establish that an event was caused by treatment, and the difference between group entries cannot be interpreted as a treatment effect because the groups were nonrandomized and not like-for-like.

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

The registry record establishes that a phase II, nonrandomized study evaluated treatment pathways for children who had undergone surgery for localized ependymoma. It documents a primary endpoint of event-free survival at 5 years, defines which events count, specifies a Kaplan-Meier estimation method, and identifies all eligible patients as the intended primary analysis population. It also records safety counts for two groups and names several serious-event terms.

Those details do not reveal the primary outcome's numerical result. The event-free survival estimate and 95% confidence interval are not reported in the registry record. Without them, the reader cannot determine the observed probability of remaining event-free or how precisely it was estimated. The enrollment figure does not substitute for an outcome estimate, and the description of an analysis method does not indicate what that analysis found.

The record also does not establish that one management pathway outperformed another. Treatment groups were nonrandomized and defined by clinical characteristics. Any differences between the groups could reflect those characteristics, the treatments, or other factors. The registry explicitly cautions that the arms are not a like-for-like control comparison.

The safety entries provide counts of affected participants as recorded, but not a full comparative safety analysis. They do not establish causation or show how event frequency changed over time. The serious-event term list lacks a clearly stated group assignment in the record. It should not be generalized beyond the information recorded.

Overall, the registry supports a description of the study design, endpoint definition, treatment pathways, and selected safety entries. It does not support a numerical conclusion about event-free survival, a causal comparison among pathways, or a comprehensive safety judgment. Those conclusions would require results and methods not reported in the registry record.

13. Why This Trial Matters Statistically

This study is statistically notable because the endpoint is time-to-event while the clinical pathways are nonrandomized and tied to patient and tumor characteristics. Reading those features together helps distinguish an outcome description from a comparative treatment effect. A Kaplan-Meier estimate can summarize event-free survival over time, but it cannot by itself remove confounding or make groups comparable when treatment allocation was not randomized.

The endpoint definition is also important. Progression, relapse, a second neoplasm, and death from any cause all count as events. This gives the measure a broader meaning than an endpoint based on a single event type. Clear definitions allow readers to understand what an estimate would represent and what it would not represent. Here, the definition is available, but the estimate and its confidence interval are not.

The registry record illustrates why design labels and outcome results must be considered separately. It reports enrollment of 378 and describes a completed phase II study, yet it lists no analyses and does not report a numerical primary result. Enrollment and completion status do not establish statistical precision or clinical benefit. Similarly, a planning illustration is not a substitute for observed outcomes or an account of the original design justification.

The safety entries also demonstrate the limits of summary counts. Serious and other adverse events are recorded for different nonrandomized groups, with a separate list of serious-event terms and percentages. Without clear attribution, event definitions, observation periods, and a comparative analysis, these figures are descriptive rather than causal. The registry's missing result fields are therefore consequential: readers can identify what was planned and what selected safety counts were recorded, but cannot reconstruct the central efficacy analysis from this record.

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 is that the numerical primary endpoint result is absent. Although the registry describes event-free survival, its time horizon, event definition, Kaplan-Meier method, and 95% confidence interval, it reports neither the estimate nor the interval. The record also lists no analyses. As a result, the primary finding and its precision cannot be evaluated from the registry information.

The nonrandomized design limits comparisons across treatment pathways. Groups differ by tumor location, histology, and extent of resection, and the record does not report a randomized contrast or adjustment that addresses those differences. The single-group model designation appears alongside descriptions of distinct pathways, so the administrative model label does not make the groups comparable.

The registry reports enrollment of 378 but does not give outcome-specific analysis counts or the number eligible for the primary analysis. It says all eligible patients were included, but does not report eligibility counts, exclusions, missing follow-up, censoring patterns, or missing-data procedures. The full description of the subtotal-resection pathway is truncated, limiting interpretation of that treatment group's complete regimen.

Safety information is also limited. The record does not report death counts in the adverse-event entries, and it does not explain the observation period, event attribution, or whether affected counts represent unique participants or event episodes. The serious-event term list is not clearly assigned to a group. Finally, multiplicity handling, interim analyses, the statistical analysis plan, and the original sample-size justification are not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Observation or Radiation Therapy and/or Chemotherapy and Second Surgery in Treating Children Who Have Undergone Surgery for Ependymoma — NCT00027846.
• Registry record last updated 2019-08-07; status last verified 2017-01.
• Results first posted 2017-03-01.
• 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.

18. Record Summary

This completed phase II study was sponsored by Children's Oncology Group and is recorded as nonrandomized, unmasked, and single-group. The registry reports enrollment of 378, a study start of 2003-08, and primary completion on 2010-01-01. Its population was children who had undergone surgery for localized ependymoma.

The treatment pathways included observation for a specified completely resected group, conformal radiation therapy for specified clinical groups, and chemotherapy with second surgery among the questions for subtotal resection. Event-free survival at 5 years was the primary endpoint, defined from enrollment to progression, relapse, second neoplasm, or death from any cause. The record specifies a Kaplan-Meier estimate and 95% confidence interval, but does not report the numerical estimate or interval.

The record lists six outcome measures and no analyses. It reports selected serious and other adverse-event counts for the radiation and subtotal-resection groups, plus a list of serious-event terms. Because allocation was nonrandomized and the registry results are incomplete, these entries do not establish comparative treatment effects.