Bevacizumab in Treating Patients With Recurrent or Progressive Glioma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Bevacizumab in Treating Patients With Recurrent or Progressive Glioma (NCT00337207), a na phase 2 study. The primary outcome is Safety of Treatment.

2. Clinical Question

This framing matters when interpreting the record. The primary outcome concerns treatment-emergent toxicity under a specified causality rule, not whether bevacizumab reduced tumors or improved patients’ health. Although the study’s brief description says it was studying how well bevacizumab works, the primary outcome listed in the registry is safety. No primary-outcome analysis or efficacy estimate is reported in the registry record, so the posted safety counts should not be treated as an answer to an efficacy question.

3. Trial Design

This was a Phase II, single-group study of bevacizumab in patients with recurrent or progressive malignant gliomas. The registry records enrollment of 55 and identifies Northwestern University as the sponsor. The study began in 2006-03, had primary completion in 2008-11, and is marked completed. These record details describe the study’s status and scope; they do not establish how many participants ultimately contributed to each outcome analysis.

Allocation is listed as not applicable, masking as none, and the model as single group. Accordingly, there is no randomized control group to provide a concurrent comparison. The Avastin arm name is listed without an arm description. The registry identifies bevacizumab as the biological intervention.

The design should therefore be read as a single-group safety study, not as a head-to-head treatment comparison. Without a concurrent control, adverse events occurring during the study cannot by themselves distinguish effects of treatment from effects of the underlying disease, other care, or circumstances not described in the record. The primary safety definition adds a causality condition, but the registry does not report how individual events were judged against that condition.

The registry lists 55 enrolled participants, while the primary outcome description says that patients receiving at least one dose are evaluable for toxicity. The number who received a dose, and the number included in a primary-outcome analysis, are not reported in the registry record. The enrollment figure should not be assumed to be the denominator for every analysis. No protocol details beyond those in the registry record are available to explain recruitment, treatment delivery, or the planned 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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All patients that receive at least one dose of study treatment are evaluable for toxicity

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 Safety of Treatment, measured over From treatment initiation, throughout treatment and up to 30 days post-treatment, for up to 1 year..

The primary outcome is titled “Safety of Treatment” and is measured as the number of participants experiencing qualifying adverse events. Specifically, the definition includes grade 3 or grade 4 adverse events for which a causal relationship with bevacizumab cannot be completely ruled out. The registry describes grading using Common Terminology Criteria for Adverse Events v3.0: grade 3 is severe, grade 4 is life-threatening or disabling, and grade 5 is death related to an adverse event. Grade 1 is mild and grade 2 is moderate.

This kind of endpoint is used to describe serious treatment-period toxicity in a form that can be counted across participants. The severity grades distinguish the intensity of reported events, while the causality rule narrows the outcome to events that are not clearly unrelated to treatment. The observation window runs from treatment initiation, throughout treatment and up to 30 days post-treatment, for up to 1 year.

The registry identifies patients who receive at least one dose as evaluable for toxicity. It does not report a primary-outcome estimate, the number meeting this exact definition, or the details of event-by-event causality adjudication. The separate adverse-event counts in the record should not automatically be equated with this primary endpoint, because the record does not establish that they use the same severity and causality criteria.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a single-group design allow?

In a single-group study, outcomes are described among people who receive the study intervention, without a randomized concurrent control group. This can provide a record of observed events during the study and can be useful for describing safety. It cannot, on its own, show how those outcomes compare with what would have happened under another treatment or no study treatment. Differences in disease status, other care, and participant characteristics cannot be balanced by random assignment between groups when there is only one group.

Here, allocation is listed as not applicable, masking as none, and the model as single group. Consequently, the adverse-event summaries are descriptive. They do not estimate a randomized treatment contrast. The absence of a comparison group is especially important when an event could arise from the underlying condition as well as from treatment. The endpoint’s causality condition addresses whether a link can be ruled out, but the record does not report individual attribution decisions or an analysis based on those decisions.

What does a participant count represent?

A participant count is a numerator tied to a defined outcome and a population at risk. Those definitions need to match before counts can be interpreted as a rate or compared. The registry reports 55 enrolled, and its primary endpoint description says that patients who receive at least one dose are evaluable for toxicity. It does not report the number who received at least one dose or the number analyzed for the primary endpoint. The at-risk counts in the adverse-event summary are also listed as 55, but that does not establish that every outcome uses the same population definition.

Counts of participants and counts of events are not interchangeable. One person may experience more than one event, and a participant can potentially appear in more than one event-term listing. The record does not state whether the serious-event summary counts people with at least one event, individual events, or both in all relevant details. Its adverse-event fields are presented as participant counts, so they should be described as such rather than treated as a total event burden.

Why do severity and attribution definitions matter?

The primary outcome is defined by severity grade and a causality rule. Grade 3 is severe, and grade 4 is life-threatening or disabling under the stated CTCAE v3.0 descriptions. The endpoint includes events where a causal relationship with bevacizumab cannot be completely ruled out. This criterion is not identical to a count of all serious adverse events: the terms and definitions may differ, and the record does not say that the serious-event summary applies the primary endpoint’s grading and attribution rules.

Using a predefined grade threshold can make a safety outcome more interpretable by focusing attention on events of substantial severity. Requiring that a treatment relationship cannot be ruled out further focuses the count on events considered potentially related. However, the quality and reproducibility of such an outcome depend on consistent grading and attribution. The registry does not report the adjudication process, the degree of reviewer agreement, or the number of events excluded as unrelated.

How should uncertainty be understood?

A confidence interval is a statistical range expressing uncertainty around an estimated quantity under a stated method and its assumptions. It is not a guarantee that the true value lies within that range. A narrower interval generally indicates greater precision, while a wider interval indicates less precision. The registry record does not report a confidence interval for the primary safety endpoint, so its precision cannot be assessed from the registered results.

Likewise, a hypothesis test evaluates compatibility between data and a specified statistical model or null hypothesis. A p-value does not measure the probability that a treatment works, nor does it measure the clinical importance of an outcome. No primary endpoint hypothesis test or p-value is reported in the registry record. Without a comparator and a reported analysis, there is no registered treatment contrast to interpret.

Over-dispersion is a general concern when data vary more than a chosen statistical model expects. For example, event counts may cluster among some participants rather than being distributed evenly. A model that assumes less variability than the data have can understate uncertainty. This is a general statistical consideration, not a finding about this trial: the registry record does not describe a count model, dispersion estimate, or event-level dataset that would allow such an assessment.

What do the planning illustrations mean?

Sample-size planning asks how many participants a future study might require under specified assumptions about the effect to detect, the significance level, the allocation, and the analysis method. The record describes planning illustrations based on fractions of an observed effect, two-sided significance level 0.05, equal allocation between two arms, and a normal approximation. Such calculations are hypothetical planning exercises; they are not evidence that the present trial had that design or achieved a particular power.

The illustrations are described as participants per arm for non-time-to-event outcomes and as a total number of events for time-to-event outcomes. Those conventions are meaningful only under the assumptions stated for each illustration. This trial is recorded as single group, so the equal-allocation two-arm assumptions do not describe its design. The registry does not report the original sample-size justification, and the illustrations should not be interpreted as the original enrollment target, the trial’s observed power, or evidence of treatment efficacy.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 3 outcome measures and 0 analyses. It does not report a primary-outcome analysis, and the number of analyses listed as zero means that the registry record does not provide analysis entries from which to assess statistical testing. A multiplicity plan—how the interpretation of multiple outcomes or repeated tests would be handled—is not reported in the registry record.

Multiplicity is a general issue when a study examines more than one outcome or makes several statistical comparisons. Repeated testing can increase the chance of obtaining a result that appears noteworthy by chance, depending on the testing strategy. A prespecified hierarchy, adjustment, or other plan can address that concern, but no such plan is reported. The registry’s count of outcome measures alone does not establish which were formally tested, whether they were analyzed together, or whether adjustment was needed.

Missing-data handling is also not reported in the registry record. The primary outcome description identifies patients receiving at least one dose as evaluable for toxicity, but does not state how incomplete follow-up, withdrawal, or missing event information was handled. The record does not provide participant flow for the primary endpoint or an analysis population count. Therefore, the enrolled population should not automatically be treated as the analyzed population.

The registry does not report whether analyses were intention-to-treat, per-protocol, or based on another defined population. For a safety endpoint, the stated evaluability rule is exposure-based, but the actual number meeting that rule is not given. These omissions limit assessment of selection and follow-up effects and make it difficult to determine how completely the posted counts represent the intended toxicity population.

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: fatigue (9.1%), death due to disease progression (5.5%), hemorrhage - gi (rectal) (1.8%), perforation - gi (bowel) (1.8%), thrombosis/embolism (venous) (1.8%).

The adverse-event record lists 9 participants with serious adverse events among 55 at risk and 55 participants with other adverse events among 55 at risk. These are separate categories in the registry summary. The counts should not be combined: the record does not establish that the categories are mutually exclusive, and a participant could be represented in more than one category or term.

The listed serious-event terms are Fatigue (9.1%), Death due to disease progression (5.5%), Hemorrhage - GI (rectal) (1.8%), Perforation - GI (bowel) (1.8%), and Thrombosis/embolism (venous) (1.8%). The percentages are reproduced as posted. The record does not provide term-specific attribution, detail whether participants had multiple terms, or show how these terms map to the primary endpoint’s grade and causality definition.

The primary outcome is not simply “any serious adverse event”: it is the number of participants with grade 3 or grade 4 events for which a causal relationship with bevacizumab cannot be completely ruled out. The registry does not report that endpoint’s result. Also, the death-related term is identified as due to disease progression; the record does not identify it as a bevacizumab-related death. These counts are descriptive safety information from a single-group study, not comparative estimates of risk.

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

The record describes a Phase II, single-group study of bevacizumab in patients with recurrent or progressive malignant gliomas. Its primary outcome is a safety measure based on grade 3 or grade 4 adverse events that could not be completely ruled out as related to bevacizumab. The registry also lists serious and other adverse-event counts and several serious-event terms. Together, these details describe the study’s safety focus and the adverse events recorded in the registry.

The record does not report an analysis of the primary endpoint. It therefore does not show how many participants met the endpoint’s full severity and causality criteria. The adverse-event summaries cannot be assumed to be the same measure, and they do not by themselves establish whether an event was caused by bevacizumab. With no randomized comparator, these counts also cannot estimate how outcomes differed from those under another treatment or no study treatment.

The brief study description says the trial was studying how well bevacizumab works, but no efficacy analysis or result is reported in the registry record. The safety counts should not be taken as evidence for or against tumor response, survival, or any other efficacy outcome. The information available is insufficient to make a comparative conclusion about treatment benefit or harm.

Statistically, the findings are best understood as a registry-level description with important gaps: no reported primary-outcome analysis, no reported confidence interval or hypothesis test, no reported analysis population count, and no stated handling of missing data or multiplicity. These gaps leave uncertainty about the precision and completeness of the results. They do not negate the listed counts, but they constrain what those counts can support. The study record is informative about the intended endpoint and the reported adverse-event summaries, not a complete quantitative evaluation of treatment effects.

13. Why This Trial Matters Statistically

This study illustrates why the design and the outcome definition must be read together. The registry’s primary outcome is a participant count based on both event severity and a causality criterion. That is more specific than a broad count of serious adverse events, so the two should not be treated as interchangeable without evidence that they use the same rules. The distinction is central to interpreting safety reporting accurately.

The single-group design also shows the difference between documenting events and estimating a treatment effect. An adverse event recorded during treatment is a relevant observation, but a concurrent randomized comparison is generally needed to estimate how event occurrence differs between treatment strategies while limiting confounding. This record has no such comparison. Its adverse-event counts can describe what was reported in the study group, but cannot answer the counterfactual question of what would have happened to a comparable group without bevacizumab.

The record also illustrates the importance of analysis transparency. It identifies the planned primary endpoint and toxicity population rule, yet does not report a primary-outcome analysis, participant count for that analysis, confidence interval, or hypothesis test. Without these details, readers cannot assess the primary endpoint’s estimate or precision. Listing enrollment and outcome definitions is not a substitute for reporting the analyzed population and result.

Finally, planning calculations require assumptions that match the design. The registry describes sample-size illustrations using equal allocation between two arms, whereas this study is recorded as a single group. Recognizing that mismatch prevents hypothetical planning figures from being mistaken for the trial’s design, its achieved power, or a result. These are broadly useful statistical lessons in reading clinical trial records: distinguish endpoint definitions from related summaries, descriptive data from causal comparisons, and planning assumptions from observed analyses.

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 trial-specific limitation is the absence of a concurrent control group. Since the study is recorded as single group, the posted adverse-event counts cannot establish how event frequency compares with another treatment strategy. The registry’s causality criterion does not remove this limitation, particularly because the event-attribution process and results are not reported.

The registry record does not report the primary safety endpoint result, even though it defines the outcome and the toxicity-evaluable population. It also does not state how many enrolled patients received at least one dose or how many were included in the endpoint analysis. As a result, the relationship between the enrollment figure and the analyzed population is unclear. The serious and other event counts may be informative descriptively, but the record does not confirm that they correspond to the primary endpoint definition.

Further detail on adverse events is limited. The record does not report timing, duration, recurrence, event-level attribution, or how many participants contributed to more than one term. It does not provide a complete account of the grading and adjudication procedures beyond citing CTCAE v3.0 and stating the primary outcome’s causality rule. The recorded death-related term is attributed to disease progression, but the broader death count is not reported in the registry record.

Statistical reporting is also limited: no primary analysis, confidence interval, hypothesis test, missing-data approach, multiplicity plan, or analysis-population definition beyond the dose-based toxicity rule is reported in the registry record. The study’s sample-size justification is not reported in the registry record. The planning assumptions described for illustrations do not resolve these gaps and do not match the recorded single-group design when they assume equal allocation between two arms.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Bevacizumab in Treating Patients With Recurrent or Progressive Glioma — NCT00337207.
• Registry record last updated 2020-02-07; status last verified 2020-01.
• Results first posted 2012-11-19.
• 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

NCT00337207 is a completed Phase II, single-group study titled “Bevacizumab in Treating Patients With Recurrent or Progressive Glioma.” Northwestern University is listed as sponsor. The registry records enrollment of 55, a start date of 2006-03, and primary completion in 2008-11. The intervention is bevacizumab, and the listed arm is Avastin.

The primary outcome is safety, defined as the number of participants experiencing grade 3 or grade 4 adverse events for which a causal relationship with bevacizumab cannot be completely ruled out. Patients receiving at least one dose are described as evaluable for toxicity. The registry lists serious and other adverse-event counts and serious-event terms, but does not report a primary-outcome analysis. Allocation is not applicable, masking is none, and the model is single group, so the record has no randomized control comparison.