A Study of Bevacizumab in Combination With Chemotherapy for Treatment of Osteosarcoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Bevacizumab in Combination With Chemotherapy for Treatment of Osteosarcoma (NCT00667342), a non-randomized phase 2 study. The primary outcome is Number of Participants With Unacceptable Toxicity.

2. Clinical Question

In statistical terms, the primary question concerns the frequency of specified toxicities during treatment, rather than a direct randomized estimate of the effect of bevacizumab. The registry defines unacceptable toxicity using specified grades of hypertension, proteinuria, bleeding, thrombosis or embolism, and major wound complications. It also says a six-stage group sequential stopping rule was developed to monitor unacceptable toxicity. The detailed stopping boundaries and the primary endpoint result are not reported in the registry record.

The study was registered as phase 2 and was non-randomized, with parallel groups organized by disease stratum. Accrual closed early because of slow accrual. The registry says this limited the confidence with which a conclusion could be made, but the available wording does not complete the subject of that statement. It therefore does not establish a specific conclusion about the primary endpoint.

3. Trial Design

This completed phase 2 study evaluated bevacizumab in combination with chemotherapy for osteosarcoma and malignant fibrous histiocytoma of bone. The registry reports a planned enrollment of 43 and an actual enrollment of 43. The study began on 2008-06-03, and primary completion is recorded as 2014-08. It was non-randomized, unmasked, and used a parallel model.

Participants were grouped by disease status. Stratum A was localized resectable disease. Stratum B was metastatic disease, and stratum C was unresectable disease. The registry reports that early closure due to slow accrual left 31 participants in stratum A and 12 in strata B or C. These groups reflect clinical categories, not randomized treatment assignments; their outcomes should not be interpreted as results from comparable groups formed by random allocation.

The localized resectable regimen combined bevacizumab with cisplatin, doxorubicin, and high-dose methotrexate. For metastatic or unresectable disease, the described chemotherapy could also include ifosfamide and etoposide. Bevacizumab was scheduled before chemotherapy in the first cycle, with subsequent timing described in the arm summaries. Surgery and radiotherapy are also listed, but their use is not specified for every participant.

The primary endpoint was the number of participants with unacceptable toxicity, assessed after all patients had completed therapy, up to 1 year after the last patient was enrolled. Secondary outcomes included histologic response, 2-year event-free survival, 2-year overall survival, comparisons with the St. Jude OS99 protocol, and imaging measures. The registry record does not report the primary endpoint result or the results for these secondary outcomes. It also does not report randomization procedures because allocation was non-randomized, or a detailed statistical analysis plan.

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: All participants had newly diagnosed high-grade, biopsy-proven osteosarcoma, or malignant fibrous histiocytoma (MFH) of bone

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Due to slow accrual, the trial was closed to accrual early. Thus, only 31 stratum A patients and 12 stratum B or C patients were enrolled on the study. Therefore, based on the number of patients enrolled and the designed power for the study, we do not have confidence in making a conclusion regarding this feasibility objective.

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: Participants were eligible if they were ≤30 years of age on date of diagnostic biopsy confirmation of high-grade osteosarcoma or malignant fibrous histiocytoma (MFH) of bone. Participants had no previous chemotherapy or radiation therapy.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Unacceptable Toxicity, measured over After all patients have completed therapy, up to 1 year after last patient is enrolled.

The primary endpoint counts participants with unacceptable toxicity. It is a safety and feasibility measure: the study aimed to assess whether bevacizumab could be combined with the chemotherapy regimens used for the different disease strata while monitoring for specified toxicities. A count of participants with an event is not the same as a severity score or a measure of treatment benefit.

The registry definition includes grade 4 hypertension, proteinuria, or bleeding, excluding petechiae or purpura; grade 3 or grade 4 thrombosis or embolism, excluding catheter-related thrombosis; and grade 2, grade 3, or grade 4 major wound complications. The endpoint timeframe is after all patients have completed therapy, up to 1 year after the last patient is enrolled.

Monitoring matters because the endpoint concerns events that could make the treatment combination unacceptable under the study’s criteria. The registry says a six-stage group sequential stopping rule was developed for this monitoring. The rule’s boundaries, how it was applied, and the observed number meeting the primary endpoint are not reported in the registry record. Without that outcome value, the primary endpoint cannot be interpreted as evidence for or against the study’s feasibility objective.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The main endpoint is a participant count: the number who experienced toxicity meeting the registry’s unacceptable-toxicity definition. General statistical explanation: counts are direct summaries of how many participants meet a prespecified event definition. To compare event frequency across groups of different sizes, analysts often also report a proportion with an uncertainty interval. The primary endpoint’s count, proportion, and interval are not reported in the registry record, so those summaries cannot be interpreted for this study.

What does the toxicity definition capture?

The definition specifies event types and grades rather than treating every adverse event as unacceptable. It includes grade 4 hypertension, proteinuria, or bleeding, with petechiae and purpura excluded; grade 3 or 4 thrombosis or embolism, with catheter-related thrombosis excluded; and grade 2 through grade 4 major wound complications. This definition is important because changing which events or grades count would change the endpoint. The registry record does not report whether any definition changes occurred during the study, or how events were adjudicated.

The endpoint is framed as a count of participants, not a count of every event occurrence. General statistical explanation: participant-level event counts answer how many people experienced at least one qualifying event; event-level counts can be larger when an individual experiences multiple events. The registry record does not clarify the detailed handling of multiple qualifying events for the primary outcome, and it does not provide its result.

What does a stopping rule add?

The registry says a six-stage group sequential stopping rule was developed to monitor unacceptable toxicity. General statistical explanation: sequential monitoring allows safety information to be reviewed at planned stages, with prespecified criteria that can guide whether monitoring continues or stops. The boundaries, review schedule, and any decisions made under this rule are not reported in the registry record. It would be speculation to infer how often monitoring occurred or whether the rule triggered a stopping decision.

The record’s planning information describes sample-size illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says that the effect to detect is a stated fraction of the observed effect. These are planning illustrations, not observed power, and they are not evidence that the trial was designed around these assumptions. The trial itself was non-randomized, and the record does not report the design’s original sample-size justification or the assumptions used for that justification.

How should the safety counts be read?

The serious adverse-event listing gives a count at risk and a count affected for each listed group. General statistical explanation: such counts provide a straightforward description of events recorded in each group, but they do not by themselves account for differences in follow-up or establish a causal treatment effect. The registry lists 3 serious-event participants among 31 at risk for localized resectable disease and 0 among 11 at risk for metastatic tumors. The definitions, exposure time, and details of follow-up for these safety counts are not reported in the registry record.

The record also lists 31 of 31 and 11 of 11 affected in the “other” adverse-event category. This does not mean that each participant experienced the same event, or that each experienced only one event. The registry’s summary does not supply event-level detail for those category counts. Death counts are not reported in the registry record.

What analyses can be assessed?

The registry names 19 outcome measures and 3 analyses. The existence of listed measures does not reveal which statistical tests, models, or adjustment methods were applied. Those details are not reported in the registry record. Nor does the record report estimates, confidence intervals, or p-values for the primary endpoint. General statistical explanation: a confidence interval expresses the uncertainty around an estimate under the assumptions of the analysis; it is not a guarantee that the true value lies within the interval. Without the endpoint estimate and interval, the precision of the primary result cannot be assessed.

The study’s non-randomized allocation is central to interpretation. General statistical explanation: random allocation can help balance known and unknown prognostic factors between treatment groups. In this study, strata were based on disease status and treatment regimens differed accordingly, so the groups are not like-for-like controls. A difference between their outcomes could reflect disease status, regimen differences, or other factors rather than a treatment effect. The registry record does not report an adjusted comparative analysis that could address these differences.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 19 outcome measures and 3 analyses, including the primary toxicity endpoint, histologic response, survival outcomes, comparisons with the St. Jude OS99 protocol, and imaging measures. When many outcomes are assessed, the chance of at least one apparently notable result can increase if each is tested separately. General statistical explanation: multiplicity methods can control or describe that concern, but the appropriate approach depends on which outcomes were prespecified as primary or secondary and how the analyses were planned. Multiplicity handling is not reported in the registry record.

The primary endpoint description specifies a six-stage group sequential stopping rule for monitoring unacceptable toxicity. That is a planned monitoring feature, but it does not establish how multiplicity across the different outcomes was handled. The stopping boundaries, monitoring implementation, and relationship between interim safety monitoring and final analyses are not reported in the registry record. Interim efficacy analyses are not reported in the registry record.

Missing-data methods and analysis populations are also not reported in the registry record. The record does not describe how participants with incomplete outcome assessments, withdrawal, or missing follow-up were handled, nor whether analyses used an intention-to-treat, per-protocol, or other population. The registry gives enrollment numbers, but those numbers alone do not establish the denominator used for each outcome analysis.

These omissions limit assessment of whether the reported outcomes, if available elsewhere, would reflect the prespecified analysis approach. In the registry record, the primary endpoint result and the secondary outcome results are not reported, so neither missing-data handling nor multiplicity adjustments can be evaluated against observed estimates.

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: hypotension (3.2%), fever (in the absence of neutropenia, where neutropenia is defined as anc <1.0 x 10e9/l) (3.2%), infection with normal anc or grade 1 or 2 neutrophils, lung (pneumonia) (3.2%), cognitive disturbance (3.2%), encephalopathy (3.2%).

The serious adverse-event counts are presented by registry safety group. For localized resectable disease, 3 participants were affected among 31 at risk. For metastatic tumors, 0 were affected among 11 at risk. These are descriptive counts, not a randomized comparison. The registry does not report a statistical comparison between these groups or an adjusted estimate.

Six serious adverse-event terms are each listed at 3.2%: hypotension; fever in the absence of neutropenia, where neutropenia is defined as ANC <1.0 x 10e9/L; lung infection or pneumonia with normal ANC or grade 1 or 2 neutrophils; cognitive disturbance; encephalopathy; and motor neuropathy. Term-level percentages do not show whether the same participant contributed to more than one term. The registry record does not report the event-level detail needed to determine that.

The “other” adverse-event category lists 31 affected among 31 at risk in localized resectable disease, and 11 affected among 11 at risk in metastatic tumors. These category counts do not describe the severity or type of each event. Death counts are not reported in the registry record. The safety listing’s metastatic tumors group has 11 at risk, while the enrollment summary reports 12 participants in strata B or C; the reason for this difference is not reported in the registry record.

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

The record describes a phase 2, non-randomized study of bevacizumab combined with chemotherapy and identifies unacceptable toxicity as the primary endpoint. It also records selected serious and other adverse-event counts. These details allow readers to understand the intended safety question and the available descriptive safety summaries.

They do not establish whether the primary endpoint met the study’s feasibility objective. The number of participants with unacceptable toxicity, as defined for the primary endpoint, is not reported in the registry record. The listed serious adverse-event counts are not a substitute for that endpoint because the serious-event category and the unacceptable-toxicity definition are not stated to be identical.

The study also cannot be read as a randomized comparison of bevacizumab against a control treatment. Allocation was non-randomized, and treatment regimens varied according to disease stratum. The listed secondary comparisons with the St. Jude OS99 protocol do not create a randomized control group; their results and methods are not reported in the registry record.

The registry lists event-free and overall survival at 2 years, along with histologic response and imaging outcomes, but gives no results for them in the record described. No conclusion about survival, tumor response, or comparative benefit follows from the outcome names alone. Likewise, the safety counts should not be taken as proof that a treatment is safe or unsafe: they are descriptive observations, and their precision and context cannot be fully assessed without the missing outcome details.

The narrow conclusion supported by the record is that the study was conducted, accrued the reported enrollment, closed early because of slow accrual, and collected outcomes that included toxicity and survival. The registry itself notes limited confidence in making a conclusion after early closure. The primary endpoint estimate, its uncertainty, and the secondary results needed for broader interpretation are not reported in the registry record.

13. Why This Trial Matters Statistically

This study illustrates why a clinical question and a statistical comparison are not always the same thing. Its primary endpoint asks how many participants experienced specified unacceptable toxicities during a treatment strategy. That is a meaningful safety and feasibility question, but answering it requires the endpoint result and enough information to judge its precision. Those primary results are not reported in the registry record.

The trial also shows why disease strata should not be treated as randomized arms. Localized resectable, metastatic, and unresectable disease differ by definition, and the registry describes different chemotherapy regimens across these settings. Because allocation was non-randomized, an observed difference between strata would not isolate the effect of bevacizumab or any other single component.

Early closure due to slow accrual is another important statistical feature. The record reports 31 participants in stratum A and 12 in strata B or C and explicitly flags limited confidence in making a conclusion. Smaller enrollment can leave estimates less precise, but the degree of uncertainty cannot be quantified without the actual endpoint estimate and its interval. The registry’s planning illustrations should not be mistaken for observed power or for the original design justification.

Finally, the record lists a broad set of outcomes, including histologic response, survival, and imaging measures, alongside the primary toxicity question. This makes transparent reporting of the primary result, analysis methods, missing-data handling, and multiplicity approach especially important for interpreting the evidence. Those details are not reported in the registry record. The study is therefore statistically informative as an example of a safety-focused, non-randomized design with early closure, while its posted information does not support a firm conclusion about the treatment strategy’s outcomes.

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.

Early closure due to slow accrual reduced enrollment relative to what the study may have intended, but the original target rationale is not reported in the registry record. The enrollment summary gives 31 participants in stratum A and 12 in strata B or C. The record states that the enrollment and designed power did not support confidence in making a conclusion, but the sentence is incomplete and does not identify the specific conclusion.

There is no randomized control group. The disease strata are clinically distinct, and the chemotherapy regimens differ between localized resectable disease and metastatic or unresectable disease. Consequently, comparisons among strata would not provide a like-for-like estimate of treatment effect. Although secondary outcomes include comparisons against the St. Jude OS99 protocol, the methods and results of those comparisons are not reported in the registry record.

The primary endpoint result is not reported, and the available safety counts do not resolve that gap. The serious adverse-event group for metastatic tumors lists 11 at risk, whereas the enrollment summary reports 12 participants in strata B or C; the explanation is not reported in the registry record. Follow-up completeness, outcome-specific denominators, missing-data handling, and analysis populations are also not reported. These omissions make it difficult to judge whether the available safety counts represent the full enrolled groups or comparable observation periods.

The registry mentions a six-stage group sequential stopping rule but does not report its boundaries or implementation. The statistical tests, models, uncertainty intervals, multiplicity approach, and original sample-size justification are not reported in the registry record. The planning illustrations use assumptions that are not the same as evidence of achieved precision. These limitations mean that the record supports a design description and selected safety counts, but not a comprehensive assessment of the primary endpoint or secondary outcomes.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of Bevacizumab in Combination With Chemotherapy for Treatment of Osteosarcoma — NCT00667342.
• Registry record last updated 2023-08-07; status last verified 2019-05.
• Results first posted 2014-08-04.
• 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 2 study examined bevacizumab with chemotherapy for osteosarcoma and malignant fibrous histiocytoma of bone. It was non-randomized, unmasked, and parallel in model. The registry reports enrollment of 43 and early closure due to slow accrual, with 31 participants in localized resectable disease and 12 in metastatic or unresectable disease.

The primary endpoint was the number of participants with unacceptable toxicity, assessed after therapy completion, up to 1 year after the last patient was enrolled. The registry describes the toxicity criteria and a six-stage group sequential stopping rule, but the primary endpoint result is not reported in the registry record.

Selected serious adverse-event counts are listed for localized resectable disease and metastatic tumors. Secondary outcomes include histologic response, survival, comparisons with the St. Jude OS99 protocol, and imaging measures. Their results, analysis methods, missing-data handling, and multiplicity approach are not reported in the registry record. Early closure and the non-randomized design constrain interpretation of comparative conclusions.