This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Phase I/II Bevacizumab Versus Bevacizumab Plus TPI 287 for Recurrent Glioblastoma (NCT01582152), a randomized phase 1, phase 2 study. The primary outcome is Progression Free Survival (PFS).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01582152 |
| Phase | Phase 1, Phase 2 |
| Condition | Brain Neoplasms, Central Nervous System Neoplasms |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | None (open label) |
| Enrollment | 12 |
| Arms in the primary outcome | 2 |
| Lead sponsor | M.D. Anderson Cancer Center |
| Results posted | Yes |
2. Clinical Question
Population
Adults with recurrent glioblastoma were the intended population. The record describes a Phase I/II study and reports enrollment of 12. The primary endpoint description measures progression-free survival from registration, but the analyzed population is not reported in the registry record.
Intervention
The experimental treatment was TPI 287 given with bevacizumab. The Phase I goal was to identify a tolerable TPI 287 dose with bevacizumab; the posted results include combination groups at 160 mg/m2 and 140 mg/m2. The full Phase II dosing description is not reported in the registry record.
Comparator
The study summary describes a Phase II comparison with bevacizumab alone and allows crossover to the combination after disease worsening on bevacizumab. However, the posted primary outcome groups are both TPI 287 plus bevacizumab. A bevacizumab-alone outcome result is not reported in the registry record.
Primary question
The primary endpoint is progression-free survival, measured from registration until progression or death, or until the last date known alive without progression. The record reports medians and full ranges for the two combination-dose groups. It does not report a sponsor analysis estimating the treatment contrast, so these results do not establish whether one strategy differs from another.
3. Trial Design
The registry identifies this as a randomized Phase I/II trial with no masking and a crossover model. It is titled as a study of bevacizumab versus bevacizumab plus TPI 287 in adults with recurrent glioblastoma. The stated Phase I objective was to find the highest tolerable TPI 287 dose when combined with bevacizumab. The stated Phase II objective was to assess whether adding TPI 287 could control disease better than bevacizumab alone, while also studying safety.
The record describes a Phase I dose-level approach, with participants assigned to one of four TPI 287 dose levels according to when they joined the study. The posted primary results, however, list two groups: TPI 287 160 mg/m2 plus bevacizumab and TPI 287 140 mg/m2 plus bevacizumab. Each group has a reported sample size of 6. The registry's results do not include a bevacizumab-alone group for the primary endpoint, despite the comparator described in the study summary.
The bevacizumab group description says that participants could enroll in a crossover group to receive TPI 287 and bevacizumab if disease worsened during bevacizumab treatment. The timing, eligibility rules, and analysis handling for crossover are not reported in the registry record. The trial status is terminated. The start is recorded as 2012-07 and primary completion as 2016-03. The reason for termination and details of any protocol changes are not reported in the registry record.
Randomization can help balance prognostic factors between assigned groups, but it does not by itself make results comparable when the posted outcome groups do not represent the intended randomized contrast. The allocation sequence, randomization ratio, and implementation details are not reported in the registry record. Nor is it reported whether the two posted dose groups correspond to randomized arms, dose-escalation cohorts, or another analysis grouping. The record therefore supports a description of these group-specific summaries, not a reconstruction of the intended Phase II comparison.
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: Study was terminated early without advancing to the second portion (Phase II).
Overall Study
| Stage | TPI 287 160 mg/m2 + Bevacizumab | TPI 287 140 mg/m2 + Bevacizumab |
|---|---|---|
| STARTED | 6 | 6 |
| COMPLETED | 0 | 0 |
| NOT COMPLETED | 6 | 6 |
Reasons for not completing this period:
| Reason | TPI 287 160 mg/m2 + Bevacizumab | TPI 287 140 mg/m2 + Bevacizumab |
|---|---|---|
| Disease Progression | 4 | 5 |
| Adverse Event | 2 | 1 |
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.
| Characteristic | Unit | TPI 287 160 mg/m2 + Bevacizumab | TPI 287 140 mg/m2 + Bevacizumab | Total |
|---|---|---|---|---|
| Participants analysed | 6 | 6 | 12 | |
| Age, Continuous | years | 57.5 (35 to 72) | 52 (44 to 61) | 55 (35 to 72) |
| Sex: Female, Male: Female | Participants | 2 | 2 | 4 |
| Sex: Female, Male: Male | Participants | 4 | 4 | 8 |
| Region of Enrollment: United States | participants | 6 | 6 | 12 |
6. Primary Endpoint
The registry lists the primary outcome as Progression Free Survival (PFS), measured over PFS will be evaluated as a continuous variable, up to one year.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Full Range |
| Unit | weeks |
| Endpoint classification | Time-to-Event Analysis |
Progression-free survival (PFS) is the time from registration until disease progression or death, whichever occurs first. If neither event has occurred by the last assessment, the participant is censored at the last date known to be alive without progression. The registry describes PFS as a continuous variable assessed up to one year and reports the endpoint as median time with a full range.
PFS is used in clinical trials because it captures the time until a disease-related event or death, rather than only recording whether an event occurred at a single assessment. In general, time-to-event methods can incorporate follow-up that ends before an event is observed by treating that participant's observed time as censored. Censoring does not mean that progression has occurred; it means the event time is not observed beyond the recorded follow-up. Valid interpretation depends on appropriate event definitions, follow-up, and handling of censoring, details that are not fully reported in the registry record.
The median is the point at which half of the observed time-to-event distribution lies on either side, subject to the estimation method. The reported full range describes the lowest and highest reported times, not the uncertainty around the median. The record gives a median PFS of 12.5 weeks, with a full range of 6.0 to 37.0, for the TPI 287 160 mg/m2 plus bevacizumab group. For the TPI 287 140 mg/m2 plus bevacizumab group, the median is 9.5 weeks, with a full range of 7.0 to 44.0. Confidence intervals for these medians are not reported in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted results describe PFS within two TPI 287 plus bevacizumab groups. The median is 12.5 weeks in the 160 mg/m2 group, with a full range of 6.0 to 37.0, and 9.5 weeks in the 140 mg/m2 group, with a full range of 7.0 to 44.0. Each group has a reported sample size of 6. These are descriptive summaries. The registry record contains no posted analysis records and does not provide an estimated treatment contrast, confidence interval, or p-value for a comparison of these groups.
The medians alone do not establish that one dose level has a different PFS distribution from the other. The full ranges overlap, and a range is not a measure of statistical uncertainty. In general, time-to-event estimates depend on the timing of events and censoring, not only on the median and endpoints of the range. Individual event times, censoring counts, survival curves, and uncertainty intervals are not reported in the registry record, so the precision of either median cannot be judged from the posted summaries alone.
The study summary describes a Phase II question comparing bevacizumab plus TPI 287 with bevacizumab alone. That comparison cannot be evaluated from these primary endpoint results: the listed outcome groups both received the combination, and a bevacizumab-alone PFS result is not reported in the registry record. The record also does not explain how the two dose groups relate to randomization or to the dose-finding stage. A randomized label does not justify interpreting the two displayed dose cohorts as the intended randomized comparison.
The trial is marked terminated, but the reason is not reported in the registry record. Termination alone does not show why the study stopped or what conclusions should be drawn about efficacy or safety. The reported sample sizes are small, and the record gives no inferential estimate with which to quantify the uncertainty of a between-group difference. The available evidence is therefore limited to the posted group-specific PFS summaries and adverse-event counts. It does not support a conclusion that either regimen is superior, equivalent, or inferior.
There is one listed secondary outcome, objective response rate, but its result is not included in the posted outcomes. Sponsor-reported hazard ratios are not present among the posted analyses. Any comparison inferred from the two medians would be an unadjusted re-calculation from incomplete summary information, and such a calculation is not reported. The registry's own analysis, if one exists outside the posted result record, is not reported in the registry record.
9. Statistical Methods Explained
Progression-free survival is a time-to-event endpoint. For each participant, the event time is the time to progression or death, whichever happens first. A participant without a recorded event is censored at the last date known to be alive without progression. Censoring preserves information about observed follow-up up to that point; it does not assign an event at the censoring time. In general, time-to-event analysis uses the event and censoring times to describe how the event experience accumulates over follow-up.
What does a median time-to-event result convey?
A median is a summary of the distribution of event times. It does not indicate how many participants experienced the event, when the events occurred across follow-up, or how much uncertainty surrounds the estimate. The record reports median PFS and full ranges for each posted combination-dose group, but it does not report confidence intervals, survival probabilities at particular times, event counts, or censoring counts. Consequently, the displayed summaries cannot show the shape of the PFS distribution or the precision of the median estimates.
A full range is the span from the lowest to highest reported value. It should not be read as a confidence interval: a range describes the observed extremes, whereas a confidence interval expresses uncertainty about an estimated quantity under a statistical model. General statistical practice would ordinarily present uncertainty around treatment-effect estimates, especially when group sizes are small. Those intervals and a sponsor-reported treatment-effect estimate are not reported in the registry record.
Why is a group comparison unavailable?
The record lists two outcome groups, both receiving TPI 287 with bevacizumab, and no analysis records. Its study description also refers to a Phase II comparison against bevacizumab alone. Without a posted outcome for that comparator and without a sponsor-reported analysis, the stated treatment contrast cannot be estimated from the registry results. Comparing the two medians informally would not answer the comparison described in the study objective. The record does not report whether the two dose groups were randomized against one another or how they relate to the Phase I dose-finding process.
For time-to-event outcomes, a comparison commonly uses an estimate of relative event rates over follow-up, such as a hazard ratio, together with an uncertainty interval and a specified analysis. A hazard ratio is a model-based summary and is not obtained simply by dividing or subtracting medians. The posted results do not include a sponsor-reported hazard ratio, uncertainty interval, or p-value. The registry record also does not report model assumptions, covariate adjustment, or a statistical analysis plan.
How should censoring be understood?
Participants may have different lengths of follow-up. In general, censoring allows analysis to use the time observed for a participant even if progression or death has not been recorded before follow-up ends. Interpretation relies on the censoring process being handled appropriately; informative loss to follow-up can complicate results if participants who leave observation differ in their likely outcomes. The reasons for censoring, completeness of follow-up, and censoring assumptions are not reported in the registry record.
The endpoint timeframe is described as up to one year. That description does not establish that every participant was followed for that duration. The record does not report the actual follow-up distribution or the number still under observation at different times. With no survival curve or risk-set information, readers cannot assess how much data support estimates later in follow-up.
What do the safety counts represent?
The posted safety record gives the number at risk and the number affected by serious and other adverse events within each combination-dose group. These are counts, not comparative estimates of risk. The record does not provide exposure-adjusted rates, event timing, severity grading, attribution to study treatment, or a statistical comparison. In general, adverse-event summaries are descriptive unless a clearly specified comparative analysis is also reported.
Planning calculations, when available, are different from observed results. A sample-size illustration describes the number of participants or events that a hypothetical design might require under stated assumptions; it is not a measure of power achieved by a completed trial. No specific planning values are reported in the registry record. The record's general planning assumptions mention a two-sided significance level of 0.05, equal allocation, and a normal approximation, but do not provide a posted planning result for this comparison.
10. Multiplicity, Missing Data and Analysis Populations
The study lists PFS as the primary outcome and objective response rate as a secondary outcome. The record does not provide the result for objective response rate. It does not report a multiplicity strategy describing whether or how inference across outcomes, dose levels, or analyses was controlled. Without that information, a reader cannot determine whether any nominal statistical tests were intended to be interpreted as confirmatory. No inferential tests are posted for the primary outcome in any case.
Missing outcome information and censoring are different issues. In a time-to-event analysis, censoring records the last known event-free follow-up when an event has not been observed. Missing outcome data may arise when an observation or result is unavailable. The endpoint definition describes censoring, but the registry does not report the amount or reasons for missing data, completeness of progression assessments, or how missing observations were handled. These details are not reported in the registry record.
The analysis population is also not specified. The record does not state whether the posted summaries include all randomized participants, only participants who received treatment, or another population. It does not report exclusions, protocol deviations, or the treatment of participants who crossed over. Because crossover to the combination is described after worsening on bevacizumab, the handling of post-crossover observations could affect interpretation of a comparison, but that handling is not reported in the registry record.
The registry lists no analysis records. It does not report an analysis plan, interim analyses, stopping rules, or whether the termination affected outcome collection or analysis. These omissions mean the posted summaries cannot be assessed against a prespecified analysis strategy. The reported results should therefore be read as limited descriptive information rather than as a complete account of the planned comparative analysis.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| TPI 287 160 mg/m2 + Bevacizumab | 1 | 6 | 16.7% |
| TPI 287 140 mg/m2 + Bevacizumab | 2 | 6 | 33.3% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| TPI 287 160 mg/m2 + Bevacizumab | 6 of 6 |
| TPI 287 140 mg/m2 + Bevacizumab | 6 of 6 |
Most frequent serious adverse event terms
| Event term | Organ system | TPI 287 160 mg/m2 + Bevacizumab | TPI 287 140 mg/m2 + Bevacizumab |
|---|---|---|---|
| Intracranial Hemorrhage | Nervous system disorders | 0 of 6 | 1 of 6 |
| Chest Pain | Cardiac disorders | 1 of 6 | 0 of 6 |
| Hydrocephalus | Nervous system disorders | 1 of 6 | 0 of 6 |
| Wound Infection | Infections and infestations | 0 of 6 | 1 of 6 |
Most frequent non-serious adverse event terms
| Event term | Organ system | TPI 287 160 mg/m2 + Bevacizumab | TPI 287 140 mg/m2 + Bevacizumab |
|---|---|---|---|
| Fatigue | General disorders | 6 of 6 | 6 of 6 |
| Metabolism And Nutrition Disorders - (Other) | Metabolism and nutrition disorders | 5 of 6 | 6 of 6 |
| Generalized Muscle Weakness | Musculoskeletal and connective tissue disorders | 4 of 6 | 6 of 6 |
| Hyperglycemia | Metabolism and nutrition disorders | 5 of 6 | 5 of 6 |
| Memory Impairment | Nervous system disorders | 5 of 6 | 4 of 6 |
| Anemia | Blood and lymphatic system disorders | 5 of 6 | 3 of 6 |
| Gait Disturbance | Musculoskeletal and connective tissue disorders | 3 of 6 | 5 of 6 |
| Headache | Nervous system disorders | 3 of 6 | 5 of 6 |
| Seizure | Nervous system disorders | 3 of 6 | 5 of 6 |
| Constipation | Gastrointestinal disorders | 4 of 6 | 3 of 6 |
Serious events with the highest arm-level rate: intracranial hemorrhage (16.7%), chest pain (16.7%), hydrocephalus (16.7%), wound infection (16.7%).
The registry reports serious adverse events for both combination-dose groups. In the TPI 287 160 mg/m2 plus bevacizumab group, 1 of 6 participants was affected; in the TPI 287 140 mg/m2 plus bevacizumab group, 2 of 6 were affected. For other adverse events, the record reports 6 of 6 participants affected in each group. These counts describe participants recorded as affected and do not indicate the number of separate events.
The serious adverse event terms listed are intracranial hemorrhage, chest pain, hydrocephalus, and wound infection, each reported as 16.7%. The record does not specify the group attribution for those terms in the available summary. It also does not report event severity beyond the serious classification, timing, recurrence, treatment relatedness, or whether participants had multiple events. The number of deaths is not reported in the registry record.
These safety figures are descriptive. They do not establish that an event was caused by either study drug, and the counts alone do not permit a reliable comparison between dose groups. Exposure duration and a comparative safety analysis are not reported in the registry record. In general, differences in follow-up and treatment exposure can affect observed adverse-event counts, so the counts should be interpreted alongside those details when available.
12. What the Result Does, and Does Not, Mean
The results establish what the registry reports for the primary endpoint: median PFS was 12.5 weeks, with a full range of 6.0 to 37.0, for the TPI 287 160 mg/m2 plus bevacizumab group, and 9.5 weeks, with a full range of 7.0 to 44.0, for the TPI 287 140 mg/m2 plus bevacizumab group. Each group has a reported sample size of 6. These are group-specific summaries, not estimates of the effect of adding TPI 287 to bevacizumab.
The numbers do not establish that either dose level produced better or worse outcomes. The registry does not report a direct statistical comparison, an uncertainty interval, or a p-value for the posted medians. It also does not report the bevacizumab-alone PFS results described as the Phase II comparator. The study's intended question and the available outcome summaries therefore do not align closely enough to support a conclusion about the effect of adding TPI 287.
Nor do the results establish equivalence or absence of a difference. Failure to report a comparative estimate is not evidence that the treatment groups have the same outcomes. The small posted group sizes and absence of uncertainty estimates leave the precision of the observed medians unclear. The full ranges show the spread of reported values but do not quantify uncertainty about the treatment effect.
The safety record documents serious and other adverse events in the two combination groups, but it does not provide enough information to attribute events to treatment or compare safety statistically. The listed serious event terms and counts should not be treated as a complete causal assessment. The trial is marked terminated, but the reason for termination is not reported in the registry record, so termination cannot by itself be interpreted as evidence about efficacy or safety.
Overall, the registry supports a cautious description of PFS and adverse-event summaries for the two listed combination-dose groups. It does not answer the stated Phase II comparison of bevacizumab plus TPI 287 against bevacizumab alone. No treatment recommendation or claim of benefit follows from these posted summaries.
13. Why This Trial Matters Statistically
This trial illustrates why a study's stated clinical question, assigned treatment groups, and posted outcome results must be considered together. The record describes an intended comparison of bevacizumab plus TPI 287 with bevacizumab alone, but its primary outcome results list two groups receiving the combination at different TPI 287 dose levels. Group-specific medians cannot substitute for a missing comparator outcome or sponsor-reported contrast.
It also illustrates the distinction between descriptive and inferential statistics. A median gives a compact account of observed time-to-event data, while a full range describes the observed extremes. Neither provides a confidence interval or estimates the uncertainty of a treatment difference. Without the event and censoring details, survival curves, or a comparative analysis, the precision and shape of the PFS experience remain unclear.
The Phase I/II structure adds an interpretive challenge. Dose finding and comparative assessment answer different questions. The record describes a dose-level objective in Part I and a treatment-control objective in Part II, but does not explain how the posted outcome groups map to those stages or to randomization. That uncertainty limits interpretation beyond the reported group summaries.
Finally, the registry shows how incomplete analysis reporting constrains scientific conclusions. The trial is terminated, no analysis records are posted, and the reason for termination, analysis populations, multiplicity approach, and missing-data handling are not reported. These limitations do not erase the observed summaries, but they set a boundary on what those summaries can establish. Transparent reporting of the intended comparison and its uncertainty is essential for evaluating trial 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 posted primary outcome results cover two combination-dose groups, while the study description identifies bevacizumab alone as the Phase II comparator. A bevacizumab-alone PFS result and a sponsor-reported comparative analysis are not reported in the registry record. The relationship between the listed dose groups and the randomized comparison is also not reported.
Each posted group has a sample size of 6. The registry provides medians and full ranges but no confidence intervals, event counts, censoring counts, survival curves, or follow-up distribution. This makes it difficult to assess precision, the timing of events, and the contribution of censoring to the endpoint estimates.
The analysis population, treatment exposure, handling of crossover, missing-data procedures, and statistical analysis plan are not reported in the registry record. Neither are the reason for termination, the result for the listed secondary outcome, nor a multiplicity strategy. These gaps prevent a full evaluation of whether the posted summaries reflect the intended analysis and how they should be interpreted alongside the trial's stated objectives.
The safety summaries are limited to counts and listed serious-event terms. Event attribution, timing, severity details, exposure-adjusted rates, and comparative safety analyses are not reported in the registry record. The record also does not report the number of deaths. Therefore, the adverse-event information is descriptive and does not establish causation or provide a complete basis for comparing treatment safety.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Phase I/II Bevacizumab Versus Bevacizumab Plus TPI 287 for Recurrent Glioblastoma — NCT01582152.
- Registry record last updated 2019-06-05; status last verified 2019-05.
- Results first posted 2019-06-05.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
This is a randomized Phase I/II trial of bevacizumab with TPI 287 in adults with recurrent glioblastoma. The registry records no masking, a crossover model, enrollment of 12, and terminated status. Its stated objectives include identifying a tolerable combination dose and assessing the combination against bevacizumab alone.
The primary outcome is PFS, measured from registration until progression or death, or censored at the last date known alive without progression. Posted results give median PFS and full ranges for TPI 287 160 mg/m2 plus bevacizumab and TPI 287 140 mg/m2 plus bevacizumab. No sponsor-reported comparative analysis is posted, and a bevacizumab-alone primary outcome result is not reported in the registry record.
One secondary outcome, objective response rate, is listed without a posted result. Safety summaries include serious and other adverse-event counts for the two combination groups. The reason for termination, analysis population, statistical analysis plan, missing-data procedures, and handling of crossover are not reported in the registry record.