Multiple Doses and Regimens of Cabozantinib in Subjects With Grade IV Astrocytic Tumors in First or Second Relapse: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Multiple Doses and Regimens of Cabozantinib in Subjects With Grade IV Astrocytic Tumors in First or Second Relapse (NCT01068782), a randomized phase 2 study. The primary outcome is Randomized Phase: Number of Participants With Treatment Emergent Adverse Events (TEAEs).

2. Clinical Question

The study title and description place the trial in people with grade IV astrocytic tumors at first or second relapse. Its posted primary endpoint is a safety measure: participants with treatment-emergent adverse events (TEAEs). The record separately lists progression-free survival at 6 months, as assessed by an independent radiology facility, as a secondary outcome in the expansion phase. That secondary outcome is not the primary result described by the randomized-phase endpoint.

The clinical question is therefore about adverse events observed under different cabozantinib dosing schedules, rather than whether one regimen is more effective than another. The registry calls the study non-comparative, despite its randomized allocation to multiple regimens. The result should be read as arm-specific counts in a small study record, not as a formal estimate of comparative safety or efficacy.

3. Trial Design

This was a Phase 2, randomized, open-label study with a parallel-group model. Participants were assigned among four cabozantinib regimens. Masking was listed as none. The study title describes the regimens as single-agent XL184, and the intervention entry names cabozantinib.

The dosing schedules differed in starting dose and in whether dosing was continuous or interrupted. Arm 1 began at 25 mg once daily by mouth continuously. Arm 2 began at 75 mg once daily continuously. Arm 3 began at 125 mg once daily for 2 weeks and then continued at 50 mg once daily. Arm 4 initially used 125 mg once daily for 2 weeks followed by 1 week off; for the remainder of treatment, the schedule was 125 mg once daily every 3 weeks followed by 1 week of rest. These are descriptions of the recorded regimens, not evidence that the schedules were equivalent or directly compared statistically.

The registry record lists enrollment of 19 and identifies the study as terminated. The recorded start is 2010-04 and primary completion is 2011-05. The sponsor is Exelixis. The registry record does not report the reason for termination, the allocation ratio, the randomization procedure, or the planned duration of treatment for each participant.

The primary outcome was measured in the randomized phase and followed participants for up to 13 months. Its population was the safety population: participants who enrolled and received at least one dose of study drug. A separate secondary outcome was progression-free survival at 6 months in the expansion phase, assessed by an independent radiology facility. The registry record does not report a detailed expansion-phase design or the results for that outcome in the information available for this analysis.

The word “randomized” describes how the record says treatment assignment was made; “non-comparative” cautions against treating the study as a definitive head-to-head test. No statistical analysis is posted in the registry record. Details such as stratification, concealment of allocation, sample-size justification, and any planned interim analyses are not reported in the registry record.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The safety population included all participants who enrolled and received at least one dose of study drug.

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 Randomized Phase: Number of Participants With Treatment Emergent Adverse Events (TEAEs), measured over Up to 13 months.

The primary endpoint counts participants in the randomized phase who experienced at least one treatment-emergent adverse event. An adverse event was defined as an event beginning on or after the first dose, or an ongoing event that worsened in severity after the first dose. TEAEs were events starting on or after the first administration of study drug through the date of the last dose plus 30 days. The endpoint timeframe is up to 13 months.

This type of endpoint is used to describe how many participants experienced an event during treatment and the defined follow-up window. It provides a straightforward summary of observed safety events in each regimen group. Because it is a participant count, it does not describe the number of separate events per person, the duration or severity of each event, or the time until an event occurred. The primary count also does not by itself establish whether an event was caused by study drug.

The registry record specifies that the primary count includes events regardless of causality. It directs readers to the reported adverse events module for a summary of serious and non-serious adverse events. Seriousness and causality are distinct concepts: seriousness concerns the event’s classification, while causality concerns its relationship to treatment. The primary endpoint’s broad count should therefore be distinguished from detailed event-level summaries.

The result is reported as counts in the safety population. No confidence intervals, hypothesis tests, or adjusted estimates are posted for this endpoint. Accordingly, the endpoint supports a descriptive account of the observed participants with TEAEs, not a precise estimate of comparative risk between schedules.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical interpretation depends on separating what was observed from what was estimated. The primary endpoint is a count of participants with at least one treatment-emergent adverse event. The posted arm results are counts within the safety population: 5 of 5, 3 of 4, 6 of 6, and 4 of 4. These values describe the record’s observed participants. They are not model-based estimates of event risk, and there are no posted confidence intervals or hypothesis tests for the endpoint.

What does a participant count capture?

A participant-level event count treats a participant as having the endpoint if at least one qualifying event occurred in the defined period. It does not count repeated episodes as additional participants. It also does not encode event duration, time to onset, or severity. A separate event-level listing can add detail about event types and seriousness, but it answers a different descriptive question from the primary participant count.

The endpoint definition includes events starting on or after the first dose, and events already ongoing at the first dose if they worsened in severity afterward. TEAEs are bounded by the first administration and the date of the last dose plus 30 days. The primary endpoint’s timeframe is up to 13 months. Such rules determine which events qualify; they should not be confused with a statistical adjustment for differences in follow-up.

What would a confidence interval mean?

In general, a confidence interval describes the sampling uncertainty around an estimated quantity under a specified statistical method. It can show how imprecisely the data estimate a population value. It is not a guarantee that the true value lies within the interval, nor does it resolve bias, missing data, or problems with study design. No confidence interval is reported for these primary counts in the registry record, so the precision of a comparative estimate cannot be assessed from the posted primary result.

Why not treat the arms as a formal comparison?

Randomization can support comparisons when a trial is designed and analyzed for that purpose, with a defined estimand, comparison, and statistical plan. Here the record describes the study as non-comparative and reports zero analyses. The registry record does not state a primary pairwise contrast, a multiplicity approach, or an inferential model for comparing regimens. Consequently, the arm-specific counts can be described, but a formal treatment contrast should not be reconstructed from them.

The record indicates that values were missing or not estimable for a two-arm comparison of the binary endpoint. That means no such comparison is available in the posted results. It does not establish why the analysis was unavailable beyond the recorded note, and it does not justify filling in an estimate by calculation. The sponsor’s own analysis does not take precedence in this record because no discrepancy is flagged; however, no sponsor analysis is posted for the primary result.

How should planning illustrations be understood?

General sample-size planning asks how many participants might be needed under chosen assumptions to detect a specified effect with a selected power and significance level. The planning illustrations associated with this record are based on the observed effect, equal allocation between two arms, a two-sided significance level of 0.05, and a normal approximation. The effect to detect is expressed as a stated fraction of the observed effect. For non-time-to-event outcomes, such illustrations are expressed as participants per arm for 80% power or other listed power levels.

These planning values are illustrations, not results from the trial and not observed power. They do not demonstrate that the study was adequately powered, nor do they provide evidence of a treatment difference. The record does not report the trial’s own sample-size justification. In addition, the missing or non-estimable two-arm comparison limits what can be inferred from illustrations that depend on an observed effect.

Why does over-dispersion matter in general?

For count outcomes, a simple model may assume a particular relationship between the mean and variability. Over-dispersion occurs when the observed variability is greater than that assumption allows. If ignored, estimated uncertainty may be too small and inferential conclusions may be overstated. This is a general statistical consideration for count data; the primary endpoint is a count of participants with an event rather than a count of all event episodes. The registry record does not report an over-dispersion assessment or a model for this endpoint.

Overall, the appropriate statistical reading is descriptive: the registry record presents the number of safety-population participants counted with TEAEs in each regimen. It does not provide the inferential framework or precision estimates needed to make a comparative conclusion. General statistical concepts help clarify what such analyses would mean, but they cannot substitute for analyses not reported in the registry record.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Arm 1", "Arm 2", "Arm 3", "Arm 4")
n      <- c(5, 4, 6, 4)
events <- c(5, 3, 6, 4)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Arm 1  5  5  100.0  47.8  100.0  56.6  100.0  54.1  99.6
ARM  Arm 2  4  3  75.0  19.4  99.4  30.1  95.4  28.4  94.7
ARM  Arm 3  6  6  100.0  54.1  100.0  61.0  100.0  59.0  99.6
ARM  Arm 4  4  4  100.0  39.8  100.0  51.0  100.0  47.8  99.5

11. Multiplicity, Missing Data and Analysis Populations

The registry record does not report a statistical plan for handling multiplicity. Multiplicity arises when a study makes several statistical comparisons or evaluates several endpoints, because repeated testing can increase the chance of a chance finding. The trial included multiple cabozantinib regimens and lists a primary safety endpoint as well as a secondary progression-free survival outcome at 6 months in the expansion phase. However, no inferential analyses are posted, and the record does not state whether any multiplicity adjustment or testing hierarchy was planned.

Missing-data procedures are also not reported in the registry record. The primary safety population is defined as participants who enrolled and received at least one dose of study drug. The registry record does not explain how participants with incomplete follow-up, withdrawal, or unavailable event information were treated in the analysis. It also does not report the amount or pattern of missing data for the primary endpoint.

The record lists no analyses. It therefore provides no adjusted comparisons, confidence intervals, or statistical tests to evaluate. The note that values were missing or not estimable for a two-arm comparison of the binary endpoint should be read as a limitation of the posted comparison, not as evidence about the direction or size of a treatment effect.

The registry record does not report an intention-to-treat analysis population, a per-protocol population, or a separate population for the secondary outcome in the information. The defined safety population is the only stated analysis population for the primary outcome. These omissions constrain interpretation: readers can describe the recorded counts, but cannot determine from this record how missingness, multiple outcomes, or alternative analysis populations might have affected inferential results.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

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: intracranial tumour haemorrhage (25.0%), mental status changes (25.0%), nausea (25.0%), vomiting (25.0%), dehydration (25.0%).

The serious adverse-event figures are participant counts by regimen: 2 of 5 in Arm 1, 1 of 4 in Arm 2, 1 of 6 in Arm 3, and 1 of 4 in Arm 4. The non-serious adverse-event figures are 5 of 5, 3 of 4, 6 of 6, and 4 of 4, respectively. The death figures are zero in each arm. These counts describe participants recorded as affected; they are not counts of all individual event episodes.

The adverse-event record lists intracranial tumour haemorrhage, mental status changes, nausea, vomiting, and dehydration at 25.0%, and migraine at 20.0%. The listed percentages are not assigned to a particular arm in the registry record used for this account, so they should not be interpreted as arm-specific rates. Seriousness does not mean that an event was necessarily caused by study drug. The primary endpoint includes events regardless of causality, and no comparative statistical analysis of the serious-event figures is reported in the registry record.

These summaries are descriptive and apply to the participants represented in each arm. They do not establish differences between regimens, and the registry record does not provide uncertainty estimates for these safety figures.

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

The primary result answers a narrow descriptive question: how many safety-population participants in each randomized-phase regimen had at least one treatment-emergent adverse event under the registry’s definition and follow-up window. The record reports 5 of 5 in Arm 1, 3 of 4 in Arm 2, 6 of 6 in Arm 3, and 4 of 4 in Arm 4. The corresponding serious adverse-event counts are also recorded by arm, along with zero deaths in each arm.

These observations do not show that one dosing schedule is safer than another. The study is described as non-comparative, no analyses are posted, and the registry record does not report comparative estimates or their uncertainty. The raw arm counts should not be treated as evidence of a treatment difference merely because the values are not identical.

The endpoint does not establish efficacy. Although the registry record lists progression-free survival at 6 months as a secondary outcome in the expansion phase, no result for it is included in the registry record. The adverse-event count also does not determine whether an individual event was caused by cabozantinib, because the primary outcome includes events regardless of causality.

The result does not reveal the total number of event episodes, the duration of events, or their timing. Nor does it tell whether an absence of a recorded event reflects complete follow-up. Missing-data handling and detailed analysis populations beyond the stated safety population are not reported in the registry record.

The record is useful for documenting what was posted for this study and for distinguishing the primary safety endpoint from the secondary outcome. Its limits are substantial for statistical inference: there are no posted analyses, no confidence intervals, and no reported comparative statistical plan. The appropriate conclusion is therefore confined to the recorded counts and event summaries, without a claim about benefit, lack of benefit, or a ranking of regimens.

14. Why This Trial Matters Statistically

This trial is statistically notable because random allocation and a non-comparative design appear together in the registry record. Randomization describes assignment to regimens, but it does not mean that every study is intended or analyzed as a formal head-to-head comparison. The non-comparative label and absence of posted analyses are central to interpreting the arm results: they are descriptive rather than evidence of tested differences.

The primary outcome also illustrates why the endpoint definition matters. It counts participants with at least one qualifying treatment-emergent adverse event, regardless of causality. That summary is not interchangeable with an event-level count, a severity score, or a time-to-event analysis. The registry supplies a defined follow-up window and a safety-population definition, which help clarify the scope of the count, but it does not report missing-data procedures or uncertainty estimates.

Multiple dosing schedules in a small Phase 2 study can generate important descriptive safety information, but interpretation must remain within the design and reporting limits. The record lists four schedules and participant counts in each, alongside serious and non-serious adverse-event summaries. It does not provide a statistical basis for ranking those schedules.

The registry record also distinguishes the randomized-phase primary safety endpoint from a secondary progression-free survival outcome in the expansion phase. That separation matters: safety counts cannot answer the secondary tumor-control question, and the secondary outcome is not reported in the registry entries. Overall, the trial is a useful case for learning to separate randomization, descriptive reporting, and inferential comparison rather than treating them as synonymous.

15. 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 registry record identifies the study as terminated but does not report the reason. It lists enrollment of 19, but the primary result is reported across arm-specific safety populations, and the record does not explain any difference between enrollment and the participants represented in a particular analysis. Follow-up is stated as up to 13 months; individual follow-up durations are not reported.

The primary endpoint is limited to participants with at least one TEAE. It does not convey event frequency per participant, duration, timing, severity distribution, or treatment attribution. The event definition is specified, but the registry record does not report how event collection was managed or how incomplete safety information was handled.

The study is described as non-comparative and no statistical analyses are posted. There are no confidence intervals, hypothesis tests, adjusted estimates, or reported comparison plan for the primary endpoint. Random assignment should not be mistaken for evidence that formal comparisons were made. The registry record also does not report the allocation ratio, randomization method, sample-size justification, or any interim analyses.

The separate secondary outcome is progression-free survival at 6 months in the expansion phase, assessed by an independent radiology facility. Its result and associated analysis details are not reported in the registry record. The record does not state a multiplicity strategy or provide details of analysis populations beyond the safety definition for the primary outcome.

Finally, the event summaries come from a small study, and their precision cannot be assessed from the posted counts alone. These limitations mean that the record supports description of observed safety findings, but not broad claims about comparative safety or efficacy.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Multiple Doses and Regimens of Cabozantinib in Subjects With Grade IV Astrocytic Tumors in First or Second Relapse — NCT01068782.
• Registry record last updated 2026-02-25; status last verified 2026-02.
• Results first posted 2026-02-25.
• 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.

19. Record Summary

The registry record describes a Phase 2, randomized, open-label, parallel study of multiple cabozantinib regimens for grade IV astrocytic tumors in first or second relapse. It lists Exelixis as sponsor, enrollment of 19, status as terminated, a start of 2010-04, and primary completion in 2011-05.

The primary randomized-phase outcome is the number of participants with treatment-emergent adverse events, in the safety population, over up to 13 months. The record also lists progression-free survival at 6 months as a secondary outcome in the expansion phase. No statistical analyses are posted, and the study is described as non-comparative.

The primary counts and adverse-event summaries describe the participants recorded in each regimen group. They do not provide comparative estimates or establish treatment effects. The registry record does not report the reason for termination, the statistical plan, missing-data procedures, or the result for the secondary outcome.