Vismodegib in Treating Patients With Recurrent or Refractory Medulloblastoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Vismodegib in Treating Patients With Recurrent or Refractory Medulloblastoma (NCT00939484), a na phase 2 study. The primary outcome is Objective Response (CR+PR) Sustained for ≥ 8 Weeks.

2. Clinical Question

The endpoint focuses on whether tumor response was both observed and sustained. A complete response was defined as disappearance of all enhancing tumor; a partial response was defined as a reduction in tumor size of at least 50%. The registry record does not report the primary endpoint result, its confidence interval, or an analysis for any stratum. It therefore does not establish what proportion of participants met the response definition.

The study also listed progression-free survival, duration of objective response, and pharmacokinetic parameters including cerebrospinal fluid penetration as secondary outcomes. Their analysis results are not reported in the registry record. The clinical question is thus clear from the planned endpoint, but the posted record does not answer it with an outcome estimate.

3. Trial Design

This was a phase II clinical trial sponsored by the National Cancer Institute. It enrolled 31 participants and is recorded as completed. The registry lists a start date of 2009-06 and a primary completion date of 2013-05. The official title describes an evaluation of the efficacy and safety of GDC-0449 in adults with recurrent or refractory medulloblastoma; the intervention is named vismodegib.

The study used a single-group design, with no allocation to a separate comparator arm and no masking. Participants received vismodegib orally once daily on days 1 through 28. Courses repeated every 28 days, for up to 26 courses, unless disease progression or unacceptable toxicity occurred. Because there was no concurrent control group, the study does not directly estimate a treatment difference against another therapy or against placebo. Changes in outcome cannot be separated from the course the disease might otherwise have taken using this design alone.

The primary outcome was objective response, defined as complete or partial response sustained for at least 8 weeks, with assessment extending up to 12 months. The planned reporting was separate for each stratum. The registry identifies Stratum A as having PTCH/SHH pathway inactivation, Stratum B as having PTCH/SHH pathway activation, and Stratum C as having unknown PTCH/SHH pathway activation. The record does not report the operational details used to assign participants to strata.

Secondary outcomes included progression-free survival, duration of objective response, and pharmacokinetic parameters of vismodegib, including cerebrospinal fluid penetration. Four outcome measures are listed, but no analysis results are reported in the registry record. The protocol's sample-size justification, detailed statistical analysis plan, and rules for handling deviations from the planned treatment 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.

Before assignment: Total 32 patients were registered on this study, 8 in Stratum A, 21 in Stratum B and 3 in Stratum C. One patient in Stratum B was declared ineligible.

Overall Study

Reasons for not completing this period:

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: One patient in Stratum B was ineligible for the study and thus was not included in the baseline table.

6. Primary Endpoint

The registry lists the primary outcome as Objective Response (CR+PR) Sustained for ≥ 8 Weeks, measured over Up to 12 months.

The primary endpoint is the proportion of participants with an objective response sustained for at least 8 weeks, assessed for up to 12 months. Objective response combines complete and partial responses: complete response means disappearance of all enhancing tumor, while partial response means a reduction in tumor size of at least 50%. Requiring the response to persist for 8 weeks distinguishes a sustained response from a response that is observed briefly.

A proportion is a natural summary for an endpoint that classifies each participant as meeting or not meeting a defined response criterion. Its interpretation depends on both the response definition and the participants included in the denominator. The registry specifies that objective response was to be reported separately for each disease-pathway stratum. Separate estimates could describe response within each group, but they would not by themselves establish a difference between strata or a causal treatment effect.

Confidence intervals are useful for showing the uncertainty around a proportion, especially when estimates are based on limited numbers of participants. The endpoint lists a 95% confidence interval as its dispersion measure. However, the registry record does not report the number of responders, the estimated proportions, or their confidence intervals. The stated outcome definition can therefore be described, but the posted record does not permit assessment of the observed response frequency or its precision.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a single-group design estimate?

In a single-group study, participants receive the study intervention and outcomes are described within that group. For a binary endpoint such as sustained objective response, the central summary is the proportion meeting the response definition. This is a descriptive estimate of the observed group, not a randomized contrast. A treatment difference requires a comparator and a defined contrast; neither is present in this design. The registry lists allocation as not applicable and the model as single group.

Randomization helps balance measured and unmeasured characteristics across groups on average. Without random assignment and a concurrent control group, differences between observed outcomes and an external expectation can reflect many influences besides treatment. A single-group estimate can still describe what was recorded among enrolled participants, but causal interpretation is more limited. General statistical principle: the design determines which comparisons an estimate can support.

What does the endpoint proportion mean?

The primary outcome classifies a participant according to whether a complete or partial response was sustained for at least 8 weeks, with the outcome assessed for up to 12 months. A proportion uses the number meeting that definition relative to the number included in the analysis. The registry record does not report either the response count or the analysis denominator, so the proportion cannot be reconstructed from the posted information. The enrolled total of 31 is not necessarily the same as an endpoint-specific analysis denominator.

The planned reporting is separate for each stratum. For a stratified presentation, each estimate describes the corresponding group, and each interval reflects the data and method for that group. Estimates from separate groups do not automatically constitute a formal test of difference. A formal comparison would need a stated method and a defined hypothesis; those details are not reported in the registry record.

Why report a confidence interval?

A confidence interval expresses sampling uncertainty around an estimate under a specified statistical procedure. It is not a guarantee that the true value lies within the particular interval, nor does it account for every possible source of bias. Its interpretation depends on the analysis population, endpoint definition, and method used to calculate it. The registry identifies a 95% confidence interval for the primary outcome, but the actual interval and its calculation method are not reported in the registry record. General statistical principle: an estimate without an uncertainty measure can make the precision of a result difficult to judge.

For a binary outcome, the interval method matters, particularly when the number of participants or events is limited. Different methods can yield different interval widths and bounds. The record does not identify the interval procedure, so no specific method should be assumed. Nor can an interval be inferred from the enrollment count alone: the endpoint event count and analysis denominator are also needed.

What can the secondary outcomes add?

Progression-free survival concerns the time until a defined progression event or other specified endpoint event. Duration of objective response concerns how long a response lasts among participants who respond. Pharmacokinetic outcomes characterize drug exposure; cerebrospinal fluid penetration concerns drug presence in that compartment. These outcomes answer different questions and require their own definitions and analysis methods. Although they are listed as secondary outcomes, their definitions, estimates, and analysis results are not reported in the registry record. General statistical principle: outcomes should not be treated as interchangeable, because each measures a different aspect of disease course or drug exposure.

How should the serious-event summaries be read?

The registry reports, by stratum, numbers at risk and numbers affected for serious adverse events and other adverse events. These are descriptive counts. The selected serious-event terms are accompanied by percentages, but the listed information does not establish the denominator or time basis for each term-specific percentage. A percentage can be interpreted only in relation to its denominator and event definition. These summaries do not provide a randomized comparison and do not by themselves estimate excess risk attributable to treatment.

General statistical principle: adverse event summaries are affected by follow-up duration, ascertainment, definitions, and whether recurrent events or participants with at least one event are counted. The record's listed counts should therefore be read as reported group summaries, not as rates per unit of time or as a complete safety profile. Separate death counts are not reported in the registry record.

What is the role of sample-size planning?

A sample-size calculation is a planning exercise based on assumptions about the endpoint, target effect, significance level, power, allocation, and statistical method. It does not describe an observed trial result, and it is not observed power. The registry record does not report the trial's sample-size justification. The stated planning illustration assumptions describe a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and effects expressed as fractions of the observed effect; time-to-event illustrations use a total event count, while other illustrations refer to participants per arm. These assumptions do not turn this single-group trial into a two-arm comparison. The applicable comparison is not like-for-like because no randomized control arm is reported, and the numerical illustration results are not reported in the registry record.

General statistical principle: over-dispersion occurs when variability is greater than a model assumes. If a model ignores over-dispersion, uncertainty can be understated and tests can appear more precise than warranted. The registry record does not report whether over-dispersion was considered or whether it is relevant to any analysis in this study.

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("Stratum A (PTCH/SHH Pathway Inactivated)", "Stratum B (PTCH/SHH Pathway Activated)", "Stratum C (Unknown PTCH/SHH Pathway Activation)")
n      <- c(8, 20, 3)
events <- c(0, 0, 0)
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  Stratum A (PTCH/SHH Pathway Inactivated)  8  0  0.0  0.0  36.9  0.0  32.4  0.3  33.6
ARM  Stratum B (PTCH/SHH Pathway Activated)  20  0  0.0  0.0  16.8  0.0  16.1  0.1  16.1
ARM  Stratum C (Unknown PTCH/SHH Pathway Activation)  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2

11. Multiplicity, Missing Data and Analysis Populations

The primary response outcome was planned for separate reporting in each of three strata. Considering multiple outcomes or multiple subgroup estimates can raise multiplicity concerns: as the number of statistical questions increases, the chance of at least one apparently notable finding can also increase if each is assessed independently. General statistical principle: interpretation is clearer when the primary question, secondary questions, and any adjustment strategy are specified in advance. The registry record does not report a multiplicity adjustment, a hierarchy for testing outcomes, or detailed rules for interpreting multiple stratum-specific estimates.

Four outcome measures are listed, including the primary sustained-response endpoint and secondary outcomes of progression-free survival, duration of objective response, and pharmacokinetic parameters. No analysis results are reported in the registry record. The record does not describe whether all listed outcomes were analyzed, how multiple measurements were handled, or whether any outcome definitions changed during the study.

Missing-data handling is not reported in the registry record. In particular, the record does not specify how participants without an evaluable response, with incomplete follow-up, or with treatment discontinuation were handled in the primary endpoint analysis. General statistical principle: missing outcomes can affect an estimate if the participants with missing information differ systematically from those with observed information. Without the analysis denominator and missing-data rules, the extent and direction of any such effect cannot be assessed.

The analysis populations are not reported in the registry record. It is not stated whether the primary outcome used all enrolled participants, participants who received treatment, or a narrower evaluable group. The enrollment count alone does not identify the denominator for an endpoint estimate. Rules for exclusions, protocol deviations, and timing of analysis are also not reported in the registry record.

12. 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: dyspepsia (66.7%), death nos (66.7%), seizure (33.3%), syncope (33.3%), lung infection (25.0%).

The registry reports serious adverse events by three pathway strata. In Stratum A, 8 participants were at risk and 5 were affected; in Stratum B, 20 were at risk and 12 were affected; in Stratum C, 3 were at risk and 3 were affected. These are counts of participants, not a treatment-versus-control comparison. The record also reports that, in each stratum, the number at risk and affected for other adverse events was 8 and 8, 20 and 20, and 3 and 3, respectively.

Selected serious adverse event terms are listed with percentages: Dyspepsia and Death NOS are each 66.7%; Seizure and Syncope are each 33.3%; Lung infection and Headache are each 25.0%. The listed information does not identify the denominator or time basis for each term-specific percentage, so those percentages should be read as registry-reported summaries rather than independently interpretable rates. Death NOS appears as a listed serious-event term; separate death counts are not reported in the registry record.

These safety figures do not establish that an event was caused by vismodegib, and they cannot be compared with a control group in this single-group design. They also do not show how event occurrence varied over follow-up. The figures are descriptive and should be interpreted alongside the stated strata and the limits of the registry record.

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

The registry record establishes that a phase II, single-group study evaluated vismodegib in adults with recurrent or refractory medulloblastoma. It specifies a sustained objective-response endpoint, its response definitions, the planned assessment period, and the intention to report response separately across pathway strata. It also lists secondary outcomes and provides serious adverse event summaries by stratum.

The record does not report the primary response result, its confidence interval, or analysis results for the listed outcomes. As a result, the posted information does not quantify how many participants had a sustained complete or partial response, how precise an estimate might be, or what the progression-free survival and other secondary outcome findings were. The response criterion and planned follow-up window are not substitutes for those missing estimates.

The study's single-group design also limits what any within-group result could establish. Without a randomized comparator, the results cannot directly quantify a difference between vismodegib and another treatment or no study treatment. This does not make the design or its descriptive summaries meaningless; it defines the scope of conclusions they can support. General statistical principle: association or change observed within one group is not, on its own, a randomized estimate of causal effect.

The serious-event counts and listed terms describe reported safety information, but they do not establish causation or comparative risk. The registry record does not report separate death counts, the denominators for the listed term-specific percentages, or the full details needed to assess exposure time and ascertainment.

Accordingly, the posted record supports a description of the research question, treatment schedule, endpoint definitions, and certain safety counts. It does not support a numerical conclusion about objective response, its precision, or comparative efficacy. Missing outcome results should not be interpreted as evidence that no responses occurred or that an endpoint was not measured; the registry record does not report those findings.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of how design and reporting shape interpretation. A phase II study can investigate whether a prespecified response outcome is observed and can describe its frequency and uncertainty. Here, the endpoint is explicitly defined as a complete or partial response sustained for at least 8 weeks, with assessment up to 12 months. That definition clarifies what would count as a response, while the planned separate reporting by pathway stratum identifies how the primary outcome was intended to be organized.

The single-group design also makes the limits of inference especially important. A response proportion would describe the enrolled study group, but without a concurrent randomized comparator it would not directly estimate a treatment contrast. Statistical conclusions must match the design: a clearly defined endpoint does not remove the limitations arising from the absence of a control group.

Reporting precision is equally important. The endpoint specifies a 95% confidence interval, but no response estimate or interval is reported in the registry record. Readers therefore cannot judge the observed frequency or how much uncertainty surrounds it. This illustrates why outcome definitions, denominators, estimates, and uncertainty measures all matter in a clinical trial report.

The registry also lists several distinct secondary outcomes and provides safety summaries across three pathway strata. These details show why analysis populations, outcome-specific definitions, and multiplicity plans matter: different outcomes address different questions, and multiple subgroup summaries should not be treated as formal comparisons unless a suitable analysis is reported. The analysis results and these methodological details are not reported in the registry record.

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 main trial-specific limitation in the posted record is the absence of primary and secondary analysis results. Although 31 participants were enrolled and the primary endpoint is defined, the registry record does not report the number meeting the response definition, the stratum-specific estimates, or confidence intervals. It is therefore not possible to assess the frequency or precision of the primary outcome from the posted information.

The study is listed as a single-group trial with allocation not applicable and no masking. No concurrent randomized comparator is reported, so the design cannot directly estimate a treatment difference. The record does not report details of response assessment procedures, including whether assessors were independent, or the rules used to assign participants to the three pathway strata. Those details should not be inferred from the stratum labels.

The analysis denominator and analysis population are not reported in the registry record. Nor are missing-data procedures, participant-level follow-up completeness, handling of participants who stopped treatment, or the treatment of unevaluable outcomes. The enrollment total cannot be assumed to equal the number included in each endpoint analysis.

The record lists progression-free survival, duration of objective response, and pharmacokinetic parameters as secondary outcomes, but does not report their definitions or results in the posted information. It does not report a detailed analysis plan, multiplicity strategy, interim analysis procedures, or a sample-size justification. Selected serious adverse event terms are listed with percentages, but their term-specific denominators and time basis are not stated in the listed information. Separate death counts are not reported in the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Vismodegib in Treating Patients With Recurrent or Refractory Medulloblastoma — NCT00939484.
• Registry record last updated 2016-02-25; status last verified 2015-03.
• Results first posted 2016-02-25.
• 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

NCT00939484 is a completed phase II, single-group trial sponsored by the National Cancer Institute. It enrolled 31 adults with recurrent or refractory medulloblastoma and evaluated vismodegib, taken orally once daily on days 1 through 28 in repeating 28-day courses for up to 26 courses unless progression or unacceptable toxicity occurred.

The primary outcome was the proportion with complete or partial response sustained for at least 8 weeks, assessed for up to 12 months and planned for separate reporting by pathway stratum. Secondary outcomes included progression-free survival, duration of objective response, and pharmacokinetic parameters including cerebrospinal fluid penetration. The registry record does not report analysis results for the listed outcomes.

Serious adverse event counts are reported for Stratum A, Stratum B, and Stratum C, along with selected serious-event terms and percentages. The study has no concurrent randomized comparator, and the registry record does not report the primary response estimate, its confidence interval, the analysis population, or missing-data methods.