Therapy to Treat Ewing's Sarcoma, Rhabdomyosarcoma or Neuroblastoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Therapy to Treat Ewing's Sarcoma, Rhabdomyosarcoma or Neuroblastoma (NCT00923351), a non-randomized phase 1, phase 2 study. The primary outcome is Number of Participants With a Positive Immune Response as Evidenced by the Delayed Type of Hypersensitivity (DTH) Reaction Assay.

2. Clinical Question

3. Trial Design

This was a completed Phase 1 and Phase 2 pilot study sponsored by the National Cancer Institute (NCI). The registry lists enrollment of 44 and a parallel, non-randomized design with no masking. The study started on 2007-06-02 and had a primary completion date of 2012-08-30. Those dates describe the registry record; they do not establish the timing of individual participants' follow-up.

The study examined tumor vaccination and rhIL-7 following standard multimodality therapy in patients with high-risk pediatric solid tumors. The record describes two arms: Arm A participants did not receive rhIL-7, and Arm B participants did. Arm A's description refers to six participants with Ewing's sarcoma family tumors. Arm B's description lists eight participants with rhabdomyosarcoma, fifteen with Ewing's sarcoma family tumors, two with desmoplastic small round cell tumor, and one with synovial cell sarcoma. These arm descriptions are not a randomized allocation scheme, and the record does not establish that the arms had comparable baseline characteristics.

The brief summary describes collection of tumor tissue to prepare vaccine and white blood cells for immune-system rebuilding. It also describes standard chemotherapy and, if needed, additional fludarabine and cyclophosphamide before immunotherapy. The record lists tumor-purged/CD25-depleted lymphocytes, a tumor-lysate/KLH-pulsed dendritic-cell vaccine, and rhIL-7 among the interventions. Treatment details are not fully reported in the registry record, and the arm descriptions do not provide a complete account of every participant's exposure.

Because allocation was not randomized, differences between the arms are descriptive rather than a clean estimate of the effect of adding rhIL-7. The record does not report a separate statistical plan, sample-size justification, or interim-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: Forty-four patients were enrolled. Twelve patients were not treated in Arm A or B because they did not get far enough in the study to be designated for an Arm.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Five of the original six participants from Arm A were analyzed because one subject was changed to a special exemption.

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 Number of Participants With a Positive Immune Response as Evidenced by the Delayed Type of Hypersensitivity (DTH) Reaction Assay, measured over Week 8, 14, 20 (Arm A) and on Days 42, 84 and 126 (± 7 days) (Arm B).

The primary endpoint was the number of participants with a positive immune response, evidenced by a delayed-type hypersensitivity (DTH) reaction assay. The registry defines a positive response to the tumor vaccine as a positive result in at least one of two assays: an ex vivo assay of immune responses to tumor lysates or the DTH assay. For DTH, a positive reaction to tumor lysate meant induration of at least 0.5 cm in a patient who had not shown a positive DTH reaction before immunotherapy.

This endpoint is a measure of immune response to the vaccine, not a direct measure of tumor shrinkage, disease control, survival, or quality of life. In general, an immune-response assay can provide evidence that the immune system reacts to a vaccine target, but it does not by itself establish that the reaction changes the course of a disease. That distinction matters when interpreting a biological endpoint in a pilot study.

The assessment schedule was Week 8, 14, 20 for Arm A and Days 42, 84 and 126 (± 7 days) for Arm B. The registry record does not report the number of participants with a positive response, the number assessed at each time, or the separate assay results. It therefore does not permit a description of the observed primary endpoint frequency.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a binary endpoint mean?

In general, a binary endpoint classifies each participant into one of two outcome categories, such as meeting or not meeting a prespecified response definition. A count of participants with the event is often accompanied by the number assessed, because the count alone does not convey how common the event was among those eligible for evaluation. For this study, the primary endpoint is expressed as a participant count. The registry record does not report the primary result count or all relevant denominators, so a response frequency cannot be reconstructed from the posted record.

The study's definition combines evidence from two assays: a positive ex vivo response to tumor lysate or a positive DTH reaction. A participant meeting either criterion qualifies under the stated definition. Such a combined endpoint can capture more than one form of immune response, but interpretation depends on knowing how each assay was performed, how missing assay results were handled, and how many participants were assessed. Those details are not fully reported in the registry record.

What does non-randomized allocation imply?

In general, random assignment helps balance measured and unmeasured participant characteristics across treatment groups, on average. Without random assignment, treatment groups can differ for reasons associated with prognosis or treatment selection. Statistical adjustment can sometimes address measured differences, but it cannot guarantee removal of bias from unmeasured factors. The registry describes non-randomized allocation and does not report an adjusted primary-outcome analysis. Therefore, an arm-to-arm contrast should be treated as descriptive, not as an isolated causal estimate of the effect of rhIL-7.

The lack of masking is also relevant in general. Knowledge of treatment assignment can influence participant experience, clinical management, or outcome assessment, depending on how an endpoint is measured. A laboratory or skin-reaction measure may be less subjective than some outcomes, but the registry record does not report masking of assessors or other procedures that would clarify this issue. No further assumptions about assessment safeguards are warranted.

How should uncertainty be understood?

In general, a confidence interval describes a range of values reasonably compatible with the observed data under a specified statistical model and sampling assumptions. A narrow interval indicates greater precision than a wide interval; neither interval alone resolves bias from study design or measurement. A p-value, in general, measures how incompatible the observed data are with a specified null model, conditional on that model and its assumptions. It does not measure the probability that a treatment works, and it does not correct a non-randomized comparison.

The registry record does not report a primary-endpoint confidence interval, p-value, or arm-specific immune-response count. Precision of the primary endpoint cannot therefore be evaluated from the record. Where a planning illustration is based on an observed effect, it is a design calculation rather than an observed-power result. The registry's stated assumptions for such illustrations are a two-sided significance level of 0.05, equal allocation between arms, a normal approximation, and an effect to detect expressed as a fraction of the observed effect. Such illustrations do not establish that the study was designed with those assumptions, and they do not fill in a missing primary result.

Why can event counts be hard to compare?

For a binary event, counts depend on both the underlying event experience and the number of participants observed. In general, event percentages also depend on the denominator and the observation window. Comparing percentages across groups is most informative when outcome definitions, ascertainment, follow-up, and participant characteristics are sufficiently comparable. The record provides arm-specific at-risk and affected counts for selected safety summaries, but it does not report enough detail to verify equivalent follow-up or explain all event-collection procedures.

Adverse events can also occur more than once in a participant, while a count of participants affected records whether an individual experienced an event, not the total number of episodes. The registry's group summaries are framed as participants at risk and affected; the listed terms are presented separately with percentages. The record does not fully explain whether participants could contribute to more than one term or how term-specific denominators were determined. Consequently, term percentages should not be added together or treated as mutually exclusive categories.

What does over-dispersion mean in general?

Over-dispersion occurs when observed variability in count data is greater than a statistical model expects. It can make standard errors too small if a model assumes less variability than the data contain. This concern is relevant to some count outcomes, especially when event rates vary across participants. The primary endpoint in this record is a participant count with a binary response definition, and the registry does not report a model-based analysis or an over-dispersion assessment. It would be inappropriate to infer that such a model was used.

More broadly, statistical methods should match the endpoint, design, and analysis population. The registry record does not report the detailed analysis plan, missing-data method, adjustment variables, or model diagnostics. These omissions limit evaluation of how estimates would have been generated, even apart from the absence of the primary result.

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 A - Participants Who Did Not Receive rhIL-7", "Arm B - Participants Who Received rhIL-7")
n      <- c(5, 26)
events <- c(0, 15)
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 A - Participants Who Did Not Receive rhIL-7  5  0  0.0  0.0  52.2  0.0  43.4  0.4  45.9
ARM  Arm B - Participants Who Received rhIL-7  26  15  57.7  36.9  76.6  38.9  74.5  38.8  74.5

11. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures and one analysis, but does not report how the outcomes were prioritized statistically beyond identifying the primary endpoint. It does not describe a multiplicity adjustment or explain how interpretation was to account for examining more than one outcome or assay. In general, testing multiple outcomes can increase the chance of a nominally unusual result arising by chance; a prespecified analysis strategy helps readers understand which findings are confirmatory and which are exploratory. The record does not provide enough information to determine how that issue was addressed.

Missing-data handling is also not reported in the registry record. For the primary endpoint, the record notes that five of the original six participants from Arm A were analyzed because one subject was changed to a special exemption. It does not state the primary endpoint result for those analyzed, the analysis population for Arm B, or whether participants without an assessment were treated as nonresponders, excluded, or handled by another method. The outcome assessment schedule is described, but completeness at each assessment time is not.

These details matter because an analysis restricted to participants with an observed assessment can differ from one that includes all enrolled participants under a defined missing-data rule. In a non-randomized study, exclusions and missing observations may also differ between arms. The registry does not report enough information to assess whether missingness could have affected an arm comparison.

The registry record does not report a detailed statistical analysis plan, interim analyses, or the rationale for the enrollment target. Therefore, it is not possible to determine from the record whether these choices were prespecified or how they shaped interpretation. No assumptions about such procedures should be made.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: fever (in the absence of neutropenia, where neutropenia is defined as anc <1.0 x 10e9/l) (50.0%), dizziness (16.7%), mood alteration: depression (16.7%), pain::extremity-limb (16.7%), dyspnea (shortness of breath) (7.7%).

The registry lists serious-event summaries separately by arm. For Arm A, six participants were listed at risk for serious events and six as affected; zero deaths were listed. For Arm B, 26 participants were listed at risk and 12 as affected; one death was listed. For other events, Arm A had six listed at risk and six affected, while Arm B had 26 listed at risk and 24 affected. These are registry summaries, not randomized comparisons of safety.

Among the serious-event terms listed, fever in the absence of neutropenia was reported as 50.0%, while dizziness, mood alteration: depression, and pain in an extremity or limb were each reported as 16.7%. Dyspnea was reported as 7.7%, and death not associated with a CTCAE term due to disease progression was reported as 3.8%. The record does not fully explain the denominators for each term or whether one participant could contribute to multiple terms, so these percentages should not be summed.

These entries do not by themselves establish that an intervention caused an event. The arms were non-randomized, and the record does not provide enough information to compare observation time or participant risk. The death listed for Arm B is identified as disease progression, not as a CTCAE-term event; the registry does not establish a treatment-related cause.

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

The record establishes that a pilot study evaluated tumor vaccination after standard multimodality therapy, with one arm receiving rhIL-7 and the other not receiving it. It also defines the primary immune-response endpoint and provides selected arm-level safety summaries. These registry entries describe what was studied and what outcome information the registry chose to report.

The record does not report the primary immune-response count or an arm-specific result. It therefore cannot show how many participants had a positive response, how precise an estimated response frequency would be, or whether the response differed between arms. A description of the assay definition is not a substitute for the observed result.

Even if arm-specific endpoint results were available, the non-randomized design would limit causal interpretation. Participants were not assigned at random, and the record does not establish that the groups were comparable. Any difference could reflect baseline or treatment-course differences as well as the addition of rhIL-7. A descriptive contrast is not equivalent to evidence that rhIL-7 caused an outcome difference.

The immune-response endpoint is also not the same as a clinical outcome such as tumor control or survival. In general, an immune assay can indicate a biological response, but a positive assay result alone does not show whether that response benefits a participant. The registry's primary endpoint definition should therefore be interpreted on its own terms, without extending it to outcomes not reported as results.

The safety summaries provide counts and selected event terms, but they do not establish comparative safety or causation. The record does not fully report event definitions, follow-up duration, term-specific denominators, or the relationship of all events to treatment. The single death listed for Arm B is described as disease progression; the registry does not establish a treatment-related cause.

In short, the posted information supports a careful account of design and selected observations, but not a conclusion about the primary immune response or a causal comparison of the regimens. Missing endpoint results and design limitations are essential parts of the interpretation, not minor technical qualifications.

14. Why This Trial Matters Statistically

This trial is statistically notable because it combines a biologically focused endpoint with a non-randomized pilot design. The endpoint asks whether participants showed an immune response to a tumor vaccine, using a defined combination of ex vivo and DTH evidence. That can be useful for describing whether an immune signal was observed, but it is distinct from establishing a clinical benefit.

The registry's distinction between an arm that did not receive rhIL-7 and an arm that did receive it may invite a direct comparison. Statistical interpretation must resist treating that contrast as randomized evidence. Without random allocation, group differences can reflect who received each regimen and what other treatment they received, in addition to any effect of rhIL-7. The design is therefore more informative as a descriptive pilot than as a stand-alone causal comparison.

The record also illustrates why endpoint definitions and analysis populations need to be read together. A clear response definition helps explain what counts as an event, while the analyzed population and denominator are needed to understand how often it occurred. Here, the registry notes a change to a special exemption for one original Arm A participant, but does not provide the primary endpoint result or all assessment denominators. Those omissions prevent a reader from evaluating the response estimate and its precision.

Finally, the safety listings show why event summaries need context. Counts, affected participants, term percentages, and deaths answer different descriptive questions; they should not be merged into a single measure or interpreted as treatment effects without comparable groups and adequate follow-up information. This record is most useful when its design, endpoint, and reporting gaps are considered together.

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 primary limitation for interpreting efficacy-related evidence is that the registry record does not report the primary immune-response result. It gives the endpoint definition and time points, but not the observed response count, assessment denominator, or an estimate of uncertainty. The endpoint's ex vivo component is named, but its detailed criteria and results are not reported in the registry record.

The non-randomized allocation and lack of masking limit arm comparisons. The arm descriptions differ in their listed tumor types, and the record does not report baseline characteristics or adjusted analyses that could clarify comparability. The study also includes multiple treatment components, so a difference between arms could not be attributed to rhIL-7 alone from this design.

The analyzed primary-endpoint population is incompletely described. The record identifies five analyzed participants from the original six in Arm A after one subject was changed to a special exemption, but does not report corresponding primary analysis details for Arm B. Missing-data procedures and completeness of assessments are not reported in the registry record.

Safety interpretation is limited by incomplete detail on follow-up, event ascertainment, term-specific denominators, and treatment attribution. The registry presents selected serious-event terms and group summaries, but the record does not explain whether terms overlap or provide enough information to make a fair comparative risk assessment.

The detailed statistical analysis plan, sample-size justification, interim-analysis procedures, and methods for handling multiplicity are not reported in the registry record. Enrollment of 44 and completion status do not resolve those gaps. Together, these limitations mean the available record supports descriptive reporting, not a precise or causal estimate of treatment effects.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Therapy to Treat Ewing's Sarcoma, Rhabdomyosarcoma or Neuroblastoma — NCT00923351.
• Registry record last updated 2018-06-15; status last verified 2018-05.
• Results first posted 2013-09-20.
• 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

NCT00923351 is a completed Phase 1 and Phase 2 pilot study sponsored by the National Cancer Institute (NCI), with enrollment of 44. It began on 2007-06-02 and had a primary completion date of 2012-08-30. The design was parallel, non-randomized, and unmasked.

The study evaluated tumor vaccination following standard multimodality therapy, with Arm A participants not receiving rhIL-7 and Arm B participants receiving it. The primary endpoint was the number of participants with a positive immune response defined through ex vivo and DTH assays. Five of the original six Arm A participants were analyzed for that endpoint after one subject was changed to a special exemption.

The registry record does not report the primary endpoint result, detailed missing-data methods, or a complete statistical analysis plan. It lists selected serious and other adverse-event summaries by arm, but the non-randomized design and limited event detail do not support a causal safety comparison.