Randomized Efficacy Study of TPI 287 to Treat Primary Refractory or Early Relapsed Neuroblastoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Randomized Efficacy Study of TPI 287 to Treat Primary Refractory or Early Relapsed Neuroblastoma (NCT01505608), a randomized phase 1, phase 2 study. The primary outcome is Number of Participants With Adverse Events as a Measure of Safety and Tolerability.

2. Clinical Question

3. Trial Design

The registry identifies this as a randomized Phase I/II trial with no masking and a crossover model. Its official title describes irinotecan and temozolomide alone compared with those drugs in combination with TPI 287, among patients with primary refractory or early relapsed neuroblastoma. The registry reports enrollment of 14 and a terminated status. It records a start of 2011-12 and primary completion of 2014-12; the reason for termination is not reported in the registry record.

The phases contributed differently to the reported groups. All 8 Phase I patients received TPI 287; randomization began in Phase II. The posted TPI 287 group includes those 8 Phase I patients and 4 Phase II patients randomized to TPI 287, for 12 patients total. The irinotecan-and-temozolomide-only group contains 2 Phase II patients. This means the posted primary outcome does not compare two groups formed entirely by randomization: one combines patients from both phases, while the other is composed of Phase II patients.

The arm description permits patients whose disease progressed on irinotecan and temozolomide to cross over to the combination during cycles 1 to 6. Crossover can affect how outcomes are interpreted, especially when later outcomes reflect exposure to more than the initially assigned treatment. Details about how crossover was handled in analyses are not reported in the registry record. The registry also does not report the randomization procedure, allocation ratio, or additional design details beyond the randomized, unmasked, crossover descriptions.

The treatment descriptions include oral temozolomide and intravenous irinotecan in both arms, with intravenous TPI 287 added in the combination arm. The registry gives cycle and dosing details, but does not report a complete account of treatment exposure or adherence in the posted primary results. The study summary says it aimed to evaluate tumor response and safety and tolerability; the posted primary outcome is the number of participants with adverse events.

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: Phase 1- All patients assigned to TPI 287 group. Randomization began in Phase 2- Arm A patients were allowed to cross over to Arm B during cycles 1-6 if they experienced progression. 0 (zero) Arm A patients crossed over to Arm B on this study.

Phase I

Reasons for not completing this period:

Phase II

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: This group includes the 8 patients enrolled to phase 1 TPI 287 portion of the study plus the 4 additional patients randomized to TPI 287 in the Phase 2 portion = 12 patients total that received TPI 287 on this study.

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 Adverse Events as a Measure of Safety and Tolerability, measured over 6 months.

The primary outcome is the number of participants with adverse events as a measure of safety and tolerability, with a timeframe of 6 months. Its stated purpose is to assess safety and tolerability of TPI 287 with irinotecan and temozolomide in pediatric and young adult patients with primary refractory or recurrent neuroblastoma. The results report participant counts: 7 of 12 in the TPI 287 group and 0 of 2 in the irinotecan-and-temozolomide-only group. These are counts of participants, not counts of every event experienced; the registry does not provide an event-level breakdown for this primary measure.

General: A participant-level adverse-event outcome is useful for describing how many people experienced at least one adverse event during a defined observation period. It offers a broad safety summary, but does not by itself convey the event type, severity, duration, or relationship to treatment. Nor does a count alone establish that treatment caused an event. Those questions require more detailed event reporting and appropriate comparative analysis.

This endpoint is not a measure of tumor response, progression-free survival, or overall survival. The registry lists those topics among secondary outcomes, but the primary outcome posted for this analysis is adverse events. The exact definition of an adverse event used for the primary count, the rules for attribution, and the handling of participants without complete follow-up are not reported in the registry record. The unequal phase composition and small comparator group also limit how much can be concluded from the two participant counts.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a participant count tell us?

The primary result is a count of participants with adverse events, rather than a comparison statistic such as a risk difference or relative effect. A count is directly interpretable as the number of people meeting the outcome definition in each reported group. The associated group sizes give context: 7 of 12 participants in the TPI 287 group and 0 of 2 in the irinotecan-and-temozolomide-only group had an adverse event recorded. These figures should be read as descriptive observations from this trial, not as estimates with known precision.

General: For a binary outcome, each participant is classified according to whether the outcome occurred. Statistical uncertainty depends on both the observed count and the number of participants assessed. When groups are small, the data are compatible with a wide range of underlying event frequencies, even if the observed count is at one extreme. A reported zero is an observation in the enrolled group, not proof that the underlying chance of an event is zero.

Why does randomization matter?

General: Random assignment is intended to make comparison groups similar, on average, with respect to factors that could affect outcomes. It supports causal interpretation when the randomized groups are analyzed as assigned and other design and analysis requirements are met. In this trial, randomization began in Phase II, but the posted TPI 287 group also includes all 8 Phase I patients. That pooled group is therefore not composed only of Phase II randomized participants. The comparator consists of 2 Phase II patients. The posted counts consequently do not preserve a simple comparison of two groups formed solely through Phase II randomization.

The registry does not report a formal analysis restricted to the Phase II randomized participants, nor does it report an adjusted analysis addressing differences between phases. It also does not report the method used to assign participants in Phase II or details about concealment. Without those details, it is not possible to evaluate the randomization process beyond the registry's description.

What does a confidence interval add?

General: A confidence interval expresses statistical uncertainty around an estimated quantity, such as a treatment effect or event frequency. It gives readers a range of values compatible with the data under the assumptions of the chosen method. An interval does not remove bias, correct a design mismatch, or establish clinical importance. No confidence interval is reported for the primary outcome, and no comparative effect estimate is posted. The precision of a between-group effect therefore cannot be assessed from an interval in the registry results.

Why is a p-value not enough?

General: A p-value assesses how unusual data would be under a specified statistical model and null hypothesis; it is not the probability that a treatment works, and it does not measure the size or importance of an effect. A p-value is most useful alongside an effect estimate, its uncertainty, and a clear account of the analysis. The registry reports no analyses for this outcome, and no p-value is posted. No inferential conclusion should be inferred from the raw counts alone.

How does crossover affect interpretation?

The registry identifies a crossover model and says patients with progression on the irinotecan-and-temozolomide arm could cross over to the TPI 287 combination during cycles 1 to 6. General: Crossover can make later outcomes difficult to attribute to the initial assignment if participants receive another treatment after progression. The implications depend on the outcome, timing, and analysis strategy. The registry does not report how crossover was incorporated into analyses. The primary outcome is described over 6 months, but the posted record does not specify participant-level treatment exposure or the precise timing of adverse events in relation to crossover.

What does the safety outcome capture?

A participant count for adverse events provides a broad indication of whether participants experienced an event under the registry's outcome definition. General: It does not capture the total burden of events unless event counts and severity are also described. Multiple events may occur in one participant, and a participant with one recorded event may differ substantially from one with several or more serious events. The primary outcome details do not report such distinctions. The serious adverse event summary supplies selected serious-event information, but it is not a substitute for complete participant-level and event-level reporting.

What can be said about planning and multiplicity?

General: A sample-size justification explains how a trial's planned enrollment relates to the effect the investigators aimed to detect, the expected variability, and the chosen error rates. A planning calculation is not a measure of the trial's achieved power after results are known. A trial-specific sample-size justification is not reported in the registry record. The record lists 6 outcome measures and 4 secondary outcomes, but does not report a statistical plan for prioritizing outcomes, controlling multiplicity, or interpreting multiple tests. Since no analyses are posted, there are no reported inferential findings to evaluate in light of multiplicity.

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("TPI 287", "Arm A- TMZ + Irinotecan Only")
n      <- c(12, 2)
events <- c(7, 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  TPI 287  12  7  58.3  27.7  84.8  32.0  80.7  31.6  80.8
ARM  Arm A- TMZ + Irinotecan Only  2  0  0.0  0.0  84.2  0.0  65.8  0.8  70.8

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 6 outcome measures overall and 4 secondary outcomes: pharmacokinetics of TPI 287 in the Phase I population, quality of life using PedsQL questionnaires, progression-free survival using days until progression, and median overall survival. The primary outcome concerns the number of participants with adverse events. The record does not report the full descriptions of the other outcome measures or a statistical analysis plan that specifies which outcomes were confirmatory, how outcome timing was handled, or how multiple statistical tests were to be addressed.

General: When a study evaluates several outcomes or performs multiple tests, the chance of at least one apparently notable result can increase. A prespecified hierarchy, adjustment procedure, or other multiplicity strategy can clarify how evidence should be interpreted. No such strategy is reported in the registry record. There are also no analyses posted, so no multiplicity-adjusted or unadjusted inferential results are available for assessment.

Missing-data methods are not reported in the registry record. The posted primary counts do not explain whether all participants completed the 6-month assessment, how incomplete follow-up was treated, or whether participants with missing outcome information were included in a particular denominator. The registry does not report the handling of withdrawals, censoring, or deviations from assigned treatment for this primary outcome.

The analysis populations are only partly described. The registry states that the TPI 287 results include 8 Phase I patients and 4 Phase II patients randomized to TPI 287, while the Phase II arm without TPI 287 includes 2 participants. It does not report a separate analysis restricted to randomized Phase II participants or specify whether the posted counts follow an intention-to-treat, per-protocol, or other analysis population. These omissions matter because population definitions and missing-data handling affect what an outcome count represents.

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: progression of disease resulting in death (16.7%), diarrhea (8.3%).

The serious adverse event summary reports 3 affected participants among 12 at risk in the TPI 287 group and 0 among 2 at risk in Phase II Arm A participants who did not receive TPI 287. The registry also lists “Progression of Disease resulting in Death” at 16.7% and “Diarrhea” at 8.3% among the top serious event terms. These figures are presented as registry entries; the registry entries reported do not give a complete event-level account or specify how the percentages were calculated.

These serious-event figures should be kept distinct from the primary adverse-event outcome. The primary outcome reports 7 participants with adverse events in the TPI 287 group and 0 in the comparator group. The serious-event summary concerns serious adverse events and selected terms, not every adverse event. Participant counts, event counts, and percentages are not interchangeable, and the record does not provide enough detail to reconcile every serious-event term with the primary counts.

The registry does not report whether each serious event was judged related to study treatment, the event onset or duration, or the clinical course for each participant. The deaths field is not reported in the registry record. The listed term referring to death should not be treated as a complete mortality result or as evidence of treatment causation. Because the comparator group contains 2 participants and the TPI 287 group includes patients from both phases, these figures are descriptive and do not establish a comparative safety effect.

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

The posted primary outcome shows that adverse events were recorded for 7 of 12 participants in the pooled TPI 287 group and 0 of 2 participants in the irinotecan-and-temozolomide-only group over the stated 6-month timeframe. The serious adverse event summary separately reports 3 affected participants in the TPI 287 group and 0 in the Phase II group without TPI 287. These observations describe the participants included in the registry results; they do not establish that the treatments caused or prevented events.

The result does not establish that adding TPI 287 changes adverse-event risk. There is no posted comparative effect estimate, confidence interval, or p-value, and the groups are not aligned by phase: the TPI 287 group combines Phase I and Phase II patients, while the comparator group consists of Phase II participants. Only 2 participants are reported in the comparator group. These features make the counts too limited to support a precise comparative conclusion.

The result also does not answer whether TPI 287 affects tumor response, progression-free survival, overall survival, or quality of life. Those outcomes are part of the study description or listed secondary outcomes, but the primary result is a count of participants with adverse events. The registry does not post analyses for the listed outcomes in the registry record reported.

General: A trial result is strongest when its outcome definition, analysis population, treatment exposure, missing-data handling, and uncertainty are all clear. Here, several of those details are not reported in the registry record. The counts can still be read as a record of the observed participants, but they should not be generalized as a reliable estimate for a wider patient population. Nor do they establish the absence of events in the comparator treatment: a zero count among 2 participants is limited evidence about what might occur in other patients.

14. Why This Trial Matters Statistically

This trial highlights how a study's phases and analysis populations affect the meaning of a headline count. Although the registry describes Phase II randomization, the posted TPI 287 group combines patients from Phase I and Phase II. The comparison group contains Phase II participants only. Statistical interpretation depends not just on the treatment labels, but on how participants entered each group and which participants contributed to each result.

The trial also illustrates the limits of small groups for comparative inference. The posted primary outcome includes 12 participants in the TPI 287 group and 2 in the comparator group. Counts at this scale are useful for describing recorded observations, but provide limited precision for estimating how often an outcome might occur in a broader population. The absence of a confidence interval or formal analysis further means that comparative uncertainty is not quantified in the registry results.

The crossover design adds another interpretive consideration. Participants progressing on the control regimen could switch to the TPI 287 combination during cycles 1 to 6. Later outcomes may reflect exposure to more than the initially assigned regimen, depending on when they occurred and how the analysis handled crossover. That handling is not reported in the registry record.

Finally, the study's stated aims include tumor response and safety, while the posted primary outcome is a participant count for adverse events. The listed secondary outcomes span pharmacokinetics, quality of life, and survival. Without posted analyses or a reported plan for handling multiple outcomes, readers cannot determine how those objectives were evaluated statistically. These features make the record a useful example of why design, endpoint definition, analysis population, and uncertainty must be considered together, rather than interpreting a raw count in isolation.

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.

Several trial-specific limitations constrain interpretation of the posted results. First, the TPI 287 group includes all 8 Phase I patients plus 4 Phase II patients randomized to TPI 287, while the comparator group includes 2 Phase II participants. Pooling Phase I and Phase II participants on one side of the reported comparison means the result is not a simple randomized comparison between groups drawn from the same phase.

Second, the comparator group is very small. Its reported zero participants with adverse events is an observed count, not a precise estimate of the frequency that would occur in a larger group. The registry posts no confidence interval or inferential analysis to express uncertainty around the counts.

Third, the crossover option for participants with progression in the control arm may affect the interpretation of outcomes after a switch in treatment. The registry record does not report how crossover was handled in analysis, nor does it provide timing of adverse events in relation to crossover. Treatment exposure, adherence, and the complete follow-up record are also not reported in the primary results.

Fourth, details of the adverse-event definition and collection are not reported in the registry record. The primary result does not provide event counts per participant, severity grades, attribution, or event timing. Selected serious-event terms are listed, but the information does not provide a full event-level account. Missing-data methods and the precise analysis population are not reported either.

The registry lists secondary outcomes but no analyses. It also does not report a trial-specific sample-size justification, a statistical analysis plan, or a strategy for multiplicity. The reason for the terminated status is not reported in the registry record. These limitations do not negate the descriptive counts, but restrict what can be concluded from them.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Randomized Efficacy Study of TPI 287 to Treat Primary Refractory or Early Relapsed Neuroblastoma — NCT01505608.
• Registry record last updated 2024-08-06; status last verified 2024-08.
• Results first posted 2016-10-28.
• 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 describes NCT01505608 as a randomized, unmasked Phase I/II crossover trial of TPI 287 with irinotecan and temozolomide versus irinotecan and temozolomide alone for primary refractory or early relapsed neuroblastoma. Enrollment was 14, and the record lists the study status as terminated. The study summary names tumor response, safety, and tolerability as objectives.

The primary outcome is the number of participants with adverse events over 6 months. The posted counts are 7 of 12 in the TPI 287 group and 0 of 2 in the comparator group. The TPI 287 group includes 8 Phase I patients and 4 Phase II patients randomized to TPI 287; the comparator group consists of 2 Phase II participants. The record lists 4 secondary outcomes and reports no analyses. It does not report a comparative estimate or confidence interval for the primary outcome. The counts are descriptive, and the phase composition, small comparator group, crossover design, and unreported analysis details limit comparative interpretation.