A Study to Evaluate the Safety and Efficacy of Nivolumab Monotherapy and Nivolumab in Combination With Ipilimumab in Pediatric Participants With High Grade Primary Central Nervous System (CNS) Malignancies: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Safety and Efficacy of Nivolumab Monotherapy and Nivolumab in Combination With Ipilimumab in Pediatric Participants With High Grade Primary Central Nervous System (CNS) Malignancies (NCT03130959), a non-randomized phase 2 study. The primary outcome is Number of Safety Lead-In Participants With Dose Limiting Toxicities (DLTs).

2. Clinical Question

3. Trial Design

The record describes a phase 2, non-randomized, unmasked, parallel study of nivolumab monotherapy and nivolumab with ipilimumab in pediatric participants with high grade primary CNS malignancies. The study status is completed, and planned enrollment is 166. The listed sponsor is Bristol-Myers Squibb. The record gives no arm descriptions for Module A or Module B, so the treatment assignment and regimen for each named safety group cannot be reconstructed from the information reported.

Non-randomized means that treatment assignment was not described as random. This matters when interpreting differences between arms: without randomized allocation, groups may differ in ways that influence outcomes, and the registry cautions that these arms are not a like-for-like control comparison. Parallel describes groups followed under different study arms; it does not by itself establish that the arms are comparable. The record reports no masking, so treatment assignment was not masked according to the registry entry.

The primary endpoint focuses on a safety lead-in population, rather than on all enrolled participants. Its definition identifies the first DLT-evaluable participants in specified cohorts and modules. Secondary outcomes include progression-free survival, overall survival, survival at 12 months, progression-free survival at 6 months, and several adverse-event counts, with the eligible cohorts varying by outcome. The registry lists 20 outcome measures and 0 analyses. Results for the primary endpoint and the efficacy outcomes are not reported in the registry record.

The record identifies the study as phase 2, but does not report a detailed statistical analysis plan, sample-size justification, interim-analysis plan, or methods for controlling multiple comparisons. The planned enrollment should not be treated as the number evaluated for every endpoint; endpoint-specific analysis populations 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: 166 participants were treated

Only the first 6 of 10 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety Lead-in participants: In Module A, the first 6 DLT-evaluable participants in Cohort 1 and the first 10 DLT-evaluable participants in Cohorts 2-5; in Module B, the first 10 DLT-evaluable participants.

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.

Only the first 6 of 11 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Number of Safety Lead-In Participants With Dose Limiting Toxicities (DLTs), measured over up to 6 weeks post-dosing.

The primary endpoint is the number of safety lead-in participants with dose-limiting toxicities (DLTs), assessed up to 6 weeks post-dosing. A DLT is defined as a drug-related adverse event occurring in the first 6 weeks of study treatment. The endpoint is a count of participants meeting that definition, not a measure of tumor response or survival.

For DLT evaluation, a participant was considered evaluable if study treatment was delayed > 2 weeks or discontinued because of a related adverse event, or if the planned study treatment was administered and safety evaluation after 6 weeks on study was available to the study steering committee. The planned treatment described for this purpose was 3 doses of nivolumab in Module A or 2 doses of nivolumab plus ipilimumab in Module B.

The safety lead-in endpoint is used to describe early treatment-limiting toxicity in a defined group. In general, such monitoring can inform safety assessment before interpreting broader outcomes, but the registry record does not report a dose-selection rule or explain how DLT counts affected subsequent study conduct. It also does not report the observed primary endpoint count, the number evaluable for DLTs, or an uncertainty interval. The frequency of DLTs in the safety lead-in population therefore cannot be determined from the record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the primary endpoint count?

The primary outcome is a binary-event count expressed as the number of participants with DLTs. The registry definition ties DLTs to drug-related adverse events in the first 6 weeks of study treatment and specifies conditions for DLT evaluability. A count is useful for describing how many participants met a stated safety definition. For interpretation, it is also important to know the number evaluated, the rules for handling incomplete assessments, and the uncertainty around the count. The result, evaluable population size, and uncertainty interval are not reported in the registry record.

What does a non-randomized design imply?

General: Randomization aims to make treatment groups comparable, on average, with respect to both measured and unmeasured characteristics. Without random allocation, observed group differences can reflect differences in participant characteristics, treatment selection, follow-up, or other factors, as well as any treatment effect. This study is explicitly non-randomized, and the record cautions that its arms are not a like-for-like control comparison. Consequently, simple contrasts between the listed groups should not be read as causal estimates of the effects of nivolumab or ipilimumab.

The parallel design indicates that the study has multiple groups, but it does not resolve the comparability issue. The record does not provide arm descriptions linking each safety group to a treatment regimen. It also does not report adjustment methods that might address measured differences between groups. Whether any such methods were prespecified or used is not reported in the registry record.

How should counts and percentages be read?

General: A count reports the number of participants meeting an event definition. A percentage expresses a count relative to a denominator, so interpretation depends on knowing what the denominator represents and how participants were followed. For adverse events, a participant count is not necessarily the same as a count of event episodes: one participant may experience more than one event. The registry identifies participant counts for adverse events and gives group-specific numbers at risk and affected. The event definitions, follow-up details, and handling of repeated events are not fully described in the record.

A confidence interval, in general, conveys the range of values compatible with the data under a statistical model and reflects estimation uncertainty. No confidence interval for the primary endpoint is reported. The record lists no analyses, so it does not provide a reported model-based comparison or a statistical test for the outcomes described.

What do the planning assumptions mean?

The registry’s sample-size assumptions describe a planning illustration based on an observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. For time-to-event outcomes, the stated planning quantity is the total number of events; for other outcomes, it is participants per arm. These are planning illustrations, not observed power, and they are not evidence that the trial was designed with those assumptions or that the required sample size was achieved. The calculated planning values and the observed effect used for those calculations are not reported in the registry record.

General: Sample-size planning asks what amount of information might be needed to detect an effect under specified assumptions. It is not a substitute for results, and it does not show that an effect exists. A planning calculation can be sensitive to the assumed effect size, outcome variability, allocation, and analysis method. The registry record does not report the trial’s own sample-size justification or explain whether the planning illustration corresponds to a prespecified design decision.

Why does over-dispersion matter in some analyses?

General: Over-dispersion occurs when count data vary more than a chosen statistical model expects. If present, it can make standard errors too small and confidence intervals too narrow unless the model accounts for it. The record does not report a count model or an over-dispersion assessment. This general issue should not be taken to mean that over-dispersion was present in this study.

Overall, the registry provides endpoint definitions, design descriptors, and some adverse-event counts, but not the statistical analysis details or outcome results needed to reproduce or evaluate inferential analyses. The analysis population for each outcome, missing-data methods, model specifications, and prespecified decision rules are 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 A1, Safety Lead-in", "Arms A2-A5, Safety Lead-in", "Arms B2-B5, Safety Lead-in")
n      <- c(15, 10, 16)
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  Arm A1, Safety Lead-in  15  0  0.0  0.0  21.8  0.0  20.4  0.2  20.6
ARM  Arms A2-A5, Safety Lead-in  10  0  0.0  0.0  30.8  0.0  27.8  0.2  28.5
ARM  Arms B2-B5, Safety Lead-in  16  0  0.0  0.0  20.6  0.0  19.4  0.1  19.5

11. Multiplicity, Missing Data and Analysis Populations

The record lists 20 outcome measures, including the primary DLT count and secondary safety, progression-free survival, and overall survival outcomes. The secondary outcomes do not all apply to the same cohorts: the registry specifies Cohort 1 only for one progression-free survival outcome, Cohorts 1-4 for overall survival at 12 months, and Cohorts 2-5 for other survival outcomes. This means the outcome list encompasses different populations and follow-up questions rather than a single common comparison.

The registry reports 0 analyses. It does not report a statistical analysis plan, a hierarchy of hypotheses, a multiplicity adjustment, or a rule for interpreting multiple outcomes. In general, evaluating many outcomes can increase the chance of a false-positive finding if each is tested independently without an appropriate plan. This is a general statistical concern, not a claim that such testing or a false-positive result occurred in this study.

Missing-data procedures and endpoint-specific analysis populations are not reported in the registry record. The DLT endpoint has a defined evaluability rule, but the record does not state how participants with incomplete safety assessments were handled in the posted result. The registry names outcomes for specified cohorts, yet does not give the number included in each analysis or describe censoring and follow-up handling for survival outcomes.

Accordingly, neither the multiplicity strategy nor the treatment of missing outcome data can be assessed from the registry record. The planned enrollment of 166 does not establish how many participants contributed information to any particular endpoint.

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: malignant neoplasm progression (58.3%), hydrocephalus (25.0%), pyrexia (20.0%), headache (20.0%), nausea (16.7%).

The registry reports serious adverse-event participant counts with a group-specific number at risk. The figures are 17 of 23 for Arm A1, 16 of 22 for Arm B1, 13 of 16 for Arm A2, 12 of 15 for Arm B2, 7 of 15 for Arm A3, 10 of 15 for Arm B3, 11 of 12 for Arm A4, 6 of 10 for Arm B4, 14 of 19 for Arm A5, and 15 of 19 for Arm B5. These counts describe participants recorded as affected in each named group; they do not establish that one regimen caused more serious events than another.

The study was non-randomized, and the registry says that the arms are not a like-for-like control comparison. The treatment mapping for these arm labels is not reported in the registry record. Group differences may therefore reflect factors other than treatment, and the counts should be read descriptively rather than as a comparative safety estimate.

The registry lists malignant neoplasm progression (58.3%), hydrocephalus (25.0%), pyrexia (20.0%), headache (20.0%), nausea (16.7%), and vomiting (16.7%) among the top serious adverse-event terms. The denominator and calculation method for these term percentages are not reported in the registry record. The record also gives deaths by group, but does not report causes or timing in the information summarized in the registry. Serious adverse-event counts, term percentages, and death counts are different measures and should not be treated as interchangeable.

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

The record establishes that a phase 2, non-randomized study evaluated nivolumab alone and nivolumab with ipilimumab in pediatric participants with high grade primary CNS malignancies. It gives a clear definition for the primary safety lead-in endpoint and reports participant counts for serious and other adverse events by named group. These details describe the study question and some posted safety observations.

The record does not report the primary DLT result, the efficacy outcome results, or a statistical analysis of treatment differences. It therefore does not establish how many safety lead-in participants had DLTs, what the progression-free or overall survival outcomes were, or whether outcomes differed between regimens. The study’s stated purpose of evaluating safety and effectiveness is not itself a finding about safety or effectiveness.

Even where group-specific event counts are available, the non-randomized design limits causal interpretation. The registry cautions that the arms are not a like-for-like control comparison, and it does not provide an arm-to-regimen mapping. The counts should not be used to rank regimens or attribute differences to treatment.

General: An observed count is a description of events recorded under a particular definition and follow-up process. A treatment effect is a comparison that must account for the design, outcome measurement, and uncertainty. Without reported comparative analyses and the information needed to assess group comparability, the posted counts cannot answer that causal question.

In short, the registry record supports a description of the trial design, endpoint definitions, and selected safety counts. It does not provide enough reported results to conclude that either regimen improved outcomes, worsened outcomes, or was safer than the other.

14. Why This Trial Matters Statistically

This study raises statistical questions that are important in pediatric trials of serious disease: how to define early dose-limiting toxicity, how to report which participants were evaluable, and how to separate a safety description from an efficacy comparison. The registry provides a specific DLT definition and a defined safety lead-in population, making clear that the primary endpoint concerns early treatment-related toxicity rather than survival or tumor response.

The non-randomized design is central to interpretation. A set of parallel arms can describe experience under different study regimens, but without randomization and adequate information about group comparability, differences cannot automatically be attributed to treatment. The registry itself cautions against treating the arms as a like-for-like control comparison. This is a useful distinction for clinical researchers reading arm-level safety counts.

The record also illustrates why outcome definitions and outcome results must be considered separately. Several survival and adverse-event outcomes are named, but their results and analyses are not reported in the registry record. Listing an outcome does not show what was observed, and an enrollment target does not reveal how many participants contributed data to each endpoint.

Finally, the study raises questions about multiplicity, missing data, and analysis populations. These are important when a registry lists multiple outcomes across different cohorts. The relevant statistical plans and methods are not reported in the registry record, so their adequacy cannot be assessed. The study is statistically informative as an example of the limits of interpreting non-randomized group counts when comparative results and analysis details are unavailable.

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 principal limitation is incomplete outcome reporting in the registry record. The primary endpoint is defined, but its result and the number evaluable for DLTs are not reported. Results for the listed progression-free survival, overall survival, and other secondary outcomes are also not reported. This prevents assessment of the study’s stated efficacy questions and limits interpretation of its primary safety question.

The design is non-randomized, and the registry explicitly states that the arms are not a like-for-like control comparison. Arm-specific treatment descriptions are blank for Module A and Module B, and the mapping between the adverse-event group labels and treatment regimens is not reported. Consequently, the arm-level safety counts cannot support direct comparative claims about the regimens.

The registry does not report the detailed statistical analysis plan, sample-size justification, interim analyses, endpoint-specific analysis populations, multiplicity strategy, or missing-data procedures. It lists 20 outcome measures and 0 analyses. For the serious adverse-event terms, the denominator and calculation method for the listed percentages are not reported. The record also does not provide context needed to interpret the death counts, such as cause or timing.

These omissions do not establish that analyses or procedures were absent from the study itself; they mean that the relevant information is not reported in the registry record. The available information therefore supports descriptive reading of posted counts but not a comprehensive assessment of statistical validity or comparative treatment effects.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study to Evaluate the Safety and Efficacy of Nivolumab Monotherapy and Nivolumab in Combination With Ipilimumab in Pediatric Participants With High Grade Primary Central Nervous System (CNS) Malignancies — NCT03130959.
• Registry record last updated 2022-08-09; status last verified 2022-07.
• Results first posted 2021-04-01.
• 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 a completed, phase 2, non-randomized, unmasked, parallel study of nivolumab monotherapy and nivolumab with ipilimumab in pediatric participants with high grade primary CNS malignancies. Planned enrollment was 166. The primary endpoint was the number of safety lead-in participants with DLTs up to 6 weeks post-dosing, with a detailed definition of DLT evaluability.

The record lists 20 outcome measures and 0 analyses. It includes group-specific adverse-event counts and serious adverse-event terms, but the primary endpoint result and efficacy results are not reported in the registry record. The arms are not described as like-for-like controls, and the treatment mapping for the listed groups is not reported. The posted counts therefore support descriptive safety reading, not a causal comparison between regimens.