Selinexor in Treating Younger Patients With Recurrent or Refractory Solid Tumors or High-Grade Gliomas: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Selinexor in Treating Younger Patients With Recurrent or Refractory Solid Tumors or High-Grade Gliomas (NCT02323880), a na phase 1 study. The primary outcome is Number of Dose Limiting Toxicities of Selinexor.

2. Clinical Question

The clinical question concerned the safety and dose of selinexor in younger patients with solid tumors or central nervous system tumors that had returned or had not responded to treatment. The study was a phase 1 investigation, and its primary endpoint focused on dose-limiting toxicities rather than a comparison of treatment benefit against a control group. The registry record also lists antitumor effect and pharmacodynamic outcomes, as well as radiographic response and progression-free survival outcomes for specified high-grade glioma groups.

The posted information does not establish a treatment-effect contrast between selinexor and another treatment or usual care. The primary endpoint result is not reported in the registry record, so the endpoint question cannot be answered from the posted primary-outcome information. The adverse-event figures describe events across the recorded groups, but they are not a substitute for a controlled comparison of efficacy or safety.

3. Trial Design

The registry record describes NCT02323880 as a completed phase 1 study sponsored by Children's Oncology Group. It lists an enrollment of 59 and a single-group model, with allocation marked as not applicable and masking marked as none. The study began on 2015-10-30, and primary completion is recorded as 2022-09-30. The listed title concerns selinexor in younger patients with recurrent or refractory solid tumors or high-grade gliomas; the official title describes a phase 1 study in recurrent and refractory pediatric solid tumors, including central nervous system tumors.

The intervention was selinexor. Participants received it orally on either a twice-weekly schedule, on days 1, 3, 8, 10, 15, and 17, or a once-weekly schedule, on days 1, 8, 15, and 22. Treatment repeated every 28 days for up to 24 cycles if disease progression or unacceptable toxicity did not occur. The record includes multiple parts and dose levels, including daily schedule labels for dose levels and weekly dose levels. These are study groups within the single-group design, not randomized treatment arms that can be treated as equivalent to independent control groups.

The primary outcome was the number of toxicity-evaluable participants experiencing a dose-limiting toxicity at least possibly attributable to selinexor, reported by study part, schema, and dose level. Its timeframe was up to 28 days. The record also lists secondary outcomes relating to antitumor effect and pharmacodynamics, with additional radiographic response and progression-free survival outcomes for high-grade glioma patients, including surgical high-grade glioma patients.

The registry record does not report a randomized allocation process or masking. It also does not report a sample-size justification, a detailed statistical analysis plan, or interim analyses. No statistical analyses are listed in the record. The design and posted information therefore support descriptive interpretation of safety counts by group, but do not support a randomized treatment comparison.

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.

Only the first 6 of 8 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Toxicity evaluable patients

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 9 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Number of Dose Limiting Toxicities of Selinexor, measured over Up to 28 days.

The primary endpoint counts toxicity-evaluable participants who experienced a dose-limiting toxicity at least possibly attributable to selinexor. The outcome is defined as a count of participants, and the timeframe is up to 28 days. The registry says the count is considered by study part, schema, and dose level. It does not give the primary endpoint result in the record.

In general, dose-limiting toxicity is used in early-phase studies to describe toxicity judged sufficiently important to affect dose evaluation. Looking at these events in relation to study dose and schedule can help characterize tolerability and inform dose-selection decisions. A count is directly interpretable as the number of evaluable participants meeting the endpoint definition, but by itself it does not convey the precision of an underlying event rate or establish how the event risk compares with another treatment. This record does not include the primary endpoint counts or confidence intervals needed to assess the observed frequency precisely.

The phrase “at least possibly attributable” indicates that the endpoint includes an attribution judgment rather than simply counting every event during observation. The record does not provide the detailed criteria used to classify a toxicity as dose-limiting or to judge its relationship to selinexor. It also does not report the number of toxicity-evaluable participants in each analysis group alongside primary endpoint results. Those details are not reported in the registry record.

The primary endpoint should be kept distinct from the listed secondary outcomes. Antitumor effect, pharmacodynamics, radiographic response, and progression-free survival address different questions and may have different measurement rules and analysis populations. Their presence in the registry does not supply a primary dose-limiting toxicity result, and no values for those outcomes are reported in the registry record described by the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The design determines what conclusions the posted statistics can support. This is described as a single-group phase 1 study, with allocation marked not applicable and masking marked none. The record includes study parts and dose or schedule groups, but does not describe random assignment to a selinexor group and a control group. In general, a randomized comparison is designed to make groups comparable in expectation. Without that design, differences among the listed groups may reflect their study part, dose or schedule, participant mix, or other factors, rather than the effect of a treatment contrast. The registry explicitly cautions that the groups are not a like-for-like control comparison.

What does a count measure?

A count reports how many members of a specified population had an event under a defined outcome rule. Here, the primary endpoint counts toxicity-evaluable participants with a dose-limiting toxicity judged at least possibly attributable to selinexor. The safety data also present counts for participants at risk and participants affected by serious or other events, together with deaths. Counts need their relevant denominators for context: the same count can represent a different event frequency in populations of different sizes. The record reports safety denominators by group, but does not report primary endpoint results or their group-specific denominators.

A count is not itself a rate, a risk difference, or a ratio. Those are distinct statistical summaries and require defined populations and calculations. No such comparative effect estimates are listed in the registry record. In general, an estimate of an event frequency is more informative when accompanied by an interval expressing uncertainty. A confidence interval describes a range of values compatible with the data under a statistical model; it is not a guarantee that the true value lies within that range. No confidence intervals for the listed primary endpoint or safety counts are reported in the registry record.

Why does the study grouping matter?

The safety figures are separated into Part A, Part B, and Part PK groups and carry dose-level or schedule labels. These categories reflect the way observations were organized in the registry record. Their counts should be interpreted within their own group labels. A comparison across such groups is not equivalent to a randomized comparison, and a numerical difference does not alone establish a dose effect. Participant characteristics, eligibility, or follow-up could vary across groups; those details are not reported in the registry record. The record also does not give the allocation mechanism for participants across the listed groups.

For a binary outcome such as whether a participant experienced an event, a common general approach is to report the number affected and the number assessed in each group, and, when appropriate, an estimate of uncertainty. For a count endpoint, the analysis population and event definition are essential. The registry's primary endpoint population is toxicity-evaluable patients, while enrollment describes the study overall. These are not necessarily identical sets. The number enrolled should not be used as a substitute for an endpoint-specific denominator unless the record establishes that equivalence, which it does not.

What can be inferred from the safety percentages?

The listed serious-event terms are accompanied by percentages. A percentage is a descriptive frequency relative to a denominator, but that denominator and the specific calculation basis are not reported in the registry record for these terms. It is therefore not possible to reconstruct their precision or make a sound comparison across terms or groups from the percentages alone. The percentages also do not indicate severity beyond the registry's serious-event classification, nor do they establish causality. In general, adverse-event attribution and event seriousness are separate judgments: an event can be recorded as serious without that record alone showing it was caused by the intervention.

Over-dispersion is a general concern when modeling counts: it means observed variation is greater than a model assumes, which can make uncertainty appear too small if the model is not appropriate. The posted information does not specify a count model or provide enough detail to assess over-dispersion. In this study, the primary endpoint is described as a count of participants, not as a recurrent-event count. No statistical model for it is reported.

What is missing from the statistical description?

The record lists 8 outcome measures and no analyses. It does not report a statistical analysis plan, analysis method, interim analysis plan, sample-size justification, or handling rules for missing outcome data. These omissions limit evaluation of how endpoints were selected, how outcomes were analyzed, and how uncertainty was handled. They do not establish that no such procedures existed outside the registry record; those details are simply not reported in the registry record.

Planning sample sizes are conceptually different from observed results. A planning illustration based on a stated fraction of an observed effect would depend on assumptions about a comparison, the target effect, significance level, allocation, and model. The registry record describes this study as a single-group study and warns against treating its groups as comparable controls. Any planning illustration using an assumed two-arm comparison would therefore be a hypothetical planning exercise, not an observed result or evidence of power achieved by this trial. Sample-size calculations do not show that an effect was present, and a study's enrollment does not establish that a particular effect could be detected reliably.

Overall, the appropriate statistical reading is descriptive and endpoint-specific. The posted figures can identify what the registry records for particular safety groups, but the absence of primary endpoint results, uncertainty estimates, and analytical details prevents a full quantitative assessment of the primary safety question.

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("Part A: Dose Level 1", "Part A: Dose Level 1 (DL 1)", "Part A: Dose Level -1 (DL -1)", "Part A: Weekly Dose Level 1 (WDL 1)", "Part A: Weekly Dose Level -1 (WDL -1)", "Part B: Weekly Dose Level -1 (WDL -1)")
n      <- c(6, 12, 6, 6, 6, 7)
events <- c(0, 4, 1, 2, 1, 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  Part A: Dose Level 1  6  0  0.0  0.0  45.9  0.0  39.0  0.4  41.0
ARM  Part A: Dose Level 1 (DL 1)  12  4  33.3  9.9  65.1  13.8  60.9  13.9  61.4
ARM  Part A: Dose Level -1 (DL -1)  6  1  16.7  0.4  64.1  3.0  56.4  3.7  57.9
ARM  Part A: Weekly Dose Level 1 (WDL 1)  6  2  33.3  4.3  77.7  9.7  70.0  9.9  71.0
ARM  Part A: Weekly Dose Level -1 (WDL -1)  6  1  16.7  0.4  64.1  3.0  56.4  3.7  57.9
ARM  Part B: Weekly Dose Level -1 (WDL -1)  7  0  0.0  0.0  41.0  0.0  35.4  0.3  36.9

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 8 outcome measures, including the primary dose-limiting toxicity endpoint and secondary outcomes concerning antitumor effect, pharmacodynamics, radiographic response, and progression-free survival. When a study examines multiple outcomes, general statistical practice considers the possibility that some apparent findings may arise by chance, especially if many tests are performed. A multiplicity strategy may specify which outcomes are primary, how tests are ordered, or how significance is interpreted across outcomes. A multiplicity adjustment or other strategy is not reported in the registry record. No statistical analyses are listed.

The registry does identify the primary outcome and its population: toxicity-evaluable patients. It does not give a detailed definition of the evaluable population or report the number evaluable for that endpoint. The overall enrollment is 59, but that figure does not establish the denominator for each endpoint or safety group. The safety record gives group-specific at-risk counts, which should remain distinct from the primary endpoint population.

Missing-data handling is not reported in the registry record. The record does not state how participants without endpoint assessments, incomplete follow-up, or unavailable measurements were treated in the analysis. It also does not describe sensitivity analyses for missing data. Without this information, it is not possible to judge whether missing observations could affect the reported endpoint estimates; the primary endpoint estimates themselves are not reported in the registry record.

The record does not describe an interim analysis plan or a statistical analysis plan. Those omissions leave the intended approach to repeated outcome assessment, multiplicity, and missingness unclear. This is a limitation of what the registry record communicates, not evidence that a particular procedure was or was not used outside the record.

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: anorexia (33.3%), depressed level of consciousness (20.0%), back pain (16.7%), blurred vision (16.7%), dehydration (16.7%).

The safety figures are organized by study group and distinguish serious events, other events, and deaths. For each group, the at-risk count supplies context for the affected count. For example, Part A: Dose Level 1 reports 1 serious event among 6 at risk and 6 other events among 6 at risk, with 0 deaths. Part A: Dose Level 1 (DL1)(Amendment 1) reports 5 serious events among 13 at risk, 13 other events among 13 at risk, and 1 death. The remaining groups are likewise reported separately, rather than as a single pooled comparison.

The group labels matter. They distinguish study parts and dose or schedule levels, and the design is single-group rather than randomized. The figures therefore describe what was recorded in each group; they do not establish that a difference between groups was caused by the dose or schedule. Death counts are reported, but attribution of deaths is not reported in the registry record.

The serious-event terms include anorexia at 33.3%, depressed level of consciousness at 20.0%, and back pain, blurred vision, dehydration, and encephalopathy at 16.7% each. The denominators underlying those term percentages are not reported in the registry record. These listings should not be treated as mutually exclusive categories or used to infer treatment causality. The registry's adverse-event information is descriptive and should be kept separate from the primary dose-limiting toxicity endpoint, whose result is not reported in the registry record.

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

The posted record describes a phase 1 investigation of selinexor in younger patients with recurrent or refractory solid tumors or central nervous system tumors. It identifies dose-limiting toxicity as the primary outcome and gives safety information across groups defined by study part, dose level, or schedule. Those details help clarify the study's safety focus and the kinds of observations recorded.

The record does not report the primary endpoint result. Consequently, the number of toxicity-evaluable patients experiencing a dose-limiting toxicity, and the precision of that result, cannot be determined from the registry information described. The adverse-event counts provide descriptive information by group, but they do not answer the primary endpoint question in place of the missing primary result. Nor do the listed secondary outcome names provide outcome findings; their values are not reported in the registry record.

The data do not establish comparative efficacy. There is no randomized control group, and the record cautions that the study groups are not like-for-like control groups. The safety counts and percentages should therefore not be read as evidence that selinexor is more or less effective or safe than another treatment. They also do not establish causality for individual adverse events or deaths.

In general, a statistical finding depends on the outcome definition, analysis population, denominator, and uncertainty around the estimate. The registry record supplies some group-specific safety counts and at-risk counts, but does not provide confidence intervals or a detailed analysis plan. It does not report the primary endpoint values or endpoint-specific evaluable counts. These gaps limit how precisely the findings can be interpreted and prevent a complete assessment of the primary safety question.

The record can support a restrained description of the study design and the safety figures that are posted. It cannot support a claim about a treatment effect, a causal safety conclusion, or a definitive dose comparison on the basis of the information reported.

14. Why This Trial Matters Statistically

This trial is statistically useful as an example of why study design and endpoint definitions must guide interpretation. A phase 1 study focused on dose-limiting toxicity asks a different question from a randomized efficacy trial. Its primary endpoint is a count among toxicity-evaluable patients, classified by study part, schema, and dose level. Readers need those group labels and the analysis population to understand what the count represents.

The registry also illustrates the distinction between overall enrollment and outcome-specific denominators. Enrollment is 59, while the adverse-event record lists separate at-risk counts for each group. These figures serve different descriptive purposes and should not be combined or treated as interchangeable. The primary endpoint's evaluable population is specified, but its result and group-specific denominator are not reported in the registry record.

The study's multiple dose and schedule groups may appear to invite direct comparisons, but the record describes a single-group design and says the groups are not a like-for-like control comparison. Statistical interpretation must respect that limitation. Differences in event counts across groups are not automatically treatment effects, and a count without an uncertainty estimate does not show how precisely an underlying event frequency is known.

The record's safety percentages also demonstrate why denominators and definitions matter. Serious-event terms are listed with percentages, yet the denominator basis is not reported in the registry record. Without that information, readers should not calculate or infer further comparisons. More broadly, the absence of primary endpoint results, statistical analyses, and missing-data details limits the conclusions that can be drawn from registry information alone.

These features make the record a reminder that registry design details, analysis populations, event definitions, and reported uncertainty are integral to statistical interpretation, not merely administrative context.

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 for interpreting the primary outcome is that its result is not reported in the registry record. The endpoint definition identifies toxicity-evaluable patients and a timeframe of up to 28 days, but the record does not give the number meeting the endpoint, the number evaluable in each group, or an uncertainty estimate. The result therefore cannot be quantitatively assessed.

The study is single-group, and the registry indicates that the listed groups are not a like-for-like control comparison. The group-specific safety counts should not be interpreted as randomized contrasts. The record does not report how participants were assigned across study parts, schemas, or dose levels, nor does it provide participant characteristics by group. This makes it difficult to assess comparability or explain observed differences between group counts.

The serious-event percentages are not accompanied by their denominators in the registry record. The record also does not report attribution for the listed deaths or provide enough detail to determine how each adverse event was classified. The serious-event term percentages should be treated as descriptive, not as causal estimates.

Further limitations include no listed statistical analyses, no reported analysis plan, no sample-size justification, and no reported approach to missing data or multiplicity. The registry lists secondary outcomes, but outcome results are not reported in the registry record described. These limitations concern the information available in the registry record and do not establish what analyses or procedures may have occurred outside it.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Selinexor in Treating Younger Patients With Recurrent or Refractory Solid Tumors or High-Grade Gliomas — NCT02323880.
• Registry record last updated 2025-01-14; status last verified 2025-01.
• Results first posted 2024-01-02.
• 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

NCT02323880 is a completed phase 1, single-group study sponsored by Children's Oncology Group. It enrolled 59 younger patients with recurrent or refractory solid tumors or central nervous system tumors. Selinexor was administered orally on either a twice-weekly or once-weekly schedule, with treatment repeating every 28 days for up to 24 cycles if disease progression or unacceptable toxicity did not occur.

The primary outcome was the number of toxicity-evaluable participants experiencing a dose-limiting toxicity at least possibly attributable to selinexor, assessed by study part, schema, and dose level over up to 28 days. The primary endpoint result is not reported in the registry record. The record includes adverse-event counts by group and lists secondary outcomes involving antitumor effect, pharmacodynamics, radiographic response, and progression-free survival. No statistical analyses are listed, and the design does not provide a randomized control comparison.