Selinexor in Patients With Advanced Thymic Epithelial Tumor Progressing After Primary Chemotherapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Selinexor in Patients With Advanced Thymic Epithelial Tumor Progressing After Primary Chemotherapy (NCT03193437), a na phase 2 study. The primary outcome is Overall Response Rate.

2. Clinical Question

In practical terms, the study sought to describe how often participants receiving selinexor met the registry’s definition of complete or partial response. A complete response required disappearance of all target lesions, with pathological lymph nodes reduced in short axis to <10 mm. A partial response required a decrease of >=30% in the sum of the longest diameters of target lesions. The registry defines overall response as complete response plus partial response.

Because there was no concurrent control group, the design does not directly compare response under selinexor with response under another treatment or with no study treatment. The registry’s brief summary describes planned thymoma and thymic carcinoma arms, but the recorded design is single-group; the record does not provide arm-specific outcome results. The analysis therefore concerns what the registry reports about the study’s question and what remains unavailable, rather than estimating a comparative treatment effect.

3. Trial Design

The registry identifies NCT03193437 as a phase 2, open-label study of selinexor in patients with advanced thymic epithelial tumor progressing after primary chemotherapy. The sponsor is Georgetown University. The recorded intervention is open-label selinexor, described in the arm information as 40 mg. The study started on 2018-04-03, had a primary completion date of 2020-07-27, and is listed as terminated. Enrollment is recorded as 8.

The design fields describe a single-group study, with allocation not applicable and no masking. Thus, participants were not assigned in a registered randomized comparison, and the study does not have a like-for-like control arm. The absence of masking also means the study was open-label. Those features matter when interpreting outcomes that involve clinical assessment: the registry’s primary endpoint specifies CT assessment using RECIST v1.1, but the record does not describe further assessment procedures or safeguards.

The brief summary describes the project as two similar phase II trials, one in the US and one in the EU, with 25 patients planned for each. It describes Arm A for thymoma and Arm B for thymic carcinoma, with Stage 1 and Stage 2 enrollment plans listed for each arm. These are descriptions of the intended study structure in the summary, not evidence that those planned numbers were enrolled or that both geographic studies completed their planned stages. The enrollment recorded for this registry entry is 8, and the record does not give an explanation for the difference between the plan described in the summary and the recorded enrollment.

The summary says the study used a Simon two-stage design. Details such as the stage-specific decision boundaries, stopping rules, or any interim analysis results are not reported in the registry record. The termination reason is also not reported in the registry record. The secondary outcomes listed are overall response rate, 6-month progression-free survival rate, 24-month overall survival rate, and adverse events. The registry lists five outcome measures, but does not provide an analysis count or a primary efficacy result.

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.

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Overall Response Rate, measured over 24 months.

The primary endpoint is overall response rate, expressed as a percentage of patients and assessed over 24 months. It uses RECIST v1.1 criteria for target lesions assessed by CT. The registry defines complete response as disappearance of all target lesions; any pathological lymph nodes, whether target or non-target, must have a short-axis measurement below 10 mm. Partial response is defined as a decrease of at least 30% in the sum of the longest diameters of target lesions. Overall response is the combination of complete and partial responses.

As a general statistical concept, a response-rate endpoint reduces individual imaging assessments to a proportion of participants meeting a predefined response definition. It can be useful in an early-phase study because it gives a direct summary of observed tumor shrinkage according to specified criteria. It does not, by itself, describe how long a response lasts, how symptoms change, or how long participants live. Those are different outcomes and require their own definitions and analyses.

The registry lists response rate among the outcomes but does not report the primary response estimate, its denominator, or an uncertainty interval. It therefore is not possible to describe the observed overall response rate or its precision from the registry record. The endpoint definition is available; the numerical primary result is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does an overall response rate describe?

As a general statistical concept, an overall response rate is a proportion: its numerator is the number of participants who meet the response definition, and its denominator is the group included in the analysis. The endpoint definition specifies which outcomes count as responses. In this study, complete and partial responses are combined under the registry’s RECIST v1.1 definition. A response-rate estimate is meaningful only alongside a clearly defined analysis population, assessment period and denominator. The registry record gives the endpoint definition and a 24-month timeframe, but does not report the primary response count, analysis denominator or estimate.

A proportion is a compact summary, but it does not show the range of possible values compatible with the data. General explanation: a confidence interval expresses uncertainty around an estimate under specified statistical assumptions. With small samples, estimates of proportions can be imprecise, and the interval can be broad. The record does not report an interval for the primary response rate, so its precision cannot be assessed from the posted results.

Why does the study design matter?

The registered model is single-group, with allocation not applicable and no masking. There is no concurrent control group for a direct comparison. General statistical principle: a single-group study can describe what happened among participants observed under a study intervention, but it does not independently establish what would have happened to a comparable group under another condition. Differences from a historical expectation, if one were used, would depend on the quality and comparability of that external information. A historical comparison and its assumptions are not described in the registry record.

The study is open-label. Knowing the intervention can potentially influence reporting or assessment, depending on the outcome and how it is measured. The endpoint’s use of CT and RECIST criteria gives a defined response framework, but the registry record does not report whether imaging was centrally reviewed, whether assessors were blinded, or what procedures were used to resolve disagreements. Those details should not be assumed.

What does the two-stage description tell us?

The brief summary says that the study used a Simon two-stage design. In general, a two-stage design can allow an early decision about whether to continue enrollment based on interim outcome data, according to prespecified rules. The registry record does not report the decision boundaries, the interim results, or whether a planned continuation criterion was met. It is therefore not possible to reconstruct the operating characteristics or decisions of the design from the posted information alone.

The summary lists planned stage enrollment for thymoma and thymic carcinoma and describes similar planned studies in the US and EU. These plans should not be treated as observed enrollment or completed analyses. The registry records enrollment of 8, but does not report the number assigned to each planned category or stage. Nor does it explain the relationship between the described plans and the registered enrollment.

How should the planning illustrations be interpreted?

General explanation: sample-size planning asks how much information a future study might need to detect an effect under stated assumptions. The planning illustrations associated with the posted results are based on a recalculated, unadjusted effect, use a two-sided significance level of 0.05, assume equal allocation between two arms and use a normal approximation. They are illustrations, not the design actually recorded for this single-group study. They also depend on the chosen fraction of the observed effect used as the target effect. Because the registry record does not report the primary response estimate, those illustrations cannot be read as a trial-specific efficacy finding.

These calculations are not observed power. General statistical principle: observed power calculated from an observed effect does not add a separate measure of evidence beyond the effect estimate and its uncertainty. Planning values instead describe hypothetical future studies under their assumptions. They should not be interpreted as the number of participants needed to validate a treatment effect, nor as a formal assessment of why this study terminated. The sample-size justification and any sponsor planning calculations for the actual study are not reported in the registry record.

Over-dispersion is relevant in some statistical models for counts or proportions when variability exceeds what a basic model assumes. General explanation: if extra variability is ignored, uncertainty can be understated and inferences can appear more precise than warranted. The registry record does not report a model-based analysis of the primary endpoint or any adjustment for over-dispersion. No such adjustment should therefore be attributed to this study.

What can be inferred from the posted safety summaries?

The registry records counts for participants at risk and affected for serious and other adverse events, alongside listed serious-event terms and percentages. These are descriptive summaries. General statistical principle: an event count or participant count does not by itself establish that an intervention caused the event. Causal interpretation depends on definitions, timing, clinical review and comparison information. The record does not report enough detail to make those assessments for each listed event.

Finally, the registry lists zero analyses. This means the record does not provide analysis results through that field; it does not establish that no analyses were conducted outside the registry. The primary response estimate, analysis population, missing-data handling and statistical analysis details are not reported in the registry record. Interpretation should remain limited to the stated design and recorded results.

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("Selinexor")
n      <- c(8)
events <- c(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  Selinexor  8  0  0.0  0.0  36.9  0.0  32.4  0.3  33.6

11. Multiplicity, Missing Data and Analysis Populations

The record lists several outcomes: overall response rate, 6-month progression-free survival rate, 24-month overall survival rate and adverse events, and it identifies overall response rate as primary. Considering multiple outcomes can raise a multiplicity question: if many hypotheses are tested, the chance of at least one apparently notable result can increase. As a general statistical principle, trial plans may address this through a hierarchy, adjustment, or an explicitly descriptive interpretation. The registry record does not report a multiplicity strategy, so it is not possible to determine whether or how such issues were handled.

The brief summary describes a Simon two-stage design, but the registry does not report its decision rules or interim analysis results. The record also does not report a statistical analysis plan, the method for calculating uncertainty, or the analysis population for the primary endpoint. The number of analyses recorded is 0. This does not establish that no analyses occurred; it means analyses are not reported through the registry’s results record.

Missing-data handling is not reported in the registry record. There is no reported account of participants who lacked an evaluable CT assessment, withdrew, or were otherwise excluded from an endpoint analysis, nor a description of how such cases would be treated. The record’s enrollment number alone cannot substitute for an endpoint-specific denominator. In general, missing outcome data can affect the interpretation of a response proportion if the availability of assessments is related to participants’ outcomes. Without the analysis denominator and missing-data approach, the response estimate and its potential sensitivity to missingness cannot be evaluated.

The registry lists safety counts, but does not report detailed rules for safety analysis populations, event attribution, or treatment-emergent event windows. These limitations do not show that the underlying study lacked such methods; those details are simply not reported in the registry 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumonitis (25.0%), pneumonia (25.0%), febrile neutropenia (12.5%).

The registry lists 8 participants at risk for serious adverse events and 5 affected. For other adverse events, it lists 8 at risk and 8 affected. It also records 1 death. These are separate entries in the registry safety summary; they should not be combined into a single count, and the record does not provide enough detail to determine overlap among categories or events.

The most frequently listed serious adverse event terms are pneumonitis and pneumonia, each reported as 25.0%, followed by febrile neutropenia at 12.5%. These percentages are quoted as recorded. The registry does not provide a full event-level listing, timing, duration, severity grades, attribution, or details about whether events were ongoing or resolved. It also does not explain the relationship between the death and the adverse-event entries.

As a general statistical principle, these figures describe safety events recorded in the study; they do not establish that selinexor caused a particular event. With no comparator group, the registry does not provide a direct basis for comparing event frequency against another treatment. The available safety summaries should therefore be interpreted descriptively and with the limitations of the reported detail in mind.

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

The registry record establishes that a phase 2, open-label, single-group study of selinexor was registered for advanced thymic epithelial tumors progressing after primary chemotherapy. It records enrollment of 8 and status as terminated. It defines overall response rate as the primary endpoint and gives the RECIST v1.1 criteria and 24-month timeframe. It also reports safety counts and lists several serious adverse event terms.

What the record does not establish is the primary efficacy result. There is no reported overall response percentage, response count, analysis denominator, or confidence interval. The registry also does not provide numerical progression-free survival or overall survival results. Consequently, the posted information cannot show how often participants met the response definition, how uncertain such an estimate would be, or how the study outcomes compare with another treatment.

The design is not a randomized comparison: the model is single-group, allocation is not applicable and masking is none. General statistical principle: results from a single group do not by themselves separate an intervention’s effect from other factors that may influence outcomes. The registry’s description of planned thymoma and thymic carcinoma arms and staged enrollment does not supply arm-specific outcome data. Nor does the registry explain why the study was terminated or whether the plans described in the brief summary were changed.

The safety entries are informative as recorded counts, but they are not a causal assessment. The registry record does not provide detailed event context or a comparison group. Accordingly, the figures should not be used to claim either that the intervention caused or did not cause particular events.

In short, this record supports a description of the study design, intended endpoint, recorded enrollment and reported safety summaries. It does not provide the primary result needed to answer the study’s effectiveness question. Absence of a result in the registry record should not be mistaken for evidence of no response or evidence that an outcome was not assessed.

14. Why This Trial Matters Statistically

This trial is statistically notable because its registry entry combines an early-phase, single-group design with a clearly defined response endpoint but does not report the primary response estimate. That distinction is important: a well-described endpoint does not substitute for the numerical result, its denominator, or its uncertainty. Without those elements, readers cannot assess either the observed response frequency or how precisely it was estimated.

The study summary describes a Simon two-stage design. Such designs are generally intended to support staged decisions in early-phase research, but meaningful evaluation requires the prespecified decision rules and interim findings. Those details are not reported in the registry record. The study’s recorded enrollment of 8, alongside the larger planned structure described in the brief summary, further highlights why planned enrollment and completed outcome analyses must not be treated as equivalent.

The single-group, open-label structure also defines the scope of possible inference. It may describe outcomes among participants receiving the intervention, but it does not provide a randomized control comparison. General statistical principle: treatment-effect claims require a comparison that supports inference about what would have occurred under an alternative condition. This registry record does not provide that comparison.

Finally, the safety summaries illustrate why event counts need context. Counts and percentages can describe what was recorded, but their interpretation depends on event definitions, timing, attribution, analysis populations and comparison data. Several of those details are not reported. The statistical value of the record is therefore chiefly in documenting the intended question, design and limited posted results, while making clear which essential efficacy and analysis details remain 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 primary limitation for interpreting efficacy is that the registry record does not report the overall response result, despite identifying it as the primary endpoint. It also does not report a response denominator, confidence interval, or other measure of precision. Numerical results for the listed progression-free survival and overall survival outcomes are not reported in the registry record.

The study is recorded as single-group, open-label, with no masking and no randomized control comparison. Therefore, the registry does not support a direct comparative estimate of treatment effect. The record does not report arm-specific results for the thymoma and thymic carcinoma groups described in the brief summary, nor does it clarify how the planned US and EU studies relate to the enrollment recorded for this entry.

Although the summary describes a Simon two-stage design, its decision boundaries, stage-specific results and interim analyses are not reported. The reason for termination, any changes to planned enrollment, the sample-size justification and the statistical analysis plan are also not reported in the registry record. The analysis population, handling of missing outcome assessments and multiplicity approach are not reported.

Safety reporting is limited to the recorded counts and listed serious adverse event terms and percentages. The registry does not provide detailed timing, severity grades, attribution, event-level context, or a comparison group. The design and reporting limitations restrict interpretation; they do not establish that unreported analyses or procedures were absent from the underlying study.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Selinexor in Patients With Advanced Thymic Epithelial Tumor Progressing After Primary Chemotherapy — NCT03193437.
• Registry record last updated 2023-02-16; status last verified 2022-04.
• Results first posted 2023-02-16.
• 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

NCT03193437 is registered as a phase 2, open-label, single-group study of selinexor in advanced thymic epithelial tumor progressing after primary chemotherapy. Georgetown University is listed as sponsor. The intervention is described as selinexor 40 mg. The record lists enrollment of 8, a start date of 2018-04-03, primary completion on 2020-07-27, and status as terminated.

The primary endpoint is overall response rate using RECIST v1.1 criteria, with a 24-month timeframe. The registry does not report the primary response result or numerical survival outcomes. Safety entries list 5 participants affected by serious adverse events among 8 at risk, other adverse events in 8 among 8 at risk, and 1 death. The record does not state why the study was terminated.