Oral ONC201 in Adult Recurrent Glioblastoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Oral ONC201 in Adult Recurrent Glioblastoma (NCT02525692), a non-randomized phase 2 study. The primary outcome is Percent of Patients With Probability of Progression-Free Survival at 6 Months.

2. Clinical Question

The study examined dordaviprone (ONC201) in patients in six clinical groups involving recurrent glioblastoma, H3 K27M-mutant glioma, or diffuse midline glioma. Its primary question was whether participants remained progression-free for more than 6 months after starting treatment, assessed using Response Assessment in Neuro-Oncology (RANO) criteria for high-grade glioma.

3. Trial Design

This was a Phase 2, open-label, multicenter study with a parallel, nonrandomized design. The registry record lists enrollment as 84 and describes six arms. Because treatment assignment was not randomized and the arms enrolled different clinical groups, comparisons between arms are not like-for-like estimates of treatment differences. The record does not describe a randomized control group.

Arms A, B, and C enrolled patients with Grade IV glioblastoma at recurrence. Arm A allowed any number of recurrences and used 625 mg once every 3 weeks. Arm B involved first recurrence and used 625 mg weekly on Days 1, 8, and 15 of each cycle. Arm C involved first recurrence, eligibility for salvage surgical resection, and the same weekly schedule. Its arm description is incomplete in the registry record, so further details about its surgical procedures are not reported in the registry record.

Arm D enrolled patients with Grade IV glioma with an H3 K27M mutation, allowing any number of recurrences, and used the weekly schedule. Arm E enrolled patients with clinical and/or radiographic evidence of midline glioma who were eligible for salvage surgical resection; recurrent disease was not required. The arm description mentions surgery approximately 24 hours after a point that is not fully stated in the registry record. Arm F enrolled patients with diffuse midline glioma involving the brainstem, thalamus, or spinal cord, without the H3 K27M mutation or with unknown H3 mutation status. It also used the weekly schedule.

The study started in 2016-01, had a primary completion date of 2023-04-17, and is listed as terminated. The registry record does not explain the reason for termination. It also does not report the timing or basis for assigning participants to arms, arm-specific enrollment, or the detailed treatment exposure and follow-up for each arm.

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:

Analysis population for the primary outcome, as defined in the registry: Note: In Arm B, one participant with recurrent H3 K27M-mutant glioblastoma and non-measurable disease experienced complete regression of enhancing lesions for \>1.5 years was censored from analysis due to use of anastrozole at baseline and throughout the 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.

Only the first 6 of 7 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Percent of Patients With Probability of Progression-Free Survival at 6 Months, measured over Progression-free survival assessments were conducted 6 months following treatment initiation..

The primary endpoint was the percentage of patients with progression-free survival at 6 months. The registry definition specifies patients who exhibited progression-free survival for >6 months, with assessments conducted 6 months following treatment initiation. RANO criteria for high-grade glioma were identified as the assessment framework.

This endpoint represents a clinically interpretable milestone: whether a participant has remained free of documented progression through a specified follow-up point. In general, a milestone percentage can communicate the proportion meeting a defined status at a chosen time. It does not, by itself, describe how progression risk changes throughout follow-up or how long each participant remained progression-free beyond that point.

The endpoint title refers to a probability, while its description defines a percentage of patients meeting the progression-free criterion. Interpreting such a measure requires details about the analysis population, handling of participants whose status cannot be determined at the assessment time, and the statistical method used to estimate the percentage. Those details and the arm-level estimates are not reported in the registry record. A 95% confidence interval is named as the dispersion measure, but interval values are not reported in the registry record.

The record notes that one participant in Arm B with recurrent H3 K27M-mutant glioblastoma and non-measurable disease had complete regression of enhancing lesions for >1.5 years and was censored because of anastrozole use at baseline and throughout the study. That note is important to understanding the endpoint population and its handling, but the broader censoring and missing-data rules are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record specifies a primary endpoint based on progression-free survival at a milestone, a 95% confidence interval as the intended dispersion measure, and RANO criteria for high-grade glioma. It does not report the statistical analysis method, the endpoint estimates, or interval limits. The explanations below describe general statistical principles rather than methods confirmed as used in this study.

What does a milestone percentage represent?

In general, a milestone percentage describes the proportion of an analysis group meeting a specified condition at a chosen time. For this study, the condition was progression-free survival for >6 months after treatment initiation. A clearly defined milestone makes the endpoint easier to interpret than an unspecified measure of disease status, but interpretation depends on how progression is assessed and how participants without a known status at the milestone are handled.

Progression-free survival is commonly understood as time until progression or death, but the registry’s endpoint description focuses on the percentage meeting the more-than-6-month criterion. The record does not state whether death was included in the progression-free survival definition, how assessment dates were handled, or whether the reported percentage was a direct proportion or an estimate from a time-to-event method. Those details should not be inferred from the endpoint title alone.

What does a confidence interval add?

In general, a confidence interval expresses the precision of an estimate under the assumptions of the analysis method. A point estimate without an interval can conceal substantial uncertainty, particularly when the analysis group is small or events and censoring limit the available information. The registry names a 95% confidence interval for the primary endpoint, but does not report any numerical intervals. Without estimates and interval limits, precision cannot be evaluated for the study arms.

A confidence interval is not a range that contains a fixed proportion of individual patient outcomes, nor does it provide the probability that the true value lies within a particular calculated interval. It describes uncertainty arising from repeated sampling under a statistical procedure. Its interpretation also depends on the method and assumptions used, which are not reported in the registry record.

How should censoring be understood?

In general, censoring occurs when an individual’s event time or endpoint status is not fully observed under the analysis rules. In time-to-event analyses, a censored observation contributes information up to the last point at which the relevant status is known, subject to assumptions about censoring. A censored participant is not automatically equivalent to a participant who experienced progression or to one who remained progression-free for the entire follow-up.

The registry specifically notes that one Arm B participant was censored because of anastrozole use at baseline and throughout the study, despite complete regression of enhancing lesions for >1.5 years. That individual note does not establish the complete censoring policy for the study. The handling of other intercurrent treatments, loss to follow-up, withdrawal, or unavailable assessments is not reported in the registry record.

What can the sample-size illustrations tell readers?

The registry describes planning illustrations based on the observed effect, with two-sided significance level 0.05, equal allocation between two arms, and a normal approximation. It states that the effect to detect is a stated fraction of the observed effect. Such illustrations are hypothetical planning calculations, not observed power and not evidence that the trial was designed with those assumptions. The record does not report the trial’s own sample-size justification, and the primary endpoint estimates needed to interpret observed-effect-based illustrations are not reported in the registry record.

In general, a sample-size calculation estimates the amount of information needed for a specified design and target effect under chosen assumptions. It does not show that a completed study had a particular probability of detecting an effect after the results are known. When the calculation uses an observed effect, it is especially important to treat the result as an illustration rather than a prospective design guarantee. The six nonrandomized arms also do not constitute equal-allocation randomized groups simply because an illustration assumes equal allocation between two arms.

Why does over-dispersion matter in general?

In general, over-dispersion means the observed variability in data is greater than expected under a statistical model. If a model assumes less variability than the data contain, standard errors can be too small and confidence intervals too narrow. The registry information does not identify a count model, report an over-dispersion assessment, or indicate that an over-dispersion adjustment was used. This concept is therefore general background, not a description of the trial’s analysis.

For this endpoint, interpretation would require the actual estimates and intervals, the analysis population, the method for dealing with censoring and missing assessments, and the rules used to determine progression under RANO criteria. Those items, along with any model diagnostics or sensitivity analyses, are not reported in the registry record. The presence of an endpoint definition and a named confidence level is not enough to reconstruct the analysis.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Arm A", "Arm B", "Arm D", "Arm E", "Arm F")
n      <- c(17, 19, 30, 3, 7)
events <- c(2, 0, 4, 0, 1)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Arm A  17  2  11.8  1.5  36.4  3.3  34.3  3.6  34.7
ARM  Arm B  19  0  0.0  0.0  17.6  0.0  16.8  0.1  16.8
ARM  Arm D  30  4  13.3  3.8  30.7  5.3  29.7  5.5  29.8
ARM  Arm E  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Arm F  7  1  14.3  0.4  57.9  2.6  51.3  3.2  52.7

11. Multiplicity, Missing Data and Analysis Populations

The study has six arms covering different clinical groups, but the registry record does not report a multiplicity strategy. It does not state whether the primary endpoint was tested separately in each arm, whether some arms were considered exploratory, or whether any adjustment was planned for multiple comparisons. In general, conducting several statistical tests can increase the chance of at least one false-positive result if each is interpreted without accounting for the testing strategy. That general concern does not establish that a particular testing procedure was used.

The registry lists one outcome measure and zero analyses. It does not provide the endpoint estimates, analysis populations, or a statistical analysis plan. Interim analyses, prespecified subgroup analyses, and procedures for selecting or prioritizing analyses are not reported in the registry record. The primary endpoint’s confidence interval level is identified as 95%, but the record does not explain how that interval relates to any multi-arm inference.

Missing-data methods are also not reported in the registry record. There is no description of how unavailable RANO assessments, withdrawal, loss to follow-up, or incomplete follow-up were handled. The record does identify one Arm B participant who was censored because of anastrozole use at baseline and throughout the study. It does not explain whether similar rules applied to other participants or specify sensitivity analyses for this decision.

In general, missingness and censoring can affect an endpoint estimate when the observed participants differ systematically from those whose outcomes are unavailable. The assumptions and analysis choices matter for judging whether the estimate represents the intended study population. Because those details and the arm-level results are not reported, the direction and extent of any effect of missing data or multiplicity cannot be assessed from 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: fall (33.3%), failure to thrive (33.3%), neoplasm malignant (33.3%), seizure (17.6%), spinal compression fracture (14.3%).

The registry reports serious adverse event figures by arm, including the number at risk and the number affected: Arm A, 17 at risk and 6 affected; Arm B, 19 and 1; Arm C, 8 and 1; Arm D, 30 and 11; Arm E, 3 and 1; and Arm F, 7 and 1. It also lists other adverse event figures: Arm A, 17 at risk and 16 affected; Arm B, 19 and 17; Arm C, 8 and 5; Arm D, 30 and 29; Arm E, 3 and 2; and Arm F, 7 and 7. These are registry counts, not comparative estimates adjusted for treatment exposure or differences between groups.

Deaths are listed as 16 in Arm A, 18 in Arm B, 7 in Arm C, 20 in Arm D, 3 in Arm E, and 6 in Arm F. The registry figures do not state whether deaths were attributed to treatment, disease, or another cause. They should not be interpreted as treatment-related deaths without that information.

The listed serious adverse event terms include Fall and Failure to thrive at 33.3% each, Neoplasm malignant at 33.3%, Seizure at 17.6%, Spinal compression fracture at 14.3%, and Headache at 12.5%. The record does not specify the denominators or calculation details for these percentages in the adverse-event term list. In general, adverse-event counts and percentages describe reported events under a study’s collection rules; they do not by themselves establish causation or permit a fair safety comparison across nonrandomized groups.

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

The record establishes that a Phase 2, open-label, nonrandomized study examined dordaviprone across six arms and specified progression-free survival at 6 months as its primary endpoint. It also gives treatment schedules, describes the intended clinical populations, and reports selected adverse-event and death counts. These details provide context for understanding the study’s intended question and the limits of its design.

The posted results do not include the primary endpoint estimates or their 95% confidence intervals. Consequently, the registry record does not allow readers to determine the observed proportion meeting the progression-free survival definition in any arm, or how precise such an estimate was. The listed enrollment cannot substitute for arm-specific analysis denominators, which are not reported in the registry record.

The nonrandomized design is important when considering comparisons. The arms were defined by clinical features and do not represent like-for-like groups assigned at random to competing treatments. Even if arm-level endpoint results were available, differences between arms could not be attributed to treatment schedule alone from this design without additional evidence and appropriate methods. The registry record does not describe a concurrent randomized control group.

Likewise, the safety counts do not establish whether events were caused by treatment. The record does not provide enough detail about attribution, exposure, or event definitions to support causal safety comparisons. The listed deaths are counts, not explanations of cause.

In general, a trial endpoint can answer only the question its definition and design support. A milestone progression-free survival measure is not, by itself, a measure of overall survival, quality of life, or comparative benefit against a control. Those outcomes are not reported in the registry record. The available information therefore supports a description of study design and selected safety reporting, but not a conclusion about the primary efficacy result or comparative treatment effects.

14. Why This Trial Matters Statistically

This study is statistically notable for combining several clinically distinct populations within a six-arm, nonrandomized Phase 2 design. The arms differ in recurrence history, mutation status, tumor location, surgical eligibility, and in one case dosing schedule. Those features matter because they shape what comparisons can be made: the arms are not randomized control groups, and a difference between them would not isolate the effect of a treatment schedule or other single factor.

The primary endpoint is a milestone measure, defined as progression-free survival for >6 months following treatment initiation. This kind of endpoint can give a focused description of disease status at a specified follow-up point. Its interpretation depends on the definition of progression, ascertainment timing, censoring, and the analysis population. The registry identifies RANO criteria and a 95% confidence interval, but does not report the estimates or the analysis details needed to evaluate those features.

The record also illustrates why enrollment and analysis denominators should be distinguished. Enrollment is listed as 84, but arm-specific primary endpoint denominators are not reported. The note about a censored Arm B participant makes clear that at least one endpoint-handling decision is relevant, while the general rules for other participants are not described.

Finally, the registry lists safety counts and selected serious adverse event terms, but does not provide the context needed to infer causality or compare risk across arms. These limits do not erase the value of a clearly defined endpoint and detailed arm descriptions; they define the scope of conclusions that can be drawn. Because the primary results and analysis methods are absent from the registry record, the statistical contribution is primarily the documented design and endpoint framework rather than a reportable estimate of the primary outcome.

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 efficacy is that the registry record does not report the primary endpoint results. It identifies the endpoint as the percentage of patients with progression-free survival at 6 months and names a 95% confidence interval, but gives no arm-level estimates or interval limits. It also lists zero analyses. Thus, the magnitude and precision of the primary outcome cannot be evaluated from the registry record.

The nonrandomized design and clinically distinct arms limit comparisons. The populations differ by recurrence status, mutation status, tumor location, surgical eligibility, and dosing schedule. The registry does not describe a control group assigned to another treatment. Results from different arms, if available elsewhere, would not automatically provide unbiased comparative treatment effects.

Arm-specific enrollment and primary analysis denominators are not reported in the registry record. The record does not describe how missing assessments, loss to follow-up, deaths, or other censoring were handled beyond the note about one censored participant in Arm B. The statistical method for the milestone percentage and its confidence interval is also not reported. These omissions prevent assessment of key assumptions and the contribution of incomplete outcome information.

The study is listed as terminated, but the reason is not reported in the registry record. The registry also does not report the trial’s own sample-size justification, interim analyses, multiplicity plan, or a statistical analysis plan. Some arm descriptions are incomplete, including details of surgical procedures in Arms C and E. Safety reporting lacks information on event attribution, definitions, exposure, and the denominators behind the listed serious-event term percentages. These limitations restrict both efficacy interpretation and causal safety assessment.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Oral ONC201 in Adult Recurrent Glioblastoma — NCT02525692.
• Registry record last updated 2024-12-24; status last verified 2024-12.
• Results first posted 2024-12-24.
• 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

NCT02525692, Oral ONC201 in Adult Recurrent Glioblastoma, is listed as a Phase 2, open-label, multicenter, parallel, nonrandomized study. Enrollment is listed as 84. The study had six arms involving recurrent glioblastoma, H3 K27M-mutant glioma, and diffuse midline glioma populations. Participants received oral dordaviprone (ONC201), with Arm A using a once-every-3-weeks schedule and the other arms using weekly dosing on Days 1, 8, and 15 of each cycle.

The primary endpoint was the percentage of patients with progression-free survival for >6 months, assessed using RANO criteria for high-grade glioma. The registry record reports one outcome measure and zero analyses; it does not report the primary endpoint estimates or confidence intervals. It lists serious and other adverse event counts and deaths by arm, alongside selected serious adverse event terms. The record is listed as terminated, and the reason for termination is not reported in the registry record.