Molecularly Determined Treatment of Diffuse Intrinsic Pontine Gliomas (DIPG): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Molecularly Determined Treatment of Diffuse Intrinsic Pontine Gliomas (DIPG) (NCT01182350), a non-randomized phase 2 study. The primary outcome is 9-month Overall Survival (OS) Rate.

2. Clinical Question

The study was designed around a treatment strategy rather than a single drug comparison. The brief description emphasizes obtaining tumor tissue and using biological targets to guide assignment to FDA-approved agents. The central statistical question in the posted primary outcome is descriptive: what proportion of the defined analysis population remained alive at the specified time point? That question differs from whether one treatment is better than another. Because treatment assignment was not randomized, differences between molecular cohorts could reflect their underlying characteristics as well as treatment. The posted record does not provide a comparative treatment-effect analysis.

3. Trial Design

This was a Phase II, non-randomized, parallel study of molecularly determined treatment for children and young adults with newly diagnosed DIPG. The registry lists an enrollment of 53 and identifies Karen D. Wright MD as the sponsor. The study began in 2011-09, had primary completion in 2016-06, and is marked terminated. A reason for termination is not reported in the registry record.

The study’s premise was to obtain a surgical biopsy at presentation and use molecular findings to assign treatment. The cohort descriptions distinguish combinations of MGMT and EGFR status. Cohort 1 is listed as MGMT-/EGFR- and includes radiation with bevacizumab. Cohort 2 is MGMT-/EGFR+ and adds erlotinib to radiation and bevacizumab. Cohort 3 is MGMT+/EGFR- and includes temozolomide with radiation and bevacizumab. Cohort 4 is MGMT+/EGFR+ and lists radiation, bevacizumab, erlotinib, and temozolomide. The registry’s descriptions also state an approximately 52-week protocol treatment period, including a 4-week interim period after radiation and a maintenance phase with 28-day cycles. Further details in the arm descriptions are incomplete in the registry record.

There was no masking, and allocation is explicitly non-randomized. Thus, the cohorts are not like-for-like control groups: molecular findings determined treatment, and the record does not describe random assignment designed to balance prognostic factors. The study’s focus on biopsy feasibility, treatment assignment, toxicity, and survival is consistent with an early-phase treatment strategy assessment, but the detailed protocol rationale and statistical plan are not reported in the registry record. The registry lists 8 outcome measures and 0 analyses. It does not report the primary endpoint estimate, its confidence interval, or comparative effect estimates.

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: The analysis population is comprised of the subset of participants who underwent biopsy, obtained a molecularly-based treatment assignment and started assigned chemoradiation therapy.

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.

Population: The analysis dataset is comprised of all enrolled participants who underwent stereotactic biopsy.

6. Primary Endpoint

The registry lists the primary outcome as 9-month Overall Survival (OS) Rate, measured over 9 months.

The primary endpoint is the 9-month overall survival (OS) rate, defined as the percentage of participants remaining alive 9 months from registration. The stated analysis population includes only participants who underwent biopsy, obtained a molecularly based treatment assignment, and started assigned chemoradiation therapy. This definition is important: the endpoint is not described as applying to every enrolled participant, but to that specified subset.

Overall survival is a time-to-event outcome that records whether participants are alive at a defined time point. In general, it is a direct outcome of interest and does not depend on a clinician’s interpretation of radiographic progression. A fixed-time survival rate can make a treatment study’s outcome easier to communicate, while a confidence interval conveys uncertainty around the estimated rate. These are general statistical properties; the registry record does not report the observed 9-month rate or its 90% confidence interval.

The listed unit is percentage of participants, and the dispersion measure is a 90% confidence interval. The registry also labels the parameter as “NUMBER”; the relationship between that label and the stated percentage unit is not explained in the record. No endpoint estimate, interval limits, or analysis details are posted. Consequently, the endpoint definition can be described, but the survival experience represented by the endpoint cannot be quantified from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The main statistical measure is a fixed-time overall survival rate. For this endpoint, each participant in the defined analysis population is classified according to whether they remain alive 9 months from registration. The rate expresses the share meeting that outcome. A confidence interval is an interval estimate that communicates uncertainty arising from sampling variability. In general, a narrower interval indicates greater precision, while a wider interval indicates less precision. The registry identifies a 90% confidence interval but does not post its limits or describe the calculation method.

What does the confidence interval convey?

In general, a confidence interval describes the range of values compatible with the data and the specified statistical procedure. It is not a guarantee that the true value lies within a particular interval, and it does not establish that a treatment caused an outcome. Its interpretation depends on the analysis population, data completeness, and method used. Since the interval itself is not reported in the registry record, the precision of the 9-month OS estimate cannot be assessed.

The endpoint is defined from registration, but the analysis population is limited to participants who underwent biopsy, received a molecular treatment assignment, and started assigned chemoradiation therapy. This is not necessarily equivalent to an analysis of all enrolled participants. In general, restricting an analysis to participants who reach specified treatment milestones can affect interpretation because inclusion depends on events after enrollment. The number meeting the stated definition and the reasons others may not have qualified are not reported in the registry record.

Why does allocation matter?

Random allocation is one way to make comparison groups similar, on average, with respect to both measured and unmeasured characteristics. This study is listed as non-randomized. Its cohorts were defined by MGMT and EGFR findings and received different treatment combinations. In general, comparisons between such groups can be influenced by differences in prognosis, clinical status, or other characteristics associated with cohort assignment. A difference in an outcome between cohorts would not automatically identify a causal effect of a drug or treatment combination. The registry record reports 0 analyses and does not describe adjusted comparisons.

The parallel structure means participants were assigned to one cohort rather than described as crossing through all treatment combinations. No masking was used. The record does not report a statistical plan explaining how the non-randomized cohorts were intended to be analyzed, how potential confounding was handled, or whether any planned subgroup comparisons were made.

What can the planning illustrations establish?

In general, sample-size calculations are planning tools based on assumptions about the outcome, effect to detect, significance level, allocation, and analysis method. They do not show what happened in a trial and are not observed power. The record describes an illustration based on an observed effect, using a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect expressed as a fraction of the observed effect. It specifies that non-time-to-event illustrations are expressed as participants per arm, while time-to-event illustrations give total events. These are planning assumptions, not the trial’s own design justification. The calculated illustration values and the design’s sample-size justification are not reported in the registry record.

For survival outcomes, follow-up and censoring can affect estimation. In general, a participant whose status at the target time is not known may contribute differently depending on the chosen method and the available follow-up. The record does not state whether survival was estimated with a method that accommodates censoring, how censoring was defined, or how missing status was handled. The listed 9-month endpoint definition alone does not settle these analytical choices.

How should the outcome list be read?

The registry lists 8 outcome measures, including the primary OS measure and secondary measures related to surgery, toxicity, feasibility, radiation timing, and progression-free survival. Listing an outcome indicates that it is part of the record; it does not provide its estimate or show that it was analyzed. The record lists 0 analyses. Therefore, no p-values, adjusted effects, or analysis-specific population counts are available for statistical interpretation. General concepts such as confidence intervals, confounding, and censoring explain how such results are often understood, but they do not substitute for the missing analysis details.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 8 outcome measures, covering survival, progression-free survival, feasibility, surgery-related outcomes, treatment-related toxicity, and radiation timing. It reports 0 analyses. The record does not state which outcomes were treated as confirmatory, whether a hierarchy was prespecified, or whether any adjustment for multiple statistical testing was planned or performed. Multiplicity methods are used in general to manage the increased chance of false-positive findings when several hypotheses are tested; whether that issue arose in the reported analyses cannot be assessed because analyses and results are not posted.

The primary endpoint population is defined as participants who underwent biopsy, received a molecularly based treatment assignment, and started assigned chemoradiation therapy. The number in this population is not reported in the registry record. The record also does not state how participants with incomplete follow-up, unavailable survival status, or other missing endpoint data were handled. Missing-data methods matter because the assumptions used to address unobserved outcomes can influence an estimate and its uncertainty.

The registry’s total enrollment is 53, but that does not establish the denominator for the primary endpoint or any safety outcome. The analysis population count, endpoint-specific denominators, and reasons for any exclusions are not reported. No imputation, censoring, or sensitivity-analysis approach is described in the registry record. These omissions limit assessment of selection into the analysis, completeness of follow-up, and robustness of the posted outcomes.

11. 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: cd4 lymphocytes decreased (33.3%), hypertension (33.3%), hypokalemia (33.3%), hyponatremia (33.3%), neutrophil count decreased (33.3%).

The registry presents serious adverse event counts separately for each molecular cohort. Cohort 1 lists 30 participants at risk and 8 affected; Cohort 2 lists 14 at risk and 9 affected; Cohort 3 lists 3 at risk and 1 affected; Cohort 4 lists 3 at risk and 2 affected. These are cohort-level counts, not randomized comparisons. The registry also reports deaths of 29, 14, 3, and 2, respectively, for the four cohorts. The record does not explain how the death counts relate to follow-up duration or to the primary endpoint population.

Six serious adverse event terms are each reported at 33.3%: CD4 lymphocytes decreased, hypertension, hypokalemia, hyponatremia, neutrophil count decreased, and platelet count decreased. The record does not clarify the denominator or explain whether these terms occurred across cohorts or in a particular cohort. They should therefore be read as registry-listed descriptive percentages, not as treatment-specific rates.

In general, adverse event counts and percentages depend on who was observed, the duration of observation, event definitions, and whether participants can contribute more than one event. The registry record does not provide enough detail to determine those features for these figures. Because treatment was assigned non-randomly and cohorts differ in size and treatment combination, these data do not establish comparative safety or causation.

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

The record describes a study of a molecularly guided treatment strategy that incorporated biopsy and assigned treatment according to tumor findings. It reports that the primary endpoint was the percentage alive 9 months from registration in a specifically defined subset. The primary endpoint result and its 90% confidence interval are not reported in the registry record. Accordingly, the posted information does not quantify survival at that time point or its precision.

The study’s non-randomized design means the treatment cohorts should not be interpreted as equivalent groups created for a fair head-to-head comparison. Their molecular profiles determined treatment combinations. In general, differences in outcomes between cohorts could reflect patient or tumor characteristics, treatment, or other factors. The registry lists 0 analyses and does not present adjusted comparisons that would address these concerns.

The study also records safety and feasibility outcomes, but listing outcome names or cohort counts does not establish the frequency, cause, or comparative likelihood of an event. The reported serious adverse event figures are descriptive, and the record does not provide context sufficient to attribute them to a treatment. Nor does the record state why the study was terminated.

These limitations do not erase the study’s documented focus on biopsy, molecular assignment, and treatment in DIPG. They define what can be concluded from this registry record: the design, planned primary endpoint, treatment cohorts, and some safety counts are described, but the main survival result and analysis details are absent. No claim about treatment effectiveness, lack of effectiveness, or comparative safety is supported by the endpoint information posted.

13. Why This Trial Matters Statistically

Statistically, this study joins several sources of uncertainty that are important to keep distinct. The primary endpoint is a fixed-time survival rate, while the intervention strategy depends on obtaining biopsy material and assigning treatment from molecular findings. The analysis population excludes participants who did not complete those specified steps. As a result, the endpoint’s interpretation depends not only on survival follow-up but also on who entered the defined analysis population.

The non-randomized design is another important feature. Molecularly determined assignment reflects a clinical strategy, but it does not create randomized treatment groups. Cohort outcomes would therefore require cautious interpretation, particularly when the groups receive different drug combinations and may differ in underlying tumor characteristics. The record lists no analyses to explain whether or how these issues were addressed.

The registry also makes clear why a registered endpoint and an outcome result are different kinds of information. It defines 9-month OS and identifies a 90% confidence interval, yet does not report the estimate or interval. Statistical interpretation requires both the estimand—the outcome and population being targeted—and the actual data-based estimate with its uncertainty. Here, the estimand is described, but the result is absent.

Finally, the record lists outcomes spanning surgical feasibility, adverse events, and survival. In general, interpreting many outcomes calls for clarity about which questions were primary and how multiplicity was handled. Neither an analysis hierarchy nor a multiplicity approach is reported. The study is therefore statistically informative as a record of design choices and limitations, while the unavailable analyses prevent assessment of its clinical outcome estimates.

14. 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 endpoint result, confidence interval limits, and endpoint analysis are not reported in the registry record. Although enrollment is listed as 53, the number of participants in the stated primary analysis population is not reported. This prevents assessment of how much of the enrolled group contributed to the endpoint and what follow-up information was available.

Allocation was non-randomized and treatment was assigned according to molecular findings. The cohort structure therefore does not support a simple like-for-like comparison of treatment combinations. No adjusted or unadjusted analysis is posted, and the registry does not describe methods for controlling confounding. The study is also marked terminated, but the reason for termination is not reported in the registry record.

Important operational details are missing, including the handling of incomplete survival follow-up, the treatment of missing data, the statistical approach for the 9-month survival rate, and the analysis plan for the listed outcomes. The record lists 8 outcomes and 0 analyses. It does not report a multiplicity strategy, sample-size justification, interim analysis plan, or criteria for the molecular treatment assignments beyond the cohort descriptions.

The serious adverse event counts and listed percentages lack details about event ascertainment, observation periods, and percentage denominators. Cohort sizes are unequal, and the cohorts received different regimens. These features limit comparisons and attribution. Some arm-description text is incomplete in the registry record, so the full treatment details cannot be reconstructed from this information.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Molecularly Determined Treatment of Diffuse Intrinsic Pontine Gliomas (DIPG) — NCT01182350.
• Registry record last updated 2019-09-04; status last verified 2019-08.
• Results first posted 2019-09-04.
• 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.

18. Record Summary

NCT01182350 is a non-randomized, unmasked, parallel Phase II study of molecularly determined treatment for children and young adults with newly diagnosed DIPG. The registry lists enrollment of 53, a start date of 2011-09, primary completion in 2016-06, and status as terminated. It describes biopsy followed by molecularly based assignment to four treatment cohorts involving radiation, bevacizumab, erlotinib, and temozolomide in different combinations.

The primary endpoint is the percentage alive 9 months from registration among participants who underwent biopsy, obtained a molecularly based treatment assignment, and started assigned chemoradiation therapy. The registry lists 8 outcome measures and 0 analyses. The primary survival estimate, its 90% confidence interval, missing-data methods, multiplicity approach, and reason for termination are not reported in the registry record. Serious adverse event counts and selected descriptive percentages are listed, but they do not constitute randomized comparisons.