External Beam Radiation With or Without Chemotherapy to Treat High Risk Prostate Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for External Beam Radiation With or Without Chemotherapy to Treat High Risk Prostate Cancer (NCT01603420), a randomized phase 2, phase 3 study. The primary outcome is Phase 2 - Assessment of Number of Freedom From Failure Events in the Chemotherapy Arm.

2. Clinical Question

At the level of the study question, the comparison concerns treatment strategies that differ in chemotherapy and the duration of LHRH treatment. It is not a comparison of chemotherapy alone against no treatment: both arms include radiation and androgen suppression. Because the LHRH durations differ, the contrast also cannot isolate the contribution of chemotherapy from the contribution of that difference in treatment duration. The registry record does not report a separate analysis designed to disentangle those components.

The posted primary outcome lists zero failure events in each arm, with 1 participant in each arm. Those entries describe the observed counts at study closure. They do not establish how the regimens would compare at the planned endpoint or in a larger population.

3. Trial Design

This was a randomized, parallel-group phase 2/3 study with no masking reported. Its title describes dose-escalated external beam radiation therapy with or without chemotherapy for high-risk adenocarcinoma of the prostate. The registry lists the Proton Collaborative Group as sponsor and prostate cancer as the condition.

The two treatment arms shared conformal radiation at 79.2 Gy(RBE). The arm without chemotherapy paired radiation with 24 months of LHRH agonist treatment. The chemotherapy arm paired radiation with docetaxel 20mg/m2 every 7 days for 8 weeks, followed by 6 months of LHRH treatment. Thus, the treatment strategies differ in both the inclusion of docetaxel and the stated duration of LHRH treatment. The registry record does not report a design feature that would separate the effects of those differences.

The study started in 2012-07, and primary completion is listed as 2014-05. Its status is terminated. Enrollment is recorded as 2, with 1 participant in each arm in the posted primary outcome. The registry record does not report why the study was terminated, nor does it give a reason for the low enrollment.

The primary outcome was originally intended to assess failure at 2 years. The record says that this endpoint was not reached and instead reports the outcome at study closure, 22 months. That timing is important: the posted counts are not a completed assessment at the originally stated endpoint. The record does not report further follow-up for this outcome.

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

Analysis population for the primary outcome, as defined in the registry: No failures were reported at the time of study termination (22 months). This outcome was originally written to assess failure at 2 years. However, that end point was not reached.

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 Phase 2 - Assessment of Number of Freedom From Failure Events in the Chemotherapy Arm, measured over No failures were reported at the time of study termination (22 months). This outcome was originally written to assess failure at 2 years. However, that end point was not reached..

The primary endpoint concerns freedom from failure in the chemotherapy arm. The registry defines failure as the first occurrence of clinical failure, biochemical failure by the Phoenix definition, or starting salvage therapy, including androgen deprivation. Clinical failure includes local recurrence, regional recurrence, or distant metastasis. The biochemical definition is described as a prostate-specific antigen threshold written as “> = 2 ng/ml” over the nadir, with PSA bounces discounted at the investigators’ discretion.

A composite endpoint of this kind can capture several routes by which a participant may be considered to have experienced failure. In general, combining events can provide a broader summary of disease-control outcomes than tracking only one type of recurrence. It also means that the endpoint depends on how each component is defined and ascertained. The registry record does not report how many events of each component type occurred.

The planned assessment was at 2 years, but the endpoint was assessed at study closure, 22 months, because the study was terminated before the planned time frame was reached. The posted results report zero failure events in each arm. These counts therefore describe what was recorded by closure, not the full planned 2-year assessment. With 1 participant per arm, the counts provide little information about the range of outcomes that could occur in the intended study population.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies a randomized, parallel-group study with no masking. It does not report a statistical analysis plan or a specific model for the primary endpoint. The posted analysis count is 0, and the comparison is described as not reduced to one comparison. As a result, the statistical interpretation must remain descriptive.

What does randomization contribute?

General: Randomization assigns participants to study arms using a chance-based process. In a sufficiently large study, this can help balance measured and unmeasured characteristics between groups on average. It does not guarantee balance in every trial, and it cannot make a very small set of observations representative of the wider population. Here, the registry reports 1 participant in each arm, so the posted counts cannot establish that the groups were otherwise comparable. Baseline characteristics are not reported in the registry record.

How should the event counts be read?

The primary outcome is expressed as a number of failure events. The posted value is zero for each arm. These are counts observed by study closure; they are not estimates of the underlying risk for all people with high-risk prostate cancer. A count also does not tell us how precisely an event rate or treatment contrast has been estimated. No such estimate or interval is reported.

General: For event outcomes, statistical methods often account for the amount of observation time, particularly when participants are followed for different lengths of time or when the outcome is time to an event. The registry record reports an assessment at closure, 22 months, but does not provide participant-level follow-up times or a time-to-event analysis. It therefore does not support reconstruction of such an analysis.

What would a confidence interval convey?

General: A confidence interval describes the uncertainty around an estimated quantity under a specified statistical method. It can show which effect sizes remain compatible with the observed data and the model. The registry record does not post a confidence interval or a primary-outcome effect estimate. A zero count in each arm is not itself a confidence interval and should not be read as proof that the true outcomes are identical.

Why does the endpoint definition matter?

The outcome combines clinical recurrence, biochemical failure, and starting salvage therapy, including androgen deprivation. These events may be identified in different ways. The registry describes the biochemical criterion using a PSA threshold of “> = 2 ng/ml” over nadir, with PSA bounces discounted at investigators’ discretion. That discretion is part of the posted definition, but the record does not provide further operational details or component-specific counts.

General: Composite endpoints can increase the range of events represented by an outcome, but interpretation depends on the components and their definitions. A composite count alone does not reveal which component occurred or whether the components have similar clinical importance. The posted record does not report those details for this endpoint.

What does the sample-size section represent?

The sample-size section uses planning illustrations based on a stated fraction of the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. The record describes participant counts per arm for non-time-to-event outcomes and total events for time-to-event tables. These are planning assumptions, not observed power. Here, a two-arm comparison is not estimable from the posted summary statistics, so a usable comparison-based sample-size illustration is not available. No sample-size justification from the trial’s own design is reported in the registry record.

General: Over-dispersion matters when event counts vary more than a simple model assumes; ignoring it can make uncertainty look smaller than it is. The record does not report a count model, dispersion estimate, or model diagnostic. It is therefore not possible to say whether over-dispersion was assessed or to characterize its relevance to a sponsor analysis.

The registry does not report interim analyses, adjustment for covariates, or an inferential procedure for the primary endpoint. These omissions mean that no claim about a particular statistical model or testing strategy can be made from the posted record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 8 secondary outcomes, including adverse-event assessments, local or distant failures, biochemical failures, survival events, salvage androgen deprivation, and quality-of-life measures using the EPIC instrument. The record lists 12 outcome measures overall and an analysis count of 0. It does not post results for these secondary outcomes in the registry record.

When a study evaluates multiple outcomes, the chance of highlighting a result that appears unusual by chance can increase if each is tested separately. General: Multiplicity procedures are methods for managing that issue, for example by adjusting how evidence is evaluated across a set of tests or by specifying a hierarchy among outcomes. The registry record does not report a multiplicity strategy, a hierarchy among outcomes, or a statistical analysis plan. There is consequently no basis for judging whether or how multiplicity was addressed.

The record also does not report missing-data handling, analysis populations, exclusions after randomization, or participant-level disposition by outcome. It does not specify whether results were analyzed according to randomized assignment or another population definition. With enrollment of 2 and 1 participant in each arm for the posted primary outcome, even a small amount of missing follow-up could materially affect what is observable; however, the extent of missing data cannot be determined from the registry record.

For the primary endpoint, the timing limitation is explicit: assessment at 2 years was not reached, and the outcome was assessed at study closure, 22 months. The record does not report a method for handling censoring or incomplete follow-up. These are important gaps for interpreting an endpoint that includes failure over time.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry’s serious-adverse-event fields show 0 affected participants among 1 at risk in each arm. The corresponding fields for other adverse events show 1 affected participant among 1 at risk in each arm. These are the posted participant counts; they should not be read as counts of distinct events, since the record does not specify event totals or event types.

The serious-event fields therefore indicate that no participant is recorded as affected by a serious adverse event in either arm. The other-adverse-event fields indicate that the participant in each arm is recorded as affected. The record does not give the types, severity details beyond the serious/other distinction, timing, duration, or relationship to treatment for those other events. It also does not provide serious-event terms in the listed top serious terms.

Deaths are not reported in the registry record. With only 1 participant per arm, these figures are very limited descriptions of the participants represented in the registry. They cannot characterize how frequently adverse events would occur in a larger population or establish a comparative safety profile.

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

The result documents what was posted for this study: zero primary-outcome failure events in each arm by study closure, with 1 participant in each arm. The record places that assessment at 22 months and states that the planned 2-year endpoint was not reached. The result is therefore a short, descriptive snapshot relative to the originally specified assessment.

It does not show that the treatment strategies are equivalent. Equivalence requires a defined margin and an analysis designed to assess whether an estimated difference is sufficiently small. The registry record does not report such a margin or analysis. Nor does the absence of observed events establish that the true failure risk is zero.

The result also does not isolate the effect of chemotherapy. One arm includes docetaxel with 6 months of LHRH treatment; the other includes 24 months of LHRH treatment without the listed chemotherapy regimen. Any comparison between those arms would be a comparison of treatment packages, not a clean estimate of docetaxel’s independent contribution.

The posted information does not answer questions about longer-term failure, the separate components of the composite endpoint, or the secondary outcomes. It also does not provide uncertainty intervals or an inferential test. The appropriate interpretation is limited to the event counts and the timing reported in the registry.

Nothing in these results establishes that either strategy works or does not work. The study’s size, early termination, and incomplete planned follow-up prevent a reliable conclusion about comparative outcomes. The record gives no reason for termination, so its cause should not be inferred.

13. Why This Trial Matters Statistically

This record illustrates why an outcome’s timing is part of its meaning. The primary endpoint was intended to be assessed at 2 years, but the study was terminated before that time frame was reached. The posted count instead reflects the assessment at study closure, 22 months. A result recorded before the planned assessment cannot simply be treated as the planned endpoint.

It also shows the difference between a numerical result and an informative comparison. Zero events are reported in both arms, but with 1 participant per arm and no inferential analysis, the posted counts cannot quantify the difference between strategies with useful precision. The absence of a reported confidence interval leaves uncertainty unexpressed, rather than removing it.

The trial question further illustrates the importance of defining the treatment contrast carefully. The arms differ in chemotherapy and in LHRH treatment duration. Statistical interpretation should therefore describe the assigned strategies as a whole and avoid attributing any hypothetical difference to one component alone.

Finally, the record makes clear that registration and posted results can leave key analytic details unavailable. The registry lists several secondary outcomes, but no analyses are recorded. The analysis plan, missing-data approach, participant disposition, and reason for termination are not reported. These gaps limit assessment of both the intended analysis and the completeness of the posted evidence. The central lesson is that event counts need context: study size, follow-up timing, endpoint definition, and analysis details all shape what can be concluded.

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 most immediate limitation is enrollment: the registry reports 2 participants, with 1 in each arm for the primary outcome. Such a limited set of observations cannot provide a stable estimate of outcomes in the intended population. The posted zero counts should not be generalized beyond the participants represented in the record.

The study status is terminated, and the reason for termination is not reported in the registry record. The primary endpoint’s planned 2-year assessment was not reached; instead, the outcome was assessed at closure, 22 months. The record does not provide longer follow-up for this endpoint or participant-level follow-up details.

The primary outcome combines clinical failure, biochemical failure, and salvage therapy. The registry does not report component-specific counts, adjudication details, or further operational definitions beyond the description posted. This limits understanding of what, if anything, was observed within each part of the composite.

The arms differ in both chemotherapy use and LHRH treatment duration, so the contrast cannot isolate chemotherapy’s contribution. The record does not report baseline characteristics, an inferential analysis, effect estimates, confidence intervals, a statistical analysis plan, or the sample-size justification. It also does not report missing-data handling or analysis populations. These omissions, alongside the limited enrollment and incomplete planned follow-up, prevent a robust comparative interpretation.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: External Beam Radiation With or Without Chemotherapy to Treat High Risk Prostate Cancer — NCT01603420.
• Registry record last updated 2016-03-14; status last verified 2016-02.
• Results first posted 2015-08-31.
• 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

The registry record describes a randomized, unmasked, parallel phase 2/3 study of radiation with or without chemotherapy for high-risk prostate cancer, sponsored by the Proton Collaborative Group. The study started in 2012-07, lists primary completion as 2014-05, and has terminated status. Enrollment is recorded as 2.

The primary outcome concerns freedom-from-failure events in the chemotherapy arm, with failure defined by clinical recurrence, biochemical failure, or starting salvage therapy. The planned 2-year assessment was not reached. At study closure, 22 months, zero failures were reported in each arm; each arm had 1 participant. The record lists 8 secondary outcomes and 12 outcome measures, but the analysis count is 0. It does not report a reason for termination, an inferential comparison, or longer-term results.