Testosterone Revival Abolishes Negative Symptoms, Fosters Objective Response and Modulates Enzalutamide Resistance: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Testosterone Revival Abolishes Negative Symptoms, Fosters Objective Response and Modulates Enzalutamide Resistance (NCT02286921), a randomized phase 2 study. The primary outcome is Progression Free Survival as Measured by Number of Months Until Clinical or Radiographic Progression.

2. Clinical Question

The study asked whether bipolar androgen therapy (BAT), using intramuscular testosterone, could be compared with enzalutamide in asymptomatic men with progressive metastatic castration-resistant prostate cancer after treatment with abiraterone acetate before chemotherapy for metastatic disease. The registry describes a randomized, open-label study and identifies enzalutamide as standard therapy in the study rationale. The primary question concerned progression-free survival, reported as the median number of months until clinical or radiographic progression.

The study was intended to evaluate the primary endpoint alongside secondary measures that included radiographic progression, prostate-specific antigen response, objective response by RECIST, time to prostate-specific antigen progression, and several quality-of-life outcomes. The registry record lists 14 outcome measures, while the posted secondary results include 13. The record does not explain the difference. The posted data therefore offer selected outcome results, but not a complete statistical account of every planned measure or a reported analysis of the primary between-arm contrast.

3. Trial Design

This was a randomized Phase II study with a parallel-group model and no masking. The registry describes it as multi-institutional and open label. Enrollment was 222. The study started in 2015-01, had a primary completion date of 2018-10-26, and is listed as completed. The sponsor was Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins.

Participants were assigned to BAT or enzalutamide. The BAT arm could receive testosterone cypionate or testosterone enanthate by intramuscular injection, at a dose of 400 mg every 28 days. The enzalutamide arm used 40 mg tablets, with instructions to take 4 tablets orally per day for 28 days per cycle. The registry reports these treatment instructions; further details about treatment duration, adherence monitoring, dose changes, or treatment discontinuation are not reported in the registry record.

The primary endpoint was progression-free survival, expressed as the median number of months until clinical or radiographic progression, with a timeframe of up to 2 years. The posted arm-level results use different participant counts: 94 for BAT and 101 for enzalutamide. The registry does not explain how these endpoint counts relate to overall enrollment or provide an account of participants excluded from this analysis.

Secondary results include counts for prostate-specific antigen response and objective response by RECIST. The participant counts differ across those outcomes, and the registry does not report why. The record lists 14 outcome measures, but results are posted for 13 secondary measures; the specific measure without posted results is not identified in the record. A detailed protocol, statistical analysis plan, allocation procedure, and analysis-population definitions are not reported in the registry record.

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.

Before assignment: 27 participants were not assigned to intervention due to screen failure, withdrawn consent, or PI decision.

Overall 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.

6. Primary Endpoint

The registry lists the primary outcome as Progression Free Survival as Measured by Number of Months Until Clinical or Radiographic Progression, measured over up to 2 years.

Progression-free survival records time from randomization until progression, making it a time-to-event endpoint. In this study, the primary outcome is expressed as the median number of months until clinical or radiographic progression, with results described for a timeframe of up to 2 years. A median is the point in the observed event-time distribution at which half the observations are at or below that time and half are at or above it. It describes the middle of the distribution rather than the experience of every participant.

The registry description gives clinical progression criteria. These include cancer pain requiring chronic opiate analgesia, a skeletal-related event such as pathologic fracture or spinal cord compression, and clinically significant symptoms from loco-regional tumor progression that require an intervention. The record’s primary endpoint title also includes radiographic progression, but a detailed definition of radiographic progression is not reported in the registry record.

Time-to-event outcomes are useful when the timing of progression matters and participants may have different follow-up times. A comparison typically needs to account for censoring, which occurs when a participant’s event time is not observed during the available follow-up. The registry posts arm-specific medians and bounds, but its comparison status says that values are missing or not estimable for a two-arm comparison. It does not post a hazard ratio or another inferential comparison for the primary endpoint. The arm medians alone therefore do not establish the magnitude or precision of a treatment contrast.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry posts an arm-specific median for the primary time-to-event outcome and counts for two secondary binary outcomes. These summaries answer different questions. A median describes a central event time in each arm, whereas a response count describes how many participants met a defined response outcome among those included in that outcome’s denominator. Neither type of summary, by itself, supplies a fully adjusted or comprehensive account of the randomized comparison.

What does a median time summarize?

For a time-to-event outcome, the median is a point on the event-time distribution. It is not the average follow-up time, and it does not show how outcomes were distributed across all participants. Participants whose progression is not observed during follow-up can be censored; survival methods account for this incomplete event-time information. The registry’s primary endpoint is described as a median time to clinical or radiographic progression, but it does not report a two-arm comparison that can be interpreted as a treatment-effect estimate.

The posted medians are 5.62 months for BAT and 5.72 months for enzalutamide. The registry lists bounds alongside each median, but does not state in the posted endpoint information what inferential quantity those bounds represent. A confidence interval, when reported, expresses uncertainty about an estimated quantity under a statistical model. It is not a range that contains a fixed share of individual participants’ event times. Since the bounds are not explicitly identified as confidence intervals in the record, that interpretation should not be assumed.

Why is a two-arm time-to-event comparison important?

Arm-specific medians do not capture every feature of the event-time distributions. A formal comparison may use a model or test that incorporates the timing of events and censoring. A hazard ratio, when reported, compares event rates over time under its model assumptions; it is not the same as a difference between medians. The registry does not report a hazard ratio or other estimable two-arm comparison for this primary outcome. Its stated reason is that values are missing or not estimable for a two-arm comparison of the median-time endpoint. The intended analysis method and the reason the comparison could not be estimated are not reported in the registry record.

How should the response comparisons be read?

The secondary response outcomes are binary: participants either meet the stated response outcome or do not. A risk difference expresses the difference between the response proportions in the compared arms, with the direction depending on the ordering of the arms. For prostate-specific antigen response, the estimated risk difference is 0.031 for enzalutamide relative to BAT. Its reported interval ranges from -0.080 to 0.141, and its unadjusted p-value is 0.59. For RECIST objective response, the corresponding estimate is -0.201, the reported interval ranges from -0.367 to -0.034, and the unadjusted p-value is 0.04. The record does not label the intervals in the endpoint text as confidence intervals, so they are described as reported intervals.

A p-value measures how compatible the data are with a specified null hypothesis under the test’s assumptions. It is not the probability that the null hypothesis is true, nor does it measure the clinical importance of a result. An unadjusted p-value is calculated without accounting for the number of outcomes or comparisons examined. When many outcomes are tested, the chance of observing at least one result that appears statistically notable by chance can rise. The registry does not report a multiplicity adjustment for the secondary outcomes.

What do the sample-size illustrations mean?

The planning illustrations are based on observed effects, rather than on the trial’s original design assumptions. They use a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation. The illustrations consider 100%, 75%, and 50% of the observed effect and report the number of participants per arm for 80% or 90% power. These are planning scenarios, not observed power and not evidence that the completed trial was designed to detect those effects. Sample-size assumptions used in the original trial design are not reported in the registry record.

Observed-effect planning can be unstable, particularly when an estimate comes from a small sample or is imprecise. A smaller assumed effect generally requires a larger planned sample to achieve a target power under the stated framework. The illustrations help show how assumptions about the effect influence planning, but they do not establish the reliability of an observed result. The posted sample-size scenarios should not be read as proof that the study had adequate or inadequate power for its registered primary endpoint.

The registry lists no reported analyses in its analysis listing. It does not describe the statistical model used for the primary endpoint, the method used to calculate the secondary intervals, covariate adjustment, handling of censoring, or any interim analyses. Those methodological details are not reported in the registry record and cannot be inferred from the arm summaries alone.

10. Multiplicity, Missing Data and Analysis Populations

The registry does not report a multiplicity strategy for the secondary outcomes. This matters because the record includes multiple outcomes across progression, response, and quality-of-life domains. If each is tested separately without adjustment, a nominally small p-value can arise by chance even when no underlying effect exists. A prespecified hierarchy, family-wise error adjustment, or other planned approach can clarify how evidence from many comparisons should be interpreted. Whether any such approach was used is not reported in the registry record.

The record reports an unadjusted p-value for each of the two posted response comparisons and lists no reported analyses in its analysis listing. It does not describe whether the secondary outcomes were exploratory, whether the primary endpoint was intended to control the overall error rate, or whether there were planned interim analyses. These elements should not be inferred from a Phase II label or from the posted results.

Missing-data handling and analysis populations are also not reported. Overall enrollment was 222, while the primary endpoint summaries use 94 and 101 participants in the respective arms. Secondary outcome denominators differ by outcome. The registry does not state whether these counts reflect evaluability, missing measurements, eligibility rules, or another reason. Nor does it define an intention-to-treat, per-protocol, or other analysis population. The distinction matters because exclusions or missing outcome data can affect how well the analyzed participants represent those randomized.

The registry does not explain how censored observations were handled for progression-free survival, how missing response assessments were classified, or how incomplete quality-of-life questionnaires were treated. It also does not identify sensitivity analyses that might test whether conclusions depend on assumptions about missing data. These omissions limit the ability to assess the robustness and reproducibility of the posted results. They do not establish that the study used inappropriate methods; they mean that the relevant methods are not reported in the registry record.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: back pain (5.9%), bone pain (1.1%), stroke (1.1%), seizure (1.0%), myocardial infarction (1.0%).

The registry reports serious adverse events in 4 of 94 participants in the BAT arm and 11 of 101 participants in the enzalutamide arm. It reports other adverse events in 22 of 94 and 29 of 101 participants, respectively. Deaths are listed as 0 in the BAT arm and 1 in the enzalutamide arm. These are descriptive counts from the registry and should not be interpreted as proof that treatment caused or prevented any event.

The listed serious adverse-event terms and percentages are Back pain, 5.9%; Bone pain, 1.1%; Stroke, 1.1%; Seizure, 1.0%; myocardial infarction, 1.0%; and Anemia, 1.0%. The registry does not provide arm-specific counts for each listed term in the information, so these percentages should not be assigned to a particular treatment arm. It also does not provide the event-level context needed to assess timing, duration, severity beyond the serious-event classification, or relationship to treatment.

Comparisons of adverse-event counts can be affected by follow-up duration, exposure, event definitions, and completeness of reporting. Those details are not reported in the registry record. The posted figures describe reported events among the listed participants; they are not a complete estimate of comparative safety or a basis for medical advice.

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

The registry supports a limited descriptive account of the trial. It identifies the randomized arms, posts medians for the primary time-to-progression outcome, and provides counts and comparisons for selected secondary response outcomes. It also reports adverse-event summaries. These results can help readers understand what was recorded for the study, but they do not constitute a complete statistical assessment of the randomized comparison.

For the primary endpoint, the arm-specific medians are 5.62 months and 5.72 months, but the registry does not report an estimable two-arm comparison. Without an effect estimate and its uncertainty for that primary contrast, the posted medians alone cannot establish whether the groups differed. The listed bounds do not resolve this issue because the record does not identify their statistical interpretation.

The RECIST objective-response comparison is a secondary result with an unadjusted p-value of 0.04 and a reported interval from -0.367 to -0.034 for the estimated risk difference. It is one endpoint among several, has participant counts that differ from the primary endpoint counts, and is not accompanied by a reported multiplicity strategy or analysis-population definition. It should not be turned into an overall claim about comparative treatment effects. The prostate-specific antigen response comparison, in turn, has a reported interval from -0.080 to 0.141 and an unadjusted p-value of 0.59. These endpoint-specific results do not supply the missing primary comparison.

The registry also does not report the original sample-size justification, statistical analysis plan, censoring methods, missing-data strategy, or interim analyses. Their absence from the record prevents assessment of whether the posted analyses followed prespecified methods and whether results are sensitive to analytic choices. The sample-size illustrations based on observed effects are planning scenarios and do not provide an observed-power calculation or establish the trial’s original design rationale.

Accordingly, the posted information describes selected observed results and their limited reported comparisons. It does not establish a comprehensive conclusion about the relative benefits or risks of the treatments. The statistical interpretation should remain bounded by the registry’s missing primary contrast, incomplete methodological detail, and unadjusted secondary analyses.

13. Why This Trial Matters Statistically

This trial is statistically notable because its registry record posts arm-specific results for a time-to-event primary endpoint but does not post an estimable comparison between the randomized groups. That distinction is important: a reader may see two medians and be tempted to treat their numerical proximity as a treatment comparison. The medians are descriptive summaries; the record’s missing comparison means that the primary treatment contrast cannot be evaluated from those summaries alone.

The trial also illustrates why outcome-specific denominators and analysis populations matter. Overall enrollment is 222, yet the endpoint summaries use different participant counts, and the secondary outcomes use still other counts. Without an explanation of those denominators, readers cannot determine whether the differences reflect missing measurements, outcome eligibility, evaluability, or another process. This is not a reason to assume bias, but it is a reason to withhold conclusions that depend on knowing who was analyzed.

Multiplicity is another central issue. A registry can include many outcomes that capture different dimensions of disease and quality of life. A nominally notable result among several unadjusted comparisons may be less persuasive than a result arising from a clearly prespecified primary analysis. Here, the primary two-arm comparison is not reported, and no multiplicity plan is available in the record. That makes it especially important to distinguish a secondary endpoint finding from the trial’s principal question.

Finally, the sample-size illustrations are based on observed effect estimates and stated planning assumptions, not on the trial’s original design rationale. This separates retrospective planning scenarios from evidence about power at trial launch. The record demonstrates the value of transparent reporting: arm summaries, denominators, analysis methods, missing-data handling, and multiplicity plans all contribute to whether results can be interpreted and reproduced.

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.

Several limitations are specific to the posted record. The primary endpoint is reported as arm-specific medians, but the registry says that a two-arm comparison is missing or not estimable. No hazard ratio, primary comparison interval, or primary p-value is posted. The record does not explain why the comparison could not be estimated, so the cause cannot be determined from the registry results.

The endpoint description defines clinical progression events but does not give a detailed radiographic progression rule. The record also lists bounds alongside the medians without identifying the statistical meaning of those bounds in the endpoint information. This leaves uncertainty about how to interpret the accompanying values and limits assessment of endpoint reproducibility.

Outcome denominators vary. The registry does not state why the primary and secondary participant counts differ, define analysis populations, or describe how missing outcome data were handled. It reports 14 outcome measures but posts 13 secondary results, without identifying the outcome for which results are absent. These issues make it difficult to assess completeness and potential selection in the reported analyses.

The record does not include a statistical analysis plan, original sample-size justification, covariate strategy, censoring method, interim-analysis information, or multiplicity approach. The secondary comparisons are unadjusted, and the posted analysis listing contains no reported analyses. The registry therefore permits a limited reading of observed summaries but does not allow a thorough audit of whether the statistical methods were prespecified or how robust conclusions are to alternative assumptions.

The trial was open label, with no masking. This is a design fact, not evidence that a particular outcome was biased. It may be relevant when considering outcomes that involve assessment or participant reporting, but the record does not describe safeguards or blinded outcome assessment. Further details are not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Testosterone Revival Abolishes Negative Symptoms, Fosters Objective Response and Modulates Enzalutamide Resistance — NCT02286921.
• Registry record last updated 2020-11-06; status last verified 2020-07.
• Results first posted 2019-05-01.
• 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

This completed randomized Phase II study compared BAT using testosterone cypionate or testosterone enanthate with enzalutamide in asymptomatic men with progressive metastatic castration-resistant prostate cancer after abiraterone acetate treatment. The study used an open-label, parallel-group design and enrolled 222 participants.

The primary endpoint was median progression-free survival, measured in months until clinical or radiographic progression. The registry posts medians of 5.62 months for BAT and 5.72 months for enzalutamide, but reports no estimable two-arm comparison for this outcome. Secondary response results include an unadjusted comparison for prostate-specific antigen response and one for objective response by RECIST; they do not replace the missing primary comparison.

Important statistical details—including the original sample-size justification, analysis populations, missing-data methods, censoring approach, and multiplicity plan—are not reported in the registry record. The posted results therefore support a bounded description of selected outcomes, not a complete inferential conclusion about the randomized treatment comparison.