A Study to Learn About Novel Hormonal Therapies (NHTs) for Metastatic Castration-sensitive Prostate Cancer (mCSPC) in People Who Were in the Armed Forces: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Learn About Novel Hormonal Therapies (NHTs) for Metastatic Castration-sensitive Prostate Cancer (mCSPC) in People Who Were in the Armed Forces (NCT06151418), a registered phase not stated study. The primary outcome is Duration of Therapy (DOT).

2. Clinical Question

3. Trial Design

Despite the general use of the word “trial,” this record describes a real-world, retrospective analysis, not a clinical trial in which researchers assign treatment. It uses information from the Veterans’ Health Administration (VA) Health Care Network about patients who received NHT as part of usual care. The listed intervention is NHT, and the cohort consists of patients initiating abiraterone, apalutamide, or enzalutamide for mCSPC.

The record reports an enrollment of 2313. It does not report how many participants were ultimately included in each outcome analysis. Eligibility is described as having mCSPC, starting NHT for that condition, and being 18 years of age or older at treatment initiation. The analysis population is described as eligible participants whose data were retrieved and evaluated for the study objectives.

The index date was the initiation date of first-line NHT, with eligible initiation dates from 1-January-2020 to 31-December-2022. The record says that retrospective VA data were retrieved for 01-Jan-2020 to 30-Sep-2023, described as 45 months. It also says that information was used from one year before NHT initiation until information was available. The primary outcome timeframe is from treatment initiation to discontinuation or censoring, during a maximum follow-up of 45 months.

The registry lists the study as completed, with a start date of 2023-11-22 and primary completion date of 2024-04-23. It does not describe a randomized allocation, a control arm, or a planned comparison between the three NHTs. A detailed statistical analysis plan, including any adjustment for differences among patients or treatments, is 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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Analysis population included eligible participants whose data was retrieved and evaluated per objectives of this 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.

Population: Analysis population included eligible participants whose data was retrieved and evaluated per objectives of this study.

6. Primary Endpoint

The registry lists the primary outcome as Duration of Therapy (DOT), measured over From start of index treatment to date of NHT discontinuation or censoring date, whichever came first, during maximum follow-up of 45 months.

The primary endpoint, Duration of Therapy (DOT), measures the time from the index date—the start of first-line NHT—to discontinuation of that NHT. Discontinuation is defined as occurring at the earliest of a treatment gap of at least 90 days, initiation of a new therapy, or death. For a gap, the stated discontinuation date is the last day with days’ supply before the gap. A new therapy includes switching to a different NHT, switching to another therapy, or adding another therapy. Patients without a recorded discontinuation are censored at the end of data availability.

The registry specifies that DOT is summarized using the median and a 95% confidence interval, in months. A time-to-event endpoint is useful for describing how long patients remain on treatment when follow-up length can vary and some patients have no observed discontinuation before their data end. Censoring allows those patients to contribute information up to the point their follow-up ends, without treating that point as a discontinuation.

The definition combines treatment gaps, treatment changes, and death as ways the observed treatment period can end. The resulting measure therefore reflects the registry’s operational definition of treatment duration, rather than simply a count of medication days. The record also lists TTNT as a secondary outcome, but its detailed definition and numerical results are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The methods below explain general statistical concepts relevant to the outcomes and definitions in the registry record. The record identifies a median and 95% confidence interval for DOT, but it does not report the statistical analysis plan or the numerical estimates.

What does a time-to-event outcome measure?

Generally, a time-to-event outcome records the time from a defined starting point to a specified event. For DOT, the starting point is the initiation of first-line NHT, and the event is discontinuation under the study definition. This approach differs from summarizing only whether an event occurred: it uses the timing of the event as well. In this study, the definition treats a sufficiently long treatment gap, a new therapy, or death as discontinuation, whichever comes first.

Why can observations be censored?

In general, a participant is censored when the event has not been observed by the end of the participant’s available follow-up. Censoring does not mean that the event never happened; it means the analysis has no recorded event time beyond the last available observation. The registry says participants without discontinuation were censored at the end of data availability. A time-to-event analysis can use their observed follow-up up to that point, provided the assumptions of the chosen method are appropriate. The registry does not identify the estimator or describe checks of those assumptions.

What do the median and confidence interval communicate?

Generally, the median time-to-event estimate marks the point at which half the estimated event-time distribution lies on either side, subject to the method used and the available follow-up. It is a population summary, not a prediction for a particular patient. A 95% confidence interval conveys statistical uncertainty about the estimated median under the analysis method. A wider interval indicates less precision than a narrower one, but an interval is not a guarantee about the true value. No DOT median or confidence interval is reported in the registry record, so its precision cannot be evaluated.

How should treatment discontinuation be understood?

The endpoint is based on recorded treatment patterns. In this definition, a gap of at least 90 days counts as discontinuation, using the last day with days’ supply before that gap as the date. A switch or augmentation with another therapy also meets the definition, as does death. This operational rule makes the outcome interpretable and reproducible in principle, but it also means DOT depends on how medication supply, treatment changes, and dates are captured in the data. Details about validation of these records are not reported in the registry record.

What is the role of the secondary outcome?

TTNT is listed as a secondary time-to-event outcome. In general, a time-to-next-therapy measure describes the time until a subsequent treatment begins, but the registry’s detailed definition is not reported. The secondary outcome is distinct from DOT: a patient’s initial therapy might be recorded as discontinued without a subsequent therapy starting at that same time, depending on the endpoint definitions and data. The record does not provide an estimate or confidence interval for TTNT.

What can an observational analysis establish?

Generally, a retrospective analysis of usual-care records describes outcomes among people whose treatment and follow-up appear in those records. Because treatment is not assigned randomly, groups receiving different therapies can differ in clinical or other characteristics. Such differences can complicate causal comparisons. The record does not describe a comparison group, an adjustment strategy, or methods to address confounding. It therefore does not support interpreting the described cohort as a randomized treatment-effect comparison.

The record reports two outcome measures and zero analyses, but does not detail a statistical plan, missing-data procedures, or multiplicity methods. It does not report an observed effect estimate from which a sample-size illustration could be interpreted. Any general planning calculation is not observed power and cannot substitute for an outcome estimate. The specific analytic choices—including the time-to-event estimator, handling of potential competing events, and any covariate adjustment—are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists two outcome measures: DOT as the primary outcome and TTNT as a secondary outcome. It reports zero analyses, and no numerical estimates or hypothesis tests are reported in the registry record. No multiplicity procedure, testing hierarchy, or adjustment for multiple outcomes is described. Without a reported analysis plan, it is not possible to determine whether multiplicity was considered or how any formal statistical claims would have been controlled.

The analysis population is described as eligible participants whose data were retrieved and evaluated for the study objectives. The record reports enrollment of 2313, but does not state how many participants contributed to each outcome analysis or whether those counts differed. Enrollment and an outcome-specific analysis population are not interchangeable, particularly in a retrospective database study where eligibility and data availability may affect inclusion.

The description says that patients without a recorded discontinuation were censored at the end of data availability. That defines one aspect of follow-up, but other missing-data procedures are not reported. The registry does not explain how incomplete medication records, missing dates, missing covariates, or gaps in VA data were handled. It also does not report whether sensitivity analyses assessed the effects of incomplete follow-up or missing information.

Generally, missingness can affect a time-to-event estimate if the reasons observations are unavailable are related to the event or to the chance of remaining under observation. Censoring methods rely on assumptions about that process. The registry record does not provide enough detail to evaluate those assumptions, the number censored, or their follow-up experience.

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

The record describes a real-world cohort of people with mCSPC who initiated one of the named NHTs and sets out how DOT is defined. It identifies a median and 95% confidence interval as the intended primary summary and lists TTNT as a secondary outcome. These details explain what the analysis was designed to describe, but they do not provide the outcome estimates.

The registry record does not report the DOT median, its confidence interval, or a TTNT result. It therefore does not show how long participants remained on NHT, how precisely that duration was estimated, or how long it took to initiate another therapy. No numerical comparison among the named therapies is reported. No control group or randomized comparison is described, so the record does not establish that one treatment caused a different duration than another.

Generally, an observational treatment-duration estimate reflects both the study’s endpoint rules and the patients and care patterns represented in the data. Here, discontinuation includes a treatment gap of at least 90 days, a new therapy, or death. The measure is therefore specific to those rules and to the available VA records. It should not be read as a direct measure of biological response or as a universal duration applicable to all people with mCSPC.

The study’s enrollment figure does not establish how many participants were analyzed for each outcome. The record does not report outcome-specific counts, detailed missing-data methods, or adjusted estimates. The listed zero adverse-event counts also do not establish that no adverse events occurred, because ascertainment and analysis details are not described. The defensible interpretation is limited to the cohort, definitions, and results actually reported: the design and endpoint definitions are described, while numerical outcome findings are not reported in the registry record.

12. Why This Trial Matters Statistically

Treatment duration is a useful real-world outcome because it describes how long an initial therapy remains in use before a recorded discontinuation under a defined rule. In this study, the rule includes a prolonged treatment gap, a new therapy, or death. That makes the endpoint more informative than an unqualified statement that someone “stayed on treatment,” while also making clear that the result depends on how treatment changes and medication supply are recorded.

The study addresses care captured in the Veterans’ Health Administration network, rather than treatment assigned by a clinical-trial protocol. Such observational data can describe usual-care patterns in the study population. They can also raise important statistical questions about variable follow-up, censoring, data completeness, and differences among people receiving different treatments. The registry record supplies the endpoint framework but does not report the analyses needed to evaluate those questions in detail.

Statistically, the distinction between a defined outcome and a reported estimate matters. A median with a confidence interval could convey both a typical event time and the uncertainty in its estimation. Without those values, readers cannot assess the observed duration or its precision. Likewise, listing TTNT as a secondary outcome does not reveal its result. The record’s absence of numerical results is therefore central to interpreting what can be learned from this registration.

The analysis is not a randomized comparison, and no control arm is described. Its statistical value lies in the question it sets out to address in routine-care data, not in establishing causal differences between therapies. Clear definitions and transparent reporting of analysis populations and uncertainty would be important for readers to understand any future posted estimates.

13. 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 registry record leaves important details unavailable for evaluating the analysis. Although the enrollment is 2313, the number included in the DOT and TTNT analyses is not reported. The record does not provide numerical outcome estimates, confidence intervals, or outcome-specific event counts, so neither the magnitude nor the precision of the findings can be assessed.

The study uses retrospective VA data, and the stated period of data retrieval ends on 30-Sep-2023. In this kind of data, observed treatment patterns depend on what is recorded and available. The registry does not describe data validation, completeness of medication records, methods for identifying treatment changes, or handling of incomplete information. It also does not report how loss of available follow-up was assessed.

The endpoint definition treats a gap of at least 90 days as discontinuation and counts switching, augmentation, and death as discontinuation events. These rules are explicit, but different operational definitions could characterize treatment duration differently. The registry does not report sensitivity analyses using alternative definitions. It also does not describe how death was handled beyond including it in the discontinuation definition.

Because treatment is received as usual care and no control group is described, the record does not establish comparability among patients receiving different NHTs. Covariate adjustment and methods for addressing confounding are not reported. The registry also does not give a detailed definition of TTNT, a statistical analysis plan, or a multiplicity strategy. These omissions limit independent assessment of the intended analyses and the generalizability of any resulting estimates.

14. Related Tutorials

15. Related Statistical Calculators

16. Sources

• ClinicalTrials.gov: A Study to Learn About Novel Hormonal Therapies (NHTs) for Metastatic Castration-sensitive Prostate Cancer (mCSPC) in People Who Were in the Armed Forces — NCT06151418.
• Registry record last updated 2025-06-04; status last verified 2025-06.
• Results first posted 2025-06-04.
• 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.

17. Record Summary

This completed registry record describes a retrospective, real-world analysis of VA data on people with mCSPC who initiated NHT. The named therapies are abiraterone, apalutamide, and enzalutamide, and the reported enrollment is 2313. The primary outcome is DOT, defined from NHT initiation to discontinuation or censoring and summarized as a median with a 95% confidence interval. TTNT is listed as a secondary outcome.

The record specifies eligibility, the treatment-duration rules, and the data period, but does not report numerical results for either outcome. It also does not report outcome-specific analysis counts, detailed missing-data procedures, or a statistical analysis plan. The design is observational and describes no randomized control group or like-for-like comparison. Thus, the registry record establishes the intended questions and definitions, but not the results needed to quantify treatment duration or its uncertainty.