This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for A Study to Learn About Novel Hormonal Therapies in People With Metastatic Castration-Sensitive Prostate Cancer (mCSPC) (NCT06072196), a registered phase not stated study. The primary outcome is Duration of NHT Treatment.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT06072196 |
| Phase | Phase not stated |
| Condition | Prostatic Neoplasms |
| Allocation | Not Stated |
| Design model | Not Stated |
| Masking | Not Stated |
| Enrollment | 3017 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Pfizer |
| Results posted | Yes |
2. Clinical Question
Population
People identified in Medicare claims as having metastatic castration-sensitive prostate cancer (mCSPC), who initiated novel hormonal therapy for mCSPC and were at least 65 years old one year before treatment began. The analysis population was eligible participants whose data were retrieved and observed.
Intervention
Initiation of novel hormonal therapy (NHT), defined as abiraterone, apalutamide, or enzalutamide. Treatment was prescribed as part of usual healthcare rather than assigned by the study.
Comparator
There is no separate comparator group. The registry describes a single cohort of people initiating NHT, so the study does not make a like-for-like comparison between treatments or against an untreated group.
Primary question
How long did eligible people take NHT for mCSPC, as measured from the initial NHT claim until the recorded end of treatment observation? The registry defines the endpoint using treatment discontinuation, treatment switch, death, or the end of available data, but does not report the resulting median duration.
The study concerns treatment duration in routine care, rather than an assigned treatment effect. Researchers planned to use Medicare insurance claims to identify people meeting the eligibility criteria and describe their time on treatment. The registry describes a cohort of 3017 participants and one primary outcome measure, Duration of NHT Treatment.
The study included retrospective information from one year before the initial NHT claim through the end of the insurance period or the end of available information. Its listed observation timeframe is 2 years, while the outcome description says retrieved retrospective data were evaluated for approximately 1.2 years. These descriptions do not establish that every participant had the same length of follow-up.
The central statistical question is descriptive: what was the estimated distribution of time from treatment initiation to the defined end of observation, and where was its median? The registry record does not report the median estimate, its 95% confidence interval, or other outcome results. It therefore does not establish a value for typical treatment duration in this cohort.
3. Trial Design
Despite the general use of the word “trial” in clinical research, this record describes a real-world observational study, not a clinical trial. Researchers used Medicare claims information to observe treatment prescribed by participants’ own doctors. The study did not assign NHT or allocate participants to different treatment arms.
The cohort included people identified as having mCSPC who initiated abiraterone, apalutamide, or enzalutamide for that condition, and who were at least 65 years old one year before their treatment start date. The registry identifies one cohort, called Novel Hormonal Therapy, and reports enrollment of 3017. It does not describe a randomized comparator or a concurrent control cohort.
The primary outcome concerns duration of NHT treatment. The start was the initial claim date for one of the listed therapies. The end of observation was defined by one of several events: data availability ending on December 31, 2022; the participant’s date of death; the day before a treatment switch; or the day treatment was discontinued. A 90-day gap in days’ supply was considered discontinuation.
The registry lists a 2 years retrospective data-observation timeframe, and also states that retrieved retrospective data were evaluated for approximately 1.2 years of the study. The study began on 2023-10-04, has a primary completion date of 2024-12-18, and is marked COMPLETED. These administrative dates do not supply an outcome estimate.
The sponsor is Pfizer. The record does not report the underlying claims database’s detailed construction, the number screened or excluded, or a separate analysis plan. It also does not report treatment-specific cohort sizes or comparisons among abiraterone, apalutamide, and enzalutamide. The design supports describing observed treatment duration in the defined cohort; it does not by itself establish what duration would have occurred under another treatment strategy.
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: Retrospective data were retrieved and evaluated per objectives of this observational study from 04-Oct-2023 to 18-Dec-2024 (approximately 1.2 years).
Overall Study
| Stage | Novel Hormonal Therapy |
|---|---|
| STARTED | 3017 |
| COMPLETED | 3017 |
| NOT COMPLETED | 0 |
Analysis population for the primary outcome, as defined in the registry: Analysis population included all eligible participants who met inclusion/exclusion criteria, whose data was retrieved and observed in 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.
Population: Analysis population included all eligible participants who met inclusion/exclusion criteria, whose data was retrieved and observed in the study.
| Characteristic | Unit | Novel Hormonal Therapy |
|---|---|---|
| Participants analysed | 3017 | |
| Age, Continuous | Years | 76.0 (SD 6.5) |
| Sex: Female, Male: Female | Participants | 0 |
| Sex: Female, Male: Male | Participants | 3017 |
| Race/Ethnicity, Customized: Race/Ethnicity, Customized – Non-Hispanic White | Participants | 2475 |
| Race/Ethnicity, Customized: Race/Ethnicity, Customized – Black or African American | Participants | 246 |
| Race/Ethnicity, Customized: Race/Ethnicity, Customized – Asian/Pacific Islander | Participants | 76 |
| Race/Ethnicity, Customized: Race/Ethnicity, Customized – Hispanic | Participants | 96 |
| Race/Ethnicity, Customized: Race/Ethnicity, Customized – Other/Unknown | Participants | 124 |
6. Primary Endpoint
The registry lists the primary outcome as Duration of NHT Treatment, measured over 2 years (retrospective data observation duration).
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | 95% Confidence Interval |
| Unit | Months |
| Endpoint classification | Time-to-Event Analysis |
The primary endpoint, Duration of NHT Treatment, is a time-to-event measure expressed in months. It starts at the initial claim date for abiraterone, apalutamide, or enzalutamide and ends when the registry’s definition of treatment observation ends. The listed reasons are the end of data availability, death, a treatment switch, or treatment discontinuation. A 90-day gap in days’ supply counted as discontinuation.
The registry identifies the median as the summary measure and a 95% confidence interval as its measure of uncertainty. A median time-to-event estimate describes the time by which half of the estimated cohort has experienced the defined endpoint, subject to the analysis approach and available follow-up. It is useful when follow-up differs among participants or some participants have not yet experienced the endpoint by the end of observation. Kaplan-Meier estimation was used, according to the record.
Interpretation depends on how the endpoint’s components were treated analytically. For example, the record lists death and end of data availability as reasons observation can end, but does not state whether each reason was counted as an endpoint event or handled as censoring. That distinction affects what the median represents. The registry record does not report the estimated median, the 95% confidence interval, or the number of endpoint events, so the endpoint’s observed result cannot be interpreted from this record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record establishes that the study aimed to describe the duration of NHT use in a retrospective cohort identified through Medicare claims. It lists enrollment of 3017, identifies a single NHT cohort, and specifies a median treatment-duration endpoint estimated using the Kaplan-Meier method. It does not report the median or its 95% confidence interval. Consequently, there is no posted numerical result from which to describe the observed duration or the precision of that estimate.
The outcome definition starts at the initial claim for abiraterone, apalutamide, or enzalutamide and ends at a recorded boundary such as a treatment switch, discontinuation, death, or the end of available data. These boundaries do not all necessarily have the same statistical interpretation. Whether an occurrence is treated as the endpoint event or as censoring can change the estimated time-to-event distribution. The registry record does not explain that classification for each reason.
In general, a 95% confidence interval expresses uncertainty around an estimate under the assumptions of the analysis. A narrower interval indicates greater precision than a wider interval, but an interval does not address every source of uncertainty, including errors in identifying treatment or eligibility in claims data. No interval is reported for this outcome, so its precision cannot be assessed.
This is a descriptive, nonrandomized study. There is no separate comparator group, and the study did not assign treatments. Any observed duration would therefore describe treatment patterns in this eligible claims cohort, not a causal contrast between therapies. The record does not report separate estimates by drug, baseline characteristics, adjusted analyses, or sponsor-reported hazard ratios. It does not support a comparison of the three medicines or a conclusion about why treatment ended.
The stated observation timeframe is 2 years, while the outcome description says retrospective retrieved data were evaluated for approximately 1.2 years. Individual follow-up can also end at the listed observation boundaries. Those details matter when interpreting a median: follow-up that ends before enough endpoint events accrue can limit whether and how the median can be estimated. The registry does not report the event count, follow-up distribution, or whether the median was estimable.
Accordingly, the record supports an account of the research question, cohort framework, and planned statistical summary, but not a conclusion about the actual duration of NHT use. The absent estimate should not be replaced with an inference from enrollment, study dates, or endpoint definition. The registry record does not report the outcome result.
9. Statistical Methods Explained
The outcome is a time-to-event measure, and the registry says Kaplan-Meier estimation was used. The method is designed to estimate the distribution of time until a defined event while accommodating participants whose observation ends before that event is recorded. In general, those participants contribute information during the time they are observed; their later event status is unknown. This handling is commonly called censoring. It relies on assumptions about why observation ends, so the definition and handling of end-of-follow-up reasons are important.
What does a Kaplan-Meier estimate describe?
In general, the Kaplan-Meier method estimates the probability of remaining without the defined event over time, using the event times and the amount of follow-up available at each time. It can account for different observation lengths rather than treating every participant as if they had identical follow-up. The median time-to-event is located where the estimated event-free proportion reaches its midpoint. If follow-up does not extend far enough, or too few events occur, a median may be difficult or impossible to estimate. The registry record does not say whether the median was estimable in this study.
For this endpoint, the event is tied to the recorded end of treatment observation. The record lists the end of available data, death, a treatment switch, and discontinuation as possible endpoints. It does not specify whether all these reasons counted as events in the Kaplan-Meier analysis, or whether some were treated as censoring. This distinction is consequential. If, for example, a reason is censored, the analysis treats the participant’s time after that point as unknown rather than as a recorded treatment-ending event. The registry record does not report the specific event-status rules, so the precise interpretation of the proposed median is limited.
What does the confidence interval add?
In general, a confidence interval describes the statistical uncertainty around an estimated quantity under the analysis model and its assumptions. The endpoint specifies a 95% confidence interval for the median. The interval would indicate how precisely the median was estimated from the observed follow-up, but it would not remove uncertainty from measurement error, selection, or unmeasured differences among participants. Neither the median nor its interval is reported in the registry record.
How should discontinuation be understood?
The registry defines treatment discontinuation using a 90-day gap in days’ supply. This is an operational rule applied to claims data, not necessarily a direct record of a participant’s or clinician’s decision. The definition gives a consistent way to identify a treatment gap from prescription claims, but the registry does not report whether sensitivity analyses used different gap definitions or how changes in claims coverage were handled. Such analyses are not reported in the registry record.
What can claims data measure?
In general, claims record healthcare transactions and coverage-related information. They can support retrospective descriptions of when treatment claims occur, but they may not capture every clinical detail relevant to treatment decisions. A claim date and days’ supply are measurements used to construct the study’s duration endpoint; the record does not report validation results for these measures. Information about the reason for switching or stopping treatment, disease severity, and treatment response is not reported in the registry record.
What does the observational design imply?
Because the treatments were prescribed in routine care and not assigned by the study, the cohort description is not a randomized treatment comparison. In general, treatment patterns can reflect clinical decisions and participant characteristics. Without a comparator and a design intended to control confounding for a causal question, a duration estimate is descriptive rather than an estimate of how a treatment changes duration relative to an alternative. The registry record does not report an adjusted comparative analysis.
The record lists one primary outcome measure and zero analyses. It does not report a formal statistical analysis plan, a sample-size justification, interim analyses, or methods for checking model assumptions. Enrollment of 3017 alone does not reveal the precision of a time-to-event estimate: precision also depends on event information, observation, and the analysis. No outcome estimate or event count is reported, so those features cannot be evaluated for this study.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists one outcome measure, Duration of NHT Treatment, and zero analyses. It does not report a multiplicity adjustment or a plan for controlling false-positive findings across multiple statistical tests. With no reported analyses or outcome estimates, there is no set of posted inferential results from which to assess multiplicity in practice. A formal statistical plan is not reported in the registry record.
The registry describes the analysis population as eligible participants who met inclusion and exclusion criteria and whose data were retrieved and observed. It does not report the number who met each criterion, the number with unavailable or incomplete claims, or how many participants contributed to the endpoint analysis. Enrollment is reported as 3017, but the registry record does not state that all enrolled participants were included in every analysis.
Missing-data methods are not reported in the registry record. In claims-based follow-up, the end of available data is explicitly one of the reasons observation can end; the record does not explain how that circumstance was represented in the analysis. It also does not report methods for handling gaps in insurance coverage, missing covariates, or incomplete treatment records.
The endpoint definition includes death, treatment switching, discontinuation, and the end of data availability, but the statistical treatment of each reason is not described. The record does not report censoring rules beyond the endpoint description, checks of assumptions, sensitivity analyses, or subgroup analyses. As a result, the amount and handling of incomplete observation, and the robustness of any estimate, cannot be assessed from the registry record.
11. What the Result Does, and Does Not, Mean
A treatment-duration estimate from this study would describe what was observed in the specified Medicare claims cohort under the endpoint definition. It would not, on its own, show that one therapy caused longer or shorter treatment duration than another, because the registry describes one observational cohort and no comparator group. Treatment was prescribed in usual care, not randomly assigned by the study.
Duration constructed from claims is not necessarily identical to the length of time a person actually took a medicine. The endpoint uses initial claim dates and days’ supply, with a 90-day gap used to identify discontinuation, and also includes other observation-ending reasons. The registry record does not report validation of the claims-based measure or the reasons behind treatment changes. Thus, a future estimate would need to be read as a measure defined from claims rather than as a complete account of clinical decision-making.
The median and its 95% confidence interval are not reported. There is therefore no posted result to interpret for typical duration, no reported precision to describe, and no basis for stating how often the endpoint occurred. Enrollment, completion status, and the stated observation timeframe do not substitute for an outcome estimate.
The record also does not report separate results for abiraterone, apalutamide, and enzalutamide. It cannot establish relative duration across these medicines, nor does it report a treatment-effect comparison, adjusted or otherwise. In general, retrospective observational findings can be affected by differences among people who receive different care, but this record does not report an analysis that addresses such differences.
The registry record is useful for understanding the study’s objective, eligibility framework, claims-based endpoint, and planned summary method. It is not enough to assess the magnitude or precision of the outcome, the causes of treatment ending, or the comparative performance of therapies. Those conclusions require results and methodological details that are not reported in the registry record.
12. Why This Trial Matters Statistically
Real-world treatment duration is a distinct descriptive question from whether a treatment changes a clinical outcome. This study’s use of Medicare claims focuses on how long eligible people remained on a defined course of NHT in routine care. That makes the endpoint and the rules for identifying treatment gaps central to interpretation: the duration is operationalized from claims dates, days’ supply, and specified observation-ending events.
The statistical approach is also important. Kaplan-Meier estimation is commonly used for time-to-event data where observation lengths differ. Its interpretation depends on a clear definition of the event and appropriate handling of participants whose observation ends without a recorded event. This registry lists several reasons for the end of observation, but does not explain how each was classified in the analysis. The record also does not report the median or its confidence interval, so the result and its precision cannot be assessed.
The study illustrates why observational design and outcome reporting need to be considered together. A cohort drawn from claims can describe treatment patterns, but without a comparator it cannot estimate a causal difference between therapies. Nor can a duration measure alone explain why treatment ended. The registry’s lack of treatment-specific results and outcome estimates leaves these questions unresolved.
The record reports enrollment of 3017 and a completed study, but those details do not establish the informativeness of the time-to-event analysis. The number and timing of endpoint events, the observation available for each participant, and the analysis rules all matter. Because these results are not reported, readers should distinguish the study’s clearly stated descriptive aim from the conclusions that an actual numerical outcome could support.
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 principal limitation for interpreting the posted record is that it does not include the primary outcome estimate. Neither the median treatment duration nor its 95% confidence interval is reported. The record also does not report the number of endpoint events or whether the median was estimable. Therefore, the magnitude and precision of the primary result cannot be assessed.
The registry’s endpoint definition combines several reasons for ending observation: data availability ending on December 31, 2022; death; a treatment switch; and discontinuation, defined using a 90-day gap in days’ supply. The record does not specify how each reason was treated in the time-to-event analysis. If some reasons were treated as censoring and others as events, that distinction would shape the estimand; the classification is not reported.
The study uses retrospective Medicare claims and a single cohort of NHT initiators. The record does not report validation of the claims-based identification of mCSPC, treatment initiation, or discontinuation. It also does not report results by individual drug, a comparator group, or an adjusted comparative analysis. As a result, the record does not support causal or between-treatment conclusions.
Further details relevant to judging the analysis are not reported: participant flow, individual follow-up distribution, missing-data handling, sensitivity analyses, subgroup analyses, and a formal statistical analysis plan. The record gives both a 2 years observation timeframe and an evaluation period of approximately 1.2 years, but does not explain how those summaries relate to individual follow-up. These gaps constrain interpretation of the study’s design and any potential outcome estimate.
14. Related Tutorials
15. Related Statistical Calculators
16. Sources
- ClinicalTrials.gov: A Study to Learn About Novel Hormonal Therapies in People With Metastatic Castration-Sensitive Prostate Cancer (mCSPC) — NCT06072196.
- Registry record last updated 2025-12-31; status last verified 2025-12.
- Results first posted 2025-12-31.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
17. Record Summary
NCT06072196 is a completed, retrospective real-world study sponsored by Pfizer. It used Medicare claims to describe how long eligible people with mCSPC took abiraterone, apalutamide, or enzalutamide in routine care. Enrollment is listed as 3017, and the record specifies one NHT cohort rather than a randomized comparison.
The primary outcome was Duration of NHT Treatment, measured from the initial NHT claim until a defined end of observation. The listed reasons include the end of data availability, death, treatment switching, and discontinuation; a 90-day gap in days’ supply counted as discontinuation. The median and 95% confidence interval were the specified summary, with Kaplan-Meier estimation.
The registry record does not report the estimated median, its confidence interval, event counts, or the classification of each observation-ending reason in the analysis. It also does not report a comparator, drug-specific results, or an adjusted treatment comparison. The record supports understanding the study’s descriptive aim and endpoint definition, but not the magnitude or precision of its outcome.