This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.
1. Trial at a Glance
S1216 was a randomized, parallel, open-label phase 3 trial evaluating LHRHa plus TAK-700 versus LHRHa plus bicalutamide in patients with prostate cancer. The primary endpoint was overall survival, defined as the time from random assignment to the date of death from any cause.
| Feature | S1216 |
|---|---|
| Trial name | S1216 |
| NCT identifier | NCT01809691 |
| Phase | Phase 3 |
| Condition | Prostate Cancer |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None |
| Primary purpose | Treatment |
| Enrollment | 1313 |
| Arms | 2 |
| Primary endpoint | Overall Survival |
| Primary endpoint type | Time-to-event |
| Primary hypothesis type | Superiority |
| Lead sponsor | SWOG Cancer Research Network |
| Status | Completed |
| Study start | March 8, 2013 |
| Primary completion | March 8, 2022 |
2. Clinical Question
The primary statistical question was whether the treatment strategy containing TAK-700, when combined with LHRHa, differed from LHRHa plus bicalutamide with respect to overall survival in metastatic prostate cancer. The registry identifies the primary hypothesis as superiority.
Population
Patients enrolled in the phase 3 S1216 study with prostate cancer.
Intervention
LHRHa + TAK-700.
Comparator
LHRHa + Bicalutamide.
Primary question
Does LHRHa + TAK-700 improve overall survival relative to LHRHa + bicalutamide?
3. Trial Design
The registry describes S1216 as a randomized, parallel-group, unmasked phase 3 treatment trial with 1313 participants and two treatment arms. Because the study was unmasked, treatment assignment was not concealed from participants and investigators after randomization.
LHRHa + TAK-700
- LHRHa
- TAK-700
LHRHa + Bicalutamide
- LHRHa
- Bicalutamide
The parallel design means that participants were assigned to one of the two treatment strategies and followed for the trial endpoints. Randomization provides the central design mechanism for comparing the groups, while the subsequent survival analysis accounts for the fact that participants can have different lengths of follow-up and may be censored before experiencing an event.
Study timeline
Study start
The registry records March 8, 2013 as the study start date.
Primary completion
The registry records March 8, 2022 as the primary completion date.
4. Primary Endpoint
| Endpoint | Registry definition | Time frame | Statistical analysis |
|---|---|---|---|
| Overall Survival | Overall survival is defined as the time from random assignment to the date of death from any cause. | Duration of treatment and follow-up until death or 9 years after study start | Cox proportional-hazards model; hazard ratio |
Overall survival is a classic time-to-event endpoint. Each participant contributes follow-up time from random assignment until death or until the observation is censored. The statistical analysis therefore differs from a simple comparison of proportions because both the timing of events and incomplete follow-up are part of the information being analyzed.
5. Secondary Endpoint
| Endpoint | Time frame | Statistical analysis | Effect measure |
|---|---|---|---|
| Progression Free Survival | Duration of treatment and follow-up until death or 9 years after study start | Cox proportional-hazards model | Hazard ratio |
The registry therefore contains two posted statistical analyses: the primary overall-survival analysis and a secondary progression-free-survival analysis. Both compare LHRHa + TAK-700 with LHRHa + bicalutamide using a Cox proportional-hazards model.
6. Statistical Methodology
Time-to-event analysis
Both overall survival and progression-free survival are time-to-event endpoints. The relevant outcome is not simply whether an event occurred, but the amount of time from random assignment until the event or censoring. This structure allows participants with different follow-up durations to contribute information without requiring every participant to experience the event during the observation period.
Cox proportional-hazards model
The registry identifies the statistical method as a Cox proportional-hazards model, reported in the analysis field as regression using Cox regression. The principal effect measure is the hazard ratio.
A hazard ratio below 1 indicates a lower estimated instantaneous event rate in the TAK-700 group relative to the bicalutamide group, under the fitted model.
Hazard ratio
The hazard ratio is a relative measure of the event rate over time. It is not a probability, a percentage of patients who benefit, a relative risk, or an absolute difference in survival at a particular time point.
For example, an HR of 0.86 corresponds to an estimated hazard approximately 14% lower in the TAK-700 group than in the bicalutamide group, because 1 − 0.86 = 0.14. Likewise, an HR of 0.58 corresponds to an estimated hazard approximately 42% lower, because 1 − 0.58 = 0.42.
Two-sided confidence intervals
The posted analyses use 95% two-sided confidence intervals. A confidence interval provides a range of values representing statistical uncertainty around the estimated hazard ratio under the model and sampling framework. It is not a range describing individual patient outcomes.
Superiority hypothesis
The registry identifies the hypothesis type as superiority. In a superiority framework, the treatment comparison asks whether the data support a difference between the randomized groups rather than whether a new treatment is merely no worse than a comparator by a prespecified non-inferiority margin.
What the registry does not report for the primary analysis
The registry record does not report a median overall survival estimate, a median progression-free survival estimate, time-specific survival percentages, event counts for the efficacy endpoints, Kaplan-Meier coordinates, or subgroup-specific hazard ratios in the statistical-analysis fields summarized here. Those quantities therefore cannot be used to characterize the magnitude or timing of survival beyond the posted hazard-ratio analyses.
7. Primary Result: Overall Survival
The posted primary analysis compares LHRHa + TAK-700 with LHRHa + bicalutamide for overall survival using a Cox proportional-hazards model.
Hazard ratio for overall survival
95% CI: 0.72–1.02 · P = .080
Comparison: LHRHa + TAK-700 vs LHRHa + Bicalutamide
| Primary endpoint | TAK-700 vs bicalutamide | 95% CI | P-value | Method |
|---|---|---|---|---|
| Overall Survival | HR 0.86 | 0.72–1.02 | .080 | Cox proportional-hazards model |
An HR of 0.86 means that the fitted Cox model estimates the instantaneous rate of death to be approximately 14% lower with LHRHa + TAK-700 than with LHRHa + bicalutamide over the analyzed follow-up. This is a relative hazard measure, not a statement that 14% fewer patients died or that each individual patient experienced a 14% reduction in risk.
The 95% CI of 0.72–1.02 expresses uncertainty around the estimated hazard ratio. Because the interval extends above 1, the reported interval includes values corresponding to a higher estimated hazard in the TAK-700 group as well as values corresponding to a lower estimated hazard.
The P-value of .080 is a measure of compatibility between the observed data and the statistical null hypothesis under the specified testing framework. It is not a measure of the size, clinical importance, or probability of the treatment effect. The HR and its confidence interval provide the more direct description of the estimated relative effect and its precision.
Because the analysis uses a Cox proportional-hazards model, interpretation of a single HR also depends on the proportional-hazards framework being a reasonable summary of the relative hazards over time. The registry result does not provide enough information here to assess that assumption directly.
Reading the primary result as a statistical estimate
The most informative way to read the result is jointly: the point estimate is below 1, the confidence interval spans 1, and the reported two-sided P-value is .080. These three pieces of information describe different aspects of the same analysis. The point estimate describes the estimated relative effect, the confidence interval describes uncertainty around that estimate, and the P-value addresses the strength of evidence against the null hypothesis under the specified testing framework.
Importantly, the result does not provide a median survival time or a time-specific survival probability. Those measures answer different questions from the hazard ratio and are not interchangeable with it.
8. Secondary Result: Progression Free Survival
The posted secondary analysis evaluates progression-free survival using the same Cox proportional-hazards framework and compares the same two treatment strategies.
Hazard ratio for progression-free survival
95% CI: 0.51–0.67 · P < 0.001
Comparison: LHRHa + TAK-700 vs LHRHa + Bicalutamide
| Secondary endpoint | TAK-700 vs bicalutamide | 95% CI | P-value | Method |
|---|---|---|---|---|
| Progression Free Survival | HR 0.58 | 0.51–0.67 | <0.001 | Cox proportional-hazards model |
An HR of 0.58 means that the fitted model estimates the instantaneous rate of the progression-free-survival event to be approximately 42% lower with LHRHa + TAK-700 than with LHRHa + bicalutamide. This does not mean that 42% of patients avoided progression, nor does it establish a 42% increase in the amount of time each individual patient remained progression-free.
The 95% CI of 0.51–0.67 is relatively narrow around the point estimate compared with the overall-survival interval of 0.72–1.02. It remains entirely below 1, indicating that all values in the reported confidence interval correspond to a lower estimated hazard in the TAK-700 group under the Cox model.
The reported P-value of <0.001 indicates strong evidence against the null hypothesis within the specified statistical framework. It should not be interpreted as the probability that the null hypothesis is true, nor does it quantify the size of the treatment effect. The effect size is communicated by the HR and confidence interval.
As with the overall-survival analysis, the Cox interpretation depends on the proportional-hazards model being a suitable representation of the relative event rates over follow-up. The registry result does not provide sufficient information to evaluate that assumption directly.
9. Comparing the Two Time-to-Event Results
| Endpoint | Hazard ratio | 95% CI | P-value | Interpretive focus |
|---|---|---|---|---|
| Overall Survival | 0.86 | 0.72–1.02 | .080 | Time to death from any cause |
| Progression Free Survival | 0.58 | 0.51–0.67 | <0.001 | Time-to-event disease-control outcome |
The two analyses illustrate why multiple time-to-event endpoints should not be collapsed into a single number. Overall survival and progression-free survival describe different clinical events. An analysis can show a different magnitude of relative effect for one endpoint than for another because the endpoints capture different stages of disease and different event processes.
The PFS hazard ratio of 0.58 is numerically farther below 1 than the OS hazard ratio of 0.86. That difference should be described as a difference between the two estimated endpoint-specific treatment effects; it should not by itself be interpreted as proof that the treatment effect on progression-free survival is statistically different from its effect on overall survival.
10. Serious Adverse Events
The registry reports serious adverse events by treatment arm using affected participants divided by participants at risk.
| Treatment arm | Serious adverse events affected / at risk |
|---|---|
| LHRHa + TAK-700 | 235 / 627 |
| LHRHa + Bicalutamide | 15 / 629 |
These figures describe the reported affected and at-risk counts for serious adverse events. They should be kept separate from the efficacy analyses because safety and efficacy answer different statistical questions and are generally based on different outcome processes.
The reported counts also illustrate why denominators matter. A numerator without its corresponding population at risk would not provide a meaningful estimate of the frequency of an adverse event. Here, the registry supplies both quantities for each arm.
11. Statistical Methods Explained
Why was a Cox proportional-hazards model used?
Overall survival and progression-free survival are time-to-event endpoints. A Cox model is designed to compare event rates over follow-up while accounting for the timing of events and censoring. It produces a hazard ratio that summarizes the relative event rate between treatment groups under the model.
What does an overall-survival HR of 0.86 mean?
It means the fitted model estimates the instantaneous rate of death in the LHRHa + TAK-700 group to be approximately 86% of the corresponding rate in the LHRHa + bicalutamide group. Expressed as a relative reduction, that corresponds to approximately 14% lower estimated hazard. It does not mean that 14% fewer patients died or that survival time increased by 14%.
What does a progression-free-survival HR of 0.58 mean?
It means the fitted model estimates the instantaneous rate of the progression-free-survival event to be approximately 58% of that in the comparator group. The complementary interpretation is approximately 42% lower estimated hazard. The HR does not directly provide a median PFS or a fixed-time PFS probability.
Why does the confidence interval matter?
The confidence interval describes the statistical precision of the estimated hazard ratio. For overall survival, the interval is 0.72–1.02, which is wider in relation to the point estimate and extends across 1. For progression-free survival, the interval is 0.51–0.67 and remains below 1. Neither interval describes the range of effects that individual patients will experience.
Why does the P-value not measure effect size?
A P-value quantifies evidence against a specified null hypothesis within the statistical framework. It is influenced by both the observed effect and the amount of information in the analysis. Consequently, two analyses can have different P-values despite similar effect estimates, or similar P-values despite different effect magnitudes. The HR and confidence interval are needed to describe the estimated effect itself.
What does proportional hazards mean?
The Cox proportional-hazards model assumes that the relative hazard represented by the treatment coefficient is adequately summarized by a constant hazard ratio over the modeled follow-up. This does not require the underlying hazard itself to be constant over time. If the relative hazards change substantially over time, a single HR may become a less complete description of the treatment effect.
Why are overall survival and progression-free survival interpreted separately?
They are distinct time-to-event outcomes. Overall survival counts death from any cause as the event, whereas progression-free survival captures an earlier disease-related event process. Their hazard ratios therefore describe different quantities and should not be treated as interchangeable measures of the same outcome.
12. Confidence Intervals and Statistical Precision
Overall survival
HR 0.86 with a 95% CI of 0.72–1.02. The interval contains values below and above 1, so the reported estimate has uncertainty that extends across the conventional no-difference value.
Progression-free survival
HR 0.58 with a 95% CI of 0.51–0.67. The interval is entirely below 1, indicating that the full reported interval corresponds to lower estimated hazard in the TAK-700 group.
A useful discipline in reading confidence intervals is to distinguish precision from clinical importance. A narrow confidence interval means the statistical estimate is relatively precise; it does not by itself establish that the effect is clinically important. Conversely, a wider interval indicates greater uncertainty but does not prove that an effect is absent.
For the S1216 results, the overall-survival interval and progression-free-survival interval tell different statistical stories. The OS interval extends from 0.72 to 1.02, while the PFS interval extends from 0.51 to 0.67. Those intervals should be read directly rather than inferred from the P-values alone.
13. P-values in Context
| Endpoint | Estimate | 95% CI | P-value | What the P-value addresses |
|---|---|---|---|---|
| Overall Survival | HR 0.86 | 0.72–1.02 | .080 | Evidence against the specified null hypothesis for the superiority comparison |
| Progression Free Survival | HR 0.58 | 0.51–0.67 | <0.001 | Evidence against the specified null hypothesis for the superiority comparison |
The two P-values should not be treated as effect-size rankings. The OS P-value of .080 and PFS P-value of <0.001 reflect the respective statistical tests, while the HRs quantify the estimated relative event rates. The confidence intervals then provide a measure of uncertainty around those estimates.
14. Interpreting Time-to-Event Endpoints
Time-to-event endpoints contain more information than a binary endpoint evaluated at one arbitrary follow-up time. A participant who experiences an event early contributes a different event time from a participant who experiences the event later. A participant who remains event-free through the available observation period contributes censored follow-up information rather than being treated as if the event never occurred.
The survival function describes the probability of remaining free of the event beyond time t. A Cox hazard ratio instead compares the modeled instantaneous event rates between treatment groups.
These are related but different quantities. A hazard ratio does not directly determine a survival probability at a particular time unless additional assumptions and information about the underlying survival distributions are available.
15. Kaplan-Meier Estimation and the Posted Results
Kaplan-Meier estimation is a standard descriptive method for time-to-event data and is naturally relevant to endpoints such as overall survival and progression-free survival. It estimates the survival function while accounting for censoring.
Here, di represents the number of events at an event time and ni represents the number at risk immediately before that time.
The statistical-analysis results posted in the registry are expressed as Cox-model hazard ratios rather than as Kaplan-Meier survival probabilities or median survival estimates. The registry record therefore provides the model-based relative treatment effects summarized above, but not the additional survival-curve quantities needed to characterize the complete shape of the two endpoint distributions.
16. Analysis Populations and Interpretation
The registry information summarized for S1216 identifies the randomized allocation, the treatment arms, the endpoints, and the posted statistical analyses. It does not provide, in the statistical-analysis information summarized here, detailed analysis-population definitions, numbers of events used in each efficacy analysis, censoring rules, or a complete statistical analysis plan.
Those details matter in survival analysis because the estimated hazard ratio depends not only on the treatment assignments but also on the observed event and censoring information and the prespecified analysis conventions.
Randomization remains the fundamental design feature supporting the comparison between the two treatment strategies. Once randomized groups are compared, however, the statistical interpretation of the resulting estimate still depends on how the time-to-event endpoint was defined and modeled.
17. Safety and Efficacy Answer Different Questions
Efficacy
The efficacy analyses estimate relative treatment effects for overall survival and progression-free survival using Cox proportional-hazards models.
Safety
The registry separately reports serious adverse events by arm as affected participants divided by participants at risk.
Different estimands
A hazard ratio for survival and an affected/at-risk safety count summarize different outcome processes and should not be combined into a single numerical treatment-effect measure.
Clinical interpretation
A complete assessment of a randomized trial requires consideration of both efficacy outcomes and adverse-event information rather than relying on a single endpoint.
18. Important Limitations and Interpretation Issues
- Hazard-ratio interpretation: a Cox HR is a model-based relative measure. It should not be translated directly into a difference in median survival, a fixed-time survival difference, or the proportion of patients who benefit.
- Proportional-hazards assumption: the Cox model provides a single HR that is most straightforward to interpret when the relative hazards are reasonably represented by the proportional-hazards framework. The posted registry result does not provide enough information to assess that assumption directly.
- Missing time-specific summaries: the registry information summarized here does not provide median overall survival, median progression-free survival, or fixed-time survival estimates, limiting interpretation of the timing and absolute magnitude of the treatment effect.
- Event information: the registry information summarized here does not provide the number or timing of efficacy events used to generate the posted Cox estimates.
- Analysis-population detail: detailed definitions of the analysis populations are not reported in the statistical-analysis information summarized here, so the exact population contributing to each posted estimate cannot be characterized further.
- Secondary endpoint interpretation: progression-free survival is a secondary endpoint in the posted analyses and should be distinguished from the registered primary endpoint of overall survival.
- Safety interpretation: the reported serious-adverse-event counts should be interpreted separately from the efficacy hazard ratios because they summarize a different outcome domain.
19. Why This Trial Matters Statistically
S1216 provides a useful teaching example of how a randomized oncology trial can be reduced to a small set of statistically meaningful quantities without losing the distinction between endpoint definitions, model estimates, uncertainty, and hypothesis testing.
| Concept | How it appears in S1216 |
|---|---|
| Randomization | The trial uses randomized allocation to two parallel treatment arms. |
| Time-to-event endpoint | Overall survival is the registered primary endpoint and is defined from random assignment to death from any cause. |
| Cox proportional-hazards model | The registry reports Cox regression for both posted efficacy analyses. |
| Hazard ratio | HR is the reported effect measure for overall survival and progression-free survival. |
| Confidence interval | Both posted analyses report two-sided 95% confidence intervals. |
| P-value | The primary OS analysis reports .080; the secondary PFS analysis reports <0.001. |
| Superiority testing | The registry identifies superiority as the hypothesis type. |
| Safety analysis | Serious adverse events are reported separately by treatment arm. |
The trial is particularly useful for demonstrating why a hazard ratio should always be accompanied by its confidence interval and P-value. The point estimate alone does not communicate precision, and the P-value alone does not communicate effect magnitude.
20. A Practical Framework for Reading the S1216 Results
Step 1: Identify the estimand
For overall survival, the event is death from any cause and the time origin is random assignment.
Step 2: Identify the model
The registry reports a Cox proportional-hazards model for the posted efficacy analyses.
Step 3: Read the HR
An HR below 1 corresponds to a lower estimated instantaneous event rate in the TAK-700 group under the model.
Step 4: Read the confidence interval
The interval describes statistical uncertainty around the HR and indicates how precisely the relative effect has been estimated.
Step 5: Read the P-value
The P-value describes evidence against the specified null hypothesis; it is not an effect-size measure.
Step 6: Check what is missing
Without median survival or time-specific survival estimates, the HR should not be used to infer absolute survival differences.
21. Primary Endpoint Interpretation in Context
The primary OS estimate was HR 0.86, with a 95% CI of 0.72–1.02 and P = .080. The point estimate is below 1, but the confidence interval extends across 1. The result therefore needs to be described using all three quantities rather than using the point estimate alone.
The secondary PFS estimate was HR 0.58, with a 95% CI of 0.51–0.67 and P < 0.001. The point estimate indicates a substantially lower modeled event rate in the TAK-700 group, and the full reported confidence interval remains below 1.
Overall survival and progression-free survival provide complementary information. OS captures death from any cause, whereas PFS is a different time-to-event outcome. A difference between their hazard ratios is not, by itself, a statistical test comparing those two treatment effects.
22. What the Hazard Ratios Do — and Do Not — Mean
The two posted estimates can be translated into relative hazard language:
| Endpoint | HR | Direct interpretation | What it does not mean |
|---|---|---|---|
| Overall Survival | 0.86 | Approximately 14% lower estimated instantaneous death hazard under the fitted model. | Not a 14% reduction in the number of deaths, not a 14% increase in survival time, and not a probability of benefit. |
| Progression Free Survival | 0.58 | Approximately 42% lower estimated instantaneous PFS-event hazard under the fitted model. | Not a 42% increase in PFS duration and not a 42-percentage-point improvement in the probability of remaining progression-free. |
This distinction is essential when interpreting oncology survival analyses. The hazard is an instantaneous event-rate concept, while survival probability is a probability of remaining event-free beyond a particular time. The two are mathematically related but are not numerically interchangeable.
23. Related Tutorials
Learn more about the methods used in this trial:
24. Related Calculators
25. Sources
- ClinicalTrials.gov: NCT01809691.
- PubMed: PMID 38980676.
- PubMed: PMID 38424319.
- PubMed: PMID 37526936.
- PubMed: PMID 35446628.
Continue through the Clinical Biostats knowledge graph
Connect trial endpoints and statistical methods to deeper biostatistics tutorials, statistical calculators, and related clinical-trial analyses.
26. Record Summary
S1216 is a randomized phase 3 trial with overall survival as its registered primary endpoint and progression-free survival as a secondary endpoint. Both posted efficacy analyses use Cox proportional-hazards models and report hazard ratios with two-sided 95% confidence intervals. The overall-survival analysis produced an HR of 0.86 (95% CI 0.72–1.02; P = .080), while the progression-free-survival analysis produced an HR of 0.58 (95% CI 0.51–0.67; P < 0.001).
The statistical interpretation depends on keeping several concepts separate: the hazard ratio describes a relative event rate under the Cox model; the confidence interval describes uncertainty around that estimate; and the P-value describes evidence against a specified null hypothesis. None of these quantities by itself provides a median survival time or an absolute survival probability at a particular time.
The trial also illustrates the importance of distinguishing primary and secondary endpoints and of keeping efficacy and safety analyses conceptually separate. The registry reports serious adverse events as 235/627 for LHRHa + TAK-700 and 15/629 for LHRHa + bicalutamide, while the efficacy analyses use time-to-event modeling. These measures answer different statistical questions and should be interpreted accordingly.