Vaccine Therapy in Treating Patients With Metastatic Prostate Cancer That Has Not Responded to Hormone Therapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Vaccine Therapy in Treating Patients With Metastatic Prostate Cancer That Has Not Responded to Hormone Therapy (NCT00005947), a randomized phase 3 study. The primary outcome is Time to Objective Disease Progression.

2. Clinical Question

The trial’s stated purpose was to determine the effectiveness of vaccine therapy in patients with metastatic prostate cancer that had not responded to hormone therapy. Its primary endpoint was time to objective disease progression, summarized as a median and compared using a log-rank analysis. Overall survival was posted as a secondary outcome. The registry record does not report the criteria used to define objective progression, nor does it give a fuller description of the treatment schedules in the two arms. The primary question is therefore best understood as a comparison of the posted time-to-event outcomes between the randomized groups, rather than as an assessment of every possible clinical outcome or of effects in other populations.

3. Trial Design

This was a randomized phase 3 trial with a placebo comparator. The registry record lists 127 participants enrolled, with 82 assigned to sipuleucel-T and 45 assigned to placebo in the posted primary and secondary outcome records. The trial status is completed. The registry lists a start of 1999-11 and a primary completion of 2004-09.

The study is described as double blind and placebo controlled in its official title; the registry’s masking field specifies quadruple masking. The record does not explain which roles were masked, how masking was maintained, or whether it was assessed during the trial. The sponsor was Dendreon. The interventions are listed as sipuleucel-T and placebo, while arm descriptions are blank, so further details of administration are not reported in the registry record.

The primary analysis concerns all randomized participants and measures time to objective disease progression over 36 months from randomization. Overall survival is the posted secondary outcome, with a timeframe from randomization to 36 months. The registry reports medians, 95% confidence intervals, hazard ratios, and log-rank p-values for both outcomes. It does not report a fuller analysis plan, allocation ratio rationale, eligibility criteria beyond the stated disease description, or the trial’s original sample-size justification. The planning calculations associated with the posted results are illustrations based on observed effects; they do not establish what assumptions were used to design the original trial.

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: Participants were screened for evaluation of subject eligibility and performance of baseline tests/procedures.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: all randomized participants

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 Time to Objective Disease Progression, measured over 36 months from randomization.

The primary endpoint is time to objective disease progression, measured in weeks. It represents the elapsed time from randomization until a participant’s disease meets the trial’s definition of objective progression. The registry gives a timeframe of 36 months from randomization and identifies all randomized participants as the analysis population. The registry record does not report the operational criteria for progression, the assessment schedule, or how an event was distinguished from a participant whose progression was not observed during follow-up.

Time-to-event endpoints are useful when the timing of an event matters and participants may have different lengths of observation. A median gives a time by which half of the observed event-time distribution has reached the event, subject to the analysis method and follow-up data. A hazard ratio compares event rates over follow-up under the model or test framework used; it is not itself a difference in median times. The log-rank method provides a group comparison that uses event timing across follow-up rather than reducing the entire comparison to a single time point.

The registry reports median time to progression of 11.7 weeks for sipuleucel-T, with a 95% confidence interval of 9.1 to 16.6, and 10.0 weeks for placebo, with a 95% confidence interval of 8.7 to 13.1. These summaries describe the posted results; they do not, by themselves, establish the meaning or direction of the hazard-ratio comparison.

7. Results Posted in the Registry

• Sipuleucel-T vs Placebo: hazard ratio 1.45 (95% CI 0.99–2.11), reported p = 0.052 (as reported in the registry).
• Sipuleucel-T vs Placebo: hazard ratio 0.69 (95% CI 0.47–1.01), reported p = 0.052 (as reported in the registry).

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Overall Survival

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor (Log Rank); it was not recalculated.
• Result:
  • Sipuleucel-T vs Placebo: hazard ratio 1.71 (95% CI 1.13–2.58), reported p = 0.010 (as reported in the registry).
  • Sipuleucel-T vs Placebo: hazard ratio 0.59 (95% CI 0.39–0.88), reported p = 0.010 (as reported in the registry).
• Interpretation:
  • The 95% confidence interval excludes no difference (a ratio of 1), so the hazard of the event is statistically higher for Sipuleucel-T vs Placebo than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference (a ratio of 1), so the hazard of the event is statistically lower for Sipuleucel-T vs Placebo than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Overall survival is the only secondary outcome posted in the registry record. Its group medians and log-rank comparison are reported separately from the primary time-to-progression outcome. The overall-survival analysis has a different pattern of reported uncertainty from the primary analysis: its hazard-ratio confidence intervals do not include 1, whereas both listed primary hazard-ratio intervals include 1. The record also reports different p-values for the two outcomes. These are observations about the posted analyses, not a basis for treating the secondary endpoint as a replacement for the primary question.

Secondary outcomes can add context by examining a different aspect of the trial, but their evidential weight depends on how they were selected and analyzed. When multiple outcomes or comparisons are tested, the chance of observing at least one apparently noteworthy result can increase. A multiplicity strategy can address that issue, but the registry record does not report whether adjustments were planned or used. It also does not describe whether overall survival was prespecified as a key confirmatory endpoint, how missing survival information was handled, or whether the analysis was adjusted for other factors.

Accordingly, the overall-survival result should be considered alongside the primary result and the limitations of the available analysis record. The secondary finding can be informative, but a nominal p-value by itself does not establish confirmatory evidence when the multiplicity plan and analysis hierarchy are not reported. Nor does the registry’s summary provide enough information to reconcile the differing patterns between the endpoints. The best-supported reading is to keep the outcomes distinct, preserve their reported uncertainty, and avoid making a broader claim than the posted analyses allow.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Overall Survival

9. Statistical Interpretation

10. Statistical Methods Explained

Both outcomes are time-to-event measures: the primary endpoint is time to objective disease progression, and the secondary endpoint is overall survival. For each, the registry reports a median for each group and a log-rank comparison. This combination gives a summary of event timing within each group and a statistical comparison that takes account of event times across follow-up. It does not provide a complete account of how follow-up, censoring, or missing outcome information was handled.

What does a median time summarize?

A median is the point in the estimated event-time distribution at which half the distribution lies on either side. For a time-to-event outcome, it is not the average time to an event and does not describe every participant’s experience. It also does not directly express the contrast between groups. For progression, the posted medians are 11.7 weeks for sipuleucel-T and 10.0 weeks for placebo. For overall survival, they are 25.9 months and 21.4 months, respectively. The registry reports confidence intervals for each median, which indicate uncertainty around those estimates.

What does the log-rank method compare?

The log-rank method compares the event experience of groups over follow-up, using the timing of events rather than only comparing the medians. It is commonly used for time-to-event outcomes. Its p-value measures how compatible the observed comparison is with a null hypothesis under the test’s assumptions; it is not the probability that the null hypothesis is true, nor does it quantify the size of a treatment effect. The record reports p = 0.052 for time to progression and p = 0.010 for overall survival. The details of the test implementation and its assumptions are not reported in the registry record.

How should the hazard ratios be read?

A hazard ratio expresses a relative comparison of event rates over time, in the direction determined by which group is the numerator and which is the reference. A value of 1 represents no relative difference under that parameterization. A value above or below 1 can point in opposite directions depending on the reference coding. The registry lists both orientations for each outcome: 1.45 and 0.69 for progression, and 1.71 and 0.586 for overall survival. It attaches the same p-value to the two orientations within each outcome but does not state which group is the reference in the displayed analyses. The medians give a separate, directly labeled group summary; the hazard ratios should not be interpreted without regard to their orientation.

In general, a hazard ratio is not the ratio of the two medians and should not be read as a difference in median time. It summarizes a relative event-rate comparison under a time-to-event analysis. If the relative event rate changes over follow-up, a single hazard ratio may not capture that changing pattern. The registry record does not describe whether proportional hazards were assessed, what model or estimation procedure was used for the reported hazard ratios, or whether additional modeling adjustments were made. The reported method for the comparison is log rank.

What do the confidence intervals add?

A confidence interval gives a range of values compatible with the observed data under the statistical procedure used. It conveys uncertainty and precision in a way that a point estimate alone cannot. For the primary hazard ratio of 1.45, the 95% interval is 0.99 to 2.11; for the other orientation, it is 0.47 to 1.01. Both intervals include 1. For overall survival, the interval for 1.71 is 1.13 to 2.58, and the interval for 0.586 is 0.388 to 0.884; neither includes 1. The interpretation of these intervals still depends on the orientation of the ratio and on the analysis plan, which the record does not fully specify.

Confidence intervals are not guarantees that the true value lies within the particular interval, and they do not indicate the probability that a result will be reproduced. Their width reflects the amount of information and variability in the analysis. The group sizes and the event counts used for these reported outcome analyses are not fully documented in the registry record; enrollment counts alone do not reveal how many events contributed to each time-to-event comparison.

Why can over-dispersion matter in general?

General statistical explanation: over-dispersion occurs when variability in data is greater than a model assumes. If a method treats observations as less variable or more independent than they are, standard errors and confidence intervals may be too small, making results appear more precise than warranted. The registry record does not identify an over-dispersion issue or report a method addressing one; this concept is included as general background, not as a claim about this trial.

The registry also includes planning illustrations based on the observed effects. Those calculations use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation, with the effect to detect set to a stated fraction of the observed effect. For time-to-event outcomes, the planning quantities are total events. The registry does not report that these were the design assumptions for the original trial. They should be understood as planning scenarios, not as evidence about the trial’s original statistical plan or the information actually available for its analyses.

11. Confidence Intervals and Precision

The 95% confidence intervals show substantial uncertainty around the posted time-to-event estimates. For time to objective disease progression, the median estimate is 11.7 weeks for sipuleucel-T, with an interval from 9.1 to 16.6, and 10.0 weeks for placebo, with an interval from 8.7 to 13.1. These intervals overlap. Overlap alone is not a formal test of the group comparison; the registry reports the log-rank p-value and hazard-ratio intervals for that purpose.

The primary hazard-ratio intervals extend across 1 in both listed orientations: 0.99 to 2.11 for the 1.45 estimate and 0.47 to 1.01 for the 0.69 estimate. Thus, the posted interval estimates include the no-relative-difference value. The p-value is 0.052. These results do not support a highly precise estimate of the primary comparison, and the registry’s lack of an explicit reference-group explanation makes the direction of the hazard ratio less straightforward to communicate.

For overall survival, the medians are 25.9 months and 21.4 months, with confidence intervals of 20.0 to 32.4 and 12.3 to 25.8. The hazard-ratio intervals are 1.13 to 2.58 and 0.388 to 0.884, each excluding 1 in its displayed orientation. The p-value is 0.010. These intervals still cover a range of possible relative effects, so the point estimate should not be treated as exact. The registry does not report event counts for these analyses, a detailed account of censoring, or further precision measures that would clarify the information available for each comparison.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures for these time-to-event comparisons are total-event targets, not participants per arm. The calculations are described as illustrations based on the observed effect, not as the trial’s own original design assumptions. They use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The trial’s original sample-size justification and design assumptions are not reported in the registry record.

The 80% power column describes the total number of events in the planning calculation for a design with an 80% chance of detecting the specified effect under its assumptions. The 90% power column describes the corresponding event target for a design with a 90% chance. These are planning probabilities under a hypothetical design, not probabilities that the observed result is correct. For a given outcome, the planning rows consider 100%, 75%, and 50% of the observed effect. The event targets increase as the effect being targeted becomes smaller: detecting a smaller effect generally requires more information, here represented by more total events.

For the primary outcome, the posted planning illustrations include 228 total events for 80% power and 305 for 90% power when using 100% of the observed effect in one listed orientation. The corresponding entries for 75% are 405 and 542, and for 50% are 910 and 1,218. The alternate orientation has its own listed values. The overall-survival planning figures are separate, because they are based on that outcome’s observed effect. These illustrations do not report how many events occurred in the trial and should not be confused with observed power.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

d is the total number of events needed (Schoenfeld's formula) with 1:1 allocation and proportional hazards.

Worked calculation for this trial

1. Hazard ratio HR = 1.450, so ln(HR) = 0.3716.
2. d = 4 × (1.96 + 0.842)² / [ln(1.450)]² = 227.4, rounded up to 228 events in total for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
hr <- 1.45                              # observed hazard ratio
events_needed <- function(power, frac = 1) ceiling(round(4 * (z_a + qnorm(power))^2 / log(hr^frac)^2, 6))   # Schoenfeld, 1:1 allocation
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: events = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(events_needed(0.80, f), big.mark = ","), format(events_needed(0.90, f), big.mark = ",")))

# Simulation: given d total events under proportional hazards, events in the treatment arm are Binomial(d, hr / (1 + hr)).
# The score (log-rank type) statistic is (D1 - d / 2) / sqrt(d / 4).
set.seed(5947)
d    <- events_needed(0.80)
reps <- 20000
D1 <- rbinom(reps, d, hr / (1 + hr))
z  <- (D1 - d / 2) / sqrt(d / 4)
cat(sprintf("Simulated power with %s events: %.3f (%d replicates)\n", format(d, big.mark = ","), mean(abs(z) > z_a), reps))

Output from running the code

100% of observed effect: events = 228 (80% power), 305 (90% power)
 75% of observed effect: events = 405 (80% power), 542 (90% power)
 50% of observed effect: events = 910 (80% power), 1,218 (90% power)
Simulated power with 228 events: 0.807 (20000 replicates)

The simulation uses the fact that, with d events in total and proportional hazards, the number of events in the treatment arm is binomial. It is an approximation that conditions on the total number of events, and it should land close to the target power.

14. Analyses Reported by the Sponsor

The registry posts sponsor-reported log-rank analyses for both outcomes. For the primary endpoint, it reports hazard ratios of 1.45 and 0.69, each with p = 0.052. For overall survival, it reports 1.71 and 0.586, each with p = 0.010. Each pair gives two orientations of the same comparison, while the record does not explain the reference-group coding. The reported results include 95% confidence intervals for both orientations.

The statistical display uses the sponsor’s reported hazard ratios for interpretation rather than substituting a newly calculated hazard ratio from summary statistics. There is no discrepancy flag in the record indicating that a recalculation conflicts with the sponsor’s analysis. A simple recalculation from posted summary values would not replace the sponsor’s time-to-event analysis, which uses information beyond the group medians. The sponsor’s own reported analysis therefore remains the relevant source for the hazard-ratio comparison.

The record does not provide further details about the sponsor’s model specification, reference coding, or any adjusted analyses. The paired hazard ratios should not be treated as separate trial findings; the record reports the same p-value for both orientations of each outcome. Interpretation should retain that ambiguity rather than silently selecting one direction.

15. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary endpoint and one secondary outcome, but it does not report a multiplicity plan. In general, testing several outcomes or conducting multiple analyses can increase the chance of a nominally notable result arising by chance. Methods such as prespecifying an outcome hierarchy or adjusting the testing procedure can address this issue. Whether such methods were planned or used in this trial is not reported in the registry record. The overall-survival p-value should therefore not be assumed to have been adjusted, or to have retained a particular confirmatory status, without that information.

The record identifies all randomized participants as the primary analysis population. It does not report whether the sponsor analyzed the primary endpoint strictly according to randomized assignment, whether exclusions occurred, or how participants without an observed progression event were treated. Details about censoring rules, withdrawals, losses to follow-up, and missing outcome data are not reported in the registry record. These details matter in time-to-event analyses because a participant may not have an observed event during the available follow-up; the treatment of that observation affects the analysis.

The record gives a 36-month timeframe for the primary endpoint and for overall survival, but does not explain how events after that period were handled or how follow-up completeness was assessed. It does not report interim analyses, stopping rules, a statistical analysis plan, or the original sample-size justification. None of these omissions establishes that a procedure was or was not used; they mean the registry record does not document it.

The posted results do not describe adjustments for baseline characteristics or other covariates, and the record does not state whether the hazard ratios are adjusted or unadjusted. It also does not give a detailed account of the log-rank implementation or any sensitivity analyses. These limits make it difficult to assess the robustness of the reported results beyond their point estimates, confidence intervals, and p-values. The results are most appropriately interpreted as the analyses documented in the registry, with the unreported analysis details kept explicit.

16. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: haematuria (4.4%), urinary retention (4.4%), dyspnoea (3.7%), cerebrovascular accident (2.4%), chills (2.4%).

The serious adverse event record lists 22 affected participants among 82 at risk in the sipuleucel-T group and 8 among 45 at risk in the placebo group. The record also lists other adverse events for 82 of 82 at risk in the sipuleucel-T group and 44 of 45 at risk in the placebo group. These are counts as posted; no new proportions are calculated. The registry does not provide further detail about the definitions or collection procedures for these categories.

The listed serious adverse event terms and percentages include haematuria, 4.4%; urinary retention, 4.4%; dyspnoea, 3.7%; cerebrovascular accident, 2.4%; chills, 2.4%; and pyrexia, 2.4%. These are reported terms, not a complete narrative of adverse events or a comparison adjusted for follow-up or exposure. The record does not provide the term-specific event counts, clarify whether participants could contribute more than one event, or describe how seriousness was determined. Death counts are not reported in the registry record. These safety entries should be read as registry reporting, not as a comprehensive evaluation of comparative safety.

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

The registry documents a randomized comparison of sipuleucel-T and placebo in asymptomatic metastatic hormone-refractory prostate cancer. The primary endpoint was time to objective disease progression, and the posted medians were 11.7 weeks and 10.0 weeks. The reported log-rank p-value was 0.052, with both hazard-ratio confidence intervals including 1. This indicates that the primary comparison is uncertain in the posted analysis; it does not establish the absence of an effect or prove that the groups are equivalent.

The secondary overall-survival analysis has a different statistical pattern: the posted medians were 25.9 months and 21.4 months, the log-rank p-value was 0.010, and the two listed hazard-ratio intervals exclude 1. This is a result reported for a secondary endpoint. Without a documented multiplicity plan, analysis hierarchy, and complete statistical methods, it should not be treated as conclusive evidence on its own. The hazard-ratio orientation is also not explained in the registry record, so claims about direction should be cautious.

Neither p-values nor confidence intervals establish clinical importance. The record does not report a threshold for a clinically meaningful difference, and medians do not describe every participant’s experience. The results apply to the randomized study population and the outcomes and follow-up described in the registry; they do not, by themselves, answer questions about other populations, other treatment settings, or outcomes not posted.

This statistical account is not medical advice. It describes the posted estimates and the uncertainty around them, while noting important details the registry does not report. A trial result is one component of a broader evidence base and should not be converted into a treatment recommendation from these registry statistics alone.

18. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates why the primary endpoint, secondary outcomes, and analysis details need to be read together. Time to objective disease progression is summarized through group medians and a log-rank comparison, while the registry also gives hazard ratios and confidence intervals. The median summaries describe event timing in each randomized group; the log-rank result compares the event experience across follow-up. These measures answer related but different questions.

The primary comparison has a p-value of 0.052 and hazard-ratio intervals that include 1. That pattern demonstrates the value of reporting estimates and confidence intervals rather than relying on a thresholded p-value alone. It also shows why the direction of a hazard ratio should be explicit: the registry lists both orientations for each comparison but does not identify the reference coding. Clear reporting of the analysis contrast is necessary for readers to interpret the numerical estimate correctly.

The posted overall-survival result has a different pattern from the primary endpoint. Considering both outcomes raises the importance of knowing whether the secondary analysis was adjusted for multiple testing or designated as confirmatory. Those details are not reported in the registry record. This is a general lesson in clinical-trial interpretation: a secondary result may add context, but its role depends on prespecification and the statistical plan, not solely on its p-value.

Finally, the registry’s planning illustrations distinguish total event targets from participant counts and show how smaller effects require more information. They are not the original design justification. Keeping that distinction clear prevents planning scenarios from being mistaken for trial conduct or observed evidence.

19. 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 trial-specific details needed for a fuller assessment are not reported in the registry record. The record does not give the operational definition of objective disease progression, the schedule of disease assessments, or the censoring rules for participants without an observed progression event. Although the primary timeframe is 36 months from randomization, the registry does not explain follow-up completeness or the handling of incomplete observations.

The hazard-ratio entries are reported in two orientations for each outcome, but the comparison reference coding is not specified. This limits straightforward interpretation of direction. The record also does not provide a detailed statistical analysis plan, the model specification, adjusted estimates, sensitivity analyses, or information about whether proportional hazards were assessed. The log-rank method is named, but its implementation is not described.

For overall survival, the registry does not report whether the endpoint was part of a confirmatory testing hierarchy or whether any multiplicity adjustment was applied. Interim analyses and stopping rules are not reported. The trial’s original sample-size justification and design assumptions are also not reported; the posted planning illustrations are based on observed effects and should not be read as the original design.

The safety record gives participant counts and selected serious adverse event terms, but not detailed event counts by term, exposure-adjusted measures, or the adverse-event collection procedures. Death counts are not reported in the registry record. These omissions constrain interpretation of both efficacy and safety summaries and do not establish that the unreported procedures or data were absent from the trial itself.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Vaccine Therapy in Treating Patients With Metastatic Prostate Cancer That Has Not Responded to Hormone Therapy — NCT00005947.
• Registry record last updated 2010-11-01; status last verified 2010-10.
• Results first posted 2010-11-01.
• 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.

23. Record Summary

The registry describes a completed randomized, placebo-controlled phase 3 trial of sipuleucel-T in asymptomatic metastatic hormone-refractory prostate cancer. It lists 127 enrolled participants and reports quadruple masking. The primary endpoint was time to objective disease progression over 36 months from randomization; overall survival was the posted secondary outcome.

The primary log-rank comparison has p = 0.052, with hazard ratios listed in both orientations and confidence intervals that include 1. The secondary overall-survival comparison has p = 0.010, also with both hazard-ratio orientations listed. The registry does not explain the reference coding or report a multiplicity plan. These results are best read as the posted analyses, with uncertainty and reporting limitations kept in view.