Sunitinib Plus Prednisone In Patients With Metastatic Castration-Resistant Prostate Cancer After Failure Of Docetaxel Chemotherapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Sunitinib Plus Prednisone In Patients With Metastatic Castration-Resistant Prostate Cancer After Failure Of Docetaxel Chemotherapy (NCT00676650), a randomized phase 3 study. The primary outcome is Overall Survival (OS).

2. Clinical Question

The trial was registered as a multicenter, randomized, double-blind, phase 3 study, with quadruple masking recorded in the registry. Its stated purpose was to compare safety and efficacy in this post-docetaxel population. The primary outcome was time to death, rather than a measure of tumor response or symptoms. Secondary outcomes included progression-free survival, objective response, duration of response, pain severity, prostate cancer-related quality of life, and health-state utility. The posted outcome results include overall survival, progression-free survival, and objective response; results for the other listed outcomes are not reported in the registry record.

3. Trial Design

The registry describes a randomized, parallel-group phase 3 study with quadruple masking. Participants were assigned to sunitinib plus prednisone or placebo plus prednisone. The official study title identifies the trial as multicenter and double-blind. The registry records enrollment of 873 participants and lists the study status as terminated. The reason for termination is not reported in the registry record.

The primary outcome was overall survival in months, measured from randomization to death. For participants who were alive, survival was censored at their last contact. The stated timeframe was baseline up to 32 months. The primary analysis population was the full analysis population, defined as all randomized participants, classified by their initial randomized assignment regardless of whether they received study drug according to the randomization schedule. The registry lists 584 participants in the sunitinib-and-prednisone group and 289 in the placebo-and-prednisone group for this outcome.

Progression-free survival and objective response were also posted. The registry lists six secondary outcomes, including duration of response and patient-reported measures of pain and quality of life. Results for those other listed outcomes are not reported in the registry record. The original sample-size rationale, allocation ratio, detailed treatment schedule, and reason for study termination are not reported in the registry record. Equal allocation appears among the assumptions for a planning illustration, but that is not evidence about the trial's original design.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis (FA) population: all participants who were randomized, with study drug assignment designated according to initial randomization, regaradless of whether participant received study drug according to the randomization schedule.

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 Overall Survival (OS), measured over Baseline up to 32 months.

Overall survival measures the time from randomization until death. In this study, the registry defines survival time as the difference between the date of death and the date of randomization, with one day added, divided by 30.4 to express the result in months. Participants who were alive had their survival time censored at their last contact. Censoring means that the precise time of death is not observed during the available follow-up; the participant contributes information up to the last known contact.

Overall survival is used as a time-to-event endpoint because it directly records the timing of death, rather than relying on an intermediate measure such as tumor response. A median summarizes the time by which half of the analyzed group has experienced the event, subject to the survival analysis and censoring. It does not describe the survival time of every participant, and it does not by itself quantify the relative event rate between groups. The registry also reports a hazard ratio, which compares event rates over follow-up under the time-to-event analysis.

The posted timeframe is baseline up to 32 months. That is the stated outcome timeframe, not a claim that every participant was followed for the same length of time. The record does not report further details about the handling of missing follow-up beyond censoring participants who were alive at last contact.

7. Results Posted in the Registry

• Sunitinib and Prednisone vs Placebo and Prednisone: hazard ratio 0.91 (95% CI 0.76–1.10), reported p = 0.168 (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.

Progression-Free Survival (PFS)

• 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:
  • Sunitinib and Prednisone vs Placebo and Prednisone: hazard ratio 0.72 (95% CI 0.59–0.89), reported p < 0.001 (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 lower for Sunitinib and Prednisone vs Placebo and Prednisone than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percent of Participants With Objective Response (OR)

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Sunitinib and Prednisone: 6.2% (36/584) versus 1.7% (5/289) on control; risk difference 4.4 percentage points (95% CI 2.0 to 6.9), risk ratio 3.56 (1.41–8.98); unadjusted two-proportion z-test p = 0.004.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Sunitinib and Prednisone than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary findings have a generally favorable direction for sunitinib plus prednisone on progression-free survival and objective response. These outcomes address disease progression and tumor response, rather than death from any cause. They therefore provide context about outcomes other than the primary endpoint, but cannot resolve the uncertainty in the overall-survival comparison.

Only progression-free survival and objective response have posted results among the registry's six listed secondary outcomes. Results for duration of response, pain severity, FACT-P, and EQ-5D health-state utility are not reported in the registry record. The record therefore does not permit a complete assessment of the secondary outcome pattern, including patient-reported outcomes.

General statistical explanation: testing multiple outcomes creates more opportunities for a nominally positive finding to occur by chance. Unless a multiplicity strategy is prespecified and accounted for, individual secondary p-values generally carry less confirmatory weight than a single prespecified primary test. The registry record does not report whether the secondary analyses were adjusted for multiplicity or how their order was planned. The posted secondary findings should therefore be viewed as supportive or exploratory rather than as substitutes for the primary result.

The secondary outcomes also differ in how they are measured and analyzed. Progression-free survival is time-to-event, whereas objective response is binary. Their estimates answer distinct questions and should not be combined into a single claim about overall benefit. The posted results describe statistical patterns in this registry record, not medical advice.

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

• Progression-Free Survival (PFS)
• Percent of Participants With Objective Response (OR)
• Duration of Response (DR)
• Change From Baseline in Pain Severity
• Change From Baseline in Functional Assessment of Cancer Therapy-Prostate (FACT-P)
• Change From Baseline in Euro Quality of Life (EQ-5D)- Health State Profile Utility Score

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is a time-to-event measure: the time from randomization until death. Time-to-event analyses can use information from participants who have not experienced the event by the end of their observed follow-up by accounting for their censored follow-up time. Here, the registry states that participants alive at last contact were censored at that contact. General statistical explanation: censoring is informative only up to the last time the participant is known to be alive; it does not establish when a later event occurred.

What does the median describe?

The median overall survival is the point in the estimated survival distribution at which the estimated proportion remaining alive reaches one half. It is a group-level summary, not a claim that every participant lived for that duration. It also does not capture the entire shape of survival over time. Two groups can have medians that appear different while the full time-to-event comparison remains uncertain.

What does the hazard ratio describe?

A hazard ratio compares the estimated event rates between groups over the observed follow-up, using the time-to-event analysis. A value of 1 corresponds to equal estimated rates; a value below 1 indicates a lower estimated rate in the first-named group, and a value above 1 indicates a higher estimated rate. General statistical explanation: a hazard ratio is not a ratio of the median survival times, and it is not a direct estimate of how many participants will experience the event. Its interpretation depends on the time-to-event model and assumptions.

The registry reports a log-rank analysis for overall survival and gives the sponsor-reported hazard ratio with its confidence interval and p-value. The log-rank test compares the observed event experience over follow-up under its null hypothesis. Its result should be interpreted alongside the estimated hazard ratio and confidence interval: the p-value alone does not express the size or precision of the estimated effect.

Why is the confidence interval important?

A confidence interval conveys the statistical uncertainty around an estimate. A narrower interval indicates greater precision under the analysis; a wider interval indicates less precision. For the primary hazard ratio, the interval includes 1, meaning that values consistent with the data include both a lower event rate in the sunitinib group and no difference or a higher rate. The interval should not be read as a range containing a known true value with certainty.

For objective response, the registry reports a sponsor analysis using Fisher's exact test and an odds ratio. The posted recalculation uses an unadjusted risk difference from the arm-level summaries. General statistical explanation: a risk difference contrasts response probabilities, whereas an odds ratio contrasts the odds of response. These are different effect measures, and their associated tests and intervals need not match. The reported sponsor analysis is the registered inferential result; the simple unadjusted recalculation is useful as a separate descriptive comparison, not a replacement for that analysis.

The registry's full analysis population includes everyone randomized, assigned according to the initial randomization regardless of whether study drug was received as scheduled. This preserves the randomized grouping for the primary comparison. The record does not report further details about analysis assumptions, missing-data procedures, or whether any alternative analysis populations were used.

General statistical explanation: over-dispersion matters when observed variation is greater than a statistical model expects; ignoring it can make uncertainty appear too small. The posted primary outcome is time-to-event, and the registry identifies the log-rank method. It does not report an over-dispersion model or information that would support applying one to these results. Therefore, no additional model adjustment can be inferred from the record.

11. Confidence Intervals and Precision

The primary median estimates are 13.1 months for sunitinib plus prednisone and 11.8 months for placebo plus prednisone. Their 95% confidence intervals are 12.0 to 14.1 months and 10.8 to 14.2 months, respectively. These intervals show uncertainty around each median estimate; they are not a direct confidence interval for the difference between medians.

For the primary comparative analysis, the hazard ratio is 0.914 and its 95% confidence interval is 0.762 to 1.097. The interval includes 1, so the estimated relative event rate is not precise enough to rule out no difference. It also spans values on either side of 1, leaving uncertainty about the direction of the comparative effect. The reported log-rank p-value is 0.168. Taken together, the estimate, interval, and test do not provide clear statistical evidence of an overall-survival difference.

Confidence intervals are not guarantees about the true value, nor do they describe the range of outcomes for individual participants. They quantify uncertainty in an estimated group-level quantity under the analysis assumptions. A confidence interval that includes a no-difference value does not prove that the groups are identical; it indicates that the data are compatible with a range of possibilities that includes no difference.

The registry reports no dispersion estimate for the median outcomes. The original sample-size justification and assumptions used for the trial's own design are not reported in the registry record. Thus, the observed precision can be described from the posted intervals, but the record does not establish whether the trial was originally planned to estimate the primary effect to a specified precision.

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 are illustrations based on the observed effects, not the trial's original design calculations. For the time-to-event outcomes, the figures refer to total events, not participants per arm. For the other outcomes, they refer to participants per arm. The calculation assumptions are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. Each planning scenario uses the stated fraction of the observed effect as the effect to detect. The original trial's own sample-size assumptions are not reported in the registry record.

The 80% power column represents a planning scenario with an 80% chance of detecting the specified effect if that effect is the true one and the calculation assumptions hold. The 90% power column represents the corresponding scenario with a higher planned chance of detection. These are design-planning quantities, not guarantees about what a study will find.

Within each outcome, the scenarios consider planning for 100%, 75%, and 50% of the observed effect. Smaller effects require more information to detect, so the illustrations call for more events or participants as the assumed effect becomes smaller. For the time-to-event outcomes, the required information is expressed as events; for the binary outcome, it is expressed as participants per arm. The event figures should not be read as total enrollment, and the per-arm figures should not be read as totals across both groups.

This is not observed power. It is a planning illustration computed from the observed effect and the stated assumptions. It does not measure the probability that the posted result is true or the probability that a future study will succeed.

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 = 0.914, so ln(HR) = -0.0899.
2. d = 4 × (1.96 + 0.842)² / [ln(0.914)]² = 3,882.5, rounded up to 3,883 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 <- 0.914                              # 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(676650)
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 = 3,883 (80% power), 5,198 (90% power)
 75% of observed effect: events = 6,903 (80% power), 9,241 (90% power)
 50% of observed effect: events = 15,530 (80% power), 20,791 (90% power)
Simulated power with 3,883 events: 0.801 (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. Group-Sequential Design Analysis

Many time-to-event trials, especially in oncology, stop or report early when interim data are overwhelming, so a single fixed-sample number understates the real design problem. The calculation below is an illustrative group-sequential design built from this trial's posted control median and hazard ratio. The registry record does not state the planned number of looks or the spending function, so the design is a planning illustration, not a reconstruction of the protocol. Assumed: one-sided α = 0.025, 90% power, three equally spaced looks on information, Lan-DeMets O'Brien-Fleming spending, 24 months of accrual, 60 months in total and a 1% monthly dropout hazard.

Under these assumptions the design needs about 6,497 participants in total, with the boundaries below.

Fixed design versus group-sequential alternatives

Group-sequential designs cost a small increase in maximum sample size and events compared with the fixed design, in exchange for the chance to stop early for overwhelming efficacy. O'Brien-Fleming spending keeps early boundaries very conservative and the final boundary close to the fixed-design value, whereas Pocock spending spends more alpha early.

Sensitivity to a smaller true effect

Planning for exactly the observed effect is optimistic, because observed effects in published trials tend to be somewhat larger than the true effect. The table repeats the main design (three looks, O'Brien-Fleming spending, 90% power) for smaller true hazard ratios.

For context, this trial enrolled 873 participants.

Cross-check, upper boundaries: gsDesign 3.71, 2.511, 1.993 | rpact 3.71, 2.511, 1.993. Agreement between the two packages confirms the boundary calculation.

R code and output

suppressWarnings(suppressMessages({ library(gsDesign); library(rpact) }))
med_control <- 11.8          # control-arm median, months (converted from the posted unit)
hr          <- 0.914          # hazard ratio used for planning
lambdaC     <- log(2) / med_control
eta         <- 0.01               # dropout hazard per month (assumption)
acc_months  <- 24                 # accrual duration (assumption)
study_months<- 60                 # total study duration (assumption)

design <- function(k, sfu, beta, h = hr) gsSurv(k = k, test.type = 1, alpha = 0.025, beta = beta, sfu = sfu,
  lambdaC = lambdaC, hr = h, hr0 = 1, eta = eta, gamma = 1, R = acc_months, T = study_months, minfup = study_months - acc_months)

# Main design: 3 looks, Lan-DeMets O'Brien-Fleming spending, one-sided alpha 0.025, 90% power
x <- design(3, sfLDOF, 0.10)
cat(sprintf("N_TOTAL|%s\n", format(ceiling(sum(x$gamma * x$R)), big.mark = ",")))
for (i in 1:3) cat(sprintf("GS|%d|%.1f|%.0f|%.3f|%.5f|%.5f\n", i, x$T[i], x$n.I[i], x$upper$bound[i], pnorm(-x$upper$bound[i]), cumsum(x$upper$spend)[i]))
cat("\n"); print(gsBoundSummary(x))

# Comparison of designs
cmp <- list(list("Fixed design, one analysis", 1, sfLDOF), list("Two looks, O'Brien-Fleming", 2, sfLDOF),
            list("Three looks, O'Brien-Fleming", 3, sfLDOF), list("Three looks, Pocock", 3, sfLDPocock))
cat("\n")
for (b in c(0.20, 0.10)) for (d in cmp) {
  y <- design(d[[2]], d[[3]], b)
  cat(sprintf("DES|%s|%d|%.0f|%s|%.1f\n", d[[1]], round(100 * (1 - b)), tail(y$n.I, 1), format(ceiling(sum(y$gamma * y$R)), big.mark = ","), tail(y$T, 1)))
}

# Sensitivity: what if the true effect is smaller than the observed one? (3 looks, O'Brien-Fleming, 90% power)
cat("\n")
for (fr in c(1, 0.75, 0.5)) {
  y <- design(3, sfLDOF, 0.10, exp(log(hr) * fr))
  cat(sprintf("SENS|%.2f|%.3f|%.0f|%s\n", fr, exp(log(hr) * fr), tail(y$n.I, 1), format(ceiling(sum(y$gamma * y$R)), big.mark = ",")))
}

# Independent cross-check of the boundaries with rpact
r <- getDesignGroupSequential(kMax = 3, alpha = 0.025, beta = 0.10, sided = 1, typeOfDesign = "asOF", informationRates = x$timing)
cat(sprintf("\nCross-check, upper boundaries: gsDesign %s | rpact %s\n", paste(round(x$upper$bound, 3), collapse = ", "), paste(round(r$criticalValues, 3), collapse = ", ")))

Method: LachinFoulkes 
     Analysis               Value Efficacy
    IA 1: 33%                   Z   3.7103
      N: 4930         p (1-sided)   0.0001
 Events: 1753        ~HR at bound   0.8376
    Month: 18    P(Cross) if HR=1   0.0001
              P(Cross) if HR=0.91   0.0338
    IA 2: 67%                   Z   2.5114
      N: 6498         p (1-sided)   0.0060
 Events: 3506        ~HR at bound   0.9187
    Month: 29    P(Cross) if HR=1   0.0060
              P(Cross) if HR=0.91   0.5603
        Final                   Z   1.9930
      N: 6498         p (1-sided)   0.0231
 Events: 5258        ~HR at bound   0.9465
    Month: 60    P(Cross) if HR=1   0.0250
              P(Cross) if HR=0.91   0.9000



Cross-check, upper boundaries: gsDesign 3.71, 2.511, 1.993 | rpact 3.71, 2.511, 1.993

15. Analyses Reported by the Sponsor

For overall survival, the registry reports the sponsor's log-rank analysis: a hazard ratio of 0.914, a 95% confidence interval from 0.762 to 1.097, and a p-value of 0.168. The time-to-event interpretation uses the sponsor-reported hazard ratio, as the registry record indicates. The median survival values are descriptive summaries and do not substitute for that comparative analysis.

For objective response, the sponsor reports Fisher's exact test with an odds ratio of 3.561, a 95% confidence interval from 1.0 to 19.0, and a p-value of 0.040. A separate unadjusted recalculation from the posted arm summaries expresses the contrast as a risk difference and gives a p-value of 0.004. These analyses use different effect measures and methods, so their estimates and p-values are not interchangeable. The sponsor's reported analysis is the registered inferential result; the recalculation provides a separate, unadjusted summary comparison and should not replace it.

For progression-free survival, the registry likewise reports the sponsor's log-rank hazard ratio and associated interval and p-value. The recalculated descriptive median comparisons do not replace the sponsor's time-to-event analysis. No discrepancy flag is recorded for these results.

16. Multiplicity, Missing Data and Analysis Populations

The registry identifies overall survival as the primary outcome and lists multiple secondary outcomes. It reports analyses for the primary outcome, progression-free survival, and objective response. The original statistical analysis plan, including the planned hierarchy of endpoints and tests, is not reported in the registry record. Whether multiplicity adjustments were planned or applied is also not reported in the registry record.

General statistical explanation: when several endpoints or comparisons are tested, the chance of at least one apparently positive result can increase if each is judged separately at the same threshold. A prespecified hierarchy or adjustment can address this issue, but neither should be assumed without documentation. Accordingly, the secondary results should not be treated as if they were independent confirmatory tests, and the reported p-values should be interpreted in light of the missing multiplicity details.

The registry defines the primary full analysis population as all randomized participants, classified according to initial assignment regardless of whether study drug was received as scheduled. For overall survival, it specifies censoring at last contact for participants who were alive. This provides some information about how ongoing follow-up was handled, but additional missing-data details are not reported in the registry record. The record does not describe the amount or pattern of missing outcome data, how missing secondary measures were handled, or whether sensitivity analyses were performed.

Interim analyses, stopping rules, and any sample-size re-estimation are not reported in the registry record. The trial status is terminated, but the reason for termination is not reported. These gaps mean that the posted results can be described, but the record alone does not establish the full decision process behind stopping or the intended control of statistical error across analyses.

17. 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: disease progression (6.7%), anaemia (3.6%), pulmonary embolism (3.3%), dehydration (2.9%), vomiting (2.8%).

The registry reports serious adverse events for participants at risk in each group. In the sunitinib-and-prednisone group, 275 participants were recorded as affected among 581 at risk; in the placebo-and-prednisone group, 86 were affected among 285 at risk. The registry also lists other adverse events separately, with 573 affected among 581 at risk in the sunitinib group and 256 among 285 at risk in the placebo group. These are registry counts, not adjusted comparisons.

The listed serious adverse event terms include disease progression, anaemia, pulmonary embolism, dehydration, vomiting, and urinary tract infection, each with a reported percentage. The registry does not provide enough detail in this record to infer from those percentages alone how the event terms overlap within participants or to compare their severity across groups. Death counts for the adverse-event display are not reported in the registry record.

These safety figures should be read as descriptive event reporting. They do not by themselves establish that a particular event was caused by study treatment, and the record does not provide a formal comparative safety analysis in the material reported.

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

The primary overall-survival analysis does not provide clear statistical evidence of a difference between sunitinib plus prednisone and placebo plus prednisone. The estimated hazard ratio is close to 1, and its 95% confidence interval includes 1. The median estimates differ, but their uncertainty and the comparative time-to-event analysis need to be considered together. The result does not prove that the regimens are identical; it leaves uncertainty about the magnitude and direction of any overall-survival difference.

The secondary results point toward a difference in progression-free survival and objective response, but these are not the primary outcome. They measure different aspects of disease course and response, and their interpretation is limited by the absence of reported multiplicity procedures. Results for several other listed secondary outcomes are not reported in the registry record.

The registry describes a specific population: participants with progressive metastatic castration-resistant prostate cancer after failure of docetaxel-based chemotherapy. Findings from this record should not be generalized automatically to people with different disease settings or treatment histories. The registry record also does not report the reason the study was terminated, the full statistical analysis plan, or additional missing-data procedures.

These statistical results are not medical advice and do not determine an individual treatment choice. They describe the posted comparisons and the uncertainty around them. The randomized design supports a comparison between assigned groups, while the available registry record leaves important questions about analysis planning and unreported outcomes unresolved.

19. Why This Trial Matters Statistically

This trial illustrates why a primary endpoint should be interpreted through its estimated effect, uncertainty, and analysis method together. The median overall-survival estimates alone do not settle the randomized comparison. The hazard ratio and its confidence interval provide a comparative time-to-event summary, and the interval's inclusion of 1 signals that the estimate remains uncertain about whether the relative event rate differs from no difference.

It also illustrates the distinction between a primary endpoint and secondary outcomes. The posted progression-free survival and objective response findings have a favorable direction for the experimental group, but they do not answer the same question as overall survival. When multiple outcomes are examined, multiplicity plans matter; those plans are not reported in the registry record.

The endpoint-specific methods matter as well. Time-to-event analyses account for event timing and censoring, whereas a binary response comparison concerns whether a response was recorded. Different summary measures and tests can produce different numerical estimates without being direct contradictions. The objective-response sponsor analysis and the unadjusted recalculation, for example, use different effect measures and methods.

Finally, the registry record shows how important reporting completeness is to statistical interpretation. It gives the primary population definition and survival censoring rule, but the original sample-size rationale, interim-analysis details, multiplicity strategy, and reason for termination are not reported. Those omissions limit assessment of the trial's design decisions while leaving the posted results interpretable within their stated scope.

20. 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 does not report why the study was terminated, so the context for termination cannot be assessed from this information. It also does not report the original sample-size justification or the assumptions behind the trial's own design. The planning illustrations based on observed effects should not be mistaken for those original design calculations.

The record does not report whether interim analyses, stopping rules, or sample-size re-estimation were used. It does not specify a multiplicity strategy for the listed secondary outcomes. As a result, the inferential weight of secondary findings cannot be fully judged from the registry record alone.

Overall survival censoring at last contact is described for participants who were alive, but further information about follow-up completeness, missing data, and sensitivity analyses is not reported. Results for several listed secondary outcomes, including patient-reported measures, are also not reported in the registry record. The adverse-event display provides counts and selected terms, but death counts are not reported, and the record does not establish causal attribution for individual events.

Finally, the results apply to the registered post-docetaxel population and the comparison of the two assigned regimens. The registry record does not support extension to other populations or comparisons without additional evidence.

21. Related Tutorials

22. Related Statistical Calculators

23. Sources

• ClinicalTrials.gov: Sunitinib Plus Prednisone In Patients With Metastatic Castration-Resistant Prostate Cancer After Failure Of Docetaxel Chemotherapy — NCT00676650.
• Registry record last updated 2013-03-08; status last verified 2013-02.
• Results first posted 2013-03-08.
• 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.

24. Record Summary

NCT00676650 is a randomized, parallel-group phase 3 study of sunitinib plus prednisone versus placebo plus prednisone in progressive metastatic castration-resistant prostate cancer after failure of docetaxel-based chemotherapy. The registry records enrollment of 873 and study status as terminated. Its primary outcome was overall survival, measured from randomization to death, with censoring at last contact for participants alive.

The sponsor-reported overall-survival hazard ratio was 0.914, with a 95% confidence interval from 0.762 to 1.097 and a log-rank p-value of 0.168. The interval includes 1, leaving the primary comparison uncertain. Posted secondary results include progression-free survival and objective response; results for other listed secondary outcomes are not reported in the registry record. The reason for termination and the original statistical planning details are not reported.