Efavirenz in Treating Patients With Metastatic Prostate Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efavirenz in Treating Patients With Metastatic Prostate Cancer (NCT00964002), a na phase 2 study. The primary outcome is Percentage of Participants Without Prostate-specific Antigen Progression at 3 Months.

2. Clinical Question

The trial addresses an activity question in a specified disease setting: whether participants receiving efavirenz met the registry’s definition of PSA non-progression at the stated assessment time. It does not, by design, estimate a randomized treatment contrast. Any observed rate would describe outcomes in this single group under the trial’s eligibility, assessment, and follow-up procedures; without a concurrent comparator, it would not establish what would have happened to comparable participants under another treatment or no study treatment.

The registry also names percentage without PSA progression at 6 months as a secondary outcome. Its result is not reported in the registry record. Both endpoints concern PSA progression rather than a direct comparison of treatments, and neither result can be inferred from the study title, rationale, enrollment, or safety information.

3. Trial Design

This was a phase II, single-group study of efavirenz in patients with metastatic prostate cancer. The sponsor was Institut Bergonié. The registry lists a start date of 2008-05, primary completion on 2011-09-19, and a completed status. Enrollment is recorded as 61. These details describe the study’s conduct and scope; they do not reveal how many participants contributed to the primary endpoint analysis.

The intervention was efavirenz. The arm description specifies 600 mg daily as oral tablets at bedtime and in a fast condition, described as 1-2 hours far from dinner. Treatment was to continue until objective biological, radiological or clinical disease progression or study discontinuation. The registry record does not report additional details about treatment exposure, adherence, dose changes, or the handling of participants who stopped treatment.

Allocation is listed as not applicable, masking as none, and the model as single group. There is consequently no randomized assignment or masked treatment comparison described in the record. These design features matter when interpreting outcomes: an endpoint rate can characterize the participants observed under the study conditions, but the design does not isolate the treatment effect from other explanations for the observed course.

The primary endpoint is the percentage without PSA progression at 3 months. The registry defines progression using PSA change from baseline or nadir, depending on whether PSA had decreased, and requires a specified absolute increase and confirmation. Participants who died within the first 3 months were to be considered progressions for that assessment. PSA assessment was scheduled every 28 days in the first 6 months after inclusion, then every 3 months until progression or end of treatment. The registry record lists 0 analyses and does not report a statistical analysis plan, analysis population, or primary endpoint result.

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:

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 Percentage of Participants Without Prostate-specific Antigen Progression at 3 Months, measured over 3 months.

The primary endpoint is the percentage of participants without prostate-specific antigen progression at 3 months. PSA is a laboratory measure used in monitoring prostate cancer. In this trial, the endpoint classifies participants according to whether they met the registry’s defined PSA progression criteria by the assessment time. It is a binary outcome at that time: each participant is treated as having either progression or no progression under the stated definition.

The progression definition distinguishes participants whose PSA did not decrease from those whose PSA decreased but did not meet response criteria. In the first case, progression requires a 25% increase over baseline and an absolute PSA increase of at least 5 ng/mL, confirmed by a second value. In the second, progression is defined by a 25% increase over the nadir, again with an increase of at least 5 ng/mL and confirmation. Death within the first 3 months is counted as progression for this evaluation. These rules specify how a participant’s endpoint status is assigned; they are not an endpoint result.

In general, a percentage endpoint makes the frequency of a defined event or status interpretable as a proportion of the participants assessed. Its meaning depends on the event definition, assessment schedule, and population included in the calculation. The registry’s specified PSA schedule and confirmation rules are therefore part of the endpoint’s interpretation, not administrative details that can be ignored.

The registry record does not report the observed percentage without progression, the number classified in either category, or the 95% confidence interval. As a result, the endpoint’s observed level and precision cannot be evaluated from the record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies a binary percentage endpoint and gives a 95% confidence interval as its dispersion measure. It does not report an estimate, interval, p-value, or analysis procedure. The following statistical concepts are general explanations of how such outcomes are commonly interpreted; they should not be mistaken for the methods actually used in this trial.

What does a percentage endpoint estimate?

For a binary endpoint, each participant is classified into a defined outcome category at the specified assessment time. The percentage without PSA progression would describe the share of participants in that category among the analysis population. The interpretation depends on which participants form that population and how participants without an observed assessment are treated. The registry record defines PSA progression and specifies that deaths within the first 3 months count as progressions, but it does not report the endpoint counts, denominator, or analysis population.

A percentage is descriptive unless it is tied to a clearly defined population and analysis. It does not by itself show why participants had the observed outcomes. In a single-group study, the rate may describe outcomes among participants receiving the study intervention, but it cannot isolate the effect of the intervention from the natural course of disease, participant characteristics, or other factors. This distinction remains important even when an endpoint has a precise operational definition.

What does a confidence interval mean?

In general, a confidence interval expresses statistical uncertainty around an estimate under a specified sampling and analysis framework. For this endpoint, the registry names a 95% confidence interval. The interval would ordinarily be read alongside the estimated percentage: it conveys a range of values compatible with the data and method, rather than guaranteeing that the true value lies within a particular range. The interval’s width is influenced by the amount of information in the analysis and by the method used to calculate it.

Precision and validity are different considerations. A narrow interval can indicate limited sampling uncertainty, but it cannot correct biased participant selection, incomplete outcome ascertainment, or an endpoint definition that does not answer the intended clinical question. An interval around a single-group rate also does not create a treatment comparison. No interval or point estimate is reported in the registry record, so neither the magnitude nor precision of the primary endpoint can be evaluated.

Why does the endpoint definition matter?

The PSA progression rule uses a relative increase as well as an absolute increase of at least 5 ng/mL and requires confirmation by a second value. The rule also treats participants differently depending on whether PSA had decreased. This approach makes classification depend on the participant’s prior PSA course, the timing of measurements, and whether a confirming value is available. The registry says that PSA assessments were to occur every 28 days in the first 6 months after inclusion and every 3 months thereafter until progression or end of treatment. It also specifies use of the same local laboratory and PSA assay for eligibility assessment.

These details support consistent classification in principle, but the record does not say how often assessments were actually completed, whether the planned laboratory procedures were followed, or how unconfirmed or unavailable measurements were handled. The endpoint description is therefore more complete than the available information about outcome ascertainment.

What can the design establish?

The trial is recorded as single group, with allocation not applicable and no masking. In general, a randomized comparator helps distinguish changes associated with treatment from changes that could occur for other reasons. A single-group study lacks that direct contrast. Its descriptive result, if reported, would answer what was observed in the enrolled group under the study conditions, not the causal difference between efavirenz and an alternative.

Masking can reduce the influence of expectations on some study procedures or outcome assessments. The record lists masking as none. PSA is a laboratory measure, but the absence of masking remains part of the trial context; the registry does not describe procedures used to minimize other forms of measurement or analysis bias.

How should planning calculations be read?

The planning assumptions described for illustrative sample-size calculations use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The target effect is framed as a stated fraction of the observed effect. These are planning illustrations, not observed power and not a description of the actual single-group trial’s design. The trial itself has no two-arm allocation in the registry record. The primary endpoint result and observed effect are not reported, so numerical claims about the trial’s endpoint or the implications of an effect-based planning illustration cannot be grounded in the registry record.

In general, a sample-size calculation estimates how much information a future design might need under chosen assumptions. It is not evidence that a study had a particular probability of detecting an effect after the results are known. Its relevance depends on the target effect, variability assumptions, outcome definition, analysis method, and design. The registry record does not report the trial’s own sample-size justification or statistical plan.

Why can over-dispersion matter?

In general, over-dispersion occurs when observed variability exceeds what a statistical model assumes. For binary data, this may arise when outcomes are more heterogeneous than a simple model allows, potentially making model-based uncertainty estimates too optimistic if that extra variability is ignored. This is a general consideration, not a reported feature or analysis of this trial. The registry record does not report a model, an over-dispersion assessment, or any fitted analysis.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Efavirenz")
n      <- c(53)
events <- c(15)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Efavirenz  53  15  28.3  16.8  42.3  18.0  41.6  18.0  41.6

11. Multiplicity, Missing Data and Analysis Populations

The registry record names a primary endpoint of percentage without PSA progression at 3 months and a secondary endpoint of percentage without PSA progression at 6 months. It lists 2 outcome measures and 0 analyses. The record does not report results for either endpoint or describe a statistical plan for prioritizing or interpreting multiple outcomes.

Multiplicity arises when several outcomes, time points, subgroups, or statistical tests are considered. In general, testing or selecting among many results can increase the chance of finding an apparently notable result by chance. Prespecified priorities or adjustment procedures can address some multiplicity concerns, but the registry record does not report whether such procedures were planned or used. It would be inappropriate to infer an adjustment method from the endpoint listing alone.

The endpoint description states that participants who died within the first 3 months are considered progressions for the primary evaluation. This is a specified classification rule, not a general account of how other missing data were handled. The registry record does not report the number of missing PSA assessments, the number of participants included in each endpoint analysis, methods for imputation, or sensitivity analyses. It also does not state whether participants who discontinued treatment remained in endpoint follow-up or how withdrawal affected analysis.

The registry record does not report an analysis population such as all enrolled participants, treated participants, or participants with an evaluable PSA assessment. Enrollment of 61 therefore cannot be treated as the endpoint denominator. Nor can the safety listing’s 59 participants at risk be assumed to define an efficacy population. Without these details and the endpoint counts, the possibility of bias related to missing assessments or exclusions cannot be assessed.

More broadly, the record does not report an interim analysis plan, stopping rule, or statistical analysis plan. Those matters are not inferable from the completed status or endpoint definitions. The available information supports describing the intended outcomes and some classification rules, but not reconstructing the analysis choices or the handling of incomplete data.

12. 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 0% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: psychosis (3.4%), confusion (1.7%), abdominal pain (1.7%), back pain (1.7%), sepsis (1.7%).

The safety listing records 59 participants at risk in the efavirenz group and 11 participants with serious adverse events. It reports psychosis at 3.4%, and confusion, abdominal pain, back pain, sepsis, and urinary tract obstruction at 1.7% each among the listed serious-event terms. These figures describe the registry’s safety entries; they are not estimates of comparative risk because the trial has no comparator group.

A serious adverse event is a safety classification, not a statement that the event was caused by the study drug. The listing does not establish causation, and it does not report how event rates compare with another treatment or with an untreated group. The percentages should be read as reported, rather than recalculated or interpreted as treatment-attributable risks.

The registry record does not report the number of deaths in the safety listing. A missing death value should not be read as evidence that no deaths occurred. Nor does the safety information establish the frequency of all adverse events: the listed terms concern the serious-event entries, not a complete account of every possible symptom or event. Safety findings also do not substitute for the unreported primary PSA endpoint.

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

The record establishes that a phase II, single-group study of efavirenz in metastatic prostate cancer was completed, and that its primary endpoint was defined as the percentage without PSA progression at 3 months. It also documents a secondary PSA non-progression endpoint at 6 months, an assessment schedule, and criteria for classifying progression. These features explain what the study set out to measure, but they do not reveal the outcome.

The primary percentage, its 95% confidence interval, and the secondary endpoint result are not reported in the registry record. The record lists 0 analyses. Consequently, it is not possible to state from this record how often participants met the non-progression definition, how uncertain an estimate would be, or whether any statistical comparison was made. The study title and rationale are not outcome evidence.

Even if an endpoint estimate were available, the single-group design would limit causal interpretation. A rate among participants receiving efavirenz would not, by itself, establish how outcomes would differ under another treatment. The absence of a concurrent comparator means that effects of treatment cannot be separated from other possible influences by a randomized contrast. The record’s lack of masking is another design feature to consider, although the endpoint uses a laboratory measure and the record does not detail assessment safeguards.

The safety entries offer a partial description of serious adverse events, including 11 affected participants among 59 at risk and the listed event terms. They do not establish that efavirenz caused those events or indicate how the safety experience compares with another option. The record also does not report a death count in the safety listing.

Accordingly, the registry record supports a description of the trial’s design, endpoint definitions, and recorded safety entries. It does not support a conclusion about the magnitude or precision of the primary result, a treatment benefit or lack of benefit, or comparative safety. These distinctions separate what the registry documents from what remains unavailable.

14. Why This Trial Matters Statistically

This trial illustrates why endpoint definitions and outcome reporting need to be read together. The registry gives a detailed operational definition of PSA progression, including separate rules based on whether PSA decreased, an absolute increase requirement, confirmation, and treatment of deaths within the primary assessment period. It also describes the intended PSA measurement schedule. Such details determine how a binary endpoint is classified and help readers understand what a reported percentage would mean.

At the same time, an endpoint definition is not an endpoint result. Without the observed percentage, its denominator, and its confidence interval, readers cannot assess the rate or its precision. A study record may contain substantial information about intent and measurement while still leaving the principal statistical finding unavailable. The distinction is especially important when enrollment is reported, because enrollment alone does not show how many participants contributed usable endpoint data.

The study also demonstrates the limits of interpreting single-group data. A descriptive rate can characterize a group but cannot supply the counterfactual comparison needed to estimate what would have happened under another intervention. This remains true regardless of how clearly the outcome is defined. Design and analysis determine which questions the evidence can answer.

Finally, safety counts and efficacy endpoints answer different questions. The serious adverse event listing offers some information about events recorded during the trial, while the primary PSA outcome is not reported. Neither type of information can stand in for the other, and neither establishes comparative effects without an appropriate design. Statistically, the record is most informative about the intended measurement framework and the limits imposed by absent outcome analyses.

15. 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 omits the primary endpoint percentage and its 95% confidence interval, as well as the secondary endpoint result. It lists 0 analyses. Without the endpoint counts and denominators, the analysis population, observed rate, and precision cannot be evaluated. Enrollment of 61 does not establish the number assessed at 3 months or included in an endpoint calculation.

The single-group design has no concurrent control, and masking is listed as none. A reported outcome rate would therefore be descriptive rather than a randomized treatment contrast. The record does not report the trial’s statistical analysis plan, sample-size justification, interim analyses, or stopping rules. It also does not explain how missing PSA assessments, treatment discontinuation, or withdrawals were handled, apart from classifying deaths within the first 3 months as progressions for the primary endpoint.

Although PSA monitoring procedures are described, the record does not report whether the planned assessment schedule and laboratory procedures were followed in practice. It does not state how many participants had confirmed progression, how unconfirmed increases were treated in the analysis, or whether any sensitivity analyses were conducted.

The safety entry reports serious adverse events among participants at risk and lists selected serious-event terms, but the death count is not reported. The record does not provide a comparator safety profile, a causal assessment, or a complete account of all adverse events. These omissions limit interpretation of both the outcome and the safety information without permitting assumptions about unreported results.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Efavirenz in Treating Patients With Metastatic Prostate Cancer — NCT00964002.
• Registry record last updated 2022-05-16; status last verified 2021-11.
• Results first posted 2022-05-16.
• 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.

19. Record Summary

This completed phase II trial evaluated efavirenz in metastatic prostate cancer using a single-group design. Institut Bergonié is listed as sponsor; the registry records a start date of 2008-05, primary completion on 2011-09-19, and enrollment of 61.

The primary endpoint was percentage without PSA progression at 3 months, with a 95% confidence interval specified. The registry describes progression criteria and an assessment schedule, and names percentage without PSA progression at 6 months as a secondary outcome. It lists 0 analyses and reports neither endpoint result.

The safety listing records 11 participants with serious adverse events among 59 at risk and names several serious-event terms with percentages. The record does not report a death count. With no comparator arm and no outcome estimates, the registry supports description of the trial design and recorded safety entries, but not a treatment-effect or comparative-safety conclusion.