Preoperative Fulvestrant With or Without Enzalutamide in ER+/Her2- Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Preoperative Fulvestrant With or Without Enzalutamide in ER+/Her2- Breast Cancer (NCT02955394), a randomized phase 2 study. The primary outcome is Number of Patients With a PEPI Score Equal to Zero at Post Treatment.

2. Clinical Question

The trial is an active-treatment comparison: both groups received fulvestrant, while enzalutamide was the additional drug in one group. Its primary endpoint is a short-term, pathology-based measure assessed after treatment, rather than a direct comparison of long-term recurrence or survival. Disease-free survival, clinical benefit rate, overall response rate, and androgen receptor expression are listed among secondary outcomes, but their detailed results are not included in the posted outcome results described in the registry record.

The primary contrast is therefore about the observed frequency of a zero PEPI score in the two groups at the stated assessment time. The registry describes zero as the lowest PEPI score and as the low-risk category; this endpoint focuses on zero versus any non-zero score, not on differences among the non-zero scores. The posted data can describe the trial participants and the uncertainty around the group comparison, but do not alone establish a treatment effect that would apply to other populations or clinical settings.

3. Trial Design

This is a randomized, parallel-group phase II trial with no masking reported. Its stated purpose is to further evaluate fulvestrant plus enzalutamide against single-agent fulvestrant before surgery in postmenopausal women with locally advanced AR+/ER+/Her2- breast cancer. The registry lists 61 enrolled participants. The primary endpoint is reported for 59 evaluable patients: one patient was unable to participate in treatment after baseline and another came off study for a treatment change.

The fulvestrant-only arm is the comparator. Its regimen is described as 500 mg intramuscularly on days 1, 15, and 28, then every 4 weeks. The combination arm uses that fulvestrant schedule and adds daily enzalutamide at 160 mg. The planned preoperative treatment period is described as about 4 months in the brief summary; the primary outcome timeframe is 16 weeks. The registry lists local surgery after treatment but does not report further details of the surgery schedule or procedures.

The posted primary endpoint counts are based on 26 participants in the fulvestrant-only arm and 33 in the combination arm. These arm-specific endpoint denominators sum to the 59 evaluable patients. Safety figures use different denominators: 27 and 34 participants at risk in the respective arms. The registry record does not explain the difference between the endpoint and safety denominators, so those populations should be kept distinct when interpreting the results.

The trial began on 2017-09-21, and the primary completion date is 2023-02-17. Its current registry status is active, not recruiting. The registry lists three secondary outcomes, while detailed secondary outcome results and analysis records are not posted. The original statistical design, including its own sample-size justification, is not reported in the registry record. The displayed sample-size illustrations are based on the observed primary comparison and should not be mistaken for 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

Analysis population for the primary outcome, as defined in the registry: 59 patients were evaluable. One patient was unable to participate in treatment after baseline and the other came off study for treatment change.

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 Number of Patients With a PEPI Score Equal to Zero at Post Treatment, measured over 16 Weeks.

The primary endpoint is the number of patients with a PEPI score equal to zero at post-treatment assessment, with a timeframe of 16 weeks. PEPI, the preoperative endocrine prognostic index, combines estrogen receptor level, pathologic tumor site, nodal status, and Ki67 score at surgery to predict subsequent recurrence risk. The registry describes PEPI as a validated measure of pathologic response to endocrine therapy.

PEPI scores are grouped into zero, 1–3, and at least 4, corresponding in the record to low, intermediate, and high risk categories. This study’s primary question uses a simpler division: zero versus non-zero. Zero is the minimum possible score, and lower scores are described as better; the registry does not state a maximum score. Thus, the endpoint counts patients who reached the zero category rather than estimating a mean score or comparing all score categories separately.

A pathology-based measure assessed after preoperative treatment offers a way to characterize the response observed by the time of surgery. It is not itself a direct count of later recurrences or deaths. Disease-free survival is listed as a secondary outcome, but its detailed results are not reported in the registry record. The primary endpoint therefore addresses the short-term PEPI category, while longer-term outcomes remain a separate question.

7. Results Posted in the Registry

• Fulvestrant With Enzalutamide: 24.2% (8/33) versus 7.7% (2/26) on control; risk difference 16.6 percentage points (95% CI -1.3 to 34.4), risk ratio 3.15 (0.73–13.59); unadjusted two-proportion z-test p = 0.093.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary: each evaluable participant either had a PEPI score of zero at the 16-week assessment or did not. For a binary endpoint, a count is interpreted with its denominator. The posted counts are 2 of 26 in the fulvestrant-only group and 8 of 33 in the combination group. The denominator is essential: a count alone would not show how frequently the event occurred in each group. Because the outcome is a count of participants with an event, it is not a continuous measurement such as a mean score.

The main comparative measure is the risk difference. It expresses the difference between the proportions with a zero score in the combination arm and the fulvestrant-only arm. A risk difference of zero represents equal proportions; positive values indicate a higher proportion in the combination arm under this ordering, and negative values indicate a lower proportion. The posted estimate is 0.166. Its confidence interval, -0.013 to 0.344, spans zero, so the data do not narrowly identify the direction of the underlying difference.

The registry also posts a risk ratio and an odds ratio. A risk ratio compares the event proportions between groups, with one representing equal proportions. An odds ratio compares event odds, also with one representing no contrast. Both can be useful descriptions of a binary outcome, but they answer the comparison on different scales. When an event is uncommon, these relative measures may be especially sensitive to a small change in the event counts. The posted relative-effect confidence intervals are broad and include one, indicating substantial uncertainty in those estimates.

A confidence interval describes a range of values compatible with the observed data under the statistical method used. In repeated use of a method under its assumptions, intervals at a stated confidence level have a long-run coverage interpretation; an individual interval is not a probability statement that the true effect lies within its endpoints. A narrow interval indicates greater precision, while a wide interval indicates less precision. Here, the risk-difference interval crosses zero and extends over both directions, so the estimate does not resolve whether the group contrast is positive or negative with much precision.

The posted p-value is unadjusted and equals 0.093. A p-value measures how unusual the observed data, or data more extreme, would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of the result. The registry record does not report a detailed testing plan, the particular inferential procedure used for the primary comparison, or any adjustment for multiple testing. Consequently, the p-value should be read as the posted unadjusted result, without assuming an undocumented analysis procedure.

Random allocation is intended to make treatment groups comparable on average at the start of a trial. This is a general feature of randomized designs, not proof that every baseline characteristic was balanced in these particular groups. The registry lists the allocation as randomized and the design as parallel, but does not report the allocation sequence method, concealment procedure, or stratification factors. Masking is listed as none. These details can affect the risk of bias or interpretation, and additional information is not reported in the registry record.

The endpoint denominators are 26 and 33, whereas the safety denominators are 27 and 34. These are reported separately rather than assumed to describe the same analysis population. The record identifies two reasons participants were not evaluable for the primary endpoint, but does not give further missing-data methods, imputation procedures, or sensitivity analyses. General statistical practice distinguishes an analysis of all randomized participants from one restricted to participants with an observed endpoint; without a stated analysis plan, the population definition should not be inferred beyond the posted counts.

Over-dispersion is a concern mainly for count data modeled as arising from a rate or repeated events, when variability exceeds that assumed by a simple model. It is not directly the issue in these patient-level binary counts, where each participant contributes a zero-or-non-zero outcome. The key sources of uncertainty are the limited group sizes and event counts, along with any unreported features of the analysis and missing-data handling. General explanations of these methods do not imply that a particular model was used in this trial.

Finally, statistical significance and estimation answer different questions. A p-value summarizes evidence against a null model under specified assumptions; a confidence interval communicates the range and precision of the estimated contrast. Neither alone determines clinical importance. The registry describes PEPI zero as its low-risk category, but does not supply a threshold for a clinically meaningful difference between arms. The observed counts and estimates therefore support a cautious description, not a definitive clinical conclusion.

10. Confidence Intervals and Precision

The risk-difference estimate is 0.166, with a 95% confidence interval from -0.013 to 0.344. The interval crosses zero, the value representing equal proportions. Its endpoints also span both a small negative contrast and a positive contrast, showing that the direction and size of the underlying difference are not precisely determined by these data.

Only 2 participants in the fulvestrant-only group and 8 in the combination group had a PEPI score of zero. With event counts of this size, each observed event contributes meaningfully to the group proportions, and the resulting comparative estimates can be uncertain. The relative-effect confidence intervals are also wide and include the no-association value of one. Together, these interval results indicate that the evidence is not precise enough to pin down a single narrow estimate of the treatment-group contrast.

The p-value of 0.093 gives another summary of the data under a null comparison, but it does not resolve the uncertainty represented by the confidence interval. A p-value just below or above a threshold would not, by itself, change the observed counts or make an estimate more precise. Interpretation should retain the magnitude and uncertainty of the estimate rather than reducing the result to a significant/not-significant label.

Precision is limited in part by the modest number of evaluable participants in each arm. The registry record does not report the original sample-size justification, target power, or expected effect used when designing the trial, so the adequacy of the original enrollment target cannot be assessed from this record. The planning illustrations are separate calculations based on the observed contrast and do not provide a retrospective measure of how likely the trial was to detect a true effect.

11. 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 participants per arm, not total participants or total events. The assumptions listed for these illustrations are 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. For time-to-event outcomes, the record says planning figures would represent total events, but the primary outcome is binary, so the figures are participants per arm.

The 80% power column gives the planned number of participants per arm for a design with an 80% chance of detecting the specified effect under those assumptions, if that effect were the true effect and the assumptions held. The 90% power column describes the corresponding planning target with a higher chance of detection under the same assumptions. These are design calculations, not predictions about what happened in the trial.

The rows consider planning for 100%, 75%, and 50% of the observed effect. The listed participant requirements increase as the target effect gets smaller: detecting a smaller difference generally requires more participants to distinguish it from random variation. The values are illustrations based on this trial’s observed estimate, not the original design assumptions. The registry record does not report the trial’s own sample-size justification or planned power. These calculations are not observed power.

Because the observed effect is uncertain, basing a future design on it alone can be misleading: the observed contrast may not equal the underlying effect. The planning estimates should therefore be read as conditional scenarios tied to the stated assumptions and effect fractions, rather than as a recommended enrollment target or a guarantee of a future result.

12. 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

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.077; treatment proportion p₁ = 0.242; average p̄ = 0.160.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5180 = 1.0153.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5046 = 0.4247.
4. n = (1.0153 + 0.4247)² / (0.1655)² = 75.7, rounded up to 76 participants per arm 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
p0 <- 0.0769230769230769                          # control proportion
p1 <- 0.242424242424242                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(2955394)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 76 (80% power), 101 (90% power)
 75% of observed effect: n per arm = 121 (80% power), 162 (90% power)
 50% of observed effect: n per arm = 238 (80% power), 319 (90% power)
power.prop.test: 75.7 
Simulated power at n = 76 per arm: 0.823 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists three secondary outcomes: disease-free survival; correlation between PEPI score and disease-free survival, clinical benefit rate, and overall response rate; and androgen receptor expression. It also lists four outcome measures, but the record does not provide detailed results for the secondary outcomes in the posted results. No analysis records are listed. The primary comparison can therefore be interpreted from its posted count and estimates, while the results of the secondary questions cannot be assessed from the available registry record.

When a study evaluates several outcomes or makes several comparisons, the chance of at least one apparently unusual result can increase. Statistical methods sometimes control or account for this multiplicity. The registry record does not report whether such an adjustment was planned or used, nor does it describe a hierarchy of outcomes or a prespecified testing sequence. The posted p-value is identified as unadjusted, so it should not be treated as a multiplicity-adjusted finding.

The primary endpoint population includes 59 evaluable patients out of 61 enrolled. The record states that one patient could not participate in treatment after baseline and another came off study for a treatment change. It does not report how missing endpoint data were handled beyond the evaluable count, whether an intention-to-treat analysis was performed, or whether sensitivity analyses addressed the two participants without primary endpoint results. These methods cannot be inferred from the difference between enrollment and evaluability alone.

The safety denominators are 27 and 34, unlike the primary endpoint denominators of 26 and 33. The registry gives no explanation for this difference. These counts should not be silently combined or treated as interchangeable analysis populations. The record also does not describe the handling of missing secondary outcomes or define the populations used for each secondary analysis. These omissions limit evaluation of the full analysis plan and the robustness of results beyond the posted primary comparison.

14. Safety Results

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

Deaths

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

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: abdominal pain (3.7%), stroke (2.9%), coronary artery disease (2.9%), myocardial infarction (2.9%), hematoma (2.9%).

The serious adverse event figures use safety groups of 27 participants at risk in the fulvestrant-only arm and 34 in the combination arm. One participant in the fulvestrant-only group and three in the combination group were reported with serious adverse events. The registry reports zero deaths in both groups. These are counts, and the denominators differ from those used for the primary efficacy endpoint.

The record lists abdominal pain at 3.7%, and stroke, coronary artery disease, myocardial infarction, and hematoma at 2.9% each among the top serious event terms. The listing does not establish which treatment arm each term occurred in or whether events were judged related to a study drug. No causal attribution should be inferred from these figures.

Serious adverse event counts are distinct from counts of other adverse events. The registry reports 26 participants with other adverse events among 27 at risk in the fulvestrant-only arm and 34 among 34 at risk in the combination arm. These aggregate counts do not describe event severity, duration, timing, or recurrence. Differences between groups should be interpreted cautiously, particularly because the safety denominators are limited and the record does not provide a detailed comparative safety analysis.

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

The posted primary endpoint shows that 2 of 26 evaluable participants receiving fulvestrant alone and 8 of 33 receiving fulvestrant plus enzalutamide had a PEPI score of zero at 16 weeks. The risk-difference estimate favors a higher observed proportion in the combination arm, but its confidence interval includes zero and the unadjusted p-value is 0.093. The result is therefore uncertain: it describes the trial’s observed groups, but does not establish the size or direction of an underlying difference with precision.

A zero PEPI score is described in the registry as the low-risk category and the minimum score. However, the endpoint is not the same as observing long-term recurrence or survival. Disease-free survival and other outcomes are listed as secondary outcomes, but their results are not reported in the registry record. The primary result cannot answer whether the groups later differed on those outcomes.

The findings also do not establish whether the combination is preferable for a particular patient, and they should not be used as medical advice. The trial is a randomized phase II study of preoperative treatment in a specific population. Its posted results may help characterize a treatment comparison and inform further research, but the registry record does not report a confirmatory clinical conclusion or evidence from other studies.

Uncertainty arises from the limited number of evaluable participants and the small number with a zero score, as well as from details of the analysis plan and missing-data handling that are not reported. The safety counts likewise describe reported events within the recorded safety populations; they do not prove that an event was caused by treatment or that risks are equivalent between groups. Overall, the data support a careful account of observed counts and broad uncertainty, not a definitive claim about efficacy or safety.

16. Why This Trial Matters Statistically

This trial illustrates how a binary endpoint can be reported with both raw counts and comparative estimates. The counts make the event definition tangible: a participant either had a PEPI score of zero or did not. The risk difference then expresses the group contrast on an absolute scale, while the risk ratio and odds ratio express relative contrasts. Because these measures use different scales, showing their uncertainty helps prevent a single estimate from carrying more weight than the data can support.

The comparison also demonstrates why an observed direction is not enough to settle an interpretation. More zero scores were recorded in the combination arm, and the risk-difference estimate is positive, but its confidence interval crosses zero and the unadjusted p-value is 0.093. A statistically responsible reading keeps the estimate, its uncertainty, and the limited event counts together rather than treating a point estimate as a conclusive answer.

The endpoint itself has a clear boundary: this result concerns PEPI zero at 16 weeks, not a complete comparison of long-term clinical outcomes. That distinction is important when a short-term pathology measure is used in a preoperative study and longer-term outcomes are listed separately. The registry record does not report detailed results for those secondary outcomes.

Finally, the sample-size illustrations show how planning requirements depend on the effect a future study aims to detect. Smaller target effects require more participants under the stated assumptions. Since those calculations are derived from the observed effect and the original design assumptions are not reported, they are best understood as conditional planning examples, not as a judgment of the trial’s success or failure.

17. 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 primary endpoint is based on 59 evaluable patients, rather than all 61 enrolled participants. The registry identifies two reasons for non-evaluability but does not report a fuller missing-data analysis, an intention-to-treat result, or sensitivity analyses. The endpoint comparison therefore needs to be read in the context of the reported evaluable population.

The arm sizes and event counts are limited: the primary endpoint counts are 2 of 26 and 8 of 33. The resulting confidence intervals are broad, limiting precision. The registry record does not report the original sample-size rationale or the effect assumed for the trial’s design, so it is not possible to judge the original design against its planned operating characteristics from this record.

The posted primary endpoint captures PEPI zero at 16 weeks. It does not, on its own, establish an effect on disease-free survival or other longer-term outcomes. Although secondary outcomes are listed, detailed results and analysis records are not posted. The registry also does not report whether multiplicity was addressed, the precise primary analysis method, or key implementation details such as allocation concealment and stratification.

Safety outcomes have different denominators from the primary endpoint, and the record does not explain the discrepancy. The listed serious event terms do not identify treatment attribution or provide a complete comparative assessment of event timing and severity. These omissions limit the strength of conclusions about both efficacy and safety. The findings should be treated as a focused phase II comparison in the population and timeframe described, rather than generalized beyond the information in the registry record.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Preoperative Fulvestrant With or Without Enzalutamide in ER+/Her2- Breast Cancer — NCT02955394.
• Registry record last updated 2026-08-31; status last verified 2026-08.
• Results first posted 2023-06-15.
• 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.

21. Record Summary

NCT02955394 is a randomized, parallel-group phase II study of preoperative fulvestrant with or without enzalutamide in postmenopausal women with locally advanced AR+/ER+/Her2- breast cancer. The registry lists 61 enrolled participants and a current status of active, not recruiting. The primary endpoint is the number of patients with a PEPI score of zero at 16 weeks.

Among 59 evaluable participants, the posted counts were 2 of 26 in the fulvestrant-only arm and 8 of 33 in the combination arm. The risk difference was 0.166, with a 95% confidence interval from -0.013 to 0.344; the unadjusted p-value was 0.093. Safety counts use separate denominators of 27 and 34 participants at risk, with one and three serious adverse events reported, respectively, and zero deaths in both groups.

The registry does not report the original sample-size justification, detailed secondary results, or several aspects of the statistical analysis plan. The posted primary result is therefore best interpreted as an uncertain comparison of the observed PEPI-zero counts at the stated assessment time.