Study of Nivolumab Versus Placebo in Combination With Neoadjuvant Chemotherapy and Adjuvant Endocrine Therapy in Participants With High-risk, Estrogen Receptor-Positive (ER+), Human Epidermal Growth Factor Receptor 2-Negative (HER2-) Primary Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Nivolumab Versus Placebo in Combination With Neoadjuvant Chemotherapy and Adjuvant Endocrine Therapy in Participants With High-risk, Estrogen Receptor-Positive (ER+), Human Epidermal Growth Factor Receptor 2-Negative (HER2-) Primary Breast Cancer (NCT04109066), a randomized phase 3 study. The primary outcome is Pathological Complete Response (pCR) Rate.

2. Clinical Question

3. Trial Design

This was a randomized, multicenter, double-blind, placebo-controlled phase 3 study with a parallel-group design. The registry reports enrollment of 521 and a completed status. The recorded start date is 2019-11-18, and the primary completion date is 2023-01-16.

The comparison was between nivolumab and nivolumab placebo, each given with neoadjuvant chemotherapy and followed by adjuvant endocrine therapy. In both arms, chemotherapy included paclitaxel followed by anthracycline plus cyclophosphamide; surgery followed the neoadjuvant treatment. Endocrine therapy after surgery was selected by the investigator. The registry’s arm descriptions do not state that every treatment detail was identical beyond the listed sequence and placebo substitution.

The primary endpoint was pCR, expressed as a percentage of participants. The primary assessment timeframe was up to approximately 37 months. The primary population was all randomized participants with sufficient follow-up for pCR assessment. The posted pCR results use denominators of 257 for Arm A and 253 for Arm B. The registry does not explain the difference between those endpoint denominators and the enrollment figure.

The registry also describes a pCR analysis in participants with PD-L1 >=1% and a count of participants who died, among other secondary outcomes. The pCR subgroup result uses a different population from the primary analysis, while the death outcome has its own denominators and timeframe. These outcome-specific populations should not be treated as interchangeable.

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: 2 Participants planned for Arm B treatment received Arm A treatment.

Pre-treatment

Reasons for not completing this period:

Treatment

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized participants with sufficient follow-up for pCR assessment

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 Pathological Complete Response (pCR) Rate, measured over Up to approximately 37 months.

Pathological complete response (pCR) is a measure of whether invasive cancer remains in the breast and lymph nodes after neoadjuvant treatment. In this trial, pCR was defined using local pathology assessment. The criteria include pCR in the breast, axillary lymph nodes and non-axillary sentinel node, with no histologic evidence of invasive tumor cells. The registry’s description also specifies pCR in the breast.

The endpoint is reported as a percentage of participants who achieved pCR, with a 95% confidence interval. It captures a pathological finding after pre-surgery treatment rather than a direct measure of how long participants live or remain free of disease. Its timing makes it useful for comparing how often the two assigned treatment strategies are followed by no invasive residual disease at surgery, but it does not by itself describe later recurrence, survival, or quality of life.

The assessment timeframe was up to approximately 37 months, and the primary population consisted of randomized participants with sufficient follow-up for pCR assessment. These definitions matter: the percentage describes the specified endpoint in that analysis population, not every possible clinical outcome or every enrolled participant.

7. Results Posted in the Registry

• Arm B: 13.8% (35/253) versus 24.5% (63/257) on control; risk difference -10.7 percentage points (95% CI -17.4 to -3.9), risk ratio 0.56 (0.39–0.82); unadjusted two-proportion z-test p = 0.002.

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.

Pathological Complete Response (pCR) Rate (PD-L1 >=1%)

• 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:
  • Arm B: 20.2% (17/84) versus 44.3% (39/88) on control; risk difference -24.1 percentage points (95% CI -37.6 to -10.6), risk ratio 0.46 (0.28–0.74); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Arm B than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Number of Participants Who Died

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Arm B: 3.1% (8/255) versus 5.7% (15/262) on control; risk difference -2.6 percentage points (95% CI -6.1 to 0.9), risk ratio 0.55 (0.24–1.27); unadjusted two-proportion z-test p = 0.154.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Arm B and the control. The interval gives the range of effects that remain compatible with the data.

The secondary results that are numerically detailed include pCR among participants with PD-L1 >=1% and the number of participants who died. The subgroup pCR result follows the same direction as the primary pCR finding, while the death comparison is less conclusive: its recalculated confidence interval includes zero. These outcomes answer different questions and use different populations and timeframes, so their estimates should not be combined into a single measure of benefit or risk.

The registry lists six secondary outcomes, including residual cancer burden (RCB), adverse events, serious adverse events and deaths. Numerical results for the RCB outcomes are not reported in the registry record. Safety counts are described separately, but complete outcome-level analysis details for all listed secondary measures are not reported in the registry record. The absence of a posted numerical result should not be interpreted as evidence that an outcome did or did not occur.

Secondary outcomes help describe whether the primary finding is accompanied by patterns in other measures, and they can raise questions for further analysis. They do not automatically confirm the primary result. In general, testing many outcomes increases the chance that at least one result may appear notable by chance. Adjustment for multiple comparisons can address that issue, but a multiplicity plan or adjustment is not reported in the registry record. The secondary results should therefore be read as supportive or exploratory evidence rather than as a set of independent confirmatory conclusions.

The subgroup result also warrants care because it concerns a selected population rather than all participants in the primary analysis. The registry does not report a formal interaction analysis establishing whether the treatment comparison differs by PD-L1 status. A subgroup estimate’s direction alone does not show that PD-L1 status changes the treatment effect.

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

• Pathological Complete Response (pCR) Rate (PD-L1 >=1%)
• Number of Participants With Residual Cancer Burden (RCB)
• Number of Participants With Residual Cancer Burden (RCB) PD-L1 >=1%
• Number of Participants With Adverse Events (AEs)
• Number of Participants With Serious Adverse Events (SAEs)
• Number of Participants Who Died

9. Statistical Interpretation

10. Statistical Methods Explained

For a binary endpoint such as pCR, each participant either meets the endpoint definition or does not. The reported percentage is the share of the analysis population who achieved pCR. In general, a risk difference expresses the difference between two such percentages on an absolute scale. A negative recalculated risk difference for Arm B compared with Arm A means the observed percentage was lower in Arm B. The interval around that estimate conveys uncertainty about the size of the difference.

The primary recalculation is an unadjusted comparison based on the posted summary results. It gives a risk difference of -0.107, with a 95% confidence interval from -0.174 to -0.039 and p=0.002. “Unadjusted” means this recalculation does not incorporate covariate adjustment. It should be distinguished from the sponsor’s reported analyses, which include an odds ratio from a Cochran-Mantel-Haenszel method and an adjusted difference of pCR rates.

An odds ratio compares the odds of the endpoint in one group with the odds in another. It is not the same quantity as a risk difference, and its numerical scale is not a percentage-point scale. An odds ratio above 1 for Arm A relative to Arm B is consistent with higher odds of pCR in Arm A. The sponsor reports an odds ratio of 2.05, with a 95% confidence interval from 1.29 to 3.27. The registry identifies the Cochran-Mantel-Haenszel method for this analysis. The specific adjustment factors and implementation details are not reported in the registry record.

The sponsor also reports an adjusted difference of pCR rates of 10.5, with a 95% confidence interval from 4.0 to 16.9. This is a separate analysis from the odds ratio and the unadjusted recalculation. The adjustment method for this difference is not reported in the registry record, so the estimate should be read as the sponsor’s posted adjusted result without assuming which covariates or modeling choices were used.

In general, a 95% confidence interval describes the uncertainty associated with an estimate under the analysis method and its assumptions. It is not a range containing 95% of individual participant outcomes. When an interval for a difference includes zero, the data are compatible with no difference as well as with values on either side of zero at the interval’s confidence level. When the interval excludes zero, the estimate is less compatible with a zero difference under that method; this does not alone establish clinical importance or rule out bias.

The registry posts confidence intervals for the arm-specific pCR percentages as well as the comparison estimates. Arm-specific intervals describe uncertainty in each percentage separately; they are not a substitute for a confidence interval calculated directly for the contrast. For this reason, the comparison interval and test are the relevant quantities when asking whether the groups differ statistically.

The primary and subgroup pCR analyses are binary-outcome analyses. The subgroup analysis is restricted to participants with PD-L1 >=1%, so its estimate addresses a narrower population. In general, a subgroup estimate may be less precise because it is based on fewer observations, and a subgroup result should not automatically be generalized to all randomized participants. The subgroup result is also one of multiple outcomes considered in the trial.

The death outcome is a count of participants who died, with the registry also giving arm-specific denominators. Its recalculated comparison is expressed as a risk difference, with a 95% confidence interval from -0.061 to 0.009 and p=0.154. The interval crosses zero. In general, a count-based binary comparison does not account for when an event occurred unless a time-to-event method is used; a time-to-event analysis for deaths is not reported in the registry record.

Several details needed to fully reproduce or evaluate the analyses are not reported in the registry record, including the trial’s original statistical analysis plan, the covariates used in the adjusted difference, and the handling of missing endpoint assessments. The primary population is described as randomized participants with sufficient pCR follow-up, but more detailed analysis-population rules are not reported in the registry record.

11. Confidence Intervals and Precision

The primary arm-specific pCR estimates are 24.5% for Arm A, with a 95% confidence interval from 19.4% to 30.2%, and 13.8% for Arm B, with a 95% confidence interval from 9.8% to 18.7%. These intervals convey uncertainty in each arm’s estimated percentage. The comparison is more directly described by the recalculated risk difference of -0.107 for Arm B compared with Arm A, with a 95% confidence interval from -0.174 to -0.039.

Because the risk-difference interval excludes zero, the primary recalculation is consistent with a difference between the arms under its unadjusted method. The interval also indicates that the estimate is not exact: a range of difference sizes remains compatible with the data and method. The confidence intervals should not be interpreted as a guarantee that the true value lies within the displayed limits.

The sponsor’s adjusted pCR difference is 10.5, with a 95% confidence interval from 4.0 to 16.9. This interval is for the sponsor’s adjusted analysis, not the same calculation as the unadjusted risk-difference interval. The sponsor’s odds-ratio interval, 1.29 to 3.27, likewise describes uncertainty on a different statistical scale.

For the PD-L1 >=1% subgroup, the arm-specific confidence intervals are 33.7% to 55.3% and 12.3% to 30.4%. The recalculated risk difference for Arm B compared with Arm A has a 95% confidence interval from -0.376 to -0.106. The death risk-difference interval, -0.061 to 0.009, includes zero and is compatible with no difference as well as with effects in either direction. Precision differs by endpoint and population; the intervals should be considered alongside the estimate and the analysis definition.

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 participants per arm for the primary binary comparison, not total events. The illustration assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It uses the observed effect as a reference and considers planning for 100%, 75%, and 50% of that observed effect. The trial’s own original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the estimated participants per arm for a future study to have an 80% chance of detecting the specified effect under the stated assumptions, if that effect and assumptions are correct. The 90% power column describes the corresponding planning target for a 90% chance. These are design calculations, not probabilities that the observed trial result is true or that a treatment will have a particular effect.

For planning at 100% of the observed effect, the illustration gives 213 participants per arm for 80% power and 284 participants per arm for 90% power. For planning at 75% of the observed effect, it gives 398 participants per arm for 80% power and 532 participants per arm for 90% power. For planning at 50% of the observed effect, it gives 939 participants per arm for 80% power and 1,257 participants per arm for 90% power. Smaller effects require more participants under these assumptions.

This is not observed power. It is a planning illustration based on the observed effect, which is itself uncertain. The figures should not be read as the trial’s actual planned sample size or as evidence that the original trial was designed using these assumptions.

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

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 = 0.245; treatment proportion p₁ = 0.138; average p̄ = 0.192.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5567 = 1.0912.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5516 = 0.4642.
4. n = (1.0912 + 0.4642)² / (-0.1068)² = 212.1, rounded up to 213 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 <- 0.245136186770428                          # control proportion
p1 <- 0.138339920948617                          # 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(4109066)
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 = 213 (80% power), 284 (90% power)
 75% of observed effect: n per arm = 398 (80% power), 532 (90% power)
 50% of observed effect: n per arm = 939 (80% power), 1,257 (90% power)
power.prop.test: 212.1 
Simulated power at n = 213 per arm: 0.804 (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.

14. Analyses Reported by the Sponsor

The sponsor’s posted primary analysis reports a Cochran-Mantel-Haenszel odds ratio of 2.05, with a 95% confidence interval from 1.29 to 3.27. It also reports an adjusted difference of pCR rates of 10.5, with a 95% confidence interval from 4.0 to 16.9. The registry gives p=0.002 for the odds-ratio analysis; the recalculated unadjusted comparison gives p=0.002.

The recalculation uses the posted arm-level summary results and estimates an unadjusted risk difference for Arm B compared with Arm A. It is therefore not a reproduction of the sponsor’s adjusted analysis. The odds ratio, adjusted rate difference and unadjusted risk difference are different measures and may reflect different analytic procedures. The adjustment details for the sponsor’s difference estimate are not reported in the registry record.

The posted estimates point in the same general direction: the arm with nivolumab has the higher observed pCR percentage, and the sponsor’s adjusted difference is positive when describing the nivolumab group relative to placebo. The registry record does not flag a discrepancy between the posted and recalculated results. The sponsor analysis remains the trial’s reported analysis; the unadjusted recalculation is a separate summary-based comparison and should not be used to imply that the sponsor’s analysis has been replicated.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when investigators assess several outcomes, subgroups, or comparisons. In general, the more statistical tests considered, the greater the chance that at least one result may appear statistically notable even when there is no underlying effect for that test. Prespecified testing hierarchies or other adjustments can help control this issue, but the registry record does not report a multiplicity strategy for the posted outcomes.

The registry lists the primary pCR endpoint and six secondary outcomes. It also reports a pCR result in the PD-L1 >=1% subgroup. The record does not specify whether the subgroup analysis was prespecified, whether secondary endpoints were ordered in a testing hierarchy, or whether any multiplicity adjustment was applied. Consequently, the primary result has a clearer confirmatory role than a collection of secondary or subgroup findings whose testing context is not reported.

The primary population is described as all randomized participants with sufficient follow-up for pCR assessment. This wording does not provide the full rules for inclusion, handling of unavailable pathology assessments, or missing data. The registry record does not report how missing or unevaluable pCR assessments were handled, whether any sensitivity analyses were conducted, or whether analysis populations differed from the randomized groups according to a formal protocol definition.

Randomization is reported, but details of stratification factors and allocation concealment are not reported in the registry record. Double masking is stated; the specific masked parties and any procedures for maintaining masking are not reported. The posted summary does not permit evaluation of all prespecified analysis decisions or selective outcome reporting. These omissions limit how fully an external reader can assess robustness, but they do not by themselves show that the analyses were biased or that data were missing.

16. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: febrile neutropenia (6.7%), covid-19 (3.7%), pyrexia (3.0%), covid-19 pneumonia (2.4%), pneumonia (2.3%).

The registry reports serious adverse event counts across four chemotherapy-schedule groups, rather than only one combined count for each randomized arm. In the two Arm A groups, serious adverse events affected 38 of 127 participants and 51 of 135 participants. In the two Arm B groups, they affected 20 of 123 participants and 26 of 132 participants. These are group-specific counts and denominators; they should not be combined into new totals.

The registry lists febrile neutropenia at 6.7%, COVID-19 at 3.7%, pyrexia at 3.0%, COVID-19 pneumonia at 2.4%, pneumonia at 2.3%, and neutropenia at 1.6% among the leading serious adverse event terms. The record’s summarized figures do not provide enough detail to interpret these terms as comparative treatment effects or to determine their timing, duration, or severity distribution.

Safety comparisons require attention to the relevant treatment groups, exposure periods, event definitions and denominators. The reported serious-event counts alone do not establish that one regimen is safer than the other. Further detail about safety analysis methods and the timeframe for these serious adverse event figures is not reported in the registry record.

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

The primary result describes a difference in the percentage of participants with no invasive residual disease in the breast and lymph nodes at pathological assessment after neoadjuvant treatment. The posted arm percentages and sponsor analysis favor a higher pCR rate in Arm A than Arm B. The unadjusted recalculation’s confidence interval excludes zero, while the sponsor reports an adjusted difference with a confidence interval from 4.0 to 16.9.

That finding is specific to the pCR definition, the randomized participants with sufficient follow-up, and the assessment timeframe of up to approximately 37 months. It does not by itself quantify effects on recurrence, long-term disease-free survival, overall survival, quality of life, or other outcomes. The death comparison has a confidence interval that includes zero, and the registry record does not provide a time-to-event analysis for deaths.

In general, pCR can provide an early pathological measure of response to pre-surgery treatment. It is not identical to a patient-centered outcome such as survival or quality of life. A difference in pCR should therefore not be translated directly into a predicted difference in later outcomes without evidence connecting those outcomes for this setting.

The trial evaluated treatment strategies that included chemotherapy, surgery and endocrine therapy, with nivolumab or placebo as the comparison. The result does not isolate the effect of nivolumab from every component of the treatment sequence, nor does it establish a result for populations or treatment regimens not represented in the registry description. The data describe outcomes in this trial and should not be interpreted as medical advice or as a guarantee for an individual participant.

18. Why This Trial Matters Statistically

This trial illustrates why statistical interpretation needs more than a comparison of two observed percentages. The primary pCR percentages are accompanied by confidence intervals, and the arm-to-arm comparison has its own uncertainty interval. The sponsor also reports adjusted analyses on different scales. Distinguishing the estimand, analysis population and method helps prevent an odds ratio, an adjusted rate difference and an unadjusted risk difference from being treated as interchangeable.

The record also highlights the difference between a primary endpoint and secondary or subgroup findings. The PD-L1 subgroup pCR result has the same direction as the primary result, but it represents a narrower population and the record does not report a formal interaction analysis. Deaths provide a different clinical outcome, with a comparison interval that includes zero. Such findings call for endpoint-specific interpretation rather than a single broad conclusion.

Finally, the planning illustration shows how sample-size needs depend on the effect a future study aims to detect and the desired power. Smaller effects require more participants under the stated assumptions. Because these are planning calculations based on the observed effect, they do not reveal the original design rationale or the trial’s observed power. The registry does not report the original sample-size justification or detailed multiplicity and missing-data plans, which limits evaluation of the full statistical strategy.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The pCR population is defined as randomized participants with sufficient follow-up for assessment, but the registry record does not report the detailed rules for sufficient follow-up, unevaluable pathology, or missing outcome data. The primary pCR denominators differ from the denominators for deaths, and the reason for those differences is not reported in the registry record. The available summaries therefore do not establish how each analysis population was selected beyond the stated endpoint definitions.

The record reports both adjusted sponsor analyses and an unadjusted recalculation, but the adjustment details for the sponsor’s difference estimate are not reported. The registry does not provide the original statistical analysis plan, detailed covariate specifications, or a full account of prespecified analyses. Multiplicity handling for the secondary and subgroup analyses is also not reported in the registry record.

Follow-up is described as up to approximately 37 months for pCR and up to approximately 41 months for deaths. These timeframes do not provide a complete account of longer-term outcomes. A survival analysis, recurrence result, or quality-of-life result is not reported in the registry record within the outcomes.

The treatment arms include chemotherapy and endocrine therapy as well as nivolumab or placebo. The safety data are divided across chemotherapy schedule groups, and the posted summary does not provide a single adjusted comparative safety analysis. The subgroup result for PD-L1 >=1% is not accompanied by a formal treatment-by-subgroup interaction analysis in the registry record. These limits constrain interpretation of subgroup differences and the broader clinical implications of the pCR result.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of Nivolumab Versus Placebo in Combination With Neoadjuvant Chemotherapy and Adjuvant Endocrine Therapy in Participants With High-risk, Estrogen Receptor-Positive (ER+), Human Epidermal Growth Factor Receptor 2-Negative (HER2-) Primary Breast Cancer — NCT04109066.
• Registry record last updated 2025-01-08; status last verified 2024-12.
• Results first posted 2024-03-12.
• 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 NCT04109066 as a completed, randomized, double-blind, parallel-group phase 3 study of nivolumab versus placebo with neoadjuvant chemotherapy and adjuvant endocrine therapy in high-risk ER-positive, HER2-negative primary breast cancer. The recorded enrollment is 521.

The primary endpoint was pCR, assessed up to approximately 37 months. The posted percentages were 24.5% for Arm A and 13.8% for Arm B. The sponsor reported an odds ratio of 2.05 and an adjusted difference of pCR rates of 10.5. The unadjusted recalculation also found a difference, with a 95% confidence interval that excludes zero.

Secondary results include a PD-L1 >=1% pCR subgroup and deaths; the death comparison is imprecise and its confidence interval includes zero. Detailed statistical plans for multiplicity, missing data and the original sample-size justification are not reported in the registry record.