Comparison of Neoadjuvant Chemotherapy With Weekly Paclitaxel or Eribulin Followed by A/C in Women With Locally Advanced HER2-Negative Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparison of Neoadjuvant Chemotherapy With Weekly Paclitaxel or Eribulin Followed by A/C in Women With Locally Advanced HER2-Negative Breast Cancer (NCT01705691), a randomized phase 2 study. The primary outcome is Pathologic Complete Response Rate (ypCR) Following Neoadjuvant Therapy in Breast and Axillary Lymph Nodes.

2. Clinical Question

This was a phase II randomized comparison of two treatment sequences before surgery. Both groups received doxorubicin and cyclophosphamide after the first drug sequence; the randomized contrast was whether the initial chemotherapy was weekly paclitaxel or eribulin. The primary endpoint therefore describes pathologic response after the full neoadjuvant regimen, not response to the initial drug alone.

The question is about the observed response proportions and the uncertainty around their between-group difference. The registry record reports a response rate of 26.3% in the paclitaxel-then-AC group and 16.7% in the eribulin-then-AC group. Those descriptive percentages do not by themselves establish that the underlying response rates differ. The comparison is based on a relatively small phase II trial, and the recorded interval for the difference includes values on both sides of no difference.

Other questions were also listed, including response in axillary nodes, clinical response assessed at different treatment times, recurrence-free interval, overall survival, and adverse events. These are useful for understanding the scope of the trial, but they do not change the primary question: whether the two randomized sequences differed in pathologic complete response in breast and axillary lymph nodes at surgery.

3. Trial Design

The registry describes a phase II, randomized, parallel-group clinical trial with no masking. It was sponsored by NSABP Foundation Inc and enrolled 50 participants. The recorded start was 2012-09, primary completion was 2014-03, and trial status is completed. The registry record does not report the original sample-size justification or a detailed statistical analysis plan.

Participants were assigned to one of two neoadjuvant treatment sequences. Arm 1 received paclitaxel 80 mg/m2 intravenously weekly for 12 doses, then doxorubicin and cyclophosphamide intravenously every 21 days for 4 cycles. Arm 2 received eribulin 1.4 mg/m2 intravenously on days 1 and 8 of a 21-day cycle for 4 cycles, then doxorubicin and cyclophosphamide intravenously every 21 days for 4 cycles. The record characterizes the control group as weekly paclitaxel followed by AC.

The trial concerned women with HER2-negative, operable and locally advanced breast cancer; the summary specifies stage IIb and III. The primary endpoint was assessed at the time of surgery, approximately 24 to 28 weeks after treatment began. One participant in Arm 1 was not analyzed for the primary endpoint because of inoperable, progressive disease. Two participants in Arm 2 had missing data for that endpoint. The reported primary-outcome denominators were 19 and 30, respectively.

Randomization supports a comparative interpretation, but the registry record does not give details about the randomization procedure, allocation concealment, stratification, or masking beyond stating that there was no masking. The results are therefore best read as a small phase II comparison of the assigned treatment sequences, with attention to the available outcome denominators and uncertainty in the estimates.

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: One patient randomly assigned to Arm 2 withdrew consent before receiving protocol therapy.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 1 patient in Arm 1 was not analyzed due to inoperable, progressive disease. 2 patients in Arm 2 are missing data.

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 Pathologic Complete Response Rate (ypCR) Following Neoadjuvant Therapy in Breast and Axillary Lymph Nodes, measured over At the time of surgery approximately 24 to 28 weeks..

The primary endpoint was the percentage of participants with no histologic evidence of cancer in either the breast or axillary lymph nodes after neoadjuvant treatment. This pathologic complete response is abbreviated ypCR in the registry record. The endpoint was assessed at surgery, approximately 24 to 28 weeks after treatment began.

A pathologic assessment uses tissue examined at surgery to determine whether cancer is present in the specified sites. This differs from a clinical response assessment, which evaluates disease before surgery using examination or imaging. The primary definition is specific to both the breast and axillary lymph nodes: a response in one site alone would not meet the endpoint as described.

Pathologic response is used in this trial as an outcome following treatment before surgery. It gives a direct measure of whether cancer was detected histologically in the target sites at the planned assessment. It is not the same outcome as recurrence-free interval or overall survival, which concern events over follow-up, and it should not be interpreted as a substitute for those outcomes.

The registry reports response percentages of 26.3% for paclitaxel followed by AC and 16.7% for eribulin followed by AC. The reported denominators differ from the total enrollment, and the endpoint description identifies one participant not analyzed in Arm 1 and missing data for two participants in Arm 2. The record does not give further detail about the handling of missing primary-endpoint information or an analysis population definition.

7. Results Posted in the Registry

• Arm 2: Eribulin Then AC: 16.7% (5/30) versus 26.3% (5/19) on control; risk difference -9.6 percentage points (95% CI -33.5 to 14.2), risk ratio 0.63 (0.21–1.90); unadjusted two-proportion z-test p = 0.414.

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.

ypCR Nodes

• 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 2: Eribulin Then AC: 30.0% (9/30) versus 42.1% (8/19) on control; risk difference -12.1 percentage points (95% CI -39.7 to 15.5), risk ratio 0.71 (0.33–1.52); unadjusted two-proportion z-test p = 0.386.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Arm 2: Eribulin Then AC and the control. The interval gives the range of effects that remain compatible with the data.

Clinical Overall Response (cOR)(Complete and Partial) Assessed by MRI at the Completion of WP or Eribulin (Before AC)

• 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 2: Eribulin Then AC: 40.0% (12/30) versus 57.9% (11/19) on control; risk difference -17.9 percentage points (95% CI -46.2 to 10.4), risk ratio 0.69 (0.39–1.24); unadjusted two-proportion z-test p = 0.221.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Arm 2: Eribulin Then AC and the control. The interval gives the range of effects that remain compatible with the data.

Clinical Complete Response (ycCR) Following Neoadjuvant Therapy Assessed by Physical Exam at the Completion of Neoadjuvant Chemotherapy

• 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 2: Eribulin Then AC: 33.3% (10/30) versus 44.4% (8/18) on control; risk difference -11.1 percentage points (95% CI -39.6 to 17.4), risk ratio 0.75 (0.36–1.55); unadjusted two-proportion z-test p = 0.441.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Arm 2: Eribulin Then AC and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover several distinct parts of the treatment course: pathologic response in nodes, clinical response before AC, clinical complete response after neoadjuvant treatment, recurrence-free interval, overall survival, and adverse events. The posted numerical comparisons for nodal response and the two clinical response outcomes have the same broad statistical feature as the primary result: their intervals include the null value, and their unadjusted p-values do not provide clear evidence of a between-group difference.

These outcomes relate to the primary endpoint but are not interchangeable with it. Nodal ypCR concerns one anatomical component of pathologic response, whereas the primary endpoint requires no histologic evidence of cancer in both breast and axillary lymph nodes. Clinical response is assessed before surgery and does not use the same tissue-based definition. Agreement or disagreement across these endpoints would not, by itself, establish a difference in long-term outcomes.

The registry also lists recurrence-free interval and two-year overall survival, but quantitative outcome results for these endpoints are not reported in the registry record. The listed outcomes therefore describe intended areas of assessment, not evidence of a comparative survival result. The same caution applies to any outcome for which no numerical result is posted.

Unadjusted secondary comparisons carry less weight because multiple outcomes and assessment times increase the opportunity for chance findings. The registry record does not report a multiplicity adjustment, a hierarchy for testing secondary outcomes, or reported analyses for them. Without such information, an isolated secondary p-value should not be treated as confirmatory evidence. The secondary findings are best considered descriptive and interpreted in the context of the primary estimate, its uncertainty, and the limited trial size.

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

• ypCR Nodes
• Clinical Overall Response (cOR)(Complete and Partial) Assessed by MRI at the Completion of WP or Eribulin (Before AC)
• Clinical Complete Response (ycCR) Following Neoadjuvant Therapy Assessed by Physical Exam at the Completion of Neoadjuvant Chemotherapy
• Recurrence Free Interval (RFI): The Time to Occurrence of Inoperable Progressive Disease and Local, Regional, and Distant Recurrence.
• 2-year Overall Survival (OS): Death From Any Cause From Time of Randomization Through 2 Years After Randomization.
• Adverse Events Experienced by Participants as a Measure of Toxicity.

9. Statistical Interpretation

10. Statistical Methods Explained

Randomization assigns participants to treatment groups by a chance-based process, with the aim of making groups comparable at the start of a trial. In this study, the randomized groups received different initial chemotherapy sequences and both then received AC. Randomization helps frame the comparison, but it does not eliminate uncertainty, especially when group sizes and outcome counts are limited. The registry record does not report the randomization procedure or allocation concealment details.

What does a response percentage represent?

For a binary endpoint, each participant either meets the response definition or does not. A response percentage expresses the observed frequency of participants who met that definition among those counted in the group. For the primary endpoint, the registry reports the percentage with no histologic evidence of cancer in breast and axillary lymph nodes. The reported values are 26.3% and 16.7% in the two groups. These are observed proportions, not guarantees about the response rate in a larger population.

The outcome denominators matter because they identify how many participants contributed to each reported percentage. The primary endpoint lists 19 participants in the paclitaxel group and 30 in the eribulin group. The registry notes that one participant in Arm 1 was not analyzed because of inoperable, progressive disease and that two participants in Arm 2 had missing data. How missing outcomes were handled in the original analysis is not reported in the registry record.

What is a risk difference?

A risk difference compares the response proportion in one group with the response proportion in the other. A value of zero represents equal observed proportions. A negative estimate in this comparison indicates that the response proportion in the eribulin group was lower than in the paclitaxel group; a positive value would indicate the reverse. The posted primary risk-difference interval crosses zero, which means the data are compatible with no difference and with differences in either direction. The interval is not a prediction of the result for an individual participant.

How should a p-value be read?

The primary unadjusted p-value is 0.414. A p-value describes how unusual the observed data, or data more extreme, would be under a specified null model of no group difference. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. The registry record does not report the trial's prespecified testing threshold or the details of the method used for the primary comparison, so the posted p-value should be read as an unadjusted result rather than as a complete account of the analysis plan.

A p-value does not replace the estimate and its interval. The estimate indicates the direction and magnitude represented by the observed data, while the interval indicates the range of values compatible with the data under the interval method. When the interval spans the null value, as it does for the primary risk difference, the data do not distinguish clearly between no difference and potentially meaningful effects in either direction.

Why do confidence intervals matter?

A confidence interval gives a range of values that are reasonably compatible with the observed data under the statistical method used. Its width reflects precision: narrower intervals indicate more precise estimates, while wider intervals indicate greater uncertainty. The primary interval includes values on both sides of no difference. It therefore does not support a confident claim that one sequence has a higher response rate, nor does it demonstrate that the sequences are the same.

Confidence intervals also help keep attention on the scale of the outcome. A risk difference is an absolute comparison of proportions. Other measures, such as a risk ratio or odds ratio, compare groups on different relative scales. Those measures can tell a similar broad story while having different numerical values and interpretations. They should not be treated as interchangeable or used to make a treatment recommendation.

What does unadjusted mean?

An unadjusted comparison is presented without statistical adjustment for other participant characteristics or additional modeling factors. It provides a direct comparison of the posted group summaries. Whether adjusted analyses were planned or performed is not reported in the registry record. The absence of reported analyses for the secondary outcomes makes it especially important not to interpret their simple group comparisons as adjusted or confirmatory findings.

These concepts are general statistical explanations. For this trial, the registry record reports the group percentages, comparison estimates, intervals, and unadjusted p-values, but does not report a detailed statistical methods plan. No conclusions about unreported modeling choices, interim analyses, or sample-size justification can therefore be drawn from the record.

11. Confidence Intervals and Precision

The primary response percentages are based on 19 participants in the paclitaxel group and 30 in the eribulin group. With these denominators, the observed percentages are subject to substantial sampling uncertainty: a different group of participants could produce different response rates. The reported risk-difference interval extends across zero, so the data are compatible with the eribulin group having a lower response rate, with little or no difference, or with a higher response rate.

This interval is important alongside the point estimate. The rates of 26.3% and 16.7% show the observed direction in this trial, but the interval shows that the comparative estimate is not precise enough to distinguish a difference from no difference. The unadjusted p-value of 0.414 is consistent with that uncertainty. Neither the interval nor the p-value establishes equivalence.

Precision is also affected by missing endpoint information. The registry states that one participant in Arm 1 was not analyzed because of inoperable, progressive disease and two participants in Arm 2 had missing data. The record does not report a sensitivity analysis or a detailed missing-data method. As a result, the possible influence of those unavailable outcomes cannot be assessed from the posted result alone.

General statistical interpretation: confidence intervals quantify uncertainty under a specified method; they do not account automatically for every design limitation, outcome-selection decision, or source of bias. Here, the randomized design supports a comparison, but the small phase II enrollment, the outcome denominators, and the unreported analysis details limit precision and the strength of conclusions.

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 sample-size figures are planning illustrations based on the observed effect, not the trial's original design calculations. They assume a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Each planning row considers detecting the stated fraction of the observed effect. The original trial's own sample-size assumptions are not reported in the registry record.

For these binary outcomes, the planning figures are participants per arm, not total event counts. The 80% power column means the number of participants per arm estimated to give an 80% chance of detecting the specified effect under those planning assumptions, if that effect were the true effect. The 90% power column describes the corresponding estimate for a 90% chance. These are design-planning quantities, not probabilities that the observed result is correct.

The rows for 100%, 75%, and 50% of the observed effect show how the estimated participant requirement changes when the target effect is reduced. Smaller effects are harder to distinguish from random variation, so they require more participants for the same planned power. The displayed figures rise as the assumed effect becomes smaller. This is a general feature of sample-size planning, not evidence that the original trial was designed to detect those effects.

These illustrations do not represent observed power. Observed power is not a useful substitute for reading the estimate, interval, and p-value after a study is completed. The original study enrolled 50 participants, whereas the planning calculations are hypothetical and use the observed effect as their starting point. They should be read as illustrations of the scale of a future study under the listed assumptions, not as a reconstruction of this trial's design.

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 = 0.263; treatment proportion p₁ = 0.167; average p̄ = 0.215.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5809 = 1.1386.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5769 = 0.4855.
4. n = (1.1386 + 0.4855)² / (-0.0965)² = 283.3, rounded up to 284 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 <- 0.263157894736842                          # control proportion
p1 <- 0.166666666666667                          # 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(1705691)
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 = 284 (80% power), 379 (90% power)
 75% of observed effect: n per arm = 525 (80% power), 702 (90% power)
 50% of observed effect: n per arm = 1,226 (80% power), 1,641 (90% power)
power.prop.test: 283.3 
Simulated power at n = 284 per arm: 0.799 (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. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates several outcomes, time points, or comparisons. Even if each statistical test is considered separately, the chance of at least one chance finding can increase as more tests are conducted. This trial listed a primary pathologic response outcome and multiple secondary outcomes, including nodal response, clinical response at different times, recurrence-free interval, overall survival, and toxicity.

The registry record does not report a multiplicity adjustment, a testing hierarchy, or a plan specifying which secondary comparisons were intended to be confirmatory. It also reports no secondary analyses. Consequently, the secondary comparisons should be interpreted cautiously and should not be viewed as independent proof of a treatment difference. Their main value is descriptive and hypothesis-generating.

Missing outcome information is relevant to the primary result. The record says one participant in Arm 1 was not analyzed for the primary outcome because of inoperable, progressive disease, while two participants in Arm 2 had missing data. It does not report the analysis population definition, the method used to handle missing values, or a sensitivity analysis assessing how alternative assumptions might affect the result. Those details are not reported in the registry record.

It is also not reported whether interim analyses were conducted, how randomization was implemented, whether any prespecified subgroup analyses were planned, or what original sample-size justification was used. These omissions do not invalidate the posted estimates, but they limit the ability to judge how the reported comparisons relate to the full analysis plan. The safest interpretation is to use the reported denominators and estimates while recognizing that unreported analysis choices cannot be reconstructed.

15. 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: febrile neutropenia (5.3%), pulmonary embolism (5.3%), colitis (5.3%), dyspnoea (3.3%), pain in extremity (3.3%).

The registry reports serious adverse events in 3 of 19 participants in the paclitaxel-then-AC group and 4 of 30 participants in the eribulin-then-AC group. Other adverse events are recorded for 19 of 19 participants in the first group and 30 of 30 in the second. These figures describe participants with reported events, not the number of distinct event episodes or the severity of every event.

The serious adverse event terms listed include febrile neutropenia, pulmonary embolism, and colitis at 5.3% each, and dyspnoea, pain in extremity, and nephrolithiasis at 3.3% each. The registry record does not assign these listed terms to a particular group in the posted summary, so they should not be attributed to either arm individually. Death counts are not reported in the registry record.

Safety counts are descriptive. They do not establish that the regimens have equal or different safety profiles, particularly in a trial of this size. The figures should be read as reported events during the study, not as a complete account of comparative risk or a basis for individual treatment decisions.

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

The posted primary result is a comparison of pathologic complete response rates after two neoadjuvant treatment sequences. The observed rate was 26.3% with paclitaxel followed by AC and 16.7% with eribulin followed by AC. The unadjusted p-value is 0.414, and the interval for the between-group difference includes no difference. The result therefore does not provide clear statistical evidence that the response rates differed, while also leaving uncertainty about the size and direction of any underlying difference.

That interpretation is not the same as concluding that the treatment sequences are equivalent. Demonstrating equivalence or non-inferiority requires a design and analysis framework that defines an acceptable margin and evaluates the result against it. Such a framework is not reported in the registry record. A non-significant comparison alone cannot show that two approaches have the same effect.

The result also does not answer whether one sequence improves recurrence-free interval or overall survival. Those outcomes are listed as secondary aims, but their quantitative results are not reported in the registry record. Pathologic response at surgery and longer-term outcomes are distinct measures. The posted response comparison should not be used as a substitute for missing survival results.

Because both groups received AC after the initial chemotherapy, the primary comparison concerns the complete assigned sequences. It does not isolate the effect of the first drug from the later treatment. The result also applies to the trial population and setting described in the record; the data do not establish how the finding would generalize to other populations or treatment settings.

The small phase II enrollment and the primary outcome denominators make the estimates imprecise. The record does not report the original sample-size justification, detailed analysis plan, or handling of all missing information. Overall, the posted data support a cautious descriptive reading: response rates were observed in both groups, but the comparative estimate is uncertain and does not establish superiority, equivalence, or a long-term outcome difference.

17. Why This Trial Matters Statistically

This trial illustrates why a randomized comparison should be read through both its point estimate and its uncertainty. The observed response percentages differ in direction, but the interval crosses the no-difference value and the unadjusted p-value is 0.414. Reporting only the percentages could make the contrast seem more conclusive than the data support; reporting only the p-value would hide the direction and scale of the observed rates.

It also shows the importance of defining the endpoint precisely. The primary outcome required no histologic evidence of cancer in both breast and axillary lymph nodes at surgery. Nodal response and clinical response are related outcomes, but each answers a different question. Keeping those definitions distinct prevents clinical assessments at one treatment time from being mistaken for the primary pathologic endpoint.

The registry record further demonstrates why denominators and missing-data information matter. The reported primary-outcome groups contain 19 and 30 participants, while the overall enrollment was 50; the record notes one participant not analyzed in Arm 1 and missing data for two participants in Arm 2. Clear reporting of these details helps readers understand which participants contributed to the estimate and what remains uncertain.

Finally, planning calculations derived from an observed effect should not be confused with the original trial design or with observed power. The registry does not report its original sample-size justification. Separating the hypothetical planning illustration from the randomized trial's actual enrollment keeps post-study interpretation aligned with what the record can establish.

18. 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.

Several limitations are specific to interpreting this registry record. It reports a phase II trial with total enrollment of 50 and primary-outcome denominators of 19 and 30. The response comparison is therefore imprecise, as reflected by an interval spanning the null value. The record does not report the original sample-size calculation or the intended effect size used in planning.

The primary endpoint has some unavailable information: one participant in Arm 1 was not analyzed because of inoperable, progressive disease, and two participants in Arm 2 are described as having missing data. The registry does not explain how these cases entered the analysis, whether any imputation or sensitivity analyses were used, or whether the reported denominators reflect a prespecified analysis population.

The analysis details are limited. The registry does not report the randomization procedure, allocation concealment, a full statistical analysis plan, or the method used to construct the comparison intervals. No reported analyses are listed for secondary outcomes. Multiplicity handling is also not reported. These gaps restrict evaluation of the analysis choices and the confirmatory status of secondary findings.

Longer-term outcomes are listed, including recurrence-free interval and two-year overall survival, but quantitative results for those endpoints are not reported in the registry record. The safety information is presented as participant counts and selected serious-event terms, without enough detail in the posted summary to assess event timing, recurrence, or group attribution for the listed terms. These limits do not change the reported response estimates, but they narrow the conclusions that can be drawn from them.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Comparison of Neoadjuvant Chemotherapy With Weekly Paclitaxel or Eribulin Followed by A/C in Women With Locally Advanced HER2-Negative Breast Cancer — NCT01705691.
• Registry record last updated 2021-10-06; status last verified 2021-10.
• Results first posted 2018-09-18.
• 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.

22. Record Summary

NCT01705691 is a completed phase II randomized, open-label, parallel-group trial sponsored by NSABP Foundation Inc. It enrolled 50 women with locally advanced HER2-negative breast cancer and compared weekly paclitaxel followed by doxorubicin and cyclophosphamide with eribulin followed by the same later chemotherapy.

The primary endpoint was pathologic complete response in breast and axillary lymph nodes at surgery, approximately 24 to 28 weeks after treatment began. The registry reports rates of 26.3% and 16.7%, respectively, with an unadjusted p-value of 0.414 and an interval for the between-group difference that includes no difference. The result is descriptive and uncertain; it does not establish superiority or equivalence.

Several secondary outcomes are listed, but quantitative results for recurrence-free interval and overall survival are not reported in the registry record. Original design assumptions, detailed analysis methods, and multiplicity procedures are also not reported.