Eribulin in HER2 Negative Metastatic BrCa: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Eribulin in HER2 Negative Metastatic BrCa (NCT01827787), a non-randomized phase 2 study. The primary outcome is Overall Response Rate (ORR).

2. Clinical Question

The study focused on eribulin in an earlier-line setting for metastatic breast cancer. The registry describes one cohort as HR+/HER2- and the other as TNBC. These are separate clinical groups, not treatment and control arms in a randomized comparison. The record does not report how participants were assigned to cohorts or whether cohort eligibility depended on additional characteristics.

Overall response rate is the primary endpoint. The registry also lists progression-free survival, time to first response, duration of overall response, treatment-related peripheral neuropathy, and changes in breast cancer and neurotoxicity patient-reported outcome subscales as secondary outcomes. Although the study title refers to efficacy, toxicity and patient-reported outcomes, the numerical results for these outcomes are not reported in the registry record.

The clinical question can therefore be stated, but the registry record does not supply the primary response estimate or enough information to quantify its uncertainty. The design describes what was studied; it does not by itself establish the size of any treatment effect or support a comparison with another therapy.

3. Trial Design

This completed phase 2 study enrolled patients with HER2-negative metastatic breast cancer. The record lists Dana-Farber Cancer Institute as sponsor, non-randomized allocation, no masking, and a parallel model. Its recorded start was 2013-05 and primary completion was 2016-05. Enrollment is listed as 83.

Participants were organized into Cohort 1: HR+/HER2- and Cohort 2: TNBC. Both cohorts received eribulin at 1.4 mg/m2 intravenously over 2-5 minutes on days 1 and 8 of each 21 day cycle. Treatment was single-agent eribulin and continued until disease progression or withdrawal for other reasons. The registry record does not report the criteria used to assign participants to cohorts, nor does it describe a separate control treatment.

Disease was evaluated radiologically at baseline and every 9 weeks on treatment. The registry reports a maximum treatment duration of 38 cycles/26 months for Cohort 1 and 17 cycles/12 months for Cohort 2. These are stated maximum durations, not summaries of how long participants typically remained on treatment.

Non-randomized allocation means the cohorts should not be interpreted as randomized treatment arms. The study can describe outcomes within the enrolled groups, but differences between those groups could reflect differences in their clinical characteristics as well as other factors. Masking was absent, so the record does not indicate that participants or study personnel were blinded to treatment. The registry does not report baseline characteristics, cohort assignment procedures, or a statistical analysis plan that would clarify how potential group differences were handled.

The phase and design identify the general structure of the investigation, but they do not establish how outcomes were analyzed. Details such as eligibility criteria, analysis populations, handling of missing observations, and any planned interim analyses are not reported in the registry record.

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

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Overall Response Rate (ORR), measured over Disease was evaluated radiologically at baseline and every 9 weeks on treatment; Maximum treatment duration was 38 cycles/26 months (Cohort 1) and 17 cycles/12 months (Cohort 2).

The primary endpoint, overall response rate, is the percentage of participants who achieved a complete response or partial response on treatment. Response was assessed using RECIST 1.1. For target lesions, a complete response means disappearance of all target lesions. A partial response means at least a 30% decrease in the sum of longest diameter of target lesions, using the baseline sum as the reference. A response also requires, at minimum, incomplete response or stable disease for non-target lesions and no new lesions.

Response rate is used to describe how often an observed tumor response meets prespecified criteria. It condenses response status into a percentage, making the endpoint straightforward to interpret as a group-level outcome. It does not, by itself, describe how long a response lasts, whether disease later progresses, or how participants feel and function. The registry lists progression-free survival, duration of overall response, and patient-reported outcomes separately, which address different aspects of outcome.

Radiological evaluations were scheduled at baseline and every 9 weeks on treatment. The primary endpoint includes a 90% confidence interval as its stated measure of dispersion. However, the numerical overall response rate and its confidence interval are not reported in the registry record, so the observed response frequency and its precision cannot be assessed from this record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does overall response rate measure?

Overall response rate is a binary outcome at the participant level: a participant either meets the response criteria or does not. At the group level, the endpoint is expressed as the percentage achieving complete or partial response. The registry defines response using RECIST 1.1 and includes requirements for target lesions, non-target lesions, and new lesions. Applying a stated response definition supports consistent classification, although the registry record does not report details of review procedures or whether response assessments were independently reviewed.

A response percentage is a summary of observed outcomes in the analyzed group. It is not the same as an estimate of treatment benefit compared with an untreated or differently treated group. A control group and a suitable comparison are generally needed to estimate comparative effects, and this study is described as non-randomized without a separate comparator.

How should a confidence interval be understood?

The primary endpoint specifies a 90% confidence interval. In general, a confidence interval describes uncertainty around an estimated quantity using a statistical procedure. It is not a guarantee about an individual participant's outcome. Its width conveys precision: narrower intervals indicate more precise estimates, while wider intervals indicate greater uncertainty. The actual interval is not reported in the registry record, so its width and the precision of the response estimate cannot be evaluated.

Confidence intervals also depend on the analysis method and the data included. For a percentage based on a binary outcome, the number of participants with outcome data and the number meeting the response definition affect the estimate and its uncertainty. Those outcome counts and analysis details are not reported in the registry record. No recalculated estimate can be described from the information recorded.

What does the design imply for comparisons?

Random allocation is one way to make treatment groups comparable, on average, with respect to measured and unmeasured factors. This study did not use random allocation. Its parallel cohorts received the same intervention, and the registry does not describe a control treatment. The cohorts therefore provide separate descriptions of outcomes in different clinical groups, not a randomized contrast. Differences between their observed outcomes, if available, could have several explanations, including differences in participant characteristics. The record does not report the information needed to assess those possibilities.

Masking was listed as none. In general, masking can reduce the influence of knowledge about treatment on some assessments or decisions. Its absence does not automatically invalidate an outcome, but it is relevant context when interpreting assessments and reporting. The registry does not provide additional detail about safeguards used in outcome assessment.

What can the planning illustrations tell readers?

The planning illustrations use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They describe planned sample sizes for detecting a stated fraction of an observed effect. The registry’s assumptions identify these as planning illustrations computed from the observed effect, not as the trial's own design. The actual numerical illustrations and observed effect are not reported in the registry record.

Such sample-size calculations are conditional on their assumptions. They ask how many participants per arm would be needed under a specified planning scenario; they do not measure the probability that this completed study detected an effect. They are not observed power. In particular, a planning illustration based on an observed effect should not be read as an independently established target or as proof that the trial could reliably detect a clinically meaningful difference.

The equal-allocation assumption belongs to the planning illustration. It should not be treated as a description of observed cohort sizes. The study itself is non-randomized and organized by clinical cohort, and the registry does not identify a randomized two-arm comparison.

What analysis details remain unclear?

The registry lists one analysis, but does not report the statistical method used for the primary outcome, the analysis population, or the handling of missing response assessments. A complete statistical interpretation would ordinarily distinguish the people enrolled, treated, and included in each endpoint analysis. Those populations and their denominators are not reported in the registry record. The timing and rules for response assessment are partly described, but the record does not explain how unavailable or unevaluable assessments were handled.

The registry names secondary outcomes spanning time-to-event outcomes, safety outcomes, and patient-reported scores. These outcome types require different summaries and assumptions. For example, time-to-event analyses account for both the time until an event and participants whose event status is not observed over the full period. The registry record does not report the analysis methods or results for these secondary outcomes, so their treatment cannot be assessed.

In general, statistical methods should be interpreted alongside outcome definitions, analysis populations, missing-data handling, and uncertainty estimates. Here, the registry gives some endpoint and design information but does not report several details needed to reproduce or fully evaluate the analysis. No further method should be assumed from the phase designation or the outcome names alone.

10. Exact Estimates for Each Arm

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

R code and output

arm    <- c("Cohort 1: HR+/HER2-", "Cohort 2: TNBC")
n      <- c(45, 38)
events <- c(16, 5)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Cohort 1: HR+/HER2-  45  16  35.6  21.9  51.2  23.2  50.2  23.2  50.2
ARM  Cohort 2: TNBC  38  5  13.2  4.4  28.1  5.8  27.3  5.9  27.4

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and several secondary outcomes, including time-to-event measures, safety outcomes, and patient-reported scores. When a study evaluates multiple outcomes, the chance of at least one apparently notable result can increase if each outcome is tested separately without adjustment. Statistical plans may address this issue through a prespecified hierarchy, adjustment procedure, or descriptive interpretation. Any multiplicity strategy for this study is not reported in the registry record.

The primary endpoint is specified, but the record does not report a statistical testing plan, whether hypothesis tests were used, or any adjustment for multiple comparisons. It also does not describe interim analyses. Those omissions prevent assessment of how the primary and secondary outcomes were prioritized or how multiple findings were intended to be interpreted.

Missing data can affect both the estimated response percentage and patient-reported outcomes. A full assessment would require knowing who was eligible for each analysis, how many had evaluable measurements, and how missing or unevaluable observations were handled. The analysis populations and missing-data procedures are not reported in the registry record. The record also does not state whether all enrolled participants were included in each outcome analysis.

Because these details are unavailable, readers should not assume that every enrolled participant contributed to every endpoint, or that missing outcomes were handled in a particular way. The registry’s endpoint definitions and design information do not replace a documented analysis plan.

12. 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: neutrophil count decreased (35.6%), peripheral motor neuropathy (6.7%), anemia (5.3%), febrile neutropenia (5.3%), peripheral sensory neuropathy (5.3%).

The registry reports serious adverse event figures separately for Cohort 1: HR+/HER2- and Cohort 2: TNBC. In Cohort 1, 22 participants were affected among 45 at risk for serious adverse events; in Cohort 2, 13 participants were affected among 38 at risk. For other adverse events, the record lists 45 affected among 45 at risk in Cohort 1 and 37 affected among 38 at risk in Cohort 2. These are reported counts, not comparisons against a control group.

The registry lists neutrophil count decreased at 35.6% among its top serious adverse event terms. It also lists peripheral motor neuropathy at 6.7%, anemia at 5.3%, febrile neutropenia at 5.3%, peripheral sensory neuropathy at 5.3%, and vomiting at 5.3%. The denominator and further details for these percentages are not reported in the registry record. The terms should not be treated as a complete description of all adverse events.

Deaths are listed as 4 for Cohort 1 and 9 for Cohort 2. The registry record does not report whether these deaths were attributed to treatment. Seriousness, treatment relatedness, and cause are distinct classifications; counts alone do not establish causation or comparative risk. The record does not provide a separate control group or the details needed to compare safety across treatments.

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

The study record documents an investigation of single-agent eribulin in two cohorts of patients with HER2-negative metastatic breast cancer. It defines the primary outcome as overall response rate using RECIST 1.1 and describes the treatment schedule and radiological assessment timing. These details clarify what the investigators intended to measure and how the treatment was administered.

The registry record does not report the numerical primary response result or its confidence interval. It also does not report numerical results for the listed secondary outcomes. As a result, the record does not allow readers to determine the observed response frequency, the precision of that estimate, or the outcomes for progression-free survival, time to response, duration of response, neuropathy, or patient-reported scores.

The design does not establish a randomized comparison. Both cohorts received eribulin, and the record does not describe a separate control treatment. Any contrast between the cohorts would not isolate a treatment effect because participants were not randomly assigned and the groups represent different clinical categories. The registry does not report the information needed to assess whether other characteristics differed between them.

The safety counts and serious event terms describe what the registry records, but they do not demonstrate that eribulin caused each event. Nor do they establish comparative safety without a control group and additional context about event attribution, exposure, and follow-up. The recorded death counts do not specify cause or treatment relationship.

Thus, the record is useful for understanding the trial question, structure, endpoint definition, and treatment regimen. It is not sufficient to quantify efficacy or to establish a causal benefit or harm. The absent results and analysis details should not be filled in by inference from the title, background description, or study design.

14. Why This Trial Matters Statistically

This trial illustrates why a clearly defined endpoint and a clearly described comparison are both important. Overall response rate has an explicit RECIST 1.1 definition, including criteria for target and non-target lesions and new lesions. That definition makes the intended outcome understandable. But an endpoint definition cannot substitute for reporting the observed result, its denominator, and its uncertainty. Those primary outcome results are not reported in the registry record.

The non-randomized structure is also important for statistical interpretation. The two cohorts represent different clinical groups and both received eribulin. They do not form randomized treatment and control arms. This distinction limits what can be learned from comparisons between groups and helps prevent descriptive differences from being mistaken for causal treatment effects.

The study also names outcomes beyond response, including progression-free survival, duration of response, neuropathy, and patient-reported changes. These measure different dimensions of outcome and need appropriate definitions and analyses. The registry record’s list of outcomes signals the breadth of questions considered, but the numerical results and analysis methods for those outcomes are not reported.

Finally, the record makes visible the difference between a planning illustration and an observed analysis. Sample-size illustrations based on a stated fraction of an observed effect are planning scenarios, not observed power and not evidence that an effect was detected. Clear separation of design assumptions, observed results, and uncertainty is central to responsible interpretation. In this record, the design and endpoint information can be described, while the missing outcome estimates limit conclusions about what the study found.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

A central limitation for interpretation is that the registry record does not report the primary overall response rate or its 90% confidence interval. It also does not report the numerical results for the listed secondary outcomes. Without these results, the magnitude and precision of observed outcomes cannot be evaluated.

The study is non-randomized and has no separate control treatment described. The two cohorts are HR+/HER2- and TNBC, so they are not like-for-like randomized comparison groups. The registry does not report cohort assignment criteria or baseline characteristics that would help readers understand differences between the groups.

Several analysis details are absent. The record does not report the analysis populations, the numbers evaluable for each outcome, how missing or unevaluable observations were handled, or the statistical methods used for the primary endpoint. Multiplicity procedures and interim analyses are also not reported in the registry record. These gaps restrict evaluation of how robust or reproducible any outcome analysis would be.

The safety record provides counts and selected serious event terms, but does not supply the context needed to interpret attribution or compare risk with another treatment. The relationship of the listed deaths to treatment is not reported. Eligibility criteria and further population characteristics are not reported either, limiting assessment of how broadly the study population represents the intended clinical population.

These limitations concern what can be concluded from the registry record; they do not establish that the study lacked methods or analyses beyond those recorded.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Eribulin in HER2 Negative Metastatic BrCa — NCT01827787.
• Registry record last updated 2024-11-06; status last verified 2024-11.
• Results first posted 2019-02-06.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

This completed phase 2 study evaluated single-agent eribulin in patients with HER2-negative metastatic breast cancer. It used non-randomized, unmasked, parallel cohorts: HR+/HER2- and TNBC. Enrollment is listed as 83, and Dana-Farber Cancer Institute is the sponsor. The recorded study start was 2013-05, with primary completion in 2016-05.

Eribulin was administered intravenously on days 1 and 8 of each 21 day cycle. The primary endpoint was overall response rate by RECIST 1.1, with a 90% confidence interval specified. The registry also lists progression, response timing and duration, neuropathy, and patient-reported outcomes.

The numerical primary and secondary outcome results are not reported in the registry record. The record does not describe a separate control treatment, analysis populations, missing-data handling, or a multiplicity strategy. Safety counts and selected serious adverse event terms are listed, but treatment attribution for deaths is not reported.