LUX-Breast 2; Afatinib in HER2 (Human Epidermal Growth Factor Receptor)-Treatment Failures: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for LUX-Breast 2; Afatinib in HER2 (Human Epidermal Growth Factor Receptor)-Treatment Failures (NCT01271725), a non-randomized phase 2 study. The primary outcome is Percentage of Participants With Objective Response (OR) Assessed by Response Evaluation Criteria in Solid Tumours Version (RECIST) 1.1.

2. Clinical Question

The study asks whether objective responses are observed during treatment with afatinib in patients with metastatic HER2-overexpressing breast cancer whose disease had failed HER2-targeted treatment in the neoadjuvant and/or adjuvant setting. It also examines safety and other outcomes. For patients whose disease progressed on afatinib monotherapy within the trial, the study allowed treatment with afatinib plus weekly paclitaxel or afatinib plus weekly vinorelbine.

This is a response question rather than a randomized treatment comparison. The monotherapy and combination-treatment groups arose at different stages of the treatment pathway: combination therapy was given after progression on monotherapy. Their outcomes therefore cannot be read as if participants had been randomly assigned to comparable treatments. The record describes the objective-response endpoint as assessed separately in each part.

3. Trial Design

LUX-Breast 2 (NCT01271725) is listed as a Phase II, non-randomized, open-label study with a single-group model. The study was sponsored by Boehringer Ingelheim and is marked completed. Its recorded start date is 2011-05-24, and its primary completion date is 2017-03-13. Enrollment was 74.

The planned treatment pathway began with afatinib monotherapy at 40mg once daily, continued until disease progression. Patients who progressed on monotherapy could additionally receive either paclitaxel 80 mg/m2 weekly or vinorelbine 25 mg/m2 weekly. The registry describes the latter treatments as combination therapy with afatinib. It reports 13 participants at risk in the vinorelbine combination group and 26 in the paclitaxel combination group in the safety information; those figures are not evidence of random allocation.

Because assignment was not randomized and treatment changes followed progression, the study does not provide a like-for-like control comparison between monotherapy and combination therapy. Participants receiving combination treatment had already progressed during monotherapy, so the treatment groups represent different points in the study pathway and potentially different clinical circumstances. The design can describe outcomes observed in the treatment groups, but differences between those groups would not isolate the effect of adding either drug.

The primary endpoint was objective response assessed by RECIST v1.1. The registry also lists response without a confirmation requirement, progression-free survival, duration of objective response, adverse events, blood pressure changes, and laboratory values among secondary outcomes. Numerical results for these outcomes are not reported in the registry record used for this account.

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: All subjects were screened for eligibility. It was not planned to compare efficacy of the combination therapies "Afatinib and Paclitaxel or Afatinib and Vinorelbine" as the study was stopped prematurely and the patient number in both treatment groups is not significant. However, for safety reporting both combination therapy are reported separately

Part A

Reasons for not completing this period:

Part B

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The treated set for monotherapy comprised all patients who received at least 1 dose of afatinib. Treated set for combination therapy comprised all patients who received at least 1 dose of each of afatinib and paclitaxel or of afatinib and vinorelbine. It was not planned to compare between the two types of combination therapies hence arm is combined

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.

Population: The treated set for monotherapy comprised all patients who received at least 1 dose of afatinib. The treated set for combination therapy comprised all patients who received at least 1 dose of each of afatinib and paclitaxel, or of afatinib and vinorelbine.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With Objective Response (OR) Assessed by Response Evaluation Criteria in Solid Tumours Version (RECIST) 1.1, measured over From the initial dose of study drug until 28 days after end of the treatment period, up to 1562 days.

The primary endpoint is the percentage of participants with an objective response assessed by RECIST v1.1. An objective response was defined as a confirmed complete response or confirmed partial response recorded after the first administration of trial medication and before the earliest of disease progression, death, or the start of the next treatment. The registry describes assessment of target lesions by magnetic resonance imaging.

Under the stated RECIST v1.1 definitions, a complete response means disappearance of all target lesions. A partial response means at least a 30% decrease in the sum of the longest diameters of target lesions. The objective-response category combines these two response types. Requiring confirmation distinguishes this primary measure from the listed secondary outcome that counts best overall response regardless of confirmation.

Response percentage is a way to describe how often a predefined tumor-response criterion was met in the assessed participants. It is different from a measure of how long disease remains controlled or how long a response lasts; progression-free survival and duration of objective response are listed separately as secondary outcomes. The primary endpoint timeframe runs from the initial study-drug dose until 28 days after the end of the treatment period, up to 1562 days.

The registry description specifies an exact 95% confidence interval using the Clopper and Pearson method. However, the actual percentage of participants with objective response and its confidence interval are not reported in the registry record. The endpoint definition alone therefore does not establish what response proportion was observed or how precisely it was estimated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a percentage endpoint with an exact 95% confidence interval using the Clopper and Pearson method. It does not report the numerical response result or interval endpoints. The following explains the statistical ideas relevant to the stated endpoint and design; general explanations are identified as such.

What does a response percentage represent?

For a binary outcome such as whether each participant met the objective-response definition, each assessed participant is classified as having or not having the event. The response percentage describes the share classified as responders in the relevant analysis population. Here, the event is a confirmed complete response or confirmed partial response by RECIST v1.1. It is not a measure of the size of tumor change for every participant, and it does not convey how long a response lasted.

The registry defines complete response as disappearance of all target lesions and partial response as at least a 30% decrease in the sum of the longest diameters of target lesions. Combining the categories produces the objective-response endpoint. Confirmation is part of the primary definition; the separately listed response outcome regardless of confirmation uses a different rule. Results from differently defined endpoints should not be treated as interchangeable.

Why report a confidence interval?

General explanation: a confidence interval expresses the uncertainty associated with an estimate. A response percentage calculated from a study group is an estimate, rather than a guarantee of the response frequency in a broader population. The interval communicates a range of values compatible with the observed data under the method's assumptions. Its width matters: greater width indicates less precision, while less width indicates greater precision. A confidence interval does not correct for selection bias, lack of a control group, or limitations in how outcomes were assessed.

The registry specifies an exact Clopper and Pearson interval at the 95% confidence level. General explanation: this method is used for a proportion based on a binary outcome and is designed to give interval coverage without relying on a normal approximation. The method's choice does not reveal the result. Because the registry record does not report the response count, percentage, or interval endpoints, neither the size of the estimate nor its precision can be reviewed.

Why does the design matter for interpretation?

General explanation: randomization helps make treatment groups comparable, on average, with respect to both measured and unmeasured factors. In a non-randomized study, group membership can be related to patient characteristics, disease course, or treatment history. Such differences can influence observed outcomes and complicate causal comparisons. This study is listed as non-randomized and single-group, with no masking.

The treatment sequence is especially important. Participants first received afatinib monotherapy, and paclitaxel or vinorelbine could be added after progression. The combination groups therefore consist of participants who reached that later treatment stage. Comparing their outcomes with outcomes during monotherapy would mix treatment effects with differences in timing and disease status. The registry does not describe a randomized comparison that could separate those influences.

How should the treatment-group counts be understood?

The overall enrollment is 74. For safety, the registry lists 74 at risk in the afatinib monotherapy group, 13 in the afatinib and vinorelbine group, and 26 in the afatinib and paclitaxel group. These are group-specific safety counts, not response results. The registry defines the treated set for monotherapy as patients who received at least one dose of afatinib. For combination therapy, it defines the treated set as patients who received at least one dose of both afatinib and the relevant combination drug.

General explanation: an analysis population is the set of participants whose data are used for a particular outcome. Different definitions can produce different denominators. A treated population, for example, may differ from an enrollment population if some enrolled participants never receive treatment. The registry's descriptions identify treated-set criteria for the primary outcome, but the numerical response denominators and response counts are not reported.

What can sample-size illustrations say?

General explanation: a sample-size calculation can illustrate how many participants might be needed to detect a specified effect under stated assumptions. It is a planning exercise, not an observed result and not evidence that a study had a particular probability of finding an effect after it was completed. The registry's planning illustration is described as based on the observed effect, with equal allocation, a two-sided significance level of 0.05, and a normal approximation; the target effect is a stated fraction of the observed effect. The illustration is not the trial's own design justification. It cannot substitute for an unavailable response estimate or confidence interval.

The registry does not report the study's sample-size justification, and no conclusion about that justification should be inferred. A planning illustration based on an observed effect also depends on the observed effect being available and may not represent a prospectively chosen target. It should be kept separate from the study's actual enrollment and from evidence about endpoint precision.

What does the record leave unresolved?

The primary endpoint description includes the time window and response criteria, but the registry does not report the primary numerical result. The statistical plan beyond the stated interval method, the handling of missing response assessments, and any multiplicity procedure are not reported in the registry record. General explanation: missing outcome data can affect a proportion if participants without an assessment are handled differently from those assessed. Without a reported rule and outcome accounting, readers cannot evaluate that possibility. Similarly, when several outcomes are examined, a multiplicity strategy can affect how inferential claims are interpreted; the record does not provide one.

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("Afatinib Monotherapy", "Afatinib and Paclitaxel or Vinorelbine Combination")
n      <- c(74, 39)
events <- c(13, 12)
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  Afatinib Monotherapy  74  13  17.6  9.7  28.2  10.6  27.8  10.6  27.8
ARM  Afatinib and Paclitaxel or Vinorelbine Combination  39  12  30.8  17.0  47.6  18.6  46.4  18.6  46.5

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary objective-response endpoint and secondary outcomes that include best overall response regardless of confirmation, progression-free survival, duration of objective response, adverse events, blood pressure changes, and laboratory values. It also states that there are nine outcome measures. The numerical results for the primary and secondary efficacy outcomes are not reported in the registry record.

Multiplicity concerns arise generally when several outcomes or analyses are considered. Testing many questions can increase the chance that at least one appears notable by chance. A prespecified hierarchy or adjustment can help define how results should be interpreted. The registry's multiplicity procedure, if any, is not reported in the registry record. The endpoint description names a confidence-interval method, but does not state a broader strategy for multiple outcomes.

The registry defines the treated set for monotherapy as patients who received at least one dose of afatinib. The combination treated set comprises patients who received at least one dose of afatinib and at least one dose of paclitaxel or vinorelbine, as applicable. The record describes the primary outcome as assessed separately in each part. The handling of participants with missing response assessments, incomplete follow-up, or other missing outcome data is not reported in the registry record.

General explanation: missing data matter because the observed participants may not represent everyone enrolled or treated. The effect depends on why data are missing and how the analysis handles them. Without the number assessed for response, reasons for missing assessments, and the analysis rule, the potential influence cannot be evaluated from the registry record.

The statistical analysis plan, interim analyses, and any adjustment for multiple comparisons are not reported in the registry record. The record lists no analyses for this account. These omissions limit appraisal of how the listed outcomes were analyzed; they do not establish that a particular method was or was not used outside the information recorded.

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 5% 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 (15.4%), pyrexia (15.4%), dyspnoea (15.4%), anaemia (7.7%), febrile neutropenia (7.7%).

The registry lists serious adverse events and other adverse events by treatment group. In the afatinib monotherapy group, 74 participants were at risk, 18 were affected by serious adverse events, 67 were affected by other adverse events, and 6 deaths are listed. In the afatinib and vinorelbine group, the corresponding counts are 13 at risk, 5 affected by serious adverse events, 13 affected by other adverse events, and 1 death. In the afatinib and paclitaxel group, the counts are 26 at risk, 10 affected by serious adverse events, 23 affected by other adverse events, and 5 deaths.

These are descriptive counts, not randomized comparisons. The combination groups received treatment after progression on afatinib monotherapy, so their participants differ by treatment sequence and disease course. The counts alone cannot establish that one regimen caused more or fewer events than another. The registry record does not provide details about event duration, recurrence, timing, or attribution, and the listed deaths should not be assumed to have been caused by study treatment.

The serious-event terms listed with percentages are abdominal pain, pyrexia, and dyspnoea, each at 15.4%; anaemia, febrile neutropenia, and nausea, each at 7.7%. These listed percentages should be read as registry-reported event information, not as a comparison between treatment groups. The record does not establish from these figures alone the clinical context or causal relationship of an event.

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

The study addresses a defined population: patients with HER2-overexpressing metastatic breast cancer whose disease had failed HER2-targeted treatment in the neoadjuvant and/or adjuvant setting. Its treatment pathway begins with afatinib monotherapy and permits the addition of weekly paclitaxel or vinorelbine after progression. This design produces descriptive evidence about treatment in that pathway, but it is not a randomized comparison.

The primary endpoint is clinically interpretable as a response proportion under RECIST v1.1: the percentage with a confirmed complete or partial response during the stated assessment window. The record specifies the response criteria and the intended confidence-interval method. However, the numerical response result and interval are not reported in the registry record. It is therefore not possible to tell from this record how many participants met the primary endpoint or how precise the estimate was.

The record also does not provide numerical results for the listed secondary efficacy outcomes. Their presence in the registry identifies questions the study intended to assess, but does not show what those assessments found. Safety counts are available, but counts across the treatment groups are not controlled comparisons and should not be interpreted as causal effects.

The design does not establish what would have happened to comparable participants under an alternative treatment. Non-random allocation and progression-dependent treatment mean that participants receiving combinations entered those groups after a different clinical course than participants considered during monotherapy. A contrast between observed groups could reflect those differences as well as treatment.

Thus, the registry record supports description of the study population, planned treatment sequence, primary endpoint, and selected adverse-event counts. It does not provide the primary response estimate, a randomized treatment effect, or sufficient information to judge the missing-data and multiplicity procedures. These limits do not erase the value of the study question, but they constrain what can be concluded from the recorded information.

14. Why This Trial Matters Statistically

This trial is statistically instructive because its treatment sequence affects what comparisons can mean. Although afatinib monotherapy and two combination regimens appear in the record, the study is non-randomized and the combinations were introduced after progression on monotherapy. A reader must distinguish treatment-group descriptions from comparative evidence. Group counts alone do not make the groups exchangeable or support a causal contrast.

The primary endpoint also illustrates the difference between a well-specified outcome and an available result. The registry defines objective response using confirmed RECIST v1.1 criteria and names an exact confidence-interval method. Yet the numerical response result and interval are not reported in the registry record. Statistical interpretation requires both an outcome definition and the data-derived estimate, together with uncertainty; a detailed definition cannot stand in for the result.

The design highlights why treatment timing and analysis populations need explicit attention. The monotherapy treated set and combination treated sets use treatment-receipt criteria, while the later combination groups are formed after progression. Those features matter when evaluating who contributes to each outcome and whether groups can be compared. The record's definitions help describe the populations, but the missing numerical response results and unreported handling of missing data limit appraisal.

Finally, the record lists multiple secondary outcomes alongside the primary response question. This makes multiplicity and endpoint hierarchy relevant statistical considerations, while the registry's approach to multiplicity is not reported. Together, these features show why careful reading separates the study's question, design, analysis population, uncertainty, and conclusions. The available information permits a precise description of several design features, but not a numerical account of primary efficacy.

15. Limitations

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

The most direct limitation for interpreting efficacy is that the registry record does not report the numerical primary objective-response result or its 95% confidence interval. The response count and analysis denominator are also not reported. Although the registry describes the endpoint and its intended Clopper and Pearson interval, the result's magnitude and precision cannot be assessed.

The non-randomized, single-group design has no randomized control arm. The monotherapy and combination treatments also occur at different stages: paclitaxel or vinorelbine could be added after progression on afatinib. Comparisons between these groups would therefore not be like-for-like treatment comparisons. The registry does not support attributing differences in outcomes to the drugs alone.

The registry lists secondary outcomes but does not report their numerical findings in the information available for this account. The analysis plan, interim analyses, sample-size justification, missing-data handling, and multiplicity procedure are not reported in the registry record. These gaps limit assessment of how outcomes were analyzed and how uncertainty across endpoints should be interpreted.

The safety information provides affected-participant counts and listed serious-event terms, but does not by itself establish event causality or comparative safety. The record also does not provide additional context about event timing or recurrence in the information. Taken together, the available information is more complete about design and endpoint definitions than about numerical clinical outcomes.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: LUX-Breast 2; Afatinib in HER2 (Human Epidermal Growth Factor Receptor)-Treatment Failures — NCT01271725.
• Registry record last updated 2019-06-17; status last verified 2019-03.
• Results first posted 2019-06-17.
• 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

NCT01271725, LUX-Breast 2, is a completed Phase II, open-label, non-randomized, single-group study sponsored by Boehringer Ingelheim. The registry reports enrollment of 74 participants. The study concerns afatinib in HER2-overexpressing metastatic breast cancer after failure of HER2-targeted treatment in the neoadjuvant and/or adjuvant setting.

Participants received afatinib 40mg once daily as monotherapy until progression. After progression, weekly paclitaxel or weekly vinorelbine could be added. The primary endpoint is the percentage with confirmed objective response by RECIST v1.1, with a stated exact 95% confidence interval. Its numerical result is not reported in the registry record.

The registry lists secondary efficacy and safety outcomes and reports selected adverse-event counts. Because treatment allocation was non-randomized and combination therapy followed progression, treatment groups are not a like-for-like control comparison. The statistical plan details, missing-data handling, and multiplicity approach are not reported in the registry record.