A Study of Pertuzumab in Combination With Trastuzumab (Herceptin) and a Taxane in First-Line Treatment in Participants With Human Epidermal Growth Factor 2 (HER2)-Positive Advanced Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Pertuzumab in Combination With Trastuzumab (Herceptin) and a Taxane in First-Line Treatment in Participants With Human Epidermal Growth Factor 2 (HER2)-Positive Advanced Breast Cancer (NCT01572038), a na phase 3 study. The primary outcome is Subgroup Analysis by Region of Enrollment: Overview of the Number of Participants With Serious Treatment-Emergent Adverse Events (TEAEs), Grade ≥3 TEAEs, and Grade ≥3 TEAEs Related to Pertuzumab.

2. Clinical Question

In practical statistical terms, this is a descriptive safety question: among participants treated with the combination, how many experienced specified adverse-event categories during the defined follow-up? The question is important to distinguish from a comparative treatment question. A count observed in a single treatment group describes what was recorded in that group; on its own, it cannot estimate how the count would differ under another treatment.

The record identifies a broad population and a flexible taxane component. Outcomes could therefore be considered across different regions and taxane choices, but the registry record does not report the detailed subgroup counts needed to compare those groups. Nor does it provide a numerical progression-free survival estimate in the posted results. The primary numerical information available for interpretation is the overall adverse-event count information and the listed serious adverse-event terms.

3. Trial Design

This was a multicenter, open-label, single-arm Phase IIIb study sponsored by Hoffmann-La Roche. The record describes participants receiving pertuzumab with trastuzumab and a taxane as first-line treatment for HER2-positive advanced breast cancer. The taxane option could be docetaxel, paclitaxel or nab-paclitaxel. All participants belonged to the same named treatment group; there was no randomized allocation and no separate comparator arm.

The planned treatment schedule used cycles of 3 weeks. Treatment was to continue until the predefined study end, unacceptable toxicity, withdrawal of consent, disease progression or death, whichever occurred first. The registry record lists study start as 2012-06-01 and primary completion as 2019-09-20, and describes the study status as completed. Enrollment was 1436.

Open-label means that masking was absent; the record lists masking as none. This matters when interpreting outcomes that depend on clinical observation or investigator judgment. The registry description states that investigators graded adverse events for severity and assessed whether events were related to study drug. The listed grading approach used NCI-CTCAE v4.0, with a general severity description for events not listed in that system. Because the study had no control group, these assessments describe events observed while participants received the regimen rather than differences attributable to the regimen compared with an alternative.

The primary outcome is titled as a subgroup analysis by region of enrollment and concerns participant counts for serious treatment-emergent adverse events, Grade >=3 treatment-emergent adverse events and Grade >=3 events related to pertuzumab. The posted overall adverse-event figures do not give a region-by-region breakdown. Progression-free survival and several subgroup analyses of progression-free survival are listed as secondary outcomes, including analyses by region, age, performance status, taxane, visceral disease, prior chemotherapy and hormone receptor status. Numerical results for these outcomes are not reported in the registry record.

The record does not report the protocol’s sample-size justification, interim analyses or statistical analysis plan. It also does not describe a randomized design or a prespecified comparative hypothesis. These omissions limit what can be inferred about planning choices and formal hypothesis testing; the design and posted outcome data support descriptive interpretation of the treated cohort.

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: A total of 1697 patients were screened: 261 failed screening and 1436 were enrolled in the study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety Population: all participants who received at least one dose of any study treatment.

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 Subgroup Analysis by Region of Enrollment: Overview of the Number of Participants With Serious Treatment-Emergent Adverse Events (TEAEs), Grade ≥3 TEAEs, and Grade ≥3 TEAEs Related to Pertuzumab, measured over From Baseline until 28 days after, or 7 months after (only for serious AEs related to study drug), the last dose of study treatment. The median (full range) duration of exposure to any study treatment was 16.2 (0.0-86.4) months..

The primary outcome focuses on the number of participants with specified treatment-emergent adverse events: serious events, Grade >=3 events, and Grade >=3 events judged related to pertuzumab. A treatment-emergent event was defined as an adverse event that began or worsened in severity on or after the first administration of study drug, through 28 days after the last dose. The record gives an additional follow-up interval of 7 months after the last dose for serious adverse events related to study drug.

Severity and seriousness are distinct concepts. Severity refers to how intense an event is; the record describes Grade 1 as mild, Grade 2 as moderate, Grade 3 as severe, Grade 4 as life-threatening or disabling, and Grade 5 as death. Seriousness is a separate classification assessed by the investigator. An event can therefore be described using both a severity grade and a seriousness assessment, and the two labels should not be treated as interchangeable.

Participant counts are useful for describing how often an event category was observed in the safety population. They are easier to interpret as a direct tally than event counts when a participant may experience multiple events. The record states that the safety population comprised participants who received at least one dose of any study treatment. The median duration of exposure to any study treatment was 16.2 months, with a full range of 0.0-86.4 months.

These counts do not by themselves describe when events occurred, how long they lasted, or the rate of events per unit of exposure. Nor do they establish whether the regimen caused an event. The investigator assessed relationship to study drug, but a single-group study cannot separate treatment effects from the underlying disease, other care or other factors.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The main statistical feature of this study is its single-group design. Every participant was assigned to the named combination group, and the record lists allocation as not applicable. In a randomized comparison, treatment groups can be compared directly under a design intended to balance other factors. Here, there is no concurrent treatment group. The analysis is therefore principally descriptive: it reports counts among participants who received the regimen rather than an estimated treatment contrast.

What does a participant count represent?

A participant count records how many people met a specified outcome definition. For adverse events, the definition matters: an event must meet the record’s criteria for being treatment-emergent, and the participant must belong to the stated safety population. Counts should not be interpreted as counts of individual episodes unless the outcome explicitly says so. A participant may have more than one event, while a participant-level count ordinarily records whether the participant experienced the category at least once.

The registry gives a safety population definition: participants who received at least one dose of any study treatment. That population is appropriate for describing events observed after treatment exposure, but it is not necessarily identical to every population that might be used for efficacy analyses. The analysis population for each secondary outcome is not reported in the registry record. The record also does not report how missing outcome information was handled.

How should severity and seriousness be distinguished?

Severity grades describe intensity. The investigator used NCI-CTCAE v4.0 and the record’s stated fallback descriptions for events not listed in that system. Seriousness is a separate judgment. The primary outcome names both serious events and events meeting a severity threshold of Grade >=3, and also names Grade >=3 events considered related to pertuzumab. These categories can overlap, but they answer different questions. A serious event is not automatically synonymous with a Grade >=3 event, and a Grade >=3 event is not automatically serious.

Relatedness is another judgment distinct from severity and seriousness. The investigator assessed whether an adverse event was related to study drug. A relatedness assessment is not a randomized causal estimate: it reflects the investigator’s assessment within this open-label, single-arm study. The registry record does not report a separate adjudication process or the detailed criteria used for assigning relatedness.

Why does exposure time matter?

Participants were followed for different treatment durations. The median exposure to any study treatment was 16.2 months, with a full range of 0.0-86.4 months. A simple participant count does not adjust for that variation. Participants observed for longer may have more opportunity for an event to be detected. A rate analysis can account for exposure time by expressing events relative to person-time, but such a rate is not reported in the registry record. No rate should be inferred from the counts alone.

The event window also affects the tally. Treatment-emergent events included events beginning or worsening on or after the first study-drug administration through 28 days after the last dose. The record specifies a longer 7-month period after the last dose only for serious adverse events related to study drug. Those time windows should be kept in view when considering which events could be counted.

What can subgroup analyses add?

The primary outcome title identifies a subgroup analysis by region of enrollment. Secondary outcomes list progression-free survival analyses by region and other participant or disease characteristics. Subgroup summaries can describe how outcomes vary across categories, but small or uneven groups can make estimates less precise. A difference between subgroup estimates does not automatically demonstrate a genuine difference between groups; a formal interaction analysis and its assumptions would be relevant. Numerical subgroup estimates and interaction tests are not reported in the registry record.

Subgroup analysis can also raise multiplicity concerns. When many outcomes or groups are examined, the chance of finding an apparently notable pattern by chance can increase. The registry record does not report a multiplicity adjustment or a hierarchy of confirmatory hypotheses. The listed analyses should therefore be treated as descriptions of planned outcome areas unless further analysis details are reported.

How should sample-size illustrations be read?

Any sample-size illustration based on an observed effect is a planning exercise, not a description of what was observed and not observed power. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. Time-to-event illustrations use a total event count; other illustrations use participants per arm. These are hypothetical two-arm planning assumptions, whereas the actual study was single-arm and had no like-for-like control comparison. Such illustrations do not establish the trial’s original design rationale. The protocol’s sample-size justification is not reported in the registry record.

Over-dispersion is a general statistical issue that can arise when outcome variability exceeds what a simple model assumes. In that situation, standard errors from the simpler model can be too small, making intervals too narrow and tests too optimistic. This is a general consideration, not a reported feature of the posted analysis. The registry record does not describe a statistical model or report whether over-dispersion was assessed.

What analysis details are unavailable?

The registry record does not report the statistical analysis plan, interim analyses, missing-data procedures, or a sample-size justification. It also does not provide confidence intervals for the aggregate adverse-event figures. Without those details, the counts can be described, but the assumptions and inferential procedures behind a more complete analysis cannot be evaluated. The appropriate reading is therefore narrow and explicit: the posted results describe event counts in the treated safety population, while comparisons, adjusted estimates and unreported analyses cannot be reconstructed from the available record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists numerous secondary outcome areas, including progression-free survival overall and subgroup analyses by region, age, ECOG performance status, taxane chemotherapy, visceral disease, prior (neo)adjuvant chemotherapy and hormone receptor status. Considering many outcomes and subgroups can create multiplicity: the more comparisons examined, the greater the chance that some patterns appear notable by chance even if there is no underlying difference. This is a general statistical concern, not evidence that any particular finding is spurious.

Whether multiplicity was controlled, and whether analyses were confirmatory or exploratory, is not reported in the registry record. The record also does not report a hierarchy of hypotheses, an adjustment method or details of interim analyses. Without those details, the inferential status of the listed outcomes cannot be assessed. The progression-free survival analyses are listed, but numerical results are not reported in the registry record.

Missing-data handling is also not described. The record does not state how participants with incomplete outcome information were treated, whether censoring rules differed across analyses, or how missing subgroup information was handled. These details can matter for time-to-event outcomes and subgroup estimates. They should not be guessed from the endpoint titles.

The record identifies the safety population as all participants who received at least one dose of any study treatment. It does not describe the analysis populations for each secondary outcome or provide a complete account of participants included in each analysis. In a single-arm study, these gaps do not erase the descriptive counts, but they limit assessment of selection, follow-up and uncertainty. The study’s sample-size justification and statistical analysis plan are not reported in the registry record.

11. 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 (4.9%), neutropenia (3.3%), diarrhoea (2.5%), pneumonia (1.8%), drug hypersensitivity (1.4%).

The registry records 535 participants in the serious adverse-event count among 1436 at risk. It also records 1394 in the other adverse-event count and 658 deaths. The aggregate figures are counts, not comparisons between treatment groups. The registry record’s primary outcome title includes serious treatment-emergent adverse events, Grade >=3 treatment-emergent adverse events and Grade >=3 events related to pertuzumab, but the posted aggregate figures do not provide a full breakdown mapping each count to every named category. That mapping is not reported in the registry record.

The listed serious adverse-event terms and percentages are febrile neutropenia, 4.9%; neutropenia, 3.3%; diarrhoea, 2.5%; pneumonia, 1.8%; drug hypersensitivity, 1.4%; and pyrexia, 1.3%. They are descriptive entries from the treated group and should not be read as comparative risks. The record does not state in this summary whether the percentages use a specific subgroup denominator; do not infer a denominator beyond the registry’s reported information.

Adverse events were graded by the investigator, and seriousness and relationship to study drug were also assessed by the investigator. Event follow-up was defined through 28 days after the last dose, with the specified 7-month period for serious events related to study drug. Exposure duration varied across participants. Counts alone do not communicate event timing, exposure-adjusted rates, uncertainty, or causation.

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

The results establish that the registry recorded adverse-event counts among participants treated with pertuzumab, trastuzumab and a taxane. The safety population contained 1436 participants. The reported aggregate counts include 535 participants for serious adverse events, 1394 for the other adverse-event count and 658 deaths. The registry also lists several serious adverse-event terms with their recorded percentages. These figures describe events in the treated group during the specified follow-up periods.

They do not establish how those counts compare with another regimen, because the study was single-arm and had no concurrent control group. They also do not establish that treatment caused a particular event. Investigator assessment of relationship to study drug is part of the record, but it is not equivalent to a randomized causal comparison. Other features of the participants, their disease and their care may also be relevant to observed events.

The counts do not give a complete account of event frequency over time. Participants had varied exposure duration, and a participant-level count does not adjust for that variation. The record does not provide confidence intervals or exposure-adjusted rates for the aggregate figures. The exact mapping of the aggregate adverse-event counts to each component of the titled primary endpoint is not reported in the registry record.

Progression-free survival is listed among the secondary outcomes, with investigator assessment using RECIST v1.1. Numerical progression-free survival results and subgroup estimates are not reported in the registry record. As a result, the posted information does not support a numerical conclusion about progression-free survival or differences across the listed subgroups. A listed outcome is not itself evidence of a particular result.

The study can therefore be read as a descriptive record of safety experience in a substantial treated cohort, with important limits on comparative and causal interpretation. It does not provide a controlled estimate of benefit or harm relative to an alternative, and the posted counts alone cannot answer how outcomes would have differed under another treatment. Those distinctions keep the interpretation aligned with the design and the results that are actually reported.

13. Why This Trial Matters Statistically

This trial illustrates why the study design determines the kind of statistical question that can be answered. A large treated cohort can provide a detailed description of events observed during treatment, but sample size does not create a comparison group. Because the study was single-arm, the central interpretation is descriptive rather than comparative. The adverse-event counts are meaningful as observations in the treated safety population, but they do not quantify differences from another treatment.

It also demonstrates the importance of keeping safety concepts separate. Severity grade, seriousness and investigator-assessed relatedness are distinct attributes. The primary outcome title names serious events and Grade >=3 events, as well as Grade >=3 events related to pertuzumab. Treating these categories as interchangeable would misrepresent what each measure captures. Outcome definitions and population definitions are essential parts of interpreting a count.

Duration of observation is another relevant statistical issue. The median exposure was 16.2 months, with a full range of 0.0-86.4 months, and the record specifies different follow-up windows for treatment-emergent events and certain serious events. A raw count does not account for varying opportunity to observe an event. The distinction between participant counts and exposure-adjusted rates is therefore important, even when only counts are posted.

Finally, the registry lists a broad set of secondary and subgroup outcomes, but numerical efficacy results and detailed subgroup results are not reported in the registry record. This highlights the difference between an outcome being named and an analyzable result being available. Missing analysis details also matter: the statistical plan, missing-data handling, multiplicity procedures and sample-size justification are not reported. A careful reading separates what the registry documents from what cannot be established from the posted record.

14. 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 important limitation for interpretation is the absence of a control group. With a single-arm design, observed adverse-event counts cannot be contrasted with concurrent outcomes under another treatment. The design therefore does not support a direct treatment-effect estimate from these figures. It also limits causal interpretation of investigator-assessed relatedness.

The study was open-label, with no masking. The record states that investigators assessed adverse-event severity, seriousness and relationship to study drug. Because those assessments were not masked, judgment-based classifications may be influenced by knowledge of treatment. The registry record does not report independent adjudication procedures.

The primary outcome title specifies a subgroup analysis by region and names three adverse-event categories, but the posted aggregate results do not give regional breakdowns or a complete mapping of counts to each category. Numerical results for progression-free survival and its listed subgroup analyses are not reported in the registry record. This prevents assessment of the intended efficacy outcomes from the information posted.

Exposure varied: median duration was 16.2 months and the full range was 0.0-86.4 months. Aggregate participant counts do not adjust for differences in exposure time. The registry record does not report exposure-adjusted event rates or confidence intervals for the counts. It also does not report missing-data procedures, a statistical analysis plan, multiplicity handling, interim analyses or the sample-size justification. The available record supports descriptive reporting, but not reconstruction of these unreported methods or additional comparative analyses.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of Pertuzumab in Combination With Trastuzumab (Herceptin) and a Taxane in First-Line Treatment in Participants With Human Epidermal Growth Factor 2 (HER2)-Positive Advanced Breast Cancer — NCT01572038.
• Registry record last updated 2020-09-25; status last verified 2020-08.
• Results first posted 2020-09-25.
• 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.

18. Record Summary

This completed, multicenter, open-label, single-arm Phase IIIb study evaluated pertuzumab with trastuzumab and a taxane in first-line treatment of HER2-positive metastatic or locally recurrent breast cancer. The taxane options were docetaxel, paclitaxel and nab-paclitaxel. Enrollment was 1436, and the safety population included participants who received at least one dose of study treatment.

The primary outcome title concerns participant counts for serious treatment-emergent adverse events, Grade >=3 treatment-emergent adverse events and Grade >=3 events related to pertuzumab, with a subgroup focus on region. The registry reports aggregate adverse-event counts and lists serious adverse-event terms, but does not report a complete regional breakdown or a full mapping of those counts to all named categories.

Progression-free survival is listed as a secondary outcome, but numerical results are not reported in the registry record. The single-arm design supports descriptive interpretation of observed safety outcomes, not a like-for-like treatment comparison. The statistical analysis plan, missing-data handling, multiplicity procedures, interim analyses and sample-size justification are not reported in the registry record.