Multimodality Treatment for Women With Stage II, Stage III, or Stage IV Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Multimodality Treatment for Women With Stage II, Stage III, or Stage IV Breast Cancer (NCT00006110), a non-randomized phase 2 study. The primary outcome is Cardiac Toxicity of Weekly Taxol Given With Weekly Herceptin When Delivered Immediately Following Four Cycles of Standard Dose AC..

2. Clinical Question

3. Trial Design

This was a phase II, parallel, nonrandomized study sponsored by UNC Lineberger Comprehensive Cancer Center. The registry records enrollment of 82 participants and a completed study. The listed start was 1998-12, and primary completion was 2012-03. Masking was listed as none.

The treatment pathways differed in trastuzumab use. The pathway titled “Neo-adjuvant Herceptin with or without radiation” included chemotherapy, paclitaxel plus trastuzumab, surgery, radiation or no radiation, and additional trastuzumab. The “Non-Herceptin with or without radiation” pathway included chemotherapy, paclitaxel, surgery, and radiation or no radiation. The intervention list also names conventional surgery and the drugs doxorubicin hydrochloride, cyclophosphamide, and paclitaxel.

The record describes four cycles of AC followed by 12 weeks of paclitaxel, with or without trastuzumab, and then local therapy. Its official title further describes weekly trastuzumab or no additional therapy after local therapy. These are treatment pathways rather than randomly assigned arms: allocation was nonrandomized. Differences between groups may therefore reflect differences in who entered each pathway, as well as treatment, and cannot be interpreted as though randomization made the groups comparable.

The registry lists overall response and disease-free survival in patients receiving and not receiving trastuzumab as secondary outcomes. It does not report the allocation process, eligibility criteria in detail, baseline characteristics by group, or a protocol-based justification for the enrollment target. It also does not report a statistical analysis plan or interim analyses. Those details 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.

Before assignment: 85 patients consented to treatment. 3 patients were consented but not treated.

Overall Study

Analysis population for the primary outcome, as defined in the registry: LVEF data during AC-TP are complete on 50 patients. 2 are incomplete because of withdrawal (1) and progressive disease (1). 43 of the 52 patients underwent LVEF determination at 1.5 years.

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 Cardiac Toxicity of Weekly Taxol Given With Weekly Herceptin When Delivered Immediately Following Four Cycles of Standard Dose AC., measured over 78 weeks (1.5 years).

The primary outcome is titled “Cardiac Toxicity of Weekly Taxol Given With Weekly Herceptin When Delivered Immediately Following Four Cycles of Standard Dose AC.” It measures participants, with the outcome described as AC-TP-associated systolic dysfunction. Systolic function was assessed using left ventricular ejection fraction (LVEF), a measure of how well the heart pumps blood and a measure used in diagnosing and tracking heart failure.

The registry specifies a timeframe of 78 weeks (1.5 years). It reports that LVEF data during AC-TP are complete on 50 patients. Two are described as incomplete because of withdrawal (1) and progressive disease (1). At 1.5 years, 43 of the 52 patients underwent LVEF determination.

In general, LVEF provides a quantitative measure of cardiac pumping function, making it useful for monitoring changes in systolic function over time. However, a count of participants with an event requires a stated event definition or threshold. The registry record does not report that threshold, the number meeting it, or a numerical estimate of change in LVEF. It therefore does not establish from the posted primary outcome information how many participants experienced the defined cardiac toxicity.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The record describes a parallel, nonrandomized phase II study with no masking. These design features determine what can be learned from comparisons between the pathways. The registry records enrollment of 82 participants and describes pathways with and without trastuzumab. Because allocation was nonrandomized, differences in outcome between groups would not by themselves isolate a treatment effect. The groups may differ in characteristics related to their treatment selection or outcomes, and the registry record does not report sufficient information to evaluate or adjust for such differences.

What does a participant count represent?

The primary outcome uses participants as its unit and is described as a count of participants with an event. In general, a count communicates how many participants meet a specified outcome definition. To interpret that count, readers need a clear definition of the event and an understanding of how many participants were assessed. Here, the registry describes LVEF measurement and identifies systolic dysfunction as the cardiac outcome, but does not report a threshold defining an event or a primary event count. No risk estimate or group contrast can be reconstructed from the outcome description alone.

Why does follow-up completeness matter?

In general, incomplete outcome measurements can make the observed participants less representative of everyone intended for follow-up. This matters especially when missingness is related to health status or treatment experience. The registry reports complete LVEF data during AC-TP on 50 patients, two incomplete records due to withdrawal and progressive disease, and LVEF determination at 1.5 years for 43 of the 52 patients. It does not describe how missing measurements were handled, whether follow-up differed by pathway, or whether an analysis accounted for the possibility that missingness was related to cardiac status.

What would a confidence interval communicate?

In general, a confidence interval expresses uncertainty around an estimated effect. A narrower interval indicates greater precision under the assumptions of the analysis; a wider interval indicates less precision. The registry record reports no primary effect estimate and no confidence interval, so the precision of a cardiac-toxicity comparison cannot be assessed. A count alone, even when available, would not show the uncertainty around a between-group difference without an appropriate analysis and its assumptions.

Why is the nonrandomized comparison limited?

In general, random allocation helps balance measured and unmeasured participant characteristics across groups on average. A nonrandomized study lacks that design protection. Statistical adjustment can sometimes address measured differences, but it cannot guarantee removal of bias from unmeasured factors. The registry explicitly flags the arms as not a like-for-like control comparison, and no analyses are listed. The record therefore does not support treating the pathways as exchangeable groups or interpreting a raw contrast as a causal effect.

How should planning calculations be understood?

In general, sample-size calculations describe what a future study might require under specified assumptions; they are not evidence of the precision or success of a completed study. The stated planning approach uses a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the effect to detect set as a stated fraction of the observed effect. For time-to-event outcomes, the stated convention is a total number of events; for other outcomes, it is participants per arm. The record characterizes these as planning illustrations rather than the trial's own design. They should not be called observed power. No numerical planning outputs or primary observed effect are reported in the registry record, so the assumptions alone do not establish what a particular calculation would show.

In general, over-dispersion means that observed variability is greater than a simple model assumes. Accounting for it can affect uncertainty estimates and sample-size requirements. The registry does not report an over-dispersion model for this study, and the primary outcome is presented as a participant count. No inference about use of such a model is warranted. More broadly, the record lists 0 analyses, so the statistical methods actually used to estimate primary or secondary outcomes are not reported.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 3 outcome measures and 0 analyses. It names cardiac toxicity as the primary outcome and lists overall response and disease-free survival in patients receiving and not receiving trastuzumab as secondary outcomes. The record does not describe a multiplicity strategy, such as how interpretation would account for examining more than one outcome. It also does not report prespecified hypotheses, a hierarchy of secondary outcomes, or adjusted significance thresholds. These details are not reported in the registry record.

Missingness is relevant to the primary LVEF assessment. The record states that LVEF data during AC-TP are complete on 50 patients, while 2 are incomplete because of withdrawal (1) and progressive disease (1). At 1.5 years, 43 of the 52 patients underwent LVEF determination. These descriptions do not specify the analysis population for the primary outcome, how incomplete records were treated, or whether results were compared for participants with and without follow-up measurements. No imputation procedure or sensitivity analysis is reported.

In general, an analysis population defines which participants contribute to a particular estimate. Different populations can answer different questions, especially when treatment is not randomly assigned or outcomes are missing. The registry record does not identify an intention-to-treat, per-protocol, or other primary analysis population. It also does not provide outcome counts for each population or explain how participants with incomplete follow-up were handled. The impact of missing data and the choice of analysis population therefore cannot be assessed from the posted 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: dyspnea (3.8%), febrile neutropenia (fever of unknown origin without clinically or microbiologically documented infe (3.3%), dizziness (1.9%), seizure (1.9%), pleuritic pain (1.9%).

The registry reports serious adverse events for two groups. In the chemotherapy-plus-trastuzumab group, 7 of 52 participants were recorded as having a serious adverse event; in the control, non-trastuzumab group, 1 of 30 were recorded. The separate counts for participants with other adverse events are 51 of 52 and 28 of 30, respectively. Deaths are recorded separately as 15 and 13. These are the registry's counts; they should not be combined because the record does not establish that the categories are mutually exclusive.

The listed serious-event terms include dyspnea (3.8%), febrile neutropenia (3.3%), dizziness (1.9%), seizure (1.9%), pleuritic pain (1.9%), and atrial fibrillation (1.9%). The record does not report whether these events were judged treatment-related, their timing, or a comparative statistical analysis. The groups were nonrandomized and were not a like-for-like control comparison, so the counts should be read descriptively rather than as a causal treatment contrast.

In general, adverse-event counts describe recorded events or affected participants according to the registry's reporting categories; they do not alone establish why an event occurred. No uncertainty intervals or adjusted safety comparisons are reported in the registry record.

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

The record documents a phase II study of multimodality treatment for women with stage II, stage III, or stage IV breast cancer. It describes treatment pathways that differed in trastuzumab use and identifies cardiac toxicity associated with AC-TP as the primary outcome. It also gives information on the timing and completeness of LVEF assessment and reports group-level adverse-event counts.

Those details do not amount to a reported estimate of the primary outcome. The registry does not give a cardiac-toxicity event count, a comparison estimate, a confidence interval, or a p-value. It names overall response and disease-free survival as secondary outcomes but does not report their results. Thus, the record cannot quantify how the groups differed on the primary or secondary outcomes or how precisely any difference was estimated.

The design also limits interpretation. Participants were not randomized, masking was absent, and the registry notes that the arms are not a like-for-like control comparison. The posted information cannot establish that differences in outcomes would be due to trastuzumab rather than differences between the groups or other factors. No statistical adjustment is reported.

The adverse-event counts are useful as descriptive registry information but do not establish causation, comparative risk, or the balance of benefits and harms. The data on LVEF follow-up indicate that measurements were not available for everyone described as at risk at 1.5 years; the record does not explain how that affected analysis. In general, treatment-effect claims require clearly defined outcomes, suitable comparisons, and estimates with uncertainty. Those elements are not reported for the main outcomes in this registry record.

13. Why This Trial Matters Statistically

This study illustrates why a registry outcome description and a reported statistical result are not the same thing. The record identifies a clinically relevant measurement—LVEF—and a follow-up timeframe of 78 weeks (1.5 years), but does not report the number meeting a cardiac-toxicity definition or an effect estimate. Without an event definition and outcome result, the primary question cannot be evaluated quantitatively from the record.

It also illustrates the importance of design when interpreting group comparisons. The treatment pathways differed in trastuzumab use, but allocation was nonrandomized. The record itself warns that the arms are not a like-for-like control comparison. This makes it important to distinguish observed group counts from evidence that treatment caused a difference. The registry lists no analyses that might address measured differences, and even adjustment would not automatically remove all potential confounding.

Follow-up completeness is another statistical concern. The record notes incomplete LVEF information during AC-TP and fewer assessments at 1.5 years than the group described as at risk. In general, missing measurements can affect both the estimate and its representativeness, especially if missingness is related to participant health or treatment. The absence of a reported missing-data approach leaves that concern unresolved.

Finally, the registry contains safety counts and secondary outcome names without corresponding inferential results. This underscores the difference between a list of endpoints and a complete analysis with definitions, estimates, and uncertainty. The record is informative about the intended clinical question and the limitations of the comparison, but it does not provide enough statistical results to quantify treatment effects.

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 registry record does not report a primary cardiac-toxicity event count, a threshold for defining systolic dysfunction, or an estimate of change in LVEF. It therefore does not show how the measured cardiac outcome was translated into the participant-level endpoint named in the record. No confidence interval, p-value, or group comparison for that endpoint is reported.

Outcome availability is incomplete as described. LVEF data during AC-TP are complete on 50 patients; 2 are incomplete because of withdrawal (1) and progressive disease (1). At 1.5 years, 43 of the 52 patients underwent LVEF determination. The registry does not explain the handling of unavailable measurements, report whether the missingness pattern differed by treatment pathway, or describe sensitivity analyses.

The nonrandomized allocation and the registry's explicit warning that the arms are not a like-for-like control comparison limit causal interpretation. Baseline characteristics by group, the reasons participants received each pathway, and any adjustment for confounding are not reported in the registry record. Masking was listed as none.

The registry lists overall response and disease-free survival as secondary outcomes, but reports neither their results nor analyses. It does not describe multiplicity handling, the analysis populations, interim analyses, or a statistical analysis plan. The serious adverse-event counts are descriptive and are not accompanied by treatment-relatedness assessments, adjusted comparisons, or uncertainty estimates. These gaps prevent a full assessment of the magnitude and precision of the trial's clinical results.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Multimodality Treatment for Women With Stage II, Stage III, or Stage IV Breast Cancer — NCT00006110.
• Registry record last updated 2017-07-31; status last verified 2017-06.
• Results first posted 2017-07-31.
• 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

NCT00006110 is a completed, nonrandomized phase II parallel study sponsored by UNC Lineberger Comprehensive Cancer Center. It enrolled 82 participants with breast cancer and examined multimodality treatment for women with stage II, stage III, or stage IV disease. The registry lists no masking.

The treatment pathways differed in trastuzumab use and included chemotherapy, surgery, and radiation or no radiation. The primary outcome concerned cardiac toxicity associated with AC-TP, assessed through LVEF over 78 weeks (1.5 years). The record reports incomplete LVEF information and follow-up counts, but no primary outcome result or statistical analysis. Overall response and disease-free survival are listed as secondary outcomes without reported results. Serious adverse-event counts are reported descriptively; the registry cautions that the arms are not a like-for-like control comparison.