Risk-Guided Cardioprotection With Carvedilol in Breast Cancer Patients Treated With Doxorubicin and/or Trastuzumab: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Risk-Guided Cardioprotection With Carvedilol in Breast Cancer Patients Treated With Doxorubicin and/or Trastuzumab (NCT04023110), a randomized phase 1 study. The primary outcome is Left Ventricular Ejection Fraction (LVEF).

2. Clinical Question

The trial’s stated purpose was to evaluate the safety, tolerability, and feasibility of a risk-guided cardioprotective strategy with carvedilol compared with usual care. The treatment context is breast cancer therapy with doxorubicin, trastuzumab, or their combination, and the outcome focus is cardiac function. LVEF is the primary measure identified in the record; secondary outcomes address other cardiac imaging measures, cardiac dysfunction, and biomarkers.

The record describes randomization, but it also says that some participants refused carvedilol after randomization and remained in follow-up. For the exploratory LVEF analysis, participants were evaluated according to treatment actually received rather than solely by randomized assignment. A low-risk, non-randomized group is also listed. These features matter when interpreting any group comparison: the groups do not all represent randomized comparisons, and an as-treated analysis can differ from an analysis based on assignment.

The registry record does not report numerical LVEF findings, a between-group effect estimate, or an uncertainty interval. It therefore does not establish the size or precision of any difference in LVEF. The described question is about cardiac function in this treatment setting; it is not, by itself, a test of cancer outcomes or a definitive evaluation of clinical benefit.

3. Trial Design

NCT04023110 is a Phase 1 pilot study with randomized allocation, no masking, and a parallel model. Its title describes risk-guided cardioprotection with carvedilol in breast cancer patients treated with doxorubicin and/or trastuzumab. The registry lists Abramson Cancer Center at Penn Medicine as sponsor. The recorded study start date is 2019-08-09, primary completion date is 2024-08-31, and status is completed. Enrollment is listed as 68.

The intervention is carvedilol, compared with usual care for the elevated-risk groups. The record also lists a low-risk, non-randomized group. For carvedilol, the arm description states a starting dose of 3.125mg twice daily, with stepwise increases as tolerated up to 25mg twice daily or until stated systolic blood pressure or heart-rate thresholds are reached. The description is cut off after the words “after fir”; the remainder of that instruction is not reported in the registry record. The usual-care description specifies collection of clinical, echocardiographic, and biomarker data at baseline and standardized time intervals during and after therapy, approximately 3, 6, 9, 12, and 24 months.

Randomization does not mean that every analysis necessarily compares participants according to their randomized assignments. The primary outcome description explicitly calls the LVEF evaluation exploratory and as treated. It also notes that three participants refused carvedilol after randomization but agreed to remain on study for follow-up. The rest of the participant-flow explanation is incomplete in the registry record, so further details about treatment receipt or group assignment cannot be inferred.

The record does not report masking, as expected from its designation of no masking, or a detailed statistical analysis plan. It does not report a sample-size justification or explain how the enrollment target was selected. Because this is a pilot study, the stated emphasis on safety, tolerability, and feasibility is important context; the record does not provide enough reported LVEF results to evaluate an efficacy estimate.

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: The cardiotoxicity risk score was calculated using Cox proportional hazards regression with race, systolic blood pressure, left ventricular ejection fraction, systemic cancer therapy, smoking status, diabetes and hypertension as selected predictors (PMID 4012740). A threshold of 4.5% risk of cardiac dysfunction at 1 year was set to prioritize sensitivity. 49 patients were classified as low risk and 19 as elevated risk. 13 elevated risk patients were randomized to carvedilol and 6 to usual care.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Exploratory cardiac efficacy by cardiac function (LVEF) was analyzed, evaluating the three groups according to the treatment the participants actually received ("as treated"). After randomization, three participants refused carvedilol but agreed to remain on study for follow-up, and one participant assigned to usual care started carvedilol for clinical indications at the baseline visit. As a result, there were 11 participants treated with carvedilol and 8 treated as per usual care.

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 Left Ventricular Ejection Fraction (LVEF), measured over up to 24 months.

The primary endpoint is left ventricular ejection fraction (LVEF), measured by echocardiogram and expressed as a percentage. It is defined as left ventricular stroke volume—end-diastolic volume minus end-systolic volume—divided by end-diastolic volume, with the fraction multiplied by 100. The registry describes estimating two-dimensional left ventricular volumes with the biplane method of disks, also called modified Simpson’s rule, and says higher LVEF values are generally considered more favorable.

LVEF is used as a measure of how much blood the left ventricle ejects relative to the volume it contains at the end of filling. In general, it offers a quantitative way to assess ventricular systolic function. It does not, on its own, capture every aspect of cardiac performance, symptoms, or clinical events. That is one reason the registry also names other imaging and biomarker outcomes, including diastolic function, ventricular-arterial coupling, cardiac strain, cardiac dysfunction, hsTnT, NTproBNP, and left ventricular mass.

The specified timeframe is up to 24 months. The primary outcome is summarized using the median and inter-quartile range, according to the registry’s outcome description. These summaries describe the center and spread of observed measurements; they do not by themselves quantify a treatment contrast or its precision. The numerical medians, inter-quartile ranges, and comparative LVEF results are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a randomized, unmasked, parallel Phase 1 pilot study and an exploratory primary-outcome analysis based on treatment actually received. Each of those design features affects what a statistical comparison can establish. The record does not report a detailed statistical analysis plan, so the explanations below describe general statistical principles rather than methods confirmed as having been used for every analysis.

What does the median and inter-quartile range describe?

The primary outcome is LVEF, reported in the registry as a median with an inter-quartile range. In general, the median is the middle of an ordered set of measurements. The inter-quartile range describes the span from the lower to the upper quartile, conveying how spread out the central portion of the observations is. These summaries can be useful when a distribution is skewed or when extreme observations make a mean and standard deviation less representative. They do not show the uncertainty around a treatment effect, and separate group summaries alone do not quantify a between-group contrast.

The registry does not give the numerical LVEF summaries or a comparative estimate. As a result, readers cannot judge the observed group distributions, the magnitude of a contrast, or its statistical precision from the record. No calculation from the median and inter-quartile range is appropriate when those values are absent.

What does randomization contribute?

In general, random allocation is intended to make treatment groups comparable at the point of assignment, on average, with respect to both measured and unmeasured characteristics. That rationale applies to comparisons analyzed according to randomized assignment. This record says the primary LVEF analysis was as treated, meaning it grouped participants by treatment received. When assignment and receipt differ, an as-treated comparison may lose some of the protection randomization offers against confounding. The registry notes that three participants refused carvedilol after randomization and stayed in follow-up. It does not report a numerical intention-to-treat LVEF analysis alongside the as-treated analysis.

The low-risk group is described as non-randomized. General comparisons between a non-randomized group and randomized groups can reflect differences in baseline risk, treatment selection, or other characteristics, as well as any treatment-related differences. The record does not report a statistical adjustment plan that would address those concerns.

How should uncertainty be understood?

In general, a confidence interval gives a range of values compatible with the data and the method used, while showing the uncertainty around an estimate. A narrow interval indicates greater precision than a wide interval, but neither alone determines whether an effect is clinically meaningful. A p-value, when reported, assesses compatibility with a specified null model under stated assumptions; it is not the probability that the null hypothesis is true. The registry does not report a confidence interval or p-value for the primary LVEF comparison. There is therefore no reported inferential result from which to judge precision or statistical evidence for a contrast.

Small studies often yield imprecise estimates, even when a point estimate can be calculated. The registry’s enrollment figure is 68, but the outcome-specific numbers contributing to each analysis are not reported in the primary outcome information. Enrollment should not be taken as the number with measured LVEF at every follow-up time. Missing measurements, follow-up, or analysis-population details can affect both precision and interpretation.

What do planning calculations mean?

The stated sample-size assumptions describe a planning illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The target effect is described as a stated fraction of the observed effect. In general, such calculations ask how many participants per arm would be needed to detect an assumed effect under specified assumptions. They are planning illustrations, not observed power and not evidence that a study had a particular probability of finding an effect after results were observed.

The record does not report a sample-size justification for the trial’s original design. The planning assumptions do not establish that the trial was designed around those assumptions. They also do not resolve the lack of reported LVEF results. No sample-size values or observed-effect value are available in the registry record, so no numerical planning conclusion can be drawn.

Why can over-dispersion matter?

In general, over-dispersion occurs when observed variability exceeds what a statistical model assumes. If a model underestimates variability, standard errors can be too small and uncertainty intervals too narrow. Appropriate methods may account for extra variability, but the choice depends on the outcome and model. The registry does not report a model for the primary LVEF outcome or state whether over-dispersion was assessed. It would therefore be inappropriate to attribute a particular modeling approach to this study.

More broadly, the registry lists ten outcome measures and no analysis entries. The outcome list indicates that multiple dimensions of cardiac function and biomarkers were considered, but it does not specify which tests were planned, which were performed, or how results were handled statistically. Those details, including any model specification, covariate adjustment, or sensitivity analysis, are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists LVEF as the primary outcome and names secondary outcomes involving diastolic function, ventricular-arterial coupling, cardiac strain, cardiac dysfunction, hsTnT, NTproBNP, and left ventricular mass. Testing multiple outcomes can increase the chance of observing an apparently unusual result by chance if each is assessed separately. In general, a prespecified hierarchy or multiplicity adjustment can help control that risk, while exploratory analyses can be interpreted as hypothesis-generating. The registry record does not report a multiplicity strategy, outcome hierarchy beyond identifying LVEF as primary, or a detailed analysis plan.

The record reports no analysis entries. That means numerical results and analysis details for the listed outcomes are not reported in the registry record. It is not possible to determine from the record whether analyses were conducted but not posted, whether all outcomes were measured, or whether any results were omitted.

Missing-data methods are also not reported. The primary outcome description notes that three participants refused carvedilol after randomization but remained in follow-up; it does not provide a complete account of follow-up completion, LVEF availability, or missing measurements. In general, missing outcomes can affect both the precision and potential bias of an estimate, depending on why observations are missing and how the analysis handles them. No assumption about the missingness mechanism or imputation method can be made from the registry record.

The record identifies an as-treated population for the exploratory primary outcome and a low-risk, non-randomized group, but does not report a full analysis-population definition or participant flow. It does not report whether a separate analysis according to randomized assignment was performed. These omissions limit the ability to assess the handling of treatment refusal, missing outcomes, and differences between randomized and non-randomized groups.

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: disease progression (9.1%), cardiac arrest (2.0%).

The serious adverse event figures are recorded by study group. In the elevated-risk / carvedilol group, 1 participant had a serious adverse event among 11 at risk. In the elevated-risk / usual-care group, 0 participants had a serious adverse event among 8 at risk. In the low-risk / non-randomized group, 1 participant had a serious adverse event among 49 at risk. The record separately lists 1 death in the elevated-risk / carvedilol group, 0 in the elevated-risk / usual-care group, and 1 in the low-risk / non-randomized group.

These are descriptive counts, not estimates of comparative risk. The record does not report uncertainty intervals or a formal statistical comparison for these figures. It also does not say whether deaths overlapped with serious adverse events or establish their causes, timing, or relationship to treatment. The counts should not be interpreted as evidence that treatment caused or prevented an event.

The registry names Disease Progression at 9.1% and Cardiac Arrest at 2.0% among the top serious adverse event terms. It does not provide enough detail in the listed details to assign those terms to a particular group or infer event timing. The absence of a listed event in a group should not be taken as proof that no event of any kind occurred; these figures address the recorded serious-event information, not a complete clinical narrative.

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

The record describes a pilot investigation of risk-guided carvedilol in the setting of breast cancer treatment with doxorubicin, trastuzumab, or their combination. Its stated aims include evaluating safety, tolerability, and feasibility compared with usual care. The primary outcome is LVEF by echocardiogram up to 24 months, with an exploratory analysis based on treatment actually received.

What the record does not establish is the size, direction, or precision of an LVEF difference. Numerical LVEF summaries and a comparative estimate are not reported in the registry record. Consequently, it is not possible to determine from this record whether the observed LVEF patterns favored one group, how uncertain any contrast was, or whether a contrast would be clinically meaningful. Missing numerical results should not be read as a finding of no difference.

The recorded serious adverse event and death counts provide descriptive safety information, but the small group counts and lack of comparative uncertainty reporting restrict interpretation. The registry does not establish causality for those events. Nor does a completed status, by itself, demonstrate that every planned outcome was reported or that the available information is conclusive.

More generally, a pilot study can contribute information about study procedures and the feasibility of a strategy, while leaving important questions about treatment effects unresolved. The record does not report feasibility results, a detailed statistical plan, missing-data methods, or a sample-size justification. It also distinguishes randomized elevated-risk groups from a low-risk non-randomized group and describes an as-treated primary outcome analysis. These design and reporting features should be kept in view when considering what comparisons can support.

The findings recorded do not provide a basis for medical decisions. They describe the study question and the information available in the registry record, while leaving the numerical primary outcome and several key analytic details unreported.

13. Why This Trial Matters Statistically

This trial is statistically notable because its question sits at the intersection of risk stratification, randomized treatment assignment, and follow-up cardiac measurements. The registry describes carvedilol for an elevated-risk population alongside usual care, while also listing a low-risk, non-randomized group. Those groups do not all answer the same causal question: randomized comparisons and descriptive comparisons with a non-randomized group have different strengths and limitations.

The primary outcome description further specifies an as-treated analysis. In general, this choice asks how outcomes relate to treatment received, rather than estimating the effect of assignment regardless of treatment uptake. The note that three participants refused carvedilol but remained in follow-up illustrates why those estimands can diverge. Without numerical results or a parallel analysis according to randomized assignment, the record does not allow readers to assess the consequences of this choice for LVEF.

Outcome reporting is also central. LVEF is a continuous cardiac measure summarized by median and inter-quartile range, while the secondary outcomes include imaging measures, cardiac dysfunction, and biomarkers. The registry identifies these measures but does not report their numerical analysis results. This is a reminder that naming an outcome is not equivalent to reporting an estimate, its precision, or the population contributing data.

Finally, the study is identified as a pilot, and its stated purpose includes safety, tolerability, and feasibility. Descriptive event counts may be relevant in that setting, but they are not a substitute for a precise comparative safety analysis. The record offers useful design context, yet the unreported LVEF results, analysis plan, and missing-data details prevent a complete statistical assessment of the primary question.

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 immediate limitation of the registry record is that numerical results for the primary LVEF outcome are not reported. Although the endpoint, measurement approach, summary statistic, and timeframe are described, the record gives no group-specific medians, inter-quartile ranges, comparative estimate, confidence interval, or p-value. The primary statistical question therefore cannot be answered quantitatively from the record.

The primary outcome population description is incomplete: it states that three participants refused carvedilol but remained on study, then ends with an unfinished reference to another participant. The remainder is not reported in the registry record. As a result, the full flow from randomization to treatment received and follow-up cannot be reconstructed from the listed information.

The as-treated analysis can complicate causal interpretation because treatment received may differ from randomized assignment. The low-risk group is expressly non-randomized, so comparisons with it may reflect group differences beyond treatment. The registry does not report adjustment methods, baseline comparability, or a separate assignment-based analysis to clarify these issues.

Other limitations are gaps in statistical reporting. The record does not describe a detailed statistical plan, sample-size justification, multiplicity strategy, missing-data approach, or outcome-specific analysis populations. It lists secondary outcomes but does not provide their numerical findings. Serious adverse event and death counts are reported without uncertainty estimates or enough information to establish event timing, overlap, or cause. These limitations concern what can be concluded from the registry record; they do not establish that analyses or information were absent from other sources.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Risk-Guided Cardioprotection With Carvedilol in Breast Cancer Patients Treated With Doxorubicin and/or Trastuzumab — NCT04023110.
• Registry record last updated 2025-10-09; status last verified 2025-09.
• Results first posted 2025-10-09.
• 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

NCT04023110 is a completed Phase 1, randomized, unmasked, parallel pilot study sponsored by Abramson Cancer Center at Penn Medicine. It lists enrollment of 68 and concerns risk-guided carvedilol for breast cancer patients receiving doxorubicin, trastuzumab, or their combination. The primary outcome is echocardiographic LVEF, summarized by median and inter-quartile range up to 24 months.

The record describes an as-treated exploratory LVEF analysis, randomized elevated-risk groups, and a low-risk non-randomized group. It lists secondary cardiac imaging and biomarker outcomes and reports group-level serious adverse event and death counts. Numerical LVEF results, comparative estimates, uncertainty intervals, a detailed statistical plan, and missing-data methods are not reported in the registry record.