PET-MR for Prediction and Monitoring of Response to Neoadjuvant Chemotherapy in Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for PET-MR for Prediction and Monitoring of Response to Neoadjuvant Chemotherapy in Breast Cancer (NCT01190566), a registered phase not stated study. The primary outcome is Patholocial Response to Chemotherapy.

2. Clinical Question

3. Trial Design

The registry record describes a completed study sponsored by Seoul National University Hospital, with enrollment of 57 participants. The listed start is 2010-05 and primary completion is 2013-05. Breast cancer is the recorded condition. The study objectives focus on imaging in patients receiving neoadjuvant chemotherapy, with the purpose of predicting and monitoring response.

The imaging approaches named in the objectives include multiparametric MRI, FDG-PET, RGD-PET, and PET-MR fusion imaging. The record also identifies MR spectroscopy, diffusion-weighted MRI, and dynamic contrast-enhanced MRI. One stated objective is to identify an optimal combination of imaging parameters. Another is to compare parametric response map analysis with pharmacokinetic parameters Ktrans, kep, and Ve for early prediction of pathological responsiveness.

The primary endpoint is recorded after operation and classifies participants as pCR or non-pCR. The record lists nine secondary outcome measures, including tumor size and volume, parametric response map signal intensity, the pharmacokinetic parameters, tumor choline measured by single voxel 1H-magnetic resonance spectroscopy, and standardized uptake value on 18F-fluoro-deoxy-glucose PET.

The design does not support a treatment-versus-control interpretation. The registry record says this is not a randomized comparison, allocation is not stated, and the arms are not a like-for-like control comparison. It does not specify the number or definition of imaging groups, the imaging schedule, or the details of chemotherapy. 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.

Overall Study

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Patholocial Response to Chemotherapy, measured over Post-operation.

The primary endpoint is pathological response to chemotherapy, recorded as pCR or non-pCR at post-operation. This is a binary outcome: each participant is classified into one of the two stated categories. Pathological response is used as the target against which the study aims to evaluate imaging measures for prediction and monitoring. In this study’s objectives, imaging measurements are intended to indicate whether a participant will have a pathological response after neoadjuvant chemotherapy.

The registry record lists 6 participants in the pCR grouping and 42 in the non-pCR grouping in its adverse-event information. These category counts do not substitute for a reported endpoint analysis: there is no estimate of imaging performance, association, or predictive accuracy, and no uncertainty measure or statistical test is reported. The enrollment is 57, while the outcome-group entries do not account for all enrolled participants; participant disposition and the reason for this difference are not reported in the registry record.

The endpoint’s timing is described only as post-operation. The record does not give a detailed pathological definition of complete response, identify who assessed response, or report how missing or indeterminate pathology results were handled. Without those details and the corresponding analysis, the endpoint definition indicates what was classified, but not how reliably or precisely the imaging objectives were answered.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is binary, with participants categorized as pCR or non-pCR. In general, a binary endpoint can be described using the number of participants in each category and the proportion in each category. For an imaging prediction study, further analyses might assess how imaging values relate to the endpoint or how well a rule based on imaging separates response categories. These are general statistical approaches, not analyses reported in this registry record. The record gives no endpoint model, predictive-performance measure, estimate, or uncertainty interval.

What would an estimate and confidence interval convey?

In general, an estimate expresses the observed size or direction of an association or difference. A confidence interval communicates the uncertainty around that estimate under the assumptions of the analysis. A narrow interval indicates greater statistical precision than a broad interval, while neither interval width nor a p-value alone establishes clinical importance. The registry record does not report an estimate or confidence interval for the primary endpoint or the imaging measures, so precision cannot be evaluated.

How should the imaging objectives be distinguished?

The objectives cover several related but distinct tasks. Prediction concerns using an imaging measurement to anticipate the later pathological category. Monitoring concerns evaluating imaging measurements over the course of treatment. Comparing imaging approaches concerns assessing their relative performance. In general, these questions call for clearly specified outcomes, measurement times, and analysis methods. The record names the imaging modalities and some parameters, but does not report the schedule, predictive criteria, model specification, or performance metrics used. It is therefore not possible to tell from the registry record how these tasks were operationalized.

The named MRI measures include parametric response map signal intensity and the pharmacokinetic parameters Ktrans, kep, and Ve. The record describes Ktrans as a transfer constant from plasma to the extracellular extravascular space, kep as the rate constant for escape from that space back into plasma, and Ve as the extracellular extravascular space per unit volume of tissue. In general, such measurements need to be interpreted in light of their measurement procedures and analysis definitions. Those procedures and any actual parameter estimates are not reported in the registry record.

Why does the comparison structure matter?

Statistical comparisons require a defined contrast: what groups or measures are being compared, in which participants, and at what time. The registry’s objectives mention comparison of dynamic contrast-enhanced MRI parametric response map analysis with pharmacokinetic parameters for early prediction. But the record does not report the comparison’s formal statistical method or a result. It also says the study is not a randomized comparison, allocation is not stated, and the arms are not a like-for-like control comparison. Accordingly, no treatment effect should be inferred from group differences.

The record includes a sample-size planning description based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation; it describes planning for a stated fraction of the observed effect. Such calculations are planning illustrations, not observed power and not evidence that an effect was established. The record does not provide the numerical planning results or the trial’s own sample-size justification. It is not reported in the registry record whether the study used a prespecified effect for design, or whether the planning illustration reflects the original design.

In general, sample-size calculations depend on assumptions about the outcome, the target effect, variability, allocation, and the chosen error rate. If those assumptions differ from the actual study conditions, the resulting planning estimate may not describe the study’s ability to estimate an effect. A calculation based on an observed effect is especially unsuitable as a retrospective claim that the study had adequate power: the observed effect is itself uncertain. No observed-power conclusion follows from the registry information.

For count outcomes, the distributional assumptions of an analysis also matter. General statistical methods for counts may need to account for over-dispersion, which occurs when variability exceeds that expected under a simple count model. The primary endpoint is listed as a binary participant outcome, not a count of recurrent events, and no model or dispersion estimate is reported. The nine secondary outcomes include continuous imaging measurements, but the record does not report their distributions or analysis methods. It is therefore not possible to identify which model assumptions were used or whether they were checked.

Multiplicity arises when several outcomes, imaging parameters, or comparisons are assessed. In general, making many statistical comparisons can increase the chance of at least one apparently notable result by chance. Researchers may address this through prespecified primary comparisons, adjusted testing, or a clearly exploratory interpretation. The record lists a primary endpoint and nine secondary outcome measures, but does not report a multiplicity strategy or identify which analyses, if any, were prespecified. No conclusion about adjustment can be drawn from the list of outcomes alone.

The record does not provide analysis populations, missing-data methods, covariate adjustments, or interim-analysis information. These details affect how results are interpreted, but cannot be reconstructed from the study objectives or outcome names. In the absence of reported estimates and methods, the sound statistical reading is descriptive: the registry states what the study planned to assess and how the endpoint was categorized, while leaving the results and inferential basis unreported.

10. Multiplicity, Missing Data and Analysis Populations

The registry record names one primary endpoint and nine secondary outcome measures. The secondary measures span tumor size and volume, parametric response map signal intensity, pharmacokinetic parameters, tumor choline amount, and PET standardized uptake value. A study that examines multiple measures or comparisons raises a general multiplicity issue: the more questions assessed, the greater the possibility of chance findings unless the analysis plan accounts for that breadth. The registry does not report whether any multiplicity adjustment was planned or used, nor does it identify a hierarchy among the secondary outcomes.

Missing-data handling is also not described. The record reports enrollment of 57 and lists 6 pCR participants and 42 non-pCR participants in the adverse-event grouping, without explaining the remaining participant disposition. It does not say whether the difference reflects missing pathology, withdrawal, an ineligible analysis record, or another reason. Those explanations are not reported in the registry record and should not be inferred.

The analysis population is not specified. The record does not state whether the endpoint analysis included all enrolled participants, only those with evaluable postoperative pathology, or another defined group. It also does not report methods for missing imaging measurements or incomplete follow-up. These omissions limit interpretation of any endpoint count and would matter when assessing possible bias, precision, and generalizability.

The registry lists zero analyses and does not report interim analyses, a statistical analysis plan, or model details. Thus, it is not possible to determine whether outcomes and methods were prespecified, how missingness was addressed, or whether multiple comparisons were controlled. These items are not reported in the registry record.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The adverse-event information is organized by pathological response category rather than by a randomized treatment arm. In the pCR grouping, 6 participants are listed at risk for serious adverse events and 0 with a serious adverse event; the corresponding other-adverse-event entries are 6 at risk and 0 affected. In the non-pCR grouping, 42 participants are listed at risk for serious adverse events and 0 with a serious adverse event; the corresponding other-adverse-event entries are 42 at risk and 0 affected.

These entries describe what is listed in the registry record. They do not establish that no adverse events occurred among all 57 enrolled participants: the displayed at-risk groups do not account for all enrollment, and the status of other participants is not explained. Deaths are recorded as null, which means death information is not reported in the registry record; it should not be read as a reported zero.

The grouping by response does not provide a treatment-safety comparison. The record states that this is not a randomized comparison, allocation is not stated, and the arms are not a like-for-like control comparison. No serious adverse event terms are listed. The adverse-event figures should therefore be read as registry counts within the recorded categories, not as evidence of a causal safety difference or a complete account of safety for the enrolled population.

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

The study addresses a relevant statistical problem: whether imaging measurements can help anticipate or monitor pathological response to neoadjuvant chemotherapy. It names several candidate methods and defines a postoperative binary response endpoint. The registry record also identifies specific MRI parameters and PET-related measures that were intended for evaluation. These details clarify the study’s scientific focus.

However, the record does not report the analyses needed to answer its stated questions. There is no estimate of predictive performance, no comparison between imaging approaches, and no uncertainty measure. The participant category counts in the adverse-event information do not establish that imaging predicted response. Without linked imaging and pathology results, the relationship between the measurements and the endpoint cannot be assessed.

The record also does not establish that any imaging approach is superior or inferior to another. The aims describe planned evaluations, but no results are given for the comparison of parametric response map analysis with Ktrans, kep, or Ve. Likewise, the listed secondary outcomes are measures of interest, not reported findings.

The design information limits causal interpretation. The record explicitly says the study is not a randomized comparison, that allocation is not stated, and that the arms are not a like-for-like control comparison. Any differences among categories or measurement groups would not, on this information, demonstrate a causal treatment effect. The record’s stated focus is imaging assessment of response rather than a randomized comparison of chemotherapy strategies.

In general, an absence of reported results is not evidence that an imaging method failed, and it is not evidence that it succeeded. It means the registry record does not permit readers to judge the direction, size, or precision of the study findings. The distinction matters: objectives describe the question; estimates and their uncertainty are needed to evaluate the answer. The available information supports a description of the study rationale and endpoint, but not a conclusion about imaging accuracy or clinical benefit.

13. Why This Trial Matters Statistically

This study is statistically interesting because it places several imaging modalities and measurement types against a clinically interpretable endpoint: postoperative pathological response categorized as pCR or non-pCR. The objectives involve prediction, monitoring, and comparison of measurements, which are related but not interchangeable statistical tasks. Prediction asks whether measurements can classify a later outcome; monitoring asks how measurements change over treatment; comparison asks whether one approach provides more information than another. Each requires a clear analysis definition.

The range of measures also highlights the challenge of evaluating many candidate predictors. The record includes MRI-derived parameters, PET uptake, tumor size and volume, and choline amount. In general, a broad set of candidates can create multiplicity and model-selection concerns. Prespecified comparisons and transparent reporting help readers distinguish confirmatory findings from exploratory ones. The registry does not report a strategy for doing so.

Another important issue is the difference between enrollment and the outcome categories listed in the adverse-event information. The record reports enrollment of 57, while its response group entries list 6 and 42 participants. The participant disposition and analysis population are not explained. That gap affects the interpretation of denominators and is a reminder that an endpoint count is most useful when its population and missing-data handling are clear.

Finally, the record’s explicit warning about the nonrandomized comparison structure matters. Imaging groups or response categories should not be treated as interchangeable treatment arms, and the information does not support causal claims about treatment. The trial’s statistical value can only be evaluated fully when analysis methods, estimates, and uncertainty are available; these are not reported in the registry 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 main limitation of the registry information is the absence of reported efficacy or imaging-analysis results. The record lists the primary endpoint and nine secondary outcomes, but provides no effect estimates, predictive-performance measures, confidence intervals, or p-values. It also lists zero analyses. Readers therefore cannot assess the magnitude or precision of any relationship between imaging and pathological response.

The participant flow is unclear. Enrollment is 57, while the adverse-event grouping lists 6 pCR participants and 42 non-pCR participants. The record does not explain the status of participants who are not represented in those entries, or define the population used for endpoint assessment. Missing pathology, incomplete imaging, exclusions, and other reasons are not reported in the registry record.

The record gives limited design and measurement detail. It does not specify eligibility criteria beyond the broad population description, the chemotherapy regimen, imaging schedules, imaging acquisition or interpretation procedures, or the pathological criteria used to define pCR. The analysis plan, covariate adjustments, handling of missing values, and multiplicity approach are also not reported. These omissions prevent evaluation of measurement consistency and analytical choices.

The registry states that this is not a randomized comparison, allocation is not stated, and the arms are not a like-for-like control comparison. Consequently, any group contrast should not be interpreted as a randomized treatment effect. The record also does not give the trial’s own sample-size justification. Its planning description is based on the observed effect and is not evidence of observed power or of a prespecified design rationale.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: PET-MR for Prediction and Monitoring of Response to Neoadjuvant Chemotherapy in Breast Cancer — NCT01190566.
• Registry record last updated 2015-07-07; status last verified 2015-06.
• Results first posted 2015-07-07.
• 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

NCT01190566 is a completed study sponsored by Seoul National University Hospital, with enrollment of 57 participants. The listed start is 2010-05 and primary completion is 2013-05. The study concerns PET-MR and related imaging approaches for prediction and monitoring of response to neoadjuvant chemotherapy in breast cancer.

The primary endpoint is postoperative pathological response, categorized as pCR or non-pCR. The record lists nine secondary outcome measures spanning tumor measurements, MRI parameters, spectroscopy, and PET uptake. Its objectives include evaluating imaging combinations and comparing dynamic contrast-enhanced MRI parametric response map analysis with pharmacokinetic parameters.

The record reports no statistical analyses or comparative outcome estimates. It states that the study is not a randomized comparison, allocation is not stated, and the arms are not a like-for-like control comparison. Participant disposition, analysis populations, missing-data methods, multiplicity handling, and statistical results are not reported in the registry record.