This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Benefit of a Diagnostic MRI in Case of Pathological Nipple Discharge With Normal Mammography / Ultrasound (NCT02819362), a registered phase not stated study. The primary outcome is The Negative Predictive Value (PVV) of MRI.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02819362 |
| Phase | Phase not stated |
| Condition | Breast Cancer |
| Allocation | Not Stated |
| Design model | Not Stated |
| Masking | Not Stated |
| Enrollment | 70 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Institut Bergonié |
| Results posted | Yes |
2. Clinical Question
Population
The intended population was patients with pathological nipple discharge and normal mammography and ultrasound. The registry describes a prospective cohort of patients with pathological nipple discharge and lists Breast Cancer as the condition. It does not report further eligibility details, such as age criteria, exclusions, or how the discharge was defined.
Intervention
The intervention was diagnostic MRI. The clinical question concerns MRI in the setting of pathological nipple discharge despite normal mammography and ultrasound. The registry states that MRI could not be performed as expected because of technical problems with the machine, unrelated to participants or the study. MRI data could not be collected, and the primary outcome could not be assessed.
Comparator
No separate control group or alternative diagnostic strategy is described. This was not a randomized comparison, and the registry does not state an allocation method. It is therefore not a like-for-like comparison of MRI with another approach. The intended endpoint is an assessment within the MRI cohort, rather than a contrast between independently assigned groups.
Primary question
The primary question was the negative predictive value (NPV) of MRI in patients with pathological nipple discharge and normal mammography and ultrasound. The registry defines this as the number of patients with a normal histological examination divided by the number of patients with a normal MRI. Because MRI data could not be collected as expected, this NPV could not be assessed. The record does not report an estimated value or a confidence interval.
3. Trial Design
The registry describes a prospective cohort evaluating MRI in patients with pathological nipple discharge. The listed intervention is MRI, and the cohort description identifies patients with pathological nipple discharge. The intended clinical setting was normal mammography and ultrasound. The record lists an enrollment of 70 and identifies Institut Bergonié as the sponsor.
The study start is recorded as 2014-04, and primary completion as 2016-12. Its status is COMPLETED. These registry entries describe the study record, but completion status does not establish that the primary endpoint was successfully measured. The primary outcome entry explicitly says that MRI could not be performed as expected because of technical problems with the machine, unrelated to participants or the study. MRI data could not be collected and the primary outcome could not be assessed.
This was not a randomized trial. The registry does not state an allocation method, and it describes no separate comparison arm. Consequently, the design cannot provide a randomized contrast between MRI and another diagnostic strategy. The stated endpoint is an NPV within the intended MRI population, not a treatment-effect comparison. The record does not report additional details about recruitment, the conduct of histological examinations, follow-up procedures, or how the cohort was handled after MRI could not be performed as expected.
The registry lists one outcome measure and zero analyses. It does not report a statistical analysis plan, a sample-size justification, or an explanation of how the enrollment target was chosen. The intended design and the available outcome record therefore need to be distinguished: the study was planned around an MRI-based endpoint, but the registry says that endpoint could not be evaluated.
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
| Stage | Prospective Cohort - MRI |
|---|---|
| STARTED | 70 |
| COMPLETED | 0 |
| NOT COMPLETED | 70 |
Reasons for not completing this period:
| Reason | Prospective Cohort - MRI |
|---|---|
| On-site MRI could not be performed | 70 |
Analysis population for the primary outcome, as defined in the registry: MRI could not performed as expected (technical problems with the machine - unrelated to the participants or the study). thus MRI data could not be collected, and as such, primary outcome could not be assessed.
5. Baseline Characteristics
The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.
Population: Population included in the study, satisfying eligibility criteria
| Characteristic | Unit | Prospective Cohort - MRI |
|---|---|---|
| Participants analysed | 70 | |
| Age, Continuous | years | 60 (30 to 87) |
| Sex: Female, Male: Female | Participants | 70 |
| Sex: Female, Male: Male | Participants | 0 |
| Race and Ethnicity Not Collected | Participants | — |
| Region of Enrollment: France | participants | 70 |
6. Primary Endpoint
The registry lists the primary outcome as The Negative Predictive Value (PVV) of MRI, measured over At baseline.
| Element | Registry entry |
|---|---|
| Result type | N/A |
| Measure of spread | not applicable |
| Unit | not stated |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was the negative predictive value of MRI for patients with pathological nipple discharge and normal mammography and ultrasound. As defined in the registry, it is the number of patients with a normal histological examination divided by the number of patients with a normal MRI. In general, NPV describes how often a negative test result corresponds to the specified reference finding in the population being assessed. It is relevant when the clinical question is whether a negative imaging result is accompanied by a normal reference assessment.
This endpoint requires both MRI results and histological findings for the relevant patients. The registry states that technical problems with the machine prevented MRI from being performed as expected, MRI data could not be collected, and the primary outcome could not be assessed. It therefore reports no NPV estimate. The record also does not provide an endpoint value, a confidence interval, or details of the histological assessment. Without the endpoint data, the intended measure cannot quantify the precision of a negative MRI result in this cohort.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The central interpretive fact is that the registry does not report an assessable primary outcome. It describes an intended evaluation of MRI's negative predictive value, but states that technical problems with the machine prevented MRI from being performed as expected. MRI data could not be collected, so the NPV could not be assessed. No numerical estimate or confidence interval is reported. There is therefore no posted endpoint result from which to judge the magnitude or precision of the intended measure.
This absence should not be read as evidence that the NPV was high, low, or equal to any particular value. It is also not a finding that MRI did or did not identify disease. The endpoint requires MRI results, and the registry says those data were unavailable. In general, a confidence interval communicates the range of estimates compatible with the observed data under the analysis method; when the primary outcome cannot be assessed, there is no reported endpoint estimate or interval to interpret.
The design also limits what the study could establish about comparative performance. It is described as a prospective cohort, not a randomized comparison, and no like-for-like control group is reported. Even if an MRI-based endpoint had been available, the design as described would address the specified NPV in the cohort rather than establish a randomized difference between diagnostic strategies. The registry does not report a comparative estimate.
The adverse-event record lists zero serious events, zero other events, and zero deaths, but also lists zero participants at risk in the reported group. Those entries do not provide an interpretable safety rate: without a reported at-risk denominator, zero event counts cannot be taken as evidence that no adverse events occurred among an evaluable cohort. The record does not provide further safety detail.
Overall, the posted record documents a completed study record and an enrollment of 70, alongside a primary outcome that could not be assessed because MRI data were not collected as expected. The registry does not report analyses or numerical primary results. Conclusions about diagnostic performance, precision, or comparative benefit are therefore not supported by the reported endpoint data.
9. Statistical Methods Explained
Several statistical ideas clarify what the planned endpoint would have meant and why the registry's missing MRI data matter. These are general explanations; the registry record does not report a completed endpoint analysis.
What does negative predictive value describe?
In general, negative predictive value (NPV) is the proportion of negative test results that are accompanied by the specified negative reference finding. Here, the registry defines the numerator as patients with a normal histological examination and the denominator as patients with a normal MRI. The intended population is patients with pathological nipple discharge and normal mammography and ultrasound. The definition ties the interpretation to this population and the reference assessment used; it is not a measure that can automatically be transferred to other clinical settings.
NPV is conditional on the group being assessed and the way test results and reference findings are established. The registry does not report details of the histological procedures or the classification rules used for MRI results. Those details matter when interpreting an NPV because the result depends on which participants contribute to the numerator and denominator and how normal findings are determined. The intended ratio is described, but the data needed to calculate it were not collected as expected.
Why is the missing endpoint decisive?
A point estimate needs observed information for its numerator and denominator. A confidence interval additionally reflects uncertainty around that estimate under a specified method. The registry states that MRI could not be performed as expected because of technical problems with the machine and that MRI data could not be collected. As a result, it says the primary outcome could not be assessed. No NPV estimate or confidence interval is reported, so neither the endpoint's direction nor its precision can be evaluated from the posted results.
In general, missing outcome information is different from an observed negative result. An unavailable MRI result cannot be classified as normal or abnormal for the endpoint, and it cannot contribute a documented denominator of normal MRI findings. The record does not describe how many participants were affected by the technical problem, whether any endpoint data were available from another source, or how any missing data would have been handled. Those details are not reported in the registry record.
What does the cohort design imply?
A prospective cohort follows participants according to a study plan, rather than assigning participants at random to competing interventions. The registry describes a prospective MRI cohort and does not report a separate control group or stated allocation method. This distinction matters: a within-cohort NPV can describe the relationship between MRI and the reference finding in the studied population, but it is not itself a randomized comparison of MRI against another strategy. A comparison requires appropriate data from the groups or strategies being contrasted, and such a comparison is not described in this record.
In general, random allocation can help balance measured and unmeasured participant characteristics between comparison groups. It does not apply to the design because the registry does not report randomized allocation. That does not prevent a cohort from addressing a descriptive diagnostic question, but it limits claims about differences attributable to one diagnostic strategy versus another. In this record, the primary obstacle is more basic: the specified MRI outcome could not be assessed.
How should planning calculations be read?
In general, sample-size calculations are planning exercises that describe the amount of information a hypothetical study might need under stated assumptions. The registry's planning assumptions describe a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. Such calculations are not observed trial results and are not observed power. In particular, an illustration based on a hypothetical two-arm comparison does not turn this prospective cohort into a randomized two-arm trial.
Planning illustrations also depend on the effect and assumptions used. The record does not report the completed endpoint estimate, and the primary outcome entry says it could not be assessed. A planning calculation cannot fill that evidence gap or serve as an estimate of diagnostic performance. It should be interpreted as a design illustration, not as proof that the study had a particular probability of detecting a real effect.
What can the event counts establish?
In general, event counts need a meaningful denominator to describe risk or frequency. The registry lists zero serious events, zero other events, and zero deaths, while the reported at-risk counts are zero. These entries do not establish an event rate for an evaluable group. The record does not report additional safety analyses, confidence intervals, or a safety population definition. The counts should therefore be read as registry entries, not as a quantified demonstration of safety.
10. Multiplicity, Missing Data and Analysis Populations
The registry record does not report a plan for handling multiplicity. It lists one outcome measure and zero analyses, but does not state whether there were secondary outcomes, multiple hypotheses, planned subgroup analyses, or adjustments for multiple testing. Without a reported analysis plan, it is not possible to determine how any potential multiplicity would have been addressed. The primary endpoint itself was not assessable, so no multiplicity-adjusted or unadjusted endpoint result is reported.
Missing data are central to interpretation. The primary outcome entry says that MRI could not be performed as expected because of technical problems with the machine, unrelated to participants or the study. It also says MRI data could not be collected and the primary outcome could not be assessed. The registry does not report how many participants were affected, whether any endpoint observations remained, or any method for handling missing results. It does not describe imputation, exclusions, sensitivity analyses, or another missing-data approach. These details are not reported in the registry record.
The analysis population is also not defined in the available record. Although enrollment is listed as 70, the registry does not state which participants would have been included in an endpoint analysis or whether an analysis population was specified. No endpoint analysis is listed. The record therefore does not support claims about an intention-to-treat population, a per-protocol population, or an evaluable population.
The record's adverse-event entries require similar care. The listed at-risk counts are zero, as are the event counts. With no reported at-risk denominator, the event counts cannot be interpreted as rates or as proof that there were no events in an assessable study population.
11. What the Result Does, and Does Not, Mean
The registry establishes the intended question: assessing MRI's negative predictive value in patients with pathological nipple discharge and normal mammography and ultrasound. It also records a prospective cohort design, an enrollment of 70, and completion status. Most importantly for statistical interpretation, the primary outcome entry says that MRI data could not be collected as expected because of machine-related technical problems, and that the outcome could not be assessed.
That record does not provide a numerical estimate of NPV, its precision, or a comparison with another diagnostic strategy. It does not show that a normal MRI was followed by a normal histological examination, nor does it show the opposite. An unavailable primary endpoint is not a negative endpoint result. The findings cannot be used to characterize the frequency of normal histological findings among patients with normal MRI results because the required endpoint data are not reported.
The enrollment figure does not resolve this limitation. It describes enrollment, not the number of participants with usable MRI and histological results, and the record does not state the size of an endpoint-analysis population. Likewise, the COMPLETED status describes the registry's study status; it does not mean that the intended outcome was successfully measured.
The record also does not establish comparative effectiveness. It describes no randomized allocation or like-for-like control group. An NPV estimate, even if reported, would answer a specific within-population diagnostic question and would not by itself establish a causal advantage over another strategy. No such estimate is available in any case.
The adverse-event entries are not a substitute for an endpoint or safety analysis: the registry lists zero events and zero at-risk counts. Those entries lack an interpretable safety denominator. The appropriate reading is therefore narrow: the registry documents the planned MRI question and the reason the primary endpoint was not assessable, but it does not report evidence that resolves that question.
12. Why This Trial Matters Statistically
This study is statistically instructive because it separates a well-defined intended endpoint from an endpoint that could not be evaluated. The registry specifies an NPV question and defines its numerator and denominator. That definition makes clear what evidence would have been needed: MRI results and histological findings for the relevant population. The record then states that MRI data could not be collected as expected, so the intended measure could not be calculated.
This distinction is important in clinical research. A study can have a stated question, enrollment, and completed status without producing an analyzable answer to its primary question. A missing endpoint is not a small or merely presentational issue: it prevents estimation of the measure itself and therefore prevents assessment of its precision. It also demonstrates why a registry's outcome description and the posted results need to be read together.
The record also illustrates the boundary between descriptive diagnostic measures and comparative claims. A cohort-based NPV concerns the correspondence between a negative MRI and a normal histological examination in a defined population. It is not, by itself, a randomized comparison of MRI with an alternative approach. The design description therefore matters when deciding what conclusions a result could support, even if the intended endpoint had been observed.
Finally, the adverse-event entries show why event counts should be interpreted alongside their denominators. Zero recorded events with zero participants listed at risk cannot quantify an event frequency. Across both diagnostic and safety outcomes, the available record supports attention to data availability, endpoint definitions, and analysis populations rather than a claim about clinical performance.
13. 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 trial-specific limitation is that the registry says the primary endpoint could not be assessed. MRI could not be performed as expected because of technical problems with the machine, unrelated to participants or the study, and MRI data could not be collected. The record does not report an NPV estimate, a confidence interval, or another result that could answer the primary question.
The extent of the missing MRI information is not described. Although enrollment is listed as 70, the registry does not report how many participants had usable MRI data, how many had histological findings, or which participants would have entered an endpoint analysis. It also does not report whether any alternative data collection or missing-data analysis was attempted. No analysis population is specified.
The record describes a prospective cohort and does not report random allocation or a separate comparison group. It therefore does not provide a like-for-like randomized comparison of MRI with another diagnostic strategy. Further details about recruitment, eligibility criteria, the MRI protocol, histological assessment, follow-up, and participant flow are not reported in the registry record.
The registry lists zero serious events, zero other events, and zero deaths, but also lists zero participants at risk. These entries do not provide an interpretable denominator for safety assessment. The record does not report a safety analysis population or additional adverse-event detail.
Finally, the registry does not report a statistical analysis plan, sample-size justification, multiplicity approach, or missing-data method. These omissions limit assessment of the planned analysis as well as interpretation of the unavailable primary result.
14. Related Tutorials
15. Related Statistical Calculators
16. Sources
- ClinicalTrials.gov: Benefit of a Diagnostic MRI in Case of Pathological Nipple Discharge With Normal Mammography / Ultrasound — NCT02819362.
- Registry record last updated 2025-10-15; status last verified 2025-09.
- Results first posted 2025-10-15.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
17. Record Summary
NCT02819362 is a completed prospective cohort record sponsored by Institut Bergonié. It lists an enrollment of 70, a study start of 2014-04, and primary completion of 2016-12. The intervention is MRI, for patients with pathological nipple discharge; the intended setting was normal mammography and ultrasound.
The primary outcome was the negative predictive value of MRI, defined in the record using normal histological examination among patients with normal MRI. The outcome entry states that MRI could not be performed as expected because of machine-related technical problems, MRI data could not be collected, and the primary outcome could not be assessed. The registry lists one outcome measure and zero analyses, and reports no endpoint estimate.
The study is not described as a randomized comparison, and no control group is reported. Adverse-event entries list zero events and deaths alongside zero at-risk counts, which do not provide an interpretable safety denominator.