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 A Validation Study of MR Lymphangiography Using SPIO, a New Lymphotropic Superparamagnetic Nanoparticle Contrast (NCT00147238), a non-randomized na study. The primary outcome is Sensitivity of MRI Per Patient.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00147238 |
| Phase | NA |
| Condition | Bladder Cancer, Genitourinary Cancer, Prostate Cancer |
| Allocation | Non-Randomized |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 10 |
| Arms in the primary outcome | n/a |
| Lead sponsor | M.D. Anderson Cancer Center |
| Results posted | Yes |
2. Clinical Question
Population
The study concerned people with bladder cancer, genitourinary cancer, or prostate cancer. Its stated purpose was to assess detection of cancer in pelvic lymph nodes, including malignant pelvic lymph nodes. The registry record does not report further eligibility details or a more specific description of the population that underwent imaging.
Intervention
The imaging approach of interest was MR lymphangiography using ferumoxtran-10, also described as USPIO. The study aimed to evaluate how well this contrast agent could support detection of cancer in pelvic lymph nodes. The registry record does not report the primary outcome as assessed.
Comparator
The comparison was between two MR images from the same patient: an image without ferumoxtran-10 contrast and a second image with contrast. The second MRI was described as occurring within 24-36 hours of contrast injection and about 24 hours after the first MRI. This is a paired imaging comparison, not a randomized comparison between separate treatment groups.
Primary question
The planned primary question was whether sensitivity for detecting disease on a per-patient basis differed between the paired MR images. The registry record specified a two-sided McNemar test for that comparison. Because the primary outcome measure was not assessed after early study termination, the record does not provide a sensitivity estimate or a test result.
3. Trial Design
This was a non-randomized, single-group study with no masking. The registry lists an enrollment of 10 and a status of terminated. The study began in 2005-07, and its primary completion date is listed as 2007-01. The sponsor was M.D. Anderson Cancer Center. The phase is listed as NA, and the study type is described as other.
The imaging plan involved MR lymphangiography before ferumoxtran-10 contrast and a repeated MRI with contrast. The outcome timeframe describes the second MRI as occurring within 24-36 hours of contrast injection and about 24 hours after the first MRI. The arm descriptions distinguish MR lymphangiography using ferumoxtran-10 from MR lymphangiography before injection. Because the images are paired within patients, the planned comparison concerns imaging conditions, not randomized assignment to separate groups.
The non-randomized designation and single-group model matter when interpreting the design. The registry record explicitly indicates that the arms are not a like-for-like control comparison. The pre-contrast image is a within-patient comparator, but it does not make this a randomized controlled trial. No masking was used, according to the record.
The primary outcome measure was not assessed due to early study termination, with the record giving patients not receiving assigned treatment as an example. The registry record does not report further details about why the study terminated, how many participants completed each imaging step, or whether any participants were excluded from the outcome assessment. It also does not report a protocol-level sample-size justification or interim-analysis plan.
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 10 patients enrolled prior to sponsor's request for early termination. Only 2 patients were evaluable for response therefore data will not be analyzed due to too small sample size.
Overall Study
| Stage | Ferumoxtran-10 MRI Contrast Agent |
|---|---|
| STARTED | 10 |
| COMPLETED | 2 |
| NOT COMPLETED | 8 |
Reasons for not completing this period:
| Reason | Ferumoxtran-10 MRI Contrast Agent |
|---|---|
| Inevaluable | 8 |
Analysis population for the primary outcome, as defined in the registry: Primary outcome measure was not assessed due to early study termination (e.g. patients did not receive assigned 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.
| Characteristic | Unit | Ferumoxtran-10 MRI Contrast Agent |
|---|---|---|
| Participants analysed | 10 | |
| Age Continuous | years | 64 (55 to 73) |
| Sex: Female, Male: Female | Participants | 8 |
| Sex: Female, Male: Male | Participants | 2 |
| Region of Enrollment: United States | participants | 10 |
6. Primary Endpoint
The registry lists the primary outcome as Sensitivity of MRI Per Patient, measured over MRI without ferumoxtran-10 contrast and second repeated MRI with contrast agent within 24-36 hours of contrast injection, about 24 hours after first MRI.
| Element | Registry entry |
|---|---|
| Result type | N/A |
| Measure of spread | not applicable |
| Unit | not stated |
| Endpoint classification | not reduced to a single comparison |
The planned primary endpoint was MRI sensitivity on a per-patient basis. Sensitivity describes the ability of an imaging assessment to identify people who have the target condition, according to the reference standard used. The registry record does not report the reference standard, so the basis for classifying a patient as having or not having pelvic lymph-node cancer is not specified in the posted information.
The endpoint was intended to compare sensitivity for two paired images: MRI without ferumoxtran-10 contrast and a repeated MRI with contrast. The record specifies a two-sided McNemar test to assess differences between the sensitivities. This method is suited to paired binary outcomes because it focuses on patients whose classifications differ between the two imaging conditions, rather than treating the images as observations from independent groups.
The outcome description says that sensitivity was to be written as a proportion, from 0.0 to 1.0. However, the registry record states that the primary outcome measure was not assessed due to early termination. It therefore gives no observed sensitivity for either imaging condition, no difference between them, and no McNemar test result. The endpoint explains the intended clinical measurement, but it does not establish how well either imaging approach performed in this study.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The main statistical fact is that the planned primary outcome was not assessed. The registry record therefore provides no sensitivity estimate for MRI without contrast, no sensitivity estimate for MRI with ferumoxtran-10, and no statistical comparison of those paired assessments. It would be inappropriate to interpret the absence of an outcome result as evidence that the two imaging conditions performed similarly or differently.
The study was designed around a within-patient comparison. Such a design can hold constant characteristics of an individual that might vary between separate groups, because each person contributes images under both conditions. That design feature does not compensate for the missing outcome data: without the paired diagnostic classifications, the sensitivity comparison cannot be estimated. The record also does not state what reference standard was used to identify pelvic lymph-node disease, an important detail for interpreting a diagnostic-accuracy endpoint.
The enrollment is listed as 10, but the record does not establish how many participants completed both imaging conditions or had data suitable for the planned analysis. It reports early termination and gives nonreceipt of assigned treatment as an example, but does not describe the extent or pattern of missing imaging or outcome data. Thus, enrollment alone cannot be treated as the analysis population for the endpoint.
The registry lists no analyses. This means that no primary statistical result is available in the record; it does not mean that a statistical test found no difference. There is no reported p-value, confidence interval, or estimate of effect for the paired sensitivity comparison. A confidence interval, in general, communicates the precision of an estimate by showing a range of values compatible with the data and analysis assumptions. No such interval is reported for this endpoint.
The study's single-group, non-randomized design also shapes what a comparison would mean. It was not a comparison of randomized arms, and the registry record cautions that the arms are not a like-for-like control comparison. The images were intended to be compared within the same patient. Any resulting interpretation would therefore concern the diagnostic performance of the paired imaging conditions in the assessed participants, not a broad causal treatment effect.
Serious adverse event figures are recorded separately from the unassessed primary outcome. The registry lists 0 serious adverse events among 10 at risk in the Ferumoxtran-10 MRI Contrast Agent group. It also lists 0 other adverse events among 10 at risk. Those recorded counts do not supply the missing diagnostic-accuracy result, nor do they establish that an event could not occur. The record gives no deaths value.
Overall, the posted information documents the planned diagnostic comparison, study status, and limited adverse-event counts, but it does not answer the primary question. The termination and lack of an assessed endpoint sharply limit statistical interpretation. No claim about the sensitivity or comparative performance of the imaging approaches can be drawn from an outcome that was not measured or reported.
9. Statistical Methods Explained
The planned primary analysis was a two-sided McNemar test comparing sensitivity from paired MR images, one without ferumoxtran-10 and one with contrast. The endpoint was defined per patient. The registry record does not report the outcome data needed to carry out this analysis, and no analyses are listed. The methods below explain the stated approach in general; they do not imply that a test was completed in this study.
Why use a paired comparison?
In a paired design, each participant contributes an assessment under both imaging conditions. The two observations from one person are related, so they should not be treated as if they came from independent groups. A paired analysis uses that relationship. In diagnostic comparisons, some patients may receive the same classification under both conditions, while others may be classified differently. The pattern of differing classifications is central to a McNemar comparison.
In general, the McNemar test evaluates whether discordant paired outcomes occur in one direction more often than the other, under the test's assumptions. Concordant pairs—those classified alike by both conditions—do not provide the same information about a difference between conditions as discordant pairs. The registry does not report any paired classifications, so neither the number nor direction of discordant pairs is known. No test statistic or p-value can therefore be recovered from the record.
What does sensitivity mean?
In general, sensitivity is the proportion of people with the target condition who are identified as positive by the test under evaluation. Estimating it requires both an imaging classification and a determination of which participants truly have the target condition. That determination ordinarily depends on a reference standard. The registry record states that sensitivity was the primary measure, but does not report the reference standard or the observed classifications. It also states that the outcome was not assessed due to early termination.
Because this was a paired comparison, the analysis would need the relevant diagnostic information for both images from the same participants. Separate sensitivity estimates alone would not necessarily be enough to reproduce a paired test, because the test depends on how the two image classifications match or differ within each participant. The record contains no such paired outcome data.
What would a two-sided test address?
In general, a two-sided test evaluates evidence for a difference in either direction, rather than testing only whether one specified condition is better. A p-value from such a test, if available, would describe how compatible the observed data are with a null hypothesis of no difference, under the test assumptions. It would not by itself measure the size or clinical importance of a difference. Here, no p-value or estimated difference is reported, because the primary outcome was not assessed.
For a complete account of diagnostic performance, an estimate should be considered with its uncertainty, and the reference standard and analysis population should be clear. A confidence interval expresses uncertainty around an estimate; narrower intervals generally indicate greater precision under the model, while wider intervals indicate less precision. The registry record does not report an estimate or confidence interval for sensitivity. Any precision assessment for the primary endpoint is consequently unavailable.
How should planning illustrations be read?
Sample-size illustrations based on an observed effect are planning exercises, not results from the trial and not evidence that a study had a particular probability of finding a result. The planning assumptions described for these illustrations are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is defined as a stated fraction of an observed effect. Those assumptions do not describe the design used for this single-group paired imaging study, and the registry does not report an observed primary effect from which to infer its performance.
In general, a planning calculation asks what sample size might be needed under specified assumptions to detect a chosen effect with a chosen power. It does not tell whether the actual study detected that effect. Planning figures are sensitive to assumptions about the effect size, variability, allocation, analysis, and outcome completeness. They should not be read as observed power, and they cannot substitute for an unreported endpoint analysis. The registry record does not report a protocol-based sample-size justification for the trial.
Some planning methods address over-dispersion, which occurs when observed variability is greater than a model assumes. In general, failing to account for over-dispersion can make uncertainty appear smaller than it is and can lead to overly optimistic planning. This concept does not supply missing data for the paired sensitivity endpoint. The registry record does not describe an over-dispersion adjustment for the planned McNemar analysis.
The important distinction is between the stated intended method and the results actually available. The record identifies a paired, two-sided McNemar test for the primary sensitivity comparison, but records that the outcome was not assessed and lists no analyses. Consequently, the intended method can be explained, but there is no reported inferential result to interpret.
10. Multiplicity, Missing Data and Analysis Populations
The registry record lists one outcome measure and describes one primary endpoint: per-patient MRI sensitivity comparing images with and without ferumoxtran-10. It does not report a completed analysis of that endpoint. The record lists no analyses, so it provides no statistical testing results from which to assess whether multiple testing occurred or whether any adjustment was made.
Multiplicity arises in general when several outcomes, comparisons, time points, or analytical choices are tested. If many tests are conducted, the chance of obtaining at least one apparently notable result can increase. Researchers may plan methods to account for this, but the registry record does not report a multiplicity strategy, a hierarchy of outcomes, or additional planned analyses. No such plan should be inferred from the single listed outcome measure.
The primary endpoint was not assessed due to early termination. The record does not report how much outcome information is missing, whether participants completed both imaging conditions, or how missing observations would have been handled. It also does not specify an analysis population, such as a defined group of all enrolled participants or a subset with complete paired images. The reason given for nonassessment includes patients not receiving assigned treatment as an example, but further details are not reported in the registry record.
These omissions matter because a paired diagnostic analysis depends on usable information from both imaging conditions and on a reference classification. If either image or the reference classification is unavailable, the relevant paired outcome may not be analyzable as intended. The registry record does not state whether any imputation, exclusion, or sensitivity analysis was planned or conducted. With the primary outcome unassessed, neither missing-data assumptions nor analysis-population choices can be evaluated from the posted information.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Ferumoxtran-10 MRI Contrast Agent | 0 | 10 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Ferumoxtran-10 MRI Contrast Agent | 0 of 10 |
The registry lists one adverse-event group titled Ferumoxtran-10 MRI Contrast Agent. For that group, 10 participants are recorded as at risk for serious adverse events, and 0 are recorded as affected. The corresponding figures for other adverse events are 10 at risk and 0 affected. These are the event counts entered in the registry record; they are not a statistical comparison between randomized groups.
A count of 0 affected means that no events of the specified type are recorded in the listed group. It should not be read as proof that the event is impossible or that the contrast agent is free of risk. In general, an observed count depends on the participants followed, the duration and completeness of observation, and how events are defined and recorded. The registry record does not describe those details sufficiently to assess event ascertainment or follow-up completeness.
The deaths value is not reported in the registry record. No serious adverse-event terms are listed. The event figures should be read separately from the primary imaging endpoint: the endpoint was not assessed, and the adverse-event counts do not provide information about comparative sensitivity. The record also does not support a like-for-like safety comparison between separate treatment groups.
12. What the Result Does, and Does Not, Mean
This record documents an intended validation study of MR lymphangiography using ferumoxtran-10, with a planned comparison against MRI before contrast in the same patient. Its primary question concerned whether per-patient sensitivity differed between the paired images. The record does not report an answer to that question: the primary outcome measure was not assessed due to early study termination.
That absence of an endpoint result has a specific interpretation. It is not evidence that the imaging conditions were equivalent, and it is not evidence that one had higher sensitivity. No sensitivity estimates, comparison estimate, confidence interval, or test result are reported. The registry record therefore cannot establish the diagnostic performance of either imaging condition in this study.
The study's design also defines the scope of any possible conclusions. It was non-randomized, unmasked, and modeled as a single group. Its planned comparison was within patient rather than between randomized groups. The registry cautions that the arms are not a like-for-like control comparison. Even if paired outcome data were available, the result would describe the measured imaging comparison in the assessed participants and would need to be interpreted in light of the reference standard and other design details.
The registry lists enrollment of 10, but does not report how many participants contributed complete paired imaging data or entered a primary analysis. It gives an early-termination explanation but not a detailed account of missing outcomes, exclusions, or analysis populations. These gaps prevent assessment of how representative any hypothetical analyzed subset would be.
The adverse-event record lists 0 serious adverse events and 0 other adverse events among 10 at risk in the named group. These counts describe what is recorded, not the certainty of safety or the absence of possible events beyond the recorded observations. The deaths value is not reported in the registry record.
In short, the record establishes the study's intended diagnostic question and reports limited status and safety information. It does not provide an efficacy finding or a completed statistical analysis. No conclusion about benefit, lack of benefit, or comparative accuracy follows from the unassessed primary endpoint.
13. Why This Trial Matters Statistically
This study illustrates a basic distinction in clinical statistics: a well-defined planned endpoint is not the same as an observed result. The registry specifies a diagnostic question—whether sensitivity differed between paired MRI images with and without ferumoxtran-10—and names a paired statistical method. But early termination meant the primary outcome was not assessed. Without the outcome data, the stated method cannot yield a result.
The paired design is statistically meaningful because images from the same person are related. A suitable paired analysis can use within-person differences rather than treating each image as independent. At the same time, the method requires the paired classifications and a way to determine the true disease status. Neither the classifications nor the reference standard is described in the outcome results. The design alone cannot create information that was not collected or reported.
The record also demonstrates why enrollment should not be confused with the analysis population. Enrollment is listed as 10, but the record does not say how many people completed both imaging conditions or had evaluable outcome data. When a study stops early, this distinction can be especially important: the number enrolled does not establish the number contributing to the primary endpoint.
Finally, safety counts and planning calculations answer different questions from an efficacy analysis. Recorded adverse-event counts do not reveal diagnostic sensitivity. Planning illustrations are not observed results and do not show what the terminated study found. Keeping these kinds of evidence separate avoids overstating what a registry entry can support.
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 consequential limitation is that the primary outcome was not assessed. The record gives no sensitivity values for either imaging condition and no paired comparison result. The reason is described as early study termination, with patients not receiving assigned treatment offered as an example, but further details about termination and participant flow are not reported in the registry record.
The diagnostic reference standard is not identified in the posted outcome description. Without knowing how pelvic lymph-node disease was established, readers cannot evaluate how the planned sensitivity measure was defined or how it would relate to clinical classification. The record also does not report the observed paired image classifications or the number of participants with complete data for both conditions.
The trial was non-randomized, unmasked, and modeled as a single group. The registry explicitly cautions that the arms are not a like-for-like control comparison. The planned pre-contrast and post-contrast images are a paired imaging comparison, but the record does not report whether image interpretation was blinded to imaging condition or other details of image assessment.
Further design and analysis details are not reported in the registry record, including a protocol-based sample-size justification, the analysis population, handling of missing outcomes, any interim analyses, and a multiplicity plan. The record lists no analyses. These absences prevent assessment of the assumptions behind a completed efficacy analysis, because no such analysis is reported.
Safety interpretation is also limited. The listed event counts are 0 serious adverse events and 0 other adverse events among 10 at risk, but the record does not describe follow-up duration or the completeness of event collection. The deaths value is not reported. These figures should therefore be understood as registry-recorded counts rather than comprehensive evidence about risk.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: A Validation Study of MR Lymphangiography Using SPIO, a New Lymphotropic Superparamagnetic Nanoparticle Contrast — NCT00147238.
- Registry record last updated 2012-08-07; status last verified 2012-07.
- Results first posted 2010-01-27.
- 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.
18. Record Summary
The registry describes a non-randomized, unmasked, single-group study of MR lymphangiography using ferumoxtran-10 for pelvic lymph-node cancer detection. Its planned primary endpoint was per-patient sensitivity comparing MRI without contrast against a repeated MRI with contrast, using a two-sided McNemar test. The second scan was described as taking place within 24-36 hours of contrast injection and about 24 hours after the first MRI.
The record lists enrollment of 10, study status as terminated, a start date of 2005-07, and a primary completion date of 2007-01. The primary outcome was not assessed due to early termination, and no analyses are listed. Accordingly, no sensitivity estimate or statistical comparison is reported.
For the Ferumoxtran-10 MRI Contrast Agent group, the record lists 0 serious adverse events and 0 other adverse events among 10 at risk. Deaths are not reported in the registry record. The available information establishes the intended study question and selected recorded counts, but does not establish a diagnostic-accuracy result.