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 Development and Validation of a Tumour Oxygenation Monitoring Probe (NCT03982355), a registered phase not stated study. The primary outcome is Count of MRIs Successfully Modelled for Oxygen Correlation.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03982355 |
| Phase | Phase not stated |
| Condition | Cancer |
| Allocation | Not Stated |
| Design model | Not Stated |
| Masking | Not Stated |
| Enrollment | 20 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Imperial College London |
| Results posted | Yes |
2. Clinical Question
Population
The study describes anonymised, pre-therapy MRI scans from people with locally advanced breast cancer and locally advanced rectal cancer. The registry record lists an enrollment of 20 and separate Breast Cancer and Rectal Cancer groups, each labelled with 10. It does not report further eligibility details or explain whether enrollment refers to people, scans, or another unit.
Intervention
The device under development is the DRS probe. The study summary describes laboratory-based development and validation of a tissue oxygenation probe, with MRI scans analysed and segmented to guide probe development. The registry record does not describe a treatment administered to participants.
Comparator
No randomized comparison is identified. Allocation is not stated, and the registry record cautions that the arms are not a like-for-like control comparison. Breast and rectal cancer are described as separate scan groups, not as intervention and control conditions.
Primary question
The registered primary outcome is the count of MRIs successfully modelled for oxygen correlation, expressed in the record as MRIs 3D modelled for oxygenation-sensor development. The underlying question is whether scans can be modelled in a way that guides development of the DRS probe. The observed count and any measure of modelling success are not reported in the registry record.
3. Trial Design
This completed study is titled “Development and Validation of a Tumour Oxygenation Monitoring Probe” and is sponsored by Imperial College London. The registry record describes it as a laboratory-based study to develop and validate a novel tissue oxygenation probe. Its stated enrollment is 20. The recorded start date is 2019-08-09, and primary completion is recorded as 2020-06-30.
The intervention is a device named the DRS probe. The two named groups are Breast Cancer and Rectal Cancer. Their descriptions refer to anonymised MRI scans taken before therapy from people with locally advanced breast cancer and locally advanced rectal cancer, respectively. Each group label includes 10. Scans were analysed and segmented to guide development of the probe. The registry record does not further specify the scan-processing workflow, how success was judged, or whether the scans were linked to individual participants in the outcome data.
This is not presented as a randomized trial or a comparison of the probe with a control intervention. The record states that allocation is not stated and that the arms are not a like-for-like control comparison. The groups therefore should not be interpreted as treatment arms or as evidence of a difference attributable to the device. The registry record does not report a protocol for assigning scans, a comparator procedure, or a clinical outcome assessment.
The record lists one outcome measure and no analyses. Although its status is completed, it does not report the primary outcome value, a statistical analysis, or a conclusion about the modelling results. A completed study status does not by itself establish that the registered outcome was achieved or that the probe was validated.
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: 20 anonymised MRI scans, including 10 each from the breast and rectal cancer groups. The inclusion and exclusion criteria are provided to a member of the radiology team who selected the appropriate cases, anonymised the MRIs and provided all 20 DICOM files. All cases were therefore appropriate for inclusion from the outset, and therefore there was other significant events.
Overall Study
| Stage | Breast Cancer (10) | Rectal Cancer (10) |
|---|---|---|
| STARTED | 10 | 10 |
| COMPLETED | 10 | 10 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: Scans analysed and segmented to guide the development of the DRS probe
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: 10 each of MRI scans from patients with locally advanced breast and rectal cancer. The MRIs were anonymised, and as such no population information is available
| Characteristic | Unit | Breast Cancer | Rectal Cancer | Total |
|---|---|---|---|---|
| Participants analysed | 10 | 10 | 20 | |
| Age, Continuous | years | NA (SD NA) | NA (SD NA) | NA (SD NA) |
| Sex/Gender, Customized: NA | Participants | NA | NA | NA |
| Race and Ethnicity Not Collected | Participants | — | — | 0 |
| Region of Enrollment: United Kingdom | participants | 10 | 10 | 20 |
6. Primary Endpoint
The registry lists the primary outcome as Count of MRIs Successfully Modelled for Oxygen Correlation, measured over MRIs were 3D modelled - single time point at time of MRI.
| Element | Registry entry |
|---|---|
| Result type | Count Of Units |
| Measure of spread | not applicable |
| Unit | MRIs analyses |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is titled “Count of MRIs Successfully Modelled for Oxygen Correlation.” Its unit is MRIs analysed, and its description says it counts MRIs 3D modelled for oxygenation-sensor development. The measurement timeframe is a single time point at the time of MRI. The record notes that the outcome was originally expressed as “Oxygenation Correlation,” but was modelled and is therefore expressed as a count.
This is a count of modelling outputs, not a direct measure of a clinical outcome or a reported numerical estimate of oxygenation accuracy. It is relevant to development because the scans were analysed and segmented to guide the DRS probe. A count can describe how many imaging inputs were modelled, but by itself it does not show how closely modelled oxygenation corresponds to an independent reference, whether performance differs across scans, or whether the probe changes clinical care. Those aspects are not reported in the registry record.
The registry record does not give the observed count, a denominator for success, a definition of successful modelling, or an uncertainty estimate. Consequently, the endpoint’s recorded title and purpose can be described, but its observed magnitude and precision cannot be evaluated.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies a feasibility-oriented modelling outcome: the number of MRI scans 3D modelled for oxygenation-sensor development. It does not report the value of that outcome. As a result, there is no posted estimate to interpret, no reported uncertainty around an estimate, and no basis in the record for judging whether modelling succeeded for all, some, or none of the scans. Those results are not reported in the registry record.
The outcome’s original wording referred to “Oxygenation Correlation,” while the registered description clarifies that the measure was modelled and expressed as a count. This distinction matters. A count of scans modelled is a measure of modelling activity or feasibility; it is not itself a correlation coefficient, a measure of agreement, or a quantified accuracy estimate. The record does not give a correlation value or other validation statistic. It therefore cannot establish how well modelled oxygenation corresponds to an external reference.
The breast and rectal cancer groups are scan groups with different clinical contexts, not a randomized comparison. The record explicitly says allocation is not stated and the arms are not a like-for-like control comparison. Any difference between groups, were one reported, could not automatically be attributed to the DRS probe or interpreted as a treatment effect. No group-specific primary outcome results are reported in the registry record.
The record lists enrollment as 20 and labels each group with 10, but provides no endpoint results to connect those figures to modelling success. It is not reported whether all enrolled units contributed an MRI, whether any scans were excluded, or whether the endpoint was assessed for every scan. The listed enrollment should therefore not be treated as the outcome denominator.
The study is marked completed, but completion status is not a substitute for results. The record lists one outcome measure and no analyses. It does not report a statistical plan, hypothesis test, confidence interval, or analysis population. Accordingly, neither statistical significance nor precision can be assessed. Nor can the record support a claim about clinical utility: it describes development work using pre-therapy scans, rather than a reported comparison of clinical outcomes.
The registry’s adverse-event fields list zero serious events, zero other events, and zero deaths for each named group. These entries describe the record’s event fields; they do not supply evidence about probe performance or establish safety in a clinical-use setting. Taken together, the posted information identifies the study’s purpose and outcome definition, while leaving its actual modelling results and validation evidence unavailable.
9. Statistical Methods Explained
The registry record does not report a statistical analysis for the primary endpoint. It lists one outcome measure and zero analyses. Therefore, no method should be attributed to the investigators beyond the outcome description: a count of MRIs 3D modelled for oxygenation-sensor development. The distinctions below explain general statistical considerations for this type of endpoint; they are not a description of analyses reported for this study.
What does a count endpoint represent?
In general, a count records how many units meet a defined condition. For this study, the registered unit is MRIs analysed, and the stated endpoint concerns scans successfully modelled. A useful interpretation depends on a clear rule for what qualifies as successful modelling and a clear account of which scans were eligible for assessment. The registry record does not report those operational details or the observed count.
Counts differ from continuous measurements such as a mean oxygenation value. A count cannot, on its own, describe the size of a correlation or the closeness of predicted and reference measurements. If a study aims to validate measurement performance, statistical summaries of agreement or prediction error may be relevant in general, but no such summaries are reported in this record. The count may still describe whether the modelling process was carried out, but it cannot establish accuracy without additional information.
How are count data often analysed?
In general, a Poisson model is one possible approach when counts arise from a process with an appropriate relationship between the mean and variance. If the observed variability is greater than that model allows, an approach that accommodates over-dispersion may be considered. Over-dispersion means that the counts vary more than the simple model expects; ignoring it can make uncertainty appear too small. These are general options, not methods reported for this study. The registry record gives no model, distributional assumption, variance estimate, or analysis plan.
Choosing a model also depends on what constitutes the unit of analysis and whether units have comparable opportunities to contribute. A count of successful scans could require an eligible-scan denominator to distinguish a high count arising from many eligible scans from a high success rate. If scans are grouped within people or otherwise related, their results may not be independent. The registry record does not describe the denominator, scan-to-participant linkage, or dependence structure, so these features cannot be evaluated.
What would uncertainty describe?
In general, an uncertainty interval describes the range of values reasonably compatible with an estimate under the assumptions of the analysis. A confidence interval is one familiar form: it communicates the precision of an estimate, rather than the probability that a particular fixed value is true. For a count endpoint, an interval around the count or around a rate could be informative, depending on the estimand and denominator. No estimate or interval is reported in the registry record, so precision cannot be assessed.
A p-value, if one were calculated for a specified hypothesis, would address compatibility between data and that hypothesis under the test’s assumptions. It would not measure the size or practical importance of a result. This record does not report a hypothesis test or p-value, and the groups are not described as a randomized, like-for-like comparison. Thus a between-group significance test should not be presumed to have been performed or to answer a causal question.
What do the planning assumptions mean?
The registry record describes 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. It says the effect to detect is a stated fraction of the observed effect. Such a calculation is a planning exercise, not an analysis of observed outcomes and not observed power. The record also states that the trial’s own design is not represented by that illustration. Because the observed primary outcome and effect are not reported, the empirical basis for such planning cannot be evaluated from this record.
For a count outcome, the suitability of a normal approximation depends on the data and the model. A planning calculation does not remove the need to define the endpoint, identify the analysis unit, assess distributional assumptions, and report uncertainty. No study-specific sample-size justification or statistical plan is reported in the registry record. The planning assumptions should therefore not be mistaken for a protocol-based design justification or evidence that the study had a particular probability of detecting an effect.
10. Multiplicity, Missing Data and Analysis Populations
The registry record lists one outcome measure, the count of MRIs 3D modelled for oxygenation-sensor development, and no analyses. It does not report whether other outcomes were measured outside the registry listing, whether any analyses were planned but not posted, or how the primary outcome was selected. The statistical plan is not reported in the registry record.
Multiplicity arises in general when several outcomes, time points, comparisons, or analyses are tested. Repeated testing can increase the chance of apparently notable findings even when no single effect is present. Methods for handling that issue depend on the prespecified questions and analysis plan. Here, no inferential analysis is reported, so there is no basis for assessing whether multiplicity adjustments were needed or used. A single listed outcome does not establish that only one analysis was considered.
Missing-data handling is also not reported. The record does not say whether any scans were unavailable, excluded, unmodelled, or omitted from the endpoint, nor does it define an analysis population. The enrollment figure and group labels do not establish how many scans contributed to the outcome. Without the number of eligible scans and an account of missing or excluded observations, the completeness of the endpoint cannot be assessed.
The record does not report interim analyses, a sample-size justification, or a protocol-defined analysis population. These elements should not be inferred from completion status or from the planning assumptions. The registry record’s adverse-event fields list zero serious events, zero other events, and zero deaths for each group, but those entries do not resolve missingness or analysis-population questions for the MRI modelling endpoint.
11. What the Result Does, and Does Not, Mean
The record describes a completed laboratory-based effort to develop and validate a tumour oxygenation monitoring probe, using anonymised pre-therapy MRI scans from breast and rectal cancer groups. It identifies the DRS probe as the device of interest and defines a primary outcome based on how many MRIs were 3D modelled for oxygenation-sensor development. These details establish the study’s stated purpose and the kind of result it intended to record.
The record does not state the observed endpoint count, report a statistical analysis, or provide a validation estimate. It therefore does not show how many scans were successfully modelled, how precise any estimate was, or how well the modelled information corresponded to oxygenation. The study’s completed status should not be read as a result or as confirmation that validation criteria were met.
The two groups are not a randomized intervention-versus-control comparison. They represent different cancer contexts, and the record says allocation is not stated and that the arms are not like-for-like controls. The information therefore cannot support a causal interpretation of differences between the groups. It also does not establish that use of the probe changes diagnosis, treatment decisions, or patient outcomes.
More generally, a modelling count can speak to whether scans were processed under a stated definition, but a count alone cannot establish measurement accuracy or clinical value. The registry record does not provide the definition of success, an eligible-scan denominator, or an independent reference comparison. The posted information is consequently useful for understanding the intended development outcome, but insufficient to quantify the probe’s performance or to draw conclusions about clinical use.
12. Why This Trial Matters Statistically
This study raises a basic statistical distinction between developing a measurement tool and demonstrating that it measures its target well. Counting how many MRI scans were modelled can describe a stage of the development process. It is not equivalent to quantifying agreement between a probe and a reference measurement, or showing that a measurement is useful in clinical decisions. Clear outcome definitions and results are important for keeping those claims distinct.
The record also illustrates why the analysis unit matters. The listed enrollment, group labels, and endpoint refer to different concepts: enrollment is recorded as 20, groups are labelled by cancer type, and the primary endpoint counts MRIs modelled. The registry does not explain the relationship between enrolled units and scans contributing to the endpoint. Without that information, the denominator and completeness of the outcome remain unclear.
Finally, development studies should not be interpreted as controlled clinical comparisons when their groups are not designed as like-for-like controls. The record explicitly states that allocation is not given and that the arms are not a control comparison. That caveat limits what can be concluded from any apparent contrast between the breast and rectal cancer groups. Since no outcome results or analyses are posted, the record chiefly demonstrates the importance of transparent reporting: readers need the observed result, the definition of success, analysis methods, and uncertainty to assess a validation claim.
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 limitation for interpretation is that the registry record does not report the primary endpoint result. It gives no count of successfully modelled MRIs, no uncertainty estimate, and no group-specific outcome values. There is therefore no quantitative result on which to base an assessment of the modelling work.
The registered outcome description does not define the criteria for a scan to count as successfully modelled. The record also does not provide the number of eligible scans, the number excluded or unavailable, or the relationship between MRI scans and enrolled units. These omissions prevent assessment of endpoint completeness and make the denominator unclear.
The two groups are not like-for-like control arms, and allocation is not stated. The group descriptions identify different cancer contexts, so they cannot be treated as randomized comparison conditions. The record does not report whether the modelling process or its evaluation differed between groups.
Further information needed to evaluate validation—including a reference measure, a numerical measure of correlation or agreement, and a statistical analysis—is not reported in the registry record. The record lists no analyses and does not give a statistical plan or sample-size justification. Completion status and the listed adverse-event values do not resolve these limitations or establish probe performance.
14. Related Tutorials
15. Related Statistical Calculators
16. Sources
- ClinicalTrials.gov: Development and Validation of a Tumour Oxygenation Monitoring Probe — NCT03982355.
- Registry record last updated 2024-09-19; status last verified 2024-04.
- Results first posted 2024-09-19.
- 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
NCT03982355 is a completed laboratory-based study sponsored by Imperial College London, titled “Development and Validation of a Tumour Oxygenation Monitoring Probe.” The recorded start date is 2019-08-09, and primary completion is recorded as 2020-06-30. Enrollment is listed as 20. The device is the DRS probe.
The groups are described as anonymised pre-therapy MRI scans from people with locally advanced breast cancer and locally advanced rectal cancer; each group label includes 10. The single listed primary outcome is the count of MRIs 3D modelled for oxygenation-sensor development. The record notes that the outcome was originally called “Oxygenation Correlation” and was expressed as a count after modelling.
No primary endpoint result or statistical analysis is reported in the registry record. Allocation is not stated, and the groups are not described as a like-for-like control comparison. The record’s adverse-event fields list zero serious events, zero other events, and zero deaths for each group.