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 Pilot Study of dMRI as a Tool for Early Evaluation of Tumor Response in Diffuse Large B-cell Lymphoma (NCT00645359), a registered phase not stated study. The primary outcome is Mean Difference in Apparent Diffusion Coefficient.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00645359 |
| Phase | Phase not stated |
| Condition | Diffuse Large B Cell Lymphoma |
| Allocation | Not Stated |
| Design model | Not Stated |
| Masking | Not Stated |
| Enrollment | 9 |
| Arms in the primary outcome | n/a |
| Lead sponsor | University of Michigan Rogel Cancer Center |
| Results posted | Yes |
2. Clinical Question
Population
People with diffuse large B-cell lymphoma were the population of interest. The registry records enrollment of 9, but states that no patients were analyzed because of insufficient resources following shifting research priorities.
Intervention
The listed study procedure was diffusion MRI (dMRI). The arm description says scans were planned at baseline and 7 days. The study summary places the research question in the setting of chemotherapy.
Comparator
No comparator is described. Allocation is not stated, and the registry record cautions that this was not a randomized comparison and that the arms are not a like-for-like control comparison.
Primary question
The primary question was whether the apparent diffusion coefficient (ADC) changed detectably from baseline to Day 8 during the early phase of chemotherapy. The broader aim was to assess whether diffusion MRI after one cycle of chemotherapy could predict individual treatment outcomes.
These are related but distinct questions. Detecting a change in ADC is a measurement question: it asks whether the imaging-derived quantity differs between the stated time points. Predicting an individual outcome is a prediction question: it asks whether an imaging result can help anticipate what happens later for a person. The registry identifies both aims, but reports no analyzed patients and no results with which to evaluate either one.
3. Trial Design
This registry record describes a pilot study of diffusion MRI as a tool for early evaluation of tumor response in diffuse large B-cell lymphoma. The sponsor is the University of Michigan Rogel Cancer Center. The study began in 2006-11, had a primary completion date of 2011-07, and is listed as TERMINATED. Enrollment is recorded as 9.
The listed intervention was diffusion MRI, classified as a procedure. The arm description states that patients would undergo dMRI at baseline and 7 days. For the primary outcome, the timeframe is given as Baseline and Day 8. Those are the time descriptions used in different parts of the registry record; no further explanation of the schedule is reported in the registry record.
The record does not describe a randomized comparison. Allocation is not stated, and it explicitly cautions that the arms are not a like-for-like control comparison. The listed study information therefore does not support interpreting the design as a comparison of assigned treatments. The study summary describes imaging in the context of chemotherapy, but chemotherapy is not listed as an intervention in the registry entries presented for this study.
The record states that no patients were analyzed because of insufficient resources, secondary to shifting research priorities. It does not describe recruitment progress, the reason for termination beyond that statement, or whether any imaging scans were completed. Those details are not reported in the registry record. Consequently, the design describes an intended early-imaging assessment, while the posted results do not provide analyzed outcome data.
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 | Diffusion MRI |
|---|---|
| STARTED | 9 |
| COMPLETED | 9 |
| NOT COMPLETED | 0 |
Analysis population for the primary outcome, as defined in the registry: No patients were analyzed due to insufficient resources, secondary to shifting research priorities.
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 | Diffusion MRI |
|---|---|---|
| Participants analysed | 9 | |
| Age, Continuous | years | 61 (35 to 77) |
| Sex: Female, Male: Female | Participants | 4 |
| Sex: Female, Male: Male | Participants | 5 |
6. Primary Endpoint
The registry lists the primary outcome as Mean Difference in Apparent Diffusion Coefficient, measured over Baseline and Day 8.
| 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 is the mean difference in apparent diffusion coefficient (ADC) between baseline and Day 8. ADC is a quantity derived from diffusion MRI. The registry description says it would be calculated at the voxel level at both time points, with the mean difference then calculated. A voxel is an image element; voxel-level measurement allows the ADC to be assessed across the imaged tissue rather than treating the scan as a single undifferentiated observation.
Comparing ADC over time is intended to assess whether an imaging-derived change can be detected during the early phase of chemotherapy. In general, an early measurement may be studied as a candidate indicator of later response, but detecting a change and predicting a later clinical outcome are not the same statistical task. The registry summary describes the broader aim as evaluating whether dMRI after one cycle of chemotherapy can predict individual treatment outcomes. The listed primary endpoint itself is the mean difference in ADC, not a reported prediction accuracy measure.
The registry supplies no endpoint value, variability estimate, confidence interval, or statistical test result. It states that no patients were analyzed due to insufficient resources following shifting research priorities. Thus, the endpoint is defined in the record, but its observed value and precision are not reported.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The central result for interpretation is the absence of an analyzed primary endpoint. The registry explicitly says that no patients were analyzed because of insufficient resources, secondary to shifting research priorities. Although enrollment is recorded as 9, that enrollment figure does not supply an estimate of change in ADC. No mean difference, measure of variability, confidence interval, or inferential result is reported.
That absence matters because the scientific question depends on observed imaging measurements at the stated time points. Without analyzed measurements, it is not possible to determine from this record whether ADC changed, how large any change was, or how precisely it might have been estimated. It is also not possible to assess whether differences across people were small or substantial. These are unavailable findings, not evidence that the true change was zero.
The study summary describes a further goal: assessing whether dMRI after one cycle of chemotherapy could predict the outcome of treatment for individual patients. A prediction claim requires information linking imaging findings to later outcomes. The registry does not report analyzed patients, prediction estimates, or prediction performance. The record therefore does not establish whether ADC can predict treatment outcomes, either positively or negatively.
The registry lists 2 outcome measures and 0 analyses. Those counts are consistent with the reported lack of analyzed results, but do not explain the intended statistical analysis in detail. The statistical analysis plan, including any model, test, handling of repeated measurements, and approach to uncertainty, is not reported in the registry record.
There is also no randomized control comparison to interpret. The registry says allocation is not stated and warns that the arms are not a like-for-like control comparison. Any observed change, if one had been reported, would therefore require careful interpretation as an imaging change over time rather than as a controlled treatment effect. In the posted record, however, even that within-person change is not estimated.
The practical statistical conclusion is limited: the registry documents an intended pilot investigation, but does not provide analyzed efficacy or prediction evidence. The lack of estimates means that neither the magnitude nor the precision of the intended primary result can be judged. It should not be translated into a claim that dMRI did or did not identify response.
9. Statistical Methods Explained
Interpreting this study begins with distinguishing its intended endpoint from an available result. The intended primary endpoint is a mean difference in ADC between baseline and Day 8, based on ADC values calculated at the voxel level. The registry does not report a calculated mean difference. It states that no patients were analyzed because of insufficient resources following shifting research priorities. In statistical terms, the endpoint is defined, but there is no posted estimate to interpret.
What would a mean difference represent?
In general, a mean difference summarizes the average change between measurements on a defined scale. It gives a direction and magnitude of change according to how the difference is defined. Interpretation depends on the measurement units, the population included, the time points, and how measurements are combined. The registry identifies ADC as the measurement and Baseline and Day 8 as the primary timeframe, but the unit and the precise sign convention are not stated. Without an observed mean difference, there is no direction or magnitude to describe.
Because ADC is calculated at the voxel level, the way voxel values are summarized can affect the result. In general, an analysis needs a clear rule for combining measurements within an image and across participants. A mean may describe a central tendency, but by itself it does not show how much observations vary. The registry does not report the aggregation details beyond saying that ADC would be calculated at the voxel level and that the mean difference would be calculated. It does not report whether the primary summary would be formed from participant-level changes, pooled voxel measurements, or another procedure.
What would uncertainty add?
In general, a confidence interval expresses the statistical uncertainty around an estimated quantity under the assumptions of the analysis. A narrow interval indicates greater precision than a wide interval, while the interval’s location indicates which values are compatible with the data and model. A confidence interval is not reported for this endpoint. Since there is no posted estimate, readers cannot judge either the size of a possible ADC change or its precision.
A test result, if reported, would answer a different question from the estimate itself. A statistical test evaluates compatibility with a specified null model under stated assumptions; it does not by itself show the size, importance, or predictive value of a change. The test, null hypothesis, significance threshold for the primary analysis, and analysis model are not reported in the registry record. The sample-size assumptions elsewhere in the record describe a planning illustration, not an observed result or observed power; they do not replace primary endpoint data.
Why is prediction a separate issue?
In general, prediction requires evaluating how well measurements made at one point anticipate outcomes observed later. An association between an imaging measurement and an outcome is not automatically evidence of reliable individual prediction. Prediction assessment may consider calibration, discrimination, and performance in data not used to build a model. These are general concepts; the registry does not report a prediction model, later outcome data, or prediction-performance estimates. Its stated aim of predicting individual treatment outcomes therefore cannot be assessed from the posted results.
What does the design allow readers to compare?
The registry says this was not a randomized comparison, allocation is not stated, and the arms are not a like-for-like control comparison. In general, a before-and-after measurement can describe change over time within the measured group, but without a suitable concurrent control it cannot by itself separate changes associated with treatment from other time-related influences. This is a general design limitation, not a claim that any particular influence occurred in this study. Here, the registry reports no analyzed change in the first place.
Overall, the statistical methods needed to interpret an endpoint estimate are mostly unavailable: no estimate, uncertainty interval, test result, or detailed analysis specification is reported. General statistical principles can explain what such quantities would mean, but cannot recover a study result that the registry does not report.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 2 outcome measures and 0 analyses. It identifies the mean difference in ADC as the primary outcome, with a Baseline and Day 8 timeframe. The other outcome measure and its analysis are not described in the registry record recorded for this study. No multiplicity strategy—such as a prespecified hierarchy, adjustment, or designation of confirmatory versus exploratory tests—is reported in the registry record. Without results or a detailed statistical plan, it is not possible to assess how multiple outcomes or analyses were intended to affect interpretation.
The record states directly that no patients were analyzed because of insufficient resources, secondary to shifting research priorities. This is different from a report that some enrolled participants had missing measurements within an analysis. The registry does not say whether scans were obtained, what measurements may have been unavailable, or whether any participants had incomplete follow-up. Those details are not reported in the registry record.
It also does not describe an analysis population, such as criteria for including participants in an endpoint analysis. The approach to missing measurements, exclusions, protocol deviations, or sensitivity analyses is not reported in the registry record. These omissions prevent readers from evaluating how the analysis population would have been defined or how incomplete data would have been handled.
Because no patients were analyzed, there is no reported analysis population from which to estimate the primary endpoint. The record’s enrollment figure should not be mistaken for an analyzed sample size. Without an analysis plan and posted results, the impact of missing data and the handling of multiple outcomes cannot be assessed.
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 |
|---|---|---|---|
| Diffusion MRI | 0 | 9 | 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 |
|---|---|
| Diffusion MRI | 0 of 9 |
The registry lists 9 participants at risk in the Diffusion MRI group for serious adverse events and 0 affected. For other adverse events, it likewise lists 9 at risk and 0 affected. These are the posted counts; they do not provide a detailed account of how events were collected, monitored, or classified.
Deaths are recorded as null, not as a reported zero. A null entry should not be read as confirmation that no deaths occurred. The list of top serious-event terms is empty. That does not provide event-level detail beyond the reported count of zero serious events in the listed group.
In general, an absence of recorded events in a registry summary is not proof that an intervention has no risks. The interpretation depends on the number of people assessed, duration and completeness of follow-up, event definitions, and reporting procedures. Those details are not reported in the registry record. The study also states that no patients were analyzed for the primary outcome; the safety figures are separately presented registry counts and should not be treated as evidence that efficacy data were analyzed.
The serious and other adverse-event counts describe what the record reports, not a comprehensive safety conclusion. In particular, the null deaths entry and absence of event terms should be distinguished from explicit evidence that no events occurred.
12. What the Result Does, and Does Not, Mean
The registry record documents an intended pilot investigation of early diffusion MRI in diffuse large B-cell lymphoma. It specifies ADC change between baseline and Day 8 as the primary endpoint and describes a broader interest in whether dMRI after one cycle of chemotherapy could predict individual treatment outcomes. It also records enrollment of 9 and a TERMINATED status.
Most importantly, the record says no patients were analyzed because of insufficient resources, secondary to shifting research priorities. It therefore contains no primary endpoint estimate and no reported prediction result. Readers cannot determine whether ADC changed, how much it changed, how precisely any change could be estimated, or whether imaging findings predicted later outcomes. The absence of an estimate is not a finding of no change, and it is not evidence that prediction failed.
The record also does not describe a randomized control comparison. Allocation is not stated, and the registry cautions that the arms are not a like-for-like control comparison. Even if an unadjusted change had been reported, that design information would be important when deciding what the change could establish. No such change is reported.
The posted adverse-event summary records zero serious and zero other affected participants among 9 listed at risk in the Diffusion MRI group, while deaths are null. These entries describe the registry’s safety reporting and do not establish that the procedure has no risks; the record does not give the follow-up and ascertainment details needed for a broader safety interpretation.
The study is therefore best understood as an incompletely reported pilot investigation with a defined imaging endpoint but no analyzed primary results. It can document the research question and the reason the record gives for the absence of analysis. It cannot answer whether dMRI detects early response or predicts treatment outcomes.
13. Why This Trial Matters Statistically
This study illustrates an important distinction in clinical research: a clearly defined measurement question does not guarantee that an estimate will be available. The registry specifies an ADC mean difference between baseline and Day 8, and explains that ADC was intended to be calculated at the voxel level. Yet it reports no analyzed patients and no endpoint result. A protocol-level objective and an empirical finding are different kinds of information.
The record also separates an early biological or imaging measurement from a prediction claim. A change in ADC would address whether the imaging quantity changed over the stated interval. Predicting an individual treatment outcome would require a relationship between that early measurement and later outcomes, along with evidence about predictive performance. The registry records both interests, but provides no analyses with which to evaluate either.
From a statistical reporting perspective, the case emphasizes why estimates and uncertainty are essential. A mean difference alone would describe magnitude and direction; a measure of uncertainty would help show how precisely it was estimated. Neither is reported. No result should be inferred from the enrollment figure or from the fact that a primary endpoint was named.
The record also highlights the importance of distinguishing safety counts from outcome analysis. The adverse-event summary reports zero affected in the listed categories, while the primary outcome has no analyzed patients and the death entry is null. These separate entries have different meanings and should not be conflated.
Finally, the stated reason for the lack of analysis—insufficient resources following shifting research priorities—shows how a study can end without answering its planned statistical question. For readers, the correct conclusion is not a judgment about the imaging approach, but recognition that the registry preserves an intended investigation without the analyzed evidence needed to evaluate its aims.
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 principal limitation is explicit: the registry states that no patients were analyzed because of insufficient resources, secondary to shifting research priorities. Consequently, the primary endpoint has no reported estimate, and the broader prediction aim has no reported performance assessment. The record does not provide enough information to characterize the precision or direction of any imaging change.
Enrollment is recorded as 9, but the record does not explain how many scans were completed or whether measurements were available at the intended time points. The number enrolled cannot be treated as the number with usable imaging data or as the analysis population. Those details are not reported in the registry record.
The design description also limits causal interpretation. Allocation is not stated, the record says the study was not a randomized comparison, and it cautions that arms are not a like-for-like control comparison. The listed intervention is diffusion MRI, while the study summary places the imaging question in the context of chemotherapy. The registry does not provide further details that would support a controlled treatment-effect interpretation.
The primary endpoint description identifies voxel-level ADC values and a mean difference, but does not report the unit, the sign convention, the exact aggregation procedure, or an analysis model. A statistical analysis plan, handling of missing data, and approach to multiple outcomes are not reported in the registry record. The timing is described as 7 days in the arm description and Day 8 in the primary outcome timeframe, without further reconciliation.
Safety interpretation is also limited by sparse reporting. The listed counts show zero affected for serious and other adverse events, but deaths are null, and follow-up and event-collection details are not reported. These limitations leave the posted safety summary insufficient for a broad assessment.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Pilot Study of dMRI as a Tool for Early Evaluation of Tumor Response in Diffuse Large B-cell Lymphoma — NCT00645359.
- Registry record last updated 2017-04-14; status last verified 2017-03.
- Results first posted 2014-10-08.
- 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 pilot study of diffusion MRI for early evaluation of tumor response in diffuse large B-cell lymphoma, sponsored by the University of Michigan Rogel Cancer Center. Enrollment is recorded as 9; the study began in 2006-11, had primary completion in 2011-07, and is listed as TERMINATED.
The primary endpoint was the mean difference in voxel-level ADC between baseline and Day 8. The registry says no patients were analyzed because of insufficient resources following shifting research priorities. It reports 2 outcome measures and 0 analyses, with no primary endpoint estimate or prediction result.
The safety summary lists zero serious and zero other adverse events among 9 at risk in the Diffusion MRI group; deaths are null. Allocation is not stated, and the registry cautions that the study was not a randomized comparison. The record documents the intended question but does not establish whether dMRI detected change or predicted treatment outcomes.