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 Colorectal Cancer Treated With Adjuvant Regorafenib Versus Placebo After Curative Treatment of Liver Metastases in a Randomized, Double-blind, Placebo-Controlled Phase-III STudy (NCT01939223), a randomized phase 3 study. The primary outcome is Disease Free Survival (DFS) as Assessed by the Investigator.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01939223 |
| Phase | Phase 3 |
| Condition | Colorectal Neoplasms |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Enrollment | 25 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Bayer |
| Results posted | Yes |
2. Clinical Question
Population
People with stage IV colorectal cancer after curative treatment of liver metastases and completion of planned chemotherapy. The registry lists colorectal neoplasms as the condition.
Intervention
Regorafenib, described as four tablets taken orally in the morning with a low-fat meal, followed by three weeks of treatment and one week off.
Comparator
Placebo, described using the same tablet schedule and meal instructions as the regorafenib arm.
Primary question
Whether disease-free survival, assessed by the investigator, differed between regorafenib and placebo after curative treatment of liver metastases.
The study was designed to evaluate efficacy and safety in this setting. Its primary endpoint was time from randomization to the first observed radiographic recurrence or death before recurrence was documented. The registry record lists overall survival as a secondary outcome, but it reports no outcome analyses. The primary endpoint has no analyzed participants because the study was terminated prematurely and endpoint data were not collected. Consequently, the registry record cannot establish an estimated difference in disease-free survival between the assigned treatments.
The clinical question is therefore a description of the intended comparison, not an answer about comparative efficacy. The record lists enrollment as 25, but that figure does not substitute for analyzed outcome data. Nor does the absence of an analysis indicate that the treatments had the same outcome. It means the primary comparison is not available in the registry record.
3. Trial Design
The registry describes a randomized, parallel-group, placebo-controlled phase 3 study with quadruple masking. The planned comparison was regorafenib versus placebo for people with colorectal cancer following curative treatment of liver metastases and completion of planned chemotherapy. The sponsor was Bayer. The registry lists enrollment as 25.
The regorafenib arm description specifies four tablets taken orally in the morning with a low-fat meal, on a schedule of three weeks of treatment followed by one week off. The placebo description specifies the same schedule and instructions. The registry record does not report the allocation ratio, the method used to generate the randomization sequence, or how allocation was concealed. The meaning of quadruple masking for the specific roles masked is not detailed in the record.
The study began on 2013-12-02, and the listed primary completion date is 2016-08-29. Its status is terminated. The record says the study was prematurely terminated and that data were not collected for the primary endpoint. It does not report the reason for termination, so no explanation for the early stop can be inferred.
Although the record identifies a phase 3 randomized design, the design label alone does not guarantee that a planned comparison can be estimated. Estimation requires relevant outcome data and an analysis. Neither is available for the primary endpoint in this record. The design describes how the trial was organized; the reported endpoint information describes what could be analyzed. Here, the record’s statement that zero participants were analyzed for disease-free survival is central to interpreting the study.
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: Overall, 65 subjects were screened, of which 40 were screen failures. The remaining 25 subjects were randomized and assigned to treatment. All 25 subjects received treatment.
Overall Study
| Stage | Regorafenib 160 mg | Placebo |
|---|---|---|
| STARTED | 14 | 11 |
| Participants Received Treatment | 14 | 11 |
| COMPLETED | 1 | 0 |
| NOT COMPLETED | 13 | 11 |
Reasons for not completing this period:
| Reason | Regorafenib 160 mg | Placebo |
|---|---|---|
| AE not assoc. with disease recurrence | 1 | 1 |
| Disease recurrence (radiological) | 1 | 3 |
| Study stopped by Sponsor | 5 | 7 |
| Withdrawal by Subject | 6 | 0 |
Analysis population for the primary outcome, as defined in the registry: Number of Participants Analyzed is 0 because this study was prematurely terminated and data were not collected for this endpoint.
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 | Regorafenib 160 mg | Placebo | Total |
|---|---|---|---|---|
| Participants analysed | 14 | 11 | 25 | |
| Age, Continuous | Years | 59.7 (SD 11.22) | 57.0 (SD 9.83) | 58.5 (SD 10.50) |
| Sex: Female, Male: Female | Participants | 9 | 5 | 14 |
| Sex: Female, Male: Male | Participants | 5 | 6 | 11 |
6. Primary Endpoint
The registry lists the primary outcome as Disease Free Survival (DFS) as Assessed by the Investigator, measured over From date of randomization to date of first observed radiographic disease recurrence (RECIST 1.1 criteria for measurable and non-measurable disease) or death due to any cause, if death occurred before disease recurrence was documented..
| 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 investigator-assessed disease-free survival. It was defined as the time from randomization until the first observed radiographic disease recurrence, using RECIST 1.1 criteria for measurable and non-measurable disease, or death from any cause if death occurred before recurrence was documented. Recurrence was assessed with CT or MRI scans. For a participant without documented recurrence or death at the time of analysis, the definition specifies censoring at the date of the last evaluable CT or MRI scan.
In general, a time-to-event endpoint captures both whether an event occurs and how long it takes to occur. It can be useful when the timing of recurrence or death matters and when participants may have different lengths of follow-up. Censoring distinguishes a participant whose event time is not observed during the available follow-up from one known to have experienced the endpoint event. It does not establish that the participant remained event-free indefinitely.
In this trial, the endpoint definition is described, but the registry says no data were collected for it because the study was terminated prematurely. The number of participants analyzed is reported as 0. Thus, the record does not provide observed disease-free-survival estimates or a treatment comparison. The endpoint’s clinical interpretation can be described from its definition, but its comparative result cannot be assessed from the posted results.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The primary endpoint is not accompanied by an efficacy estimate. The registry record explicitly states that the number of participants analyzed for investigator-assessed disease-free survival is 0 because the study was prematurely terminated and data were not collected for this endpoint. This is a lack of analyzable outcome information, not an estimate showing no difference between regorafenib and placebo.
The record lists overall survival as a secondary outcome, but it does not report an analysis of that outcome. It reports 0 analyses. No treatment-effect estimate, confidence interval, or statistical test for either outcome is reported in the registry record. Therefore, the precision of any comparative effect cannot be evaluated. In general, a confidence interval describes the range of effect estimates reasonably compatible with the observed data under the statistical model; without a reported estimate and interval, readers cannot judge how much uncertainty surrounds a treatment comparison.
The registry lists enrollment as 25, but enrollment is not interchangeable with the number contributing to an endpoint analysis. In this record, the primary endpoint analysis population is explicitly 0. The adverse-event reporting includes group-specific numbers at risk, but those safety figures do not supply disease-free-survival results. They should not be used to infer recurrence, survival, or comparative efficacy.
The record reports serious adverse events for the regorafenib group and placebo group, as well as counts affected by other adverse events. Those are descriptive safety figures, not a substitute for a full statistical comparison. In particular, the record does not report a treatment-effect estimate or uncertainty measure for these outcomes. The serious adverse-event terms listed are erythema multiforme and generalised erythema, each with a reported percentage of 7.1%. The record does not identify the arm for those terms in the information available.
The study’s terminated status and absence of primary endpoint data sharply limit what can be concluded from its intended efficacy question. A randomized design can reduce bias in a comparison when outcomes are measured and analyzed, but randomization itself does not create an estimate when endpoint information is unavailable. The record supports a description of the intended design and the reported safety counts; it does not support a conclusion about whether disease-free survival differed between treatment groups.
In general, an inconclusive or unavailable estimate should be distinguished from evidence of equivalence. Equivalence requires a prespecified framework and data precise enough to exclude differences considered important. Neither an equivalence analysis nor the data needed for one are reported in the registry record. Similarly, the early termination cannot be interpreted as evidence of benefit, lack of benefit, or harm without results and a documented reason for stopping.
9. Statistical Methods Explained
Randomization assigns participants to treatment groups using a chance-based process. In general, this can help balance measured and unmeasured prognostic factors between groups, though balance is not guaranteed in every trial, especially when group sizes are limited. The registry identifies this study as randomized and parallel-group. It does not report the allocation ratio or the detailed sequence-generation and concealment methods, so those aspects of implementation are not described in the registry record.
What does masking mean?
The record labels the study quadruple-masked. In general, masking is intended to reduce the influence of knowledge about assigned treatment on participant behavior, clinical care, outcome assessment, or analysis. The record does not specify which roles were masked or how masking was maintained. The placebo arm is described with the same tablet schedule and meal instructions as the regorafenib arm, but no further details about treatment matching are reported in the registry record.
How is disease-free survival defined?
The endpoint is a time-to-event measure, with recurrence or death as the event. The time origin is randomization. Recurrence is based on radiographic assessment by CT or MRI, using RECIST 1.1 criteria. In general, time-to-event methods use information about when events occurred and can retain follow-up information for participants who have not yet had an event by the analysis point. The registry definition says those without documented recurrence or death at analysis would be censored at the last evaluable CT or MRI scan. However, the registry also states that data were not collected for this endpoint and that no participants were analyzed. The censoring definition therefore describes the endpoint’s stated handling, not an observed analysis in this trial.
Why does the missing analysis matter?
A treatment comparison requires outcome information for the groups being compared and a statistical method that accounts for the endpoint and study design. The registry reports no analyses. It does not report the statistical model or test used for disease-free survival, nor a hazard ratio, median, event count, confidence interval, or p-value. Those methods and results are not reported in the registry record. Without an analysis, there is no effect estimate whose direction, magnitude, or precision can be interpreted.
In general, a confidence interval communicates uncertainty around an estimate. A narrow interval indicates greater precision under the assumptions of the analysis; a wide interval indicates less precision. A p-value addresses compatibility between observed data and a specified null model, under the assumptions of a test. Neither statistic can be interpreted in isolation as a measure of clinical importance. No confidence interval or p-value for the primary outcome is reported, so there is no basis for describing statistical significance or precision.
What is the role of censoring?
For time-to-event analysis, censoring records that the event time was not observed during the available follow-up. In general, valid interpretation depends on the censoring process and the analysis assumptions. Censoring is not the same as an event, and it does not mean that the event will never occur. The endpoint description gives a censoring rule based on the last evaluable scan, but it does not report how many participants would have been censored, when follow-up ended for them, or whether any censoring patterns could be assessed. Since primary endpoint data were not collected, none of those quantities can be evaluated from this record.
What can the safety counts contribute?
The safety information gives numbers at risk and numbers affected for serious and other adverse events by group. These are descriptive counts in the registry record. In general, adverse-event comparisons require clarity about event definitions, observation periods, denominators, and whether counts refer to participants or events. The record presents group-level counts, but it does not provide a comparative statistical analysis or uncertainty estimates. The listed serious-event terms have percentages, but the arm associated with each term is not identified in the information available.
Overall survival is listed as a secondary outcome, but no analysis is reported. The record does not describe a statistical analysis plan, handling of multiplicity, interim analyses, or sample-size justification. Those details are not reported in the registry record. Without them, the planned inferential framework cannot be reconstructed, and the phase designation should not be taken as evidence that the intended hypothesis tests were completed.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists investigator-assessed disease-free survival as the primary outcome and overall survival as a secondary outcome. It reports 0 analyses, and the primary endpoint description says no data were collected because the study was prematurely terminated. No outcome-specific estimates or statistical tests are reported. The statistical plan for controlling multiplicity across outcomes is not reported in the registry record.
In general, multiplicity arises when several outcomes, time points, or statistical comparisons are assessed. Testing many hypotheses can increase the chance of a chance-positive finding unless the analysis plan accounts for the set of tests. This issue cannot be evaluated for the trial’s efficacy outcomes because no analyses are reported and the record does not describe a multiplicity strategy. The registry record also does not describe interim analyses or any planned adjustments related to them.
The endpoint definition includes censoring at the last evaluable CT or MRI scan for participants without documented recurrence or death at the time of analysis. That is a stated rule for the time-to-event outcome. It does not provide an account of missing scans, incomplete follow-up, or data availability. The record says endpoint data were not collected, and it does not report a missing-data analysis or sensitivity analysis. Such methods are not reported in the registry record.
The analysis populations are also not described beyond the primary endpoint statement that the number analyzed is 0. The record does not report whether an intention-to-treat population, per-protocol population, or another analysis set was defined or used. It is therefore not possible to compare results across analysis populations or assess how deviations from assigned treatment might have been handled.
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 |
|---|---|---|---|
| Regorafenib 160 mg (BAY73-4506) | 2 | 14 | 14.3% |
| Placebo | 0 | 11 | 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 |
|---|---|
| Regorafenib 160 mg (BAY73-4506) | 13 of 14 |
| Placebo | 10 of 11 |
Most frequent serious adverse event terms
| Event term | Organ system | Regorafenib 160 mg (BAY73-4506) | Placebo |
|---|---|---|---|
| Erythema multiforme | Skin and subcutaneous tissue disorders | 1 of 14 | 0 of 11 |
| Generalised erythema | Skin and subcutaneous tissue disorders | 1 of 14 | 0 of 11 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Regorafenib 160 mg (BAY73-4506) | Placebo |
|---|---|---|---|
| Hypertension | Vascular disorders | 7 of 14 | 1 of 11 |
| Palmar-plantar erythrodysaesthesia syndrome | Skin and subcutaneous tissue disorders | 6 of 14 | 1 of 11 |
| Asthenia | General disorders | 4 of 14 | 2 of 11 |
| Constipation | Gastrointestinal disorders | 4 of 14 | 2 of 11 |
| Dysphonia | Respiratory, thoracic and mediastinal disorders | 5 of 14 | 1 of 11 |
| Blood bilirubin increased | Investigations | 5 of 14 | 0 of 11 |
| Fatigue | General disorders | 2 of 14 | 3 of 11 |
| Lipase increased | Investigations | 4 of 14 | 1 of 11 |
| Myalgia | Musculoskeletal and connective tissue disorders | 3 of 14 | 2 of 11 |
| Aspartate aminotransferase increased | Investigations | 4 of 14 | 0 of 11 |
Serious events with the highest arm-level rate: erythema multiforme (7.1%), generalised erythema (7.1%).
The registry reports serious adverse-event figures separately for regorafenib and placebo. For regorafenib, 14 were reported at risk and 2 affected by serious adverse events; for placebo, 11 were reported at risk and 0 affected. For other adverse events, the record reports 14 at risk and 13 affected in the regorafenib group, and 11 at risk and 10 affected in the placebo group. These are the reported counts; they are not an efficacy comparison.
The registry lists erythema multiforme and generalised erythema among the serious adverse-event terms, each at 7.1%. The information available does not identify which arm is associated with each listed term. Death counts are not reported in the registry record.
In general, adverse-event counts describe events or affected participants within the record’s reporting framework. They do not by themselves establish causation, and a descriptive difference should not be interpreted as a treatment effect without considering definitions, follow-up, completeness, and uncertainty. The registry does not report a statistical comparison or confidence intervals for these safety figures. The small reported groups and absence of a full comparative analysis make precise comparative safety conclusions unavailable from this record.
12. What the Result Does, and Does Not, Mean
This record documents an intended randomized comparison of regorafenib and placebo after curative treatment of liver metastases in colorectal cancer. It also documents the endpoint definition, the planned treatment schedules, the terminated status, and selected adverse-event counts. The primary disease-free-survival endpoint, however, has no analyzed participants because the study was prematurely terminated and the data were not collected. No efficacy estimate is reported.
That absence has a specific interpretation: the registry does not answer whether disease-free survival differed between the groups. It does not show that the treatments were equivalent, that either treatment was better, or that there was no effect. Those conclusions would require outcome data and an analysis. A randomized design is valuable for reducing bias in a comparison, but it cannot compensate for a missing endpoint analysis.
The safety counts offer a limited descriptive view. They report serious and other adverse events by group, plus serious-event terms with percentages. They do not include uncertainty estimates or a comparative statistical analysis, and death counts are not reported in the registry record. These data should not be used to make claims beyond the reported counts and terms.
In general, interpreting a trial involves considering the estimate, its uncertainty, the outcome definition, follow-up, analysis methods, and the population analyzed. In this record, several of those elements are unavailable for efficacy: no primary endpoint analysis, no effect estimate, and no inferential statistics. The record also does not explain why the trial was terminated. The phase 3 designation describes the planned study classification; it does not imply that conclusive phase 3 results were produced.
The most defensible reading is therefore narrow. The registry establishes that a randomized, quadruple-masked, parallel-group trial was intended and was terminated, and that its primary disease-free-survival outcome was not analyzed. It permits a factual description of the posted safety counts, but it does not support an efficacy conclusion or a definitive comparative safety conclusion.
13. Why This Trial Matters Statistically
This trial matters statistically because it illustrates the difference between a planned randomized design and an estimable treatment effect. Randomization and masking are important design features, but they do not produce a result when the primary endpoint data are unavailable. The registry identifies the trial as randomized, parallel-group, and quadruple-masked, yet also states that no participants were analyzed for disease-free survival.
It also highlights why endpoint definitions and outcome availability must be read together. The record gives a detailed time-to-event definition, including radiographic recurrence, death before documented recurrence, and censoring at the last evaluable scan for participants without an event. That definition clarifies what the planned outcome meant. It does not imply that follow-up information was collected or that a statistical comparison was performed.
The record further distinguishes enrollment from analysis. Enrollment is listed as 25, while the number analyzed for the primary endpoint is 0. These entries refer to different stages of a trial’s evidence base and should not be treated as interchangeable. The safety counts likewise cannot stand in for missing efficacy outcomes.
In general, transparent reporting of termination, analysis populations, missing outcomes, estimates, and uncertainty is necessary for readers to judge what evidence a trial contributes. Here, the termination and lack of endpoint data are essential to interpreting the registry entry. The case demonstrates that a phase label and a completed design description are not substitutes for analyzable results. It also shows why absent results should be described as unavailable rather than interpreted as a null finding.
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 central limitation is that the primary disease-free-survival endpoint has no analyzed participants. The registry says the study was prematurely terminated and endpoint data were not collected. As a result, there is no reported efficacy estimate, confidence interval, or statistical test for the primary comparison.
The reason for termination is not reported in the registry record. Without that information, readers cannot assess the circumstances leading to the stop or whether it was related to feasibility, recruitment, safety, or another consideration. No such explanation should be inferred from the terminated status alone.
The registry lists enrollment as 25 but does not report the allocation ratio or provide a detailed account of who was randomized, treated, or included in each efficacy analysis population. The primary endpoint analysis count is explicitly 0. Overall survival is listed as a secondary outcome, but no analysis is reported. The statistical analysis plan, sample-size justification, interim analysis procedures, and multiplicity strategy are not reported in the registry record.
The outcome definition describes imaging-based recurrence assessment and censoring, but the record does not report follow-up duration, numbers of recurrences or deaths, scan completeness, or censoring patterns. Therefore, the quality and completeness of outcome ascertainment cannot be evaluated.
Safety reporting is also limited to the stated counts and listed serious-event terms. The record does not provide comparative uncertainty estimates, death counts, or further detail connecting the listed terms to a particular arm. The safety figures should be treated as descriptive information rather than a complete assessment of comparative safety.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Colorectal Cancer Treated With Adjuvant Regorafenib Versus Placebo After Curative Treatment of Liver Metastases in a Randomized, Double-blind, Placebo-Controlled Phase-III STudy — NCT01939223.
- Registry record last updated 2017-10-20; status last verified 2017-09.
- Results first posted 2017-10-20.
- 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
NCT01939223 is a randomized, quadruple-masked, parallel-group, phase 3 study sponsored by Bayer. It was designed to compare regorafenib with placebo after curative treatment of colorectal cancer liver metastases and completion of planned chemotherapy. The registry lists enrollment as 25 and status as terminated.
The primary outcome was investigator-assessed disease-free survival, defined by radiographic recurrence or death before recurrence was documented. The registry states that the study was prematurely terminated and that data were not collected for this endpoint; the number analyzed is 0. Overall survival is listed as a secondary outcome, but no analyses are reported.
Safety counts are reported by group, with serious and other adverse-event figures and two serious-event terms listed at 7.1% each. The reason for termination, efficacy estimates, statistical analysis plan, and comparative safety analyses are not reported in the registry record.