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 Circulating Tumor DNA Testing in Predicting Treatment for Patients With Stage IIA Colon Cancer After Surgery (NCT04068103), a randomized phase 2, phase 3 study. The primary outcome is Clearance of Circulating Tumor Deoxyribonucleic Acid (ctDNA) (to Undetectable Levels) for the "Baseline ctDNA Detected" Patient Subset (Phase II).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT04068103 |
| Phase | Phase 2, Phase 3 |
| Condition | Colon Adenocarcinoma, Stage IIA Colon Cancer AJCC v8 |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 635 |
| Arms in the primary outcome | n/a |
| Lead sponsor | NRG Oncology |
| Results posted | Yes |
2. Clinical Question
Population
Patients with stage IIA colon cancer after surgery are the population described. The conditions listed are Colon Adenocarcinoma and Stage IIA Colon Cancer AJCC v8. The trial concerns circulating tumor deoxyribonucleic acid (ctDNA) in blood after resection; the primary endpoint is limited to the subset whose baseline ctDNA was detected.
Intervention
One group has blood tested for ctDNA at baseline. For patients in that group with ctDNA detected, chemotherapy may be given at the investigator’s discretion. The listed drugs include oxaliplatin, fluorouracil, leucovorin, capecitabine and leucovorin calcium. The record’s arm description is incomplete, so it does not establish all treatment details or exactly which regimens were used.
Comparator
The other arm stores blood for later ctDNA testing and uses active surveillance. The study is randomized, open-label and parallel. The registry record does not report the allocation procedure beyond identifying the trial as randomized, nor does it describe blinding of outcome assessment.
Primary question
Among patients with ctDNA detected at baseline, how many have ctDNA cleared to undetectable levels by up to 6 months, compared by treatment arm? The stated Phase II endpoint uses a two-by-two contingency table and a one-sided Fisher exact p-value to assess futility under a specified rule. The rule itself and an observed clearance result are not reported in the registry record.
The broader question is whether baseline blood testing can help distinguish patients for whom adjuvant chemotherapy is considered from those managed with surveillance. The registry summary describes ctDNA as tumor-derived material that may remain detectable after surgery even when residual cancer is too small to be found by usual tests. This is the rationale recorded for studying the biomarker; it is not evidence, by itself, that the test predicts treatment benefit. The study also lists recurrence-free survival, overall survival, time to recurrence, treatment or surveillance compliance, and ctDNA presence after resection as secondary outcomes.
3. Trial Design
COBRA is identified as a Phase II/III randomized, parallel-group study with no masking. Its registry title describes a study of ctDNA as a predictive biomarker for adjuvant chemotherapy in patients with stage IIA colon cancer. The listed enrollment is 635. The sponsor is NRG Oncology. The start date is 2019-12-16, the primary completion date is 2023-06-30, and the registry status is active, not recruiting.
The two arms differ in when blood is tested. In Arm I, blood is stored and tested for ctDNA later, and patients undergo active surveillance. In Arm II, blood is tested at baseline. The arm description says that patients with ctDNA detected are assigned to one of two groups and that, for the detected group, chemotherapy may be given at the investigator’s discretion. It begins to describe drug administration, but the description ends before the regimen details are complete. The full treatment assignment for the other group and the full arm-specific procedures are not reported in the registry record.
Randomization indicates that assignment to study arms is intended to use a random process, but the allocation method, stratification factors, and concealment procedures are not reported in the registry record. No masking is listed. The registry record does not report whether outcome assessors were blinded or how adherence to assigned procedures was handled. Although the study is described as Phase II/III, the record does not explain the transition between phases or report phase-specific design details.
The primary endpoint is a count of participants with clearance of ctDNA to undetectable levels in the baseline-detected subset, over baseline up to 6 months. The stated Phase II analysis uses a two-by-two contingency table by treatment arm and a one-sided Fisher exact p-value for a futility decision. The record’s endpoint description refers to a specified futility rule, but does not give the rule’s threshold or decision boundary. No results for this endpoint are reported. The registry lists seven outcome measures and one analysis, but the available descriptions do not establish the full analysis plan or results for all outcomes.
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 | Arm I (Blood Stored and Tested for ctDNA Later) | Arm II (Blood Tested for ctDNA at Baseline) |
|---|---|---|
| STARTED | 318 | 317 |
| COMPLETED | 318 | 317 |
| NOT COMPLETED | 0 | 0 |
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 | Arm I (Blood Stored and Tested for ctDNA Later) | Arm II (Blood Tested for ctDNA at Baseline) | Total |
|---|---|---|---|---|
| Participants analysed | 318 | 317 | 635 | |
| Age, Customized: Less than 60 years | Participants | 143 | 141 | 284 |
| Age, Customized: Greater than or equal to 60 years | Participants | 175 | 176 | 351 |
| Sex: Female, Male: Female | Participants | 142 | 153 | 295 |
| Sex: Female, Male: Male | Participants | 176 | 164 | 340 |
| Race/Ethnicity, Customized: American Indian or Alaska Native | Participants | 1 | 1 | 2 |
| Race/Ethnicity, Customized: Asian | Participants | 17 | 16 | 33 |
| Race/Ethnicity, Customized: Black or African American | Participants | 25 | 18 | 43 |
| Race/Ethnicity, Customized: White | Participants | 257 | 262 | 519 |
| Race/Ethnicity, Customized: Unknown or Not Reported for Race | Participants | 18 | 20 | 38 |
| Race/Ethnicity, Customized: Hispanic or Latino | Participants | 26 | 34 | 60 |
| Race/Ethnicity, Customized: Not Hispanic or Latino | Participants | 283 | 270 | 553 |
| Race/Ethnicity, Customized: Unknown or Not Reported Ethnicity | Participants | 9 | 13 | 22 |
6. Primary Endpoint
The registry lists the primary outcome as Clearance of Circulating Tumor Deoxyribonucleic Acid (ctDNA) (to Undetectable Levels) for the "Baseline ctDNA Detected" Patient Subset (Phase II), measured over Baseline up to 6 months.
| Element | Registry entry |
|---|---|
| Result type | Count Of Participants |
| Measure of spread | not applicable |
| Unit | Participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint asks whether ctDNA, detected in blood at baseline, becomes undetectable during the period from baseline up to 6 months. It is defined for the “Baseline ctDNA Detected” patient subset and is measured as a count of participants. The planned comparison is clearance by treatment arm, represented as a two-by-two contingency table.
Clearance is used as an early biomarker outcome: it records whether the blood test changes from detectable at baseline to undetectable within the stated follow-up period. The trial rationale describes ctDNA as tumor-derived material that can remain in blood after surgery and may be present when disease is too small to be identified by usual tests. Studying clearance therefore focuses on a measurable laboratory finding rather than directly measuring recurrence or survival. It should not be treated as a substitute for those clinical outcomes.
The registry describes use of a one-sided Fisher exact p-value to determine futility under a specified rule. Fisher’s exact test evaluates evidence of an association between two categorical variables—in this case, clearance status and treatment arm—using the observed counts. A futility analysis asks whether results meet a prespecified criterion for stopping or not continuing because the desired evidence is unlikely to emerge. The record does not report the exact rule, the number of patients in this subset, or observed clearance counts. It also does not report an observed endpoint estimate or p-value. Therefore, the stated endpoint and method describe the planned question, not a conclusion about clearance or treatment effect.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies a clinical question and a planned primary analysis, but it does not provide an observed result for the primary endpoint. The endpoint is clearance of ctDNA to undetectable levels among patients whose baseline test detected ctDNA, assessed from baseline up to 6 months. The record specifies that clearance will be compared by treatment arm using a two-by-two contingency table and a one-sided Fisher exact p-value for a futility decision. It does not report the arm-specific clearance counts, the p-value, or the futility threshold. Consequently, the available record does not show whether clearance differed between arms or whether the futility criterion was met.
The trial is randomized and parallel, which is a design intended to compare assigned strategies across groups. However, the registry summary alone does not establish how randomization was implemented, how many participants entered the baseline-detected subset, or whether all randomized participants were included in each outcome analysis. Those details are not reported in the registry record. The intervention description also indicates investigator discretion for chemotherapy among patients with ctDNA detected, while the complete assignment description is not available. That means the exact treatment contrast and the handling of treatment choices cannot be reconstructed from the arm descriptions alone.
The study lists recurrence-free survival, overall survival, time to recurrence, compliance with chemotherapy and/or surveillance, and ctDNA incidence after resection as secondary outcomes. No values or inferential results for these outcomes appear in the record details. Their listing does not establish that they were measured at a particular time point, analyzed in a particular population, or statistically significant. The stated enrollment of 635 is not the number of participants in the primary endpoint subset and cannot be used as its denominator without further reporting.
The safety figures record serious adverse events, other adverse events, and deaths by arm. They are counts, not estimates of a treatment effect. In Arm I, the serious- and other-event at-risk and affected counts are all recorded as 0, with 8 deaths. In Arm II, 15 participants are recorded as at risk for serious adverse events and 3 as affected; for other adverse events, 15 are recorded as at risk and 15 as affected, with 6 deaths. The record does not explain the apparent difference in the at-risk counts between arms, provide exposure time, or clarify the relationship between the safety denominators and enrollment. These figures should be interpreted as registry-recorded counts, not as proof that an arm is safer or more harmful.
Overall, the material supports describing the trial’s design, planned endpoint, and posted safety counts. It does not support a claim about ctDNA clearance, recurrence, survival, or the comparative benefit or harm of the strategies. No treatment recommendation follows from these registry entries.
9. Statistical Methods Explained
The statistical method described for the primary endpoint is a one-sided Fisher exact test applied to a two-by-two contingency table. The rows or columns of such a table represent categories—in this setting, clearance versus no clearance and treatment arm. The test evaluates whether the observed pattern of counts is compatible with no association between arm and clearance, subject to the assumptions of the test and the definition of the analyzed population. The registry record does not report the actual cell counts or the resulting p-value.
Why use Fisher’s exact test?
Fisher’s exact test is designed for categorical data and calculates probabilities from the fixed margins of a contingency table. It is often suitable when counts are small or when some cells may contain very few observations, because its calculation does not rely on the large-sample approximation used by some other tests for categorical outcomes. The baseline-detected subset may be smaller than the full enrolled study population, but its size is not reported in the registry record. It would therefore be inappropriate to infer whether sparse counts occurred or whether another test would have been preferable.
The test is described as one-sided, meaning that the planned hypothesis test addresses a directional question rather than treating departures in either direction as equally relevant. The record does not state which direction defines the alternative hypothesis. Nor does it report the significance threshold or a decision boundary for the futility rule. The endpoint description says that the p-value is used to determine futility, including a rule for degenerate tables: if no patients clear on either arm, or all patients clear on both arms, the result counts as a failure and a recommendation for early termination. These rules are part of the stated method, but the exact operational threshold for other results is not reported.
What a p-value does and does not say
In general, a p-value measures how incompatible the observed data are with a specified null hypothesis under the statistical model and test assumptions. It is not the probability that the null hypothesis is true, and it does not quantify the size or clinical importance of a difference. A one-sided p-value also depends on the direction specified in advance. Since the registry record does not report an observed p-value or the complete hypothesis and decision rule, no interpretation of statistical evidence for this trial’s primary endpoint can be made from the method description alone.
Futility analysis is intended to inform whether continuing a study or stage is worthwhile under prespecified criteria. A futility finding is not identical to proof that treatments are equivalent, and a result that does not meet a futility boundary is not by itself proof of benefit. Interpretation depends on the rule, the planned analysis timing, the endpoint data, and the population analyzed. The exact schedule of any futility assessment and whether it is the only planned interim assessment are not reported in the registry record. The endpoint description mentions early termination for degenerate tables, but does not establish that any such table occurred.
What randomization contributes
In general, random assignment can reduce systematic differences between treatment groups at baseline, on average, when properly implemented. It does not ensure that groups will be identical, and it does not resolve missing outcome data or deviations from assigned treatment. The record labels the study randomized, but does not report the allocation sequence method, concealment, stratification, or procedures for analyzing participants according to assignment. The study is also described as unmasked. This means no masking is listed in the registry; details about masking of laboratory staff or outcome assessors are not reported.
The arm descriptions raise an additional interpretation issue: chemotherapy in the baseline-detected group is described as being at the investigator’s discretion. The available description does not fully specify the regimens, the treatment for all categories of ctDNA result, or how such discretion enters the planned comparison. Accordingly, the registry description is insufficient to characterize the precise estimand—the specific treatment contrast the primary analysis is intended to estimate.
Counts, estimates, and precision
The primary outcome is expressed as a count of participants. Counts communicate how many participants meet a binary endpoint, but comparisons also depend on the number of participants assessed in each arm. The record does not report those denominators, clearance proportions, an effect estimate, or a confidence interval. In general, a confidence interval describes a range of estimates compatible with the data and method, and indicates precision; it does not guarantee that the true value lies within the interval. No interval for the primary result is reported.
The enrollment target of 635 is a study-level figure, not a substitute for endpoint-specific denominators. The number with baseline-detected ctDNA, the number with follow-up testing, and the number included in the analysis are not reported in the registry record. The stated endpoint timeframe, baseline up to 6 months, also does not establish the timing or completeness of measurements for individual participants. These omissions prevent an assessment of statistical precision, missingness, and the population contributing to the analysis.
The registry identifies one analysis and seven outcome measures, but does not describe the complete statistical analysis plan. The record does not report a sample-size justification, planned power, an interim-analysis schedule, a multiplicity strategy, or methods for missing data. These matters are especially relevant when a study has several outcomes or decision points: in general, repeated testing can increase the chance of at least one false-positive finding unless the analysis plan accounts for it. That general concern cannot establish that multiplicity occurred or was mishandled in this trial.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists seven outcome measures, including the primary ctDNA-clearance endpoint and secondary outcomes for recurrence-free survival, overall survival, time to recurrence, compliance, and ctDNA presence after resection. It identifies one analysis. The record does not report how many statistical hypotheses were planned, whether outcomes had a formal hierarchy, or whether any adjustment for multiple testing was planned. With multiple outcomes, general statistical practice distinguishes exploratory findings from confirmatory tests and considers how repeated testing affects the interpretation of p-values. No specific multiplicity approach can be attributed to this study from the available record.
The primary endpoint description includes a one-sided Fisher exact p-value used for futility and states how two degenerate tables are treated. It does not give the p-value threshold, full futility boundary, or timing of the assessment. The registry record does not report whether there were other interim analyses or stopping rules. It also does not report the sample-size justification or planned statistical power. Those details should not be inferred from the enrollment target.
Missing-data methods and analysis populations are not described. The record does not say how participants without follow-up ctDNA testing, with indeterminate results, or with incomplete outcome data would be handled. It does not identify an intention-to-treat, per-protocol, or other analysis population for the primary endpoint. For survival outcomes, censoring and follow-up procedures are not reported. The number randomized, the number in the baseline-detected subset, and the number analyzed for each outcome are also not stated. These gaps limit assessment of selection, attrition, and the precision or generalizability of any eventual estimate.
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 |
|---|---|---|---|
| Arm II (Blood Tested for ctDNA at Baseline) | 3 | 15 | 20.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Arm I (Blood Stored and Tested for ctDNA Later) | 8 of 318 |
| Arm II (Blood Tested for ctDNA at Baseline) | 6 of 317 |
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 |
|---|---|
| Arm I (Blood Stored and Tested for ctDNA Later) | 0 of 0 |
| Arm II (Blood Tested for ctDNA at Baseline) | 15 of 15 |
Most frequent serious adverse event terms
| Event term | Organ system | Arm I (Blood Stored and Tested for ctDNA Later) | Arm II (Blood Tested for ctDNA at Baseline) |
|---|---|---|---|
| Diarrhea | Gastrointestinal disorders | 0 of 0 | 2 of 15 |
| Abdominal pain | Gastrointestinal disorders | 0 of 0 | 1 of 15 |
| Hepatobiliary disorders - Other, specify | Hepatobiliary disorders | 0 of 0 | 1 of 15 |
| Neutrophil count decreased | Investigations | 0 of 0 | 1 of 15 |
| Typhlitis | Gastrointestinal disorders | 0 of 0 | 1 of 15 |
| White blood cell decreased | Investigations | 0 of 0 | 1 of 15 |
| Enterocolitis infectious | Infections and infestations | 0 of 0 | 1 of 15 |
| Non-cardiac chest pain | General disorders | 0 of 0 | 1 of 15 |
| Small intestinal mucositis | Gastrointestinal disorders | 0 of 0 | 1 of 15 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Arm I (Blood Stored and Tested for ctDNA Later) | Arm II (Blood Tested for ctDNA at Baseline) |
|---|---|---|---|
| Diarrhea | Gastrointestinal disorders | 0 of 0 | 9 of 15 |
| Fatigue | General disorders | 0 of 0 | 9 of 15 |
| Peripheral sensory neuropathy | Nervous system disorders | 0 of 0 | 9 of 15 |
| Platelet count decreased | Investigations | 0 of 0 | 8 of 15 |
| Nausea | Gastrointestinal disorders | 0 of 0 | 7 of 15 |
| Hypokalemia | Metabolism and nutrition disorders | 0 of 0 | 6 of 15 |
| Neutrophil count decreased | Investigations | 0 of 0 | 6 of 15 |
| Mucositis oral | Gastrointestinal disorders | 0 of 0 | 5 of 15 |
| Abdominal pain | Gastrointestinal disorders | 0 of 0 | 4 of 15 |
| Alkaline phosphatase increased | Investigations | 0 of 0 | 4 of 15 |
Serious events with the highest arm-level rate: diarrhea (13.3%), abdominal pain (6.7%), hepatobiliary disorders - other, specify (6.7%), neutrophil count decreased (6.7%), typhlitis (6.7%).
The serious adverse event figures are reported by arm as counts at risk and affected. Arm I has 0 recorded at risk and 0 affected for serious adverse events, and 0 at risk and 0 affected for other adverse events; deaths are recorded as 8. Arm II has 15 recorded at risk and 3 affected for serious adverse events, and 15 at risk and 15 affected for other adverse events; deaths are recorded as 6. These are the registry’s posted counts and should be read as such.
The listed serious adverse event terms include Diarrhea at 13.3%, Abdominal pain at 6.7%, Hepatobiliary disorders - Other, specify at 6.7%, Neutrophil count decreased at 6.7%, Typhlitis at 6.7%, and White blood cell decreased at 6.7%. The denominator and exact meaning of these percentages are not explained in the record details, so they should not be converted into counts or compared as rates. The record also does not clarify why Arm I has zero participants at risk in the adverse-event fields despite a death count, or how these safety data relate to overall enrollment.
Adverse-event counts alone do not establish whether an event was caused by treatment, how long participants were observed, or whether groups had comparable exposure. The registry record does not report event timing, severity grading details beyond the serious-event designation, or adjusted comparisons. These figures therefore do not support a causal or comparative safety conclusion.
12. What the Result Does, and Does Not, Mean
The registry record shows that COBRA was designed to study whether ctDNA testing after surgery could inform treatment decisions for stage IIA colon cancer. Its primary Phase II endpoint concerns whether ctDNA becomes undetectable among patients with ctDNA detected at baseline, comparing counts by treatment arm over baseline up to 6 months. A one-sided Fisher exact test is specified for a futility decision. This describes the research question and planned analysis, not an observed finding.
No clearance counts, p-value, effect estimate, or confidence interval are reported for the primary endpoint. The record therefore does not establish that clearance was more or less frequent in either arm. It also does not report results for the listed recurrence, survival, compliance, or post-resection ctDNA outcomes. The listed enrollment of 635 cannot be used to infer how many participants had detectable ctDNA at baseline or contributed data to a particular outcome.
Even a future difference in ctDNA clearance would need to be interpreted alongside the study’s full methods and clinical outcomes. A biomarker result is not automatically equivalent to a difference in recurrence or survival. The registry summary explains why ctDNA is of interest after surgery, but the record does not provide results that validate the test as a predictor of chemotherapy benefit. The chemotherapy description is incomplete and includes investigator discretion for the detected group, so the exact comparison cannot be fully reconstructed from the arm descriptions alone.
The posted safety counts are likewise not comparative estimates of risk. Their denominators are unclear, the at-risk fields differ between arms, and exposure and follow-up details are not stated. They cannot establish that one strategy caused more or fewer adverse events or deaths.
Accordingly, the record supports a description of a randomized Phase II/III design, its stated endpoint, and the reported safety counts. It does not support a conclusion that ctDNA testing or a treatment strategy improves outcomes, nor does it establish lack of benefit or equivalence. Clinical decisions should not be derived from this registry record alone.
13. Why This Trial Matters Statistically
This trial is statistically notable for separating a biomarker-focused endpoint from longer-term outcomes. The primary Phase II endpoint is ctDNA clearance in the baseline-detected subset, whereas the listed secondary outcomes include recurrence-free survival, overall survival, and time to recurrence. These measures answer different questions: a blood-test change is not the same outcome as recurrence or survival. Keeping those distinctions clear is central to interpreting biomarker studies.
The planned use of a one-sided Fisher exact test reflects a categorical endpoint analyzed through counts in a two-by-two contingency table. The futility framing also makes the decision rule important: a p-value is being used within a stopping framework, not merely as a conventional report of treatment difference. The registry record describes some degenerate-table handling but does not report the full threshold or assessment schedule. That missing detail limits independent understanding of how the statistical decision would be made.
The study is randomized and parallel, but no masking is listed, and chemotherapy for the baseline-detected group is described as being at investigator discretion. These design details matter because they shape the treatment contrast and the interpretation of any between-arm difference. The incomplete arm description and absent analysis-population details prevent a complete reconstruction of the estimand or analysis.
Finally, the registry lists multiple outcomes while giving little information about multiplicity, missing data, or precision. This illustrates why a trial’s outcome definitions and analysis plan are as important as its headline design. Without endpoint results and full methods, the record can clarify what was intended to be studied, but cannot establish what the trial found.
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 registry record does not report an observed primary-endpoint result. It gives no arm-specific clearance counts, endpoint denominators, p-value, effect estimate, or confidence interval. The endpoint is defined for the baseline-detected subset, but the size of that subset is not reported. Enrollment of 635 does not identify how many participants had detectable ctDNA or completed follow-up testing.
The treatment descriptions are incomplete. Arm I is described as active surveillance with blood stored for later testing. Arm II tests blood at baseline, and chemotherapy for patients with ctDNA detected is at investigator discretion, but the full regimen and complete assignment pathway are not stated. Randomization is reported, while allocation procedures and concealment are not. No masking is listed, and outcome-assessment procedures are not reported.
The primary endpoint description refers to a specified futility rule but does not give its threshold or complete decision boundary. The record does not explain the sample-size justification, interim-analysis schedule, missing-data methods, or the analysis population. It lists seven outcomes and one analysis but does not report how multiplicity was addressed. Follow-up details for the secondary outcomes are also not given in the record details.
The safety figures have unresolved denominator and interpretation issues. Arm I’s adverse-event at-risk counts are zero, while deaths are recorded; the record does not explain this combination. Exposure time, event timing, and the relationship of safety counts to enrollment are not stated. The listed serious-event percentages are not accompanied by a denominator in the record details. These limitations prevent valid rate comparisons or causal attribution based on the posted figures.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Circulating Tumor DNA Testing in Predicting Treatment for Patients With Stage IIA Colon Cancer After Surgery — NCT04068103.
- Registry record last updated 2025-05-07; status last verified 2025-04.
- Results first posted 2025-04-27.
- 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
NCT04068103, titled “Circulating Tumor DNA Testing in Predicting Treatment for Patients With Stage IIA Colon Cancer After Surgery,” is a Phase II/III randomized, unmasked, parallel-group study sponsored by NRG Oncology. The listed enrollment is 635, and the registry status is active, not recruiting. The start date is 2019-12-16, and the primary completion date is 2023-06-30.
The study compares an approach with baseline ctDNA testing and an approach with blood stored for later testing and active surveillance. Its Phase II primary endpoint is clearance of ctDNA to undetectable levels among patients with ctDNA detected at baseline, measured from baseline up to 6 months. The stated method is a two-by-two contingency table and a one-sided Fisher exact p-value for futility. The registry record does not report an observed primary result or the complete futility rule.
Secondary outcomes include recurrence-free survival, overall survival, time to recurrence, compliance, and ctDNA presence following resection. Posted adverse-event counts are reported by arm, but their denominators and comparative meaning are not fully explained. The registry record does not report the full statistical analysis plan, missing-data methods, or analysis populations.