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 Tiragolumab Plus Atezolizumab Versus Atezolizumab in the Treatment of Stage II Melanoma Patients Who Are ctDNA-positive Following Resection (NCT05060003), a randomized phase 2 study. The primary outcome is ctDNA Clearance Rate.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT05060003 |
| Phase | Phase 2 |
| Condition | Stage II Melanoma |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 7 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Washington University School of Medicine |
| Results posted | Yes |
2. Clinical Question
Population
The study concerned people with stage II melanoma after resection whose blood test was positive for circulating tumor DNA (ctDNA). The registry summary says consenting, eligible patients were to be tested, and those testing positive were eligible for randomization. People testing negative were to be observed off protocol.
Intervention
The randomized treatment was atezolizumab plus tiragolumab. The registry describes both medicines as intravenous infusions every 4 weeks, with treatment permitted for up to 13 cycles. The study's hypothesis concerned whether adding tiragolumab could be associated with ctDNA clearance.
Comparator
The comparison group was atezolizumab alone, administered as an intravenous infusion every 4 weeks for up to 13 cycles. The planned randomized phase used a 1:1 allocation between the combination and monotherapy groups. The posted record does not contain treatment-start or arm-specific outcome data.
Primary question
The primary endpoint was the proportion of ctDNA-positive participants with a ctDNA-negative test at Cycle 3 Day 1, estimated to be 8 weeks. The central question was whether that clearance outcome could be assessed across the randomized treatments. The registry reports that all enrolled participants did not start study treatment, and it contains no primary-outcome analyses.
3. Trial Design
This was a phase II, randomized, parallel-group study with no masking. It enrolled people with resected stage II melanoma who tested positive for ctDNA. The registry lists Washington University School of Medicine as sponsor. The study began on 2022-11-08, had a primary completion date of 2023-11-27, and is recorded as terminated. Enrollment was 7.
The registry summary describes a planned first stage in which ctDNA-positive participants would be randomized 1:1 to atezolizumab plus tiragolumab or atezolizumab alone. The combination arm was to receive both medicines intravenously every 4 weeks, and the monotherapy arm was to receive atezolizumab on the same schedule. Treatment could continue for up to 13 cycles. These are descriptions of the planned regimens; the record states that enrolled participants did not start study treatment.
The summary also describes a conditional second stage: if at least 3 participants in the combination arm were ctDNA-negative at C3D1, enrollment would begin for 25 participants assigned to the combination arm, without randomization or an atezolizumab monotherapy arm. The record does not establish whether this stage began. Its condition and proposed enrollment are protocol descriptions, not observed results.
Because no enrolled participant started treatment, the posted record does not document exposure to either regimen or a treated comparison. It records no primary-outcome analyses. The reason for termination is not reported in the registry record. The study's randomization and parallel design describe its planned comparison, but they cannot yield an estimate of treatment-group differences without post-treatment 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 | Unassigned Arm |
|---|---|
| STARTED | 7 |
| COMPLETED | 0 |
| NOT COMPLETED | 7 |
Reasons for not completing this period:
| Reason | Unassigned Arm |
|---|---|
| ctDNA negative | 7 |
Analysis population for the primary outcome, as defined in the registry: All participants enrolled did not start study treatment.
5. Baseline Characteristics
The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.
| Characteristic | Unit | Unassigned Arm |
|---|---|---|
| Participants analysed | 7 | |
| Age, Continuous | years | 63 (61 to 77) |
| Sex: Female, Male: Female | Participants | 1 |
| Sex: Female, Male: Male | Participants | 6 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 0 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 7 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 |
| Race (NIH/OMB): Asian | Participants | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 |
| Race (NIH/OMB): Black or African American | Participants | 0 |
| Race (NIH/OMB): White | Participants | 7 |
| Race (NIH/OMB): More than one race | Participants | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 |
| Region of Enrollment: United States | participants | 7 |
6. Primary Endpoint
The registry lists the primary outcome as ctDNA Clearance Rate, measured over Cycle 3 Day 1 (estimated to be 8 weeks).
| 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, ctDNA clearance rate, was defined as the proportion of ctDNA-positive participants with a ctDNA-negative test at Cycle 3 Day 1, estimated to be 8 weeks. It is a binary outcome at a specified assessment: each evaluable participant would be classified by whether the test was negative at that time. A proportion would then describe the frequency of the defined outcome in the relevant group.
Using a biological test outcome gives the study a measure that can be assessed at a planned time point, rather than relying only on later clinical events. In general, an early biomarker endpoint may be useful for studying whether a treatment strategy is associated with a change in a marker. It does not, by itself, establish a later clinical benefit. The registry lists relapse-free survival, distant metastasis-free survival, overall survival, and safety outcomes as secondary outcomes, but the primary endpoint remains the C3D1 ctDNA result.
The registry reports that all enrolled participants did not start study treatment and contains no primary-outcome analysis. Thus, it does not report an observed clearance proportion or a comparison between arms. Without post-treatment test results, the endpoint's planned definition cannot be converted into an observed trial finding.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The main statistical constraint is that the registry records 7 enrolled participants but states that all participants enrolled did not start study treatment. The record lists no primary-outcome analyses. Consequently, it does not provide a treatment-group estimate of ctDNA clearance at Cycle 3 Day 1, an uncertainty interval for that estimate, or a statistical comparison of the randomized groups. The planned primary endpoint is defined, but its value is not reported in the registry record.
The study's planned design was randomized and parallel, which would ordinarily support a comparison between the combination and monotherapy groups if participants received treatment and endpoint measurements were available. Random allocation can help balance measured and unmeasured characteristics on average across groups. It does not create outcome information when treatment and follow-up measurements are absent. Here, the record's statement that treatment was not started prevents interpretation of the planned comparison as an observed treatment effect.
The record is marked terminated. That status, together with the enrollment and treatment-start information, describes the trial's recorded course but does not explain why it ended. The reason for termination is not reported in the registry record. It would be inappropriate to infer that the study stopped because of safety, efficacy, feasibility, or any other cause.
The listed secondary outcomes include ctDNA negativity at consecutive measures, relapse-free survival, distant metastasis-free survival, overall survival, treatment-related adverse events, and treatment discontinuations due to treatment-related adverse events. These are outcome definitions, not reported findings. The registry records no analyses, and the posted results do not give event estimates, arm-specific counts, or follow-up summaries for these outcomes. The adverse-event record's unassigned group with zero participants at risk is not evidence that treatment caused no adverse events; no treatment was started according to the registry population note.
The planning illustration describes assumptions for calculations based on an observed effect, including a two-sided significance level of 0.05, equal allocation, and a normal approximation. Those assumptions do not constitute an observed analysis or establish the trial's design rationale. With no reported observed primary effect and no arm-specific endpoint results, those planning calculations cannot be interpreted as evidence about this study's treatment comparison. Any effect size, precision, or clinical implication would require outcome data that are not reported in the registry record.
Accordingly, the posted information supports a description of the study question and its termination, but not a conclusion about ctDNA clearance or later clinical outcomes under either treatment. It also cannot show whether the combination differed from atezolizumab alone. This is an absence of an interpretable outcome comparison, not evidence for or against either regimen.
9. Statistical Methods Explained
The registry describes a randomized, parallel-group phase II study without masking. In a randomized comparison, assignment is determined by chance according to the planned allocation process. In general, this approach is intended to reduce systematic differences between treatment groups at enrollment. A parallel design means participants are assigned to a group and followed under that group's planned treatment strategy rather than switching between groups. The registry's description establishes the intended design, but not that participants received the assigned interventions.
The primary endpoint is a proportion: the number of evaluable ctDNA-positive participants who have a ctDNA-negative result at Cycle 3 Day 1 relative to the relevant group denominator. In general, proportions are natural summaries for outcomes with two categories, such as negative or not negative at a specified assessment. To interpret one, readers need both the numerator and the denominator, as well as a clear rule for which participants count as evaluable. The registry states the endpoint definition but provides no outcome counts or denominator for this assessment.
For a randomized treatment comparison, the contrast would normally compare the clearance proportions between the groups. Statistical uncertainty matters because an estimated proportion based on limited observations can vary substantially from the underlying rate. A confidence interval is, in general, a range of values compatible with the observed data under the analysis assumptions. Its width conveys precision: wider intervals indicate less precise estimation. No interval can be interpreted for this primary outcome from the registry record because no analysis or post-treatment endpoint result is reported.
When a binary endpoint is analyzed, the chosen method should reflect the amount of information and the structure of the data. In general, a normal approximation can be used to approximate the sampling distribution of a proportion or a difference between proportions, but its accuracy depends on having adequate information in the relevant groups. Exact or other small-sample methods may behave differently when counts are sparse. The record's planning assumptions mention a normal approximation, but that is a setting for a planning illustration, not evidence that a particular method was used for an actual primary-outcome analysis.
The planning illustration also specifies a two-sided significance level of 0.05 and equal allocation. A significance level is, in general, a threshold used in a testing procedure to control the chance of a false positive under a stated null model. A two-sided test considers departures in either direction. These settings alone do not describe a complete analysis plan: the test statistic, model, handling of missing data, analysis population, and multiplicity approach would also matter. Those details are not reported in the registry record, and the posted results contain no analyses to evaluate.
A sample-size calculation requires assumptions about the outcome rates or effect to be detected, the intended allocation, the significance level, and desired power. Power is, in general, the probability that a specified analysis will detect an effect of a particular size under stated assumptions. It is a planning property, not a description of the probability that a completed study's conclusion is correct. A planning calculation based on an observed effect also requires an observed effect estimate. The record's primary outcome has no reported observed estimate, and the primary analysis has no reported comparison; therefore, no such calculation can be interpreted as a trial finding. Any planning illustration should be read as hypothetical and not as observed power.
The study summary describes a conditional second stage with combination treatment alone if the stated ctDNA condition was met in the first stage. In general, a multi-stage design can make later enrollment depend on an interim criterion. That feature can affect how evidence is interpreted, because the decision rule and any adjustment to error rates should be prespecified and accounted for in the analysis. The registry record does not report an analysis plan for this condition, whether the condition was assessed, or whether the later stage began. No conclusions about a stage transition can be drawn from the enrollment and termination information alone.
The record's stated population note is especially important for interpreting any endpoint: all enrolled participants did not start study treatment. A planned assessment after treatment initiation cannot be assumed to have occurred merely because the endpoint and timeframe are registered. No C3D1 results, arm-specific denominators, or missing-outcome rules are reported. General methods for handling incomplete outcomes, such as defining an analysis population or modeling missingness, require information about why measurements are absent and what was observed; those details are not available.
Finally, the registry lists 9 outcome measures and 0 analyses. A list of outcomes identifies what the study intended to evaluate; it does not establish that each outcome was collected or analyzed. The statistical interpretation must therefore remain descriptive: the design and endpoint are specified, but treatment effects and their precision are not reported.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists a primary endpoint and 9 outcome measures, including ctDNA measures, survival outcomes, and safety outcomes. When a study evaluates multiple outcomes or multiple comparisons, multiplicity can increase the chance of obtaining an apparently notable result by chance. In general, a statistical plan may address this by defining a hierarchy of outcomes, adjusting testing thresholds, or treating some results as exploratory. The multiplicity procedure is not reported in the registry record.
The record lists 0 analyses, so it does not provide analysis results for the primary or secondary outcomes. It also does not report an analysis population, rules for excluding participants from an analysis, or a method for handling missing endpoint measurements. The registry population note says all enrolled participants did not start treatment. No C3D1 measurement status or follow-up outcome information is reported, so the pattern and reasons for missing post-treatment data cannot be assessed.
These gaps prevent assessment of whether the analysis followed the planned randomized assignment, whether any participants were evaluable for the endpoint, or how outcomes were handled when unavailable. The results do not support an assumption that missing outcomes were negative, positive, or otherwise predictable. The reason for termination and the multiplicity strategy are also not reported in the registry record.
11. What the Result Does, and Does Not, Mean
The registry record establishes that this was a phase II randomized study of atezolizumab with or without tiragolumab for ctDNA-positive stage II melanoma after resection. It specifies ctDNA clearance at Cycle 3 Day 1 as the primary endpoint and describes a planned conditional stage. Those details explain the research question and intended design.
The record does not establish whether either regimen changed ctDNA status. It reports 7 enrolled participants, all of whom did not start study treatment, and 0 analyses. There is no reported primary-endpoint result, treatment-group comparison, confidence interval, or analysis of the secondary outcomes. The enrollment figure therefore cannot be treated as the number of participants with endpoint observations, nor as evidence of a treatment effect.
Likewise, the record does not show that treatment caused no adverse events. The absence of treatment initiation means the registry's unassigned adverse-event group cannot be interpreted as a safety comparison. The registry does not report why the study was terminated, whether planned follow-up occurred, or whether the conditional second stage began.
In general, an absence of reported outcome data is not a null result. A null result would require an observed estimate and an analysis that shows no clear difference, with uncertainty described. Here, the trial's treatment comparison cannot be estimated from the posted information. The record can inform readers about the planned endpoint and the study's documented status, but it cannot answer whether the combination and monotherapy groups differed in ctDNA clearance, recurrence-related outcomes, survival, or safety.
12. Why This Trial Matters Statistically
This study illustrates why registered design details and analyzed results answer different questions. The registry describes randomization, the treatment groups, a primary biomarker endpoint, and a conditional second stage. Those features communicate how investigators planned to structure the comparison. They do not substitute for outcome observations or an analysis.
The distinction is important for biomarker trials. A test result at a specified time can be summarized as a proportion, while later clinical outcomes such as relapse-free survival address different questions and generally require follow-up. A biomarker result should not be treated as interchangeable with a clinical outcome. The registry lists both types of outcomes, but reports no analyses for either.
The record also demonstrates the limits of interpreting enrollment without treatment exposure. Although 7 participants were enrolled, the population note says none started study treatment. That fact prevents the planned randomized comparison from yielding a post-treatment estimate in the posted results. It also means the record cannot be used to assess treatment-related safety outcomes.
For researchers, the missing analysis details matter as well. The registry does not report a multiplicity procedure, missing-data rules, or a complete statistical analysis plan. These details would be important for evaluating a study with several outcomes and a conditional design. Because no outcomes are analyzed, questions about model choice, precision, and the handling of multiple tests remain unresolved rather than answerable from the record.
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 most direct limitation is that the registry states all enrolled participants did not start study treatment. There are no reported post-treatment ctDNA measurements or primary-endpoint analyses, so neither arm's clearance proportion nor a comparison can be evaluated. Enrollment of 7 does not establish how many participants had endpoint assessments.
The study is recorded as terminated, but the reason for termination is not reported in the registry record. The record does not say whether follow-up continued after enrollment, whether any outcome data were collected, or whether the planned conditional second stage began. The phase structure described in the summary must therefore be distinguished from the study's documented results.
The record also lacks reported analysis populations, missing-data procedures, multiplicity handling, and details of the statistical analysis plan. With several registered outcomes, those omissions limit assessment of how results would have been prioritized and interpreted. The absence of reported analyses prevents evaluation of estimates, precision, or consistency across outcomes.
Finally, the open-label design describes the planned masking status, but there are no outcome results with which to assess whether knowledge of treatment assignment could have influenced measurement or management. This is a design consideration, not an explanation for termination or evidence of bias in observed results. The registry does not support conclusions about efficacy, safety, or the reasons the trial ended.
14. Related Tutorials
15. Related Statistical Calculators
16. Sources
- ClinicalTrials.gov: Tiragolumab Plus Atezolizumab Versus Atezolizumab in the Treatment of Stage II Melanoma Patients Who Are ctDNA-positive Following Resection — NCT05060003.
- Registry record last updated 2024-02-20; status last verified 2024-02.
- Results first posted 2024-02-15.
- 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
NCT05060003 is a phase II randomized, parallel-group study recorded as terminated. It concerned ctDNA-positive stage II melanoma after resection and compared planned atezolizumab plus tiragolumab with atezolizumab alone. The primary endpoint was the proportion with a ctDNA-negative test at Cycle 3 Day 1, estimated to be 8 weeks.
The registry reports enrollment of 7 and states that all enrolled participants did not start study treatment. It lists 0 analyses. No primary or secondary outcome results, treatment comparison, or reason for termination is reported in the registry record. The record describes a conditional combination-treatment stage, but does not establish that it began. The available information describes the study's intended design and recorded status, not an observed treatment effect.