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 Nivolumab Ipilimumab in Patients With hyperMutated Cancers Detected in Blood (NIMBLe) (NCT03461952), a randomized phase 2 study. The primary outcome is Objective Response Rate by RECIST 1.1.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03461952 |
| Phase | Phase 2 |
| Condition | Advanced Solid Tumors |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 4 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Canadian Cancer Trials Group |
| Results posted | Yes |
2. Clinical Question
Population
People with advanced solid tumors, described in the study summary as metastatic or unresectable tumors with POLE or POLD1 mutations detected by plasma cell-free DNA. The primary-outcome description says the study closed to accrual early after screening found these mutations less often than expected. It records that 4 patients were accrued, but does not report a detailed profile of the enrolled participants.
Intervention
Nivolumab alone, with a recorded schedule of 240mg Q2W. The study was designed to assess safety and effectiveness in this selected population. The registry record does not provide an estimable primary-outcome value for this arm.
Comparator
Nivolumab with ipilimumab, recorded as nivolumab 240mg Q2Wk plus ipilimumab 1mg/kg Q6W. Although allocation was randomized, the official title describes the study as non-comparative. The record therefore does not establish a formal treatment-effect comparison between the arms.
Primary question
The primary outcome was objective response rate by RECIST 1.1 over 36 months. The registry describes the outcome as a count of participants and lists CR, PR, SD, or PD in its endpoint description. Arm-level outcome values are missing or not estimable for a binary comparison, so the record does not provide an answer about response in either arm.
3. Trial Design
NCT03461952, titled Nivolumab Ipilimumab in Patients With hyperMutated Cancers Detected in Blood (NIMBLe), was a phase II, randomized, open-label, parallel-group study. Its official title describes a non-comparative trial of nivolumab alone or with ipilimumab for advanced hypermutated solid tumors detected by a blood-based assay. Masking was listed as none. The Canadian Cancer Trials Group was the sponsor.
The study began on 2019-03-11 and had a recorded primary completion date of 2021-06-01. Its current registry status is terminated. The primary-outcome description says accrual was closed prematurely on August 2, 2019, following review of screening data that found the prevalence of relevant mutations was lower than expected. Only 4 patients were accrued. The registry record does not report the reason for termination beyond this explanation of early accrual closure.
The two listed arms were nivolumab and nivolumab plus ipilimumab. The nivolumab schedule was 240mg Q2W; the combination schedule was nivolumab 240mg Q2Wk with ipilimumab 1mg/kg Q6W. Random allocation describes how participants were assigned, but does not by itself mean that the study was designed or powered to estimate a between-arm treatment effect. Here the official title explicitly calls the study non-comparative, and the posted results do not contain an estimable binary comparison.
The registry lists 6 outcome measures and names 5 secondary outcomes: objective response rate, duration of response, adverse-event number and severity using CTCAE 5.0, correlation between tumor and blood POLE or POLD1 mutations, and response by iRECIST. The available record does not include posted analysis results for these secondary outcomes. A detailed statistical analysis plan and further information about conduct beyond the listed design are not reported in the registry record.
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 | Nivolumab | Nivolumab + Ipilimumab |
|---|---|---|
| STARTED | 2 | 2 |
| COMPLETED | 2 | 2 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: The study was pre-maturely closed to accrual on August 2, 2019 following review of screening data which found that the prevalence of relavent mutations was lower than expected. Only 4 patients were accrued.
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 | Nivolumab | Nivolumab + Ipilimumab | Total |
|---|---|---|---|---|
| Participants analysed | 2 | 2 | 4 | |
| Age, Continuous | years | 53 (31 to 75) | 53 (50 to 55) | 53 (31 to 75) |
| Sex: Female, Male: Female | Participants | 0 | 1 | 1 |
| Sex: Female, Male: Male | Participants | 2 | 1 | 3 |
| Race and Ethnicity Not Collected | Participants | — | — | 0 |
6. Primary Endpoint
The registry lists the primary outcome as Objective Response Rate by RECIST 1.1, measured over 36 months.
| Element | Registry entry |
|---|---|
| Result type | Count Of Participants |
| Measure of spread | not applicable |
| Unit | Participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint is objective response rate by RECIST 1.1, with a timeframe of 36 months. The registry gives the unit as participants and the parameter as a count of participants. Its endpoint description lists CR, PR, SD, or PD. That description should be read as written: it includes complete response, partial response, stable disease, and progressive disease, rather than specifying only a subset as the event definition. The registry does not explain how this list was used to classify the count.
In general, a response endpoint aims to describe tumor status according to a defined assessment framework. A standardized framework such as RECIST 1.1 can make response classification more consistent across participants and studies. A count is a direct way to report how many participants meet a specified endpoint definition; to interpret it as a rate, the relevant denominator and event definition must also be clear.
For this record, the primary endpoint values are null in both arms, and the posted data identify values as missing or not estimable for a binary comparison. Consequently, there is no arm-level response count, denominator-based response estimate, or measure of precision reported for the primary endpoint. The record does not allow the primary endpoint to be interpreted as an observed response rate.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The principal statistical feature of the posted results is the absence of an estimable primary-outcome result. The registry lists the primary measure as a count of participants, but no values are reported for either nivolumab or nivolumab plus ipilimumab. It specifically indicates that values are missing or not estimable for a two-arm comparison of a binary endpoint. This is not evidence that the endpoint was zero in either arm; it means the registry does not provide a usable outcome estimate.
The primary-outcome description gives the reason for the study's limited accrual: screening data indicated that relevant mutations were less prevalent than expected, and accrual was closed prematurely. It records that 4 patients were accrued. With so little accrual and no posted primary endpoint counts, the registry cannot support a precise estimate of response for either regimen. It also cannot support a statistical conclusion comparing the regimens. The randomized allocation does not change that limitation, particularly because the official title describes the trial as non-comparative.
The posted adverse-event counts offer descriptive information, but they are also limited. In each arm, 2 participants were at risk for the serious adverse-event count and 0 were recorded as affected by a serious adverse event. Other adverse events were recorded for 2 participants in each arm. The death count was 1 in the nivolumab arm and 0 in the combination arm. These are small-group descriptive counts, not reliable comparative estimates. The registry does not report whether the deaths or other events were considered related to treatment, nor does it provide serious-event terms.
There are no listed analyses in the registry record. No p-values, confidence intervals, adjusted estimates, or arm-level efficacy estimates are reported. A confidence interval would ordinarily communicate the range of effect estimates compatible with the data under a stated method; without an estimate and its uncertainty, precision cannot be assessed. No claim about statistical significance or the absence of an effect can be drawn from missing or non-estimable values.
The record lists 5 secondary outcomes, but does not provide their results. The absence of posted secondary analyses means that duration of response, objective response rate as a secondary measure, adverse-event severity, tumor-to-blood mutation correlation, and iRECIST response cannot be interpreted from this record. Overall, the registry documents an early-closed study and some adverse-event counts, but does not provide an interpretable estimate of the primary efficacy endpoint or a comparative treatment effect.
9. Statistical Methods Explained
The registry record describes a phase II randomized, open-label, parallel study, but does not report a detailed statistical analysis plan. It lists the primary endpoint as a binary outcome summarized by a participant count and records no analyses. The available information therefore supports explanation of the endpoint and design, but not reconstruction of an analysis that was not posted.
What does a binary endpoint count represent?
In general, a binary endpoint classifies each participant according to whether a defined event occurred. Summing those classifications gives an event count. A rate or proportion additionally relates the event count to a clearly defined denominator. For a meaningful estimate, the endpoint definition, analysis population, time window, and handling of participants without an assessment need to be specified. The registry calls the primary outcome a count of participants and gives a 36-month timeframe, but does not provide arm-level counts or a reported denominator for the primary outcome. It also lists CR, PR, SD, or PD in its endpoint description without further detail about how those categories map to the count.
Why is a between-arm estimate unavailable?
The record states that values are missing or not estimable for a two-arm comparison of a binary endpoint. This means that no usable comparison estimate is presented in the posted results. It does not establish that the arms had identical outcomes, and it does not establish that there were no responses. When event counts are sparse or endpoint information is unavailable, a comparison may be unstable or not calculable from the available record. The registry does not state a more specific reason for the non-estimability.
In general, a treatment-effect estimate is most informative when accompanied by an uncertainty measure, such as a confidence interval. A confidence interval describes the uncertainty around an estimate under its statistical assumptions. If no outcome estimate is reported, there is no corresponding interval to interpret. The registry record reports no p-values or confidence intervals for the primary endpoint and lists zero analyses. It therefore does not support a formal inference about a difference between the treatment arms.
What does randomization contribute?
Randomization is a method for allocating participants to arms in a way intended to reduce systematic differences at assignment. Its value for estimating a treatment effect depends on the study's design, the amount of information collected, and the analysis. A randomized label alone does not make a trial comparative in its objectives or make an estimate precise. This study's official title calls it non-comparative, and the posted results do not give an estimable arm comparison. The early closure and limited accrual further constrain what can be learned, but the record does not specify the intended comparative analysis or its assumptions.
What do the planning illustrations mean?
In general, sample-size planning asks how many participants would be needed under stated assumptions to detect a target effect with a chosen error rate and power. Such calculations are prospective illustrations, not observed evidence about what a completed study detected. The registry's planning assumptions describe a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect expressed as a stated fraction of the observed effect. It specifies that non-time-to-event illustrations are expressed as participants per arm. These assumptions do not supply the missing primary outcome estimate, and no actual sample-size illustration values are reported in the registry record.
Because the illustrations are based on an observed effect, their interpretation depends on the existence and stability of that effect estimate. The primary outcome values are unavailable, so the record does not provide the empirical result needed to assess such a planning illustration. Any sample-size value should be read as a planning scenario rather than as observed power. Observed power is not a substitute for an effect estimate and its uncertainty, and it cannot recover the efficacy result missing from the registry.
Why can sparse data be difficult to analyze?
In general, sparse binary outcomes can lead to unstable estimates. A small change in the event count may substantially alter an estimated rate, and some common estimation methods can behave poorly when events are absent or very uncommon. Alternative methods may be considered in an analysis plan, but the registry record does not report any chosen method, model, correction, or estimation strategy. No such approach should be assumed from the design label alone.
Over-dispersion is a general concern when observations vary more than a chosen statistical model expects. It is relevant to methods for count data and clustered outcomes, but the posted primary outcome is described as a binary participant count, and the record does not report an over-dispersion assessment or a model that would require one. No quantitative claim about dispersion can be made from the posted information.
In sum, the statistical description available is limited to the trial structure, the endpoint definition and timeframe, the indication that the binary comparison was not estimable, and the absence of analyses. A protocol, analysis plan, missing-data procedure, and inferential results are not reported in the registry record.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 6 outcome measures and names 5 secondary outcomes. When a study evaluates multiple outcomes, multiplicity can arise because repeated statistical tests increase the chance of obtaining an apparently notable finding by chance. In general, a statistical plan may specify a primary outcome, testing hierarchy, adjustment procedure, or descriptive interpretation for secondary measures. These details are not reported in the registry record. There are no posted analyses, so the record does not show whether any multiplicity procedure was used.
Missing outcome values are central to interpreting this record. The primary measure has no reported value for either arm, and the registry indicates that values are missing or not estimable for a binary two-arm comparison. The record does not explain whether this reflects missing participant assessments, insufficient information for a calculation, or another reason. It also does not report an imputation method, sensitivity analysis, or approach to participants without an evaluable response assessment. No assumption about those participants' outcomes is justified.
The record does not identify the analysis population for the primary or secondary outcomes. In general, an analysis population defines which participants contribute to a particular estimate; different definitions can affect the result. The number accrued is reported as 4, but the registry does not say which participants were included in each outcome analysis, whether all accrued participants were evaluable, or how participants were handled after early closure.
Consequently, neither multiplicity control nor missing-data handling can be assessed from the posted record. The absence of such information is distinct from evidence that a particular method was or was not used; the method is not reported in the registry record.
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 |
|---|---|---|---|
| Nivolumab | 0 | 2 | 0.0% |
| Nivolumab + Ipilimumab | 0 | 2 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Nivolumab | 1 of 2 |
| Nivolumab + Ipilimumab | 0 of 2 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Nivolumab | 2 of 2 |
| Nivolumab + Ipilimumab | 2 of 2 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Nivolumab | Nivolumab + Ipilimumab |
|---|---|---|---|
| Constipation | Gastrointestinal disorders | 1 of 2 | 1 of 2 |
| Diarrhea | Gastrointestinal disorders | 1 of 2 | 0 of 2 |
| Abdominal pain | Gastrointestinal disorders | 0 of 2 | 1 of 2 |
The registry reports adverse-event counts separately for the nivolumab and nivolumab plus ipilimumab arms. For serious adverse events, 2 were at risk in each arm and 0 were affected in each arm. For other adverse events, 2 were at risk in each arm and 2 were affected in each arm. The recorded death count was 1 in the nivolumab arm and 0 in the combination arm.
These are descriptive counts from very small groups, not precise estimates of comparative risk. The registry does not provide event-level details for the serious adverse events, and the list of top serious terms is empty. It also does not report whether deaths or other adverse events were judged related to either treatment, the timing of events, or the duration of safety follow-up. A recorded death should not be assumed to have been caused by treatment. Likewise, the absence of a recorded serious adverse event in an arm does not establish that serious risk is absent. The counts describe what the registry posted, with substantial limits on interpretation.
12. What the Result Does, and Does Not, Mean
The posted record establishes that NIMBLe was a randomized, open-label phase II study of nivolumab alone or with ipilimumab in advanced solid tumors selected for POLE or POLD1 mutations detected through a blood-based assay. It also documents that accrual closed prematurely after screening found the relevant mutations less often than expected, with 4 patients accrued. These registry entries describe the study's conduct and the reason given for early closure; they do not constitute an efficacy finding.
The primary outcome is not reported as an estimable result. No response count, response rate, treatment-effect estimate, confidence interval, or p-value is available in the registry record. It is therefore not possible to conclude from these data that either regimen had a particular response profile, or that the regimens differed. Non-estimability is not the same as a finding of no effect, no response, or equivalence.
The adverse-event data show recorded counts, including 0 serious adverse events in each arm and a death count of 1 in the nivolumab arm and 0 in the combination arm. These observations should remain descriptive. The record does not explain attribution, follow-up duration, or the circumstances of events, and the groups are too limited in size to support a reliable comparison. No treatment benefit or harm conclusion follows from these counts alone.
More generally, a randomized study can reduce some sources of bias in assignment, but randomization cannot compensate for unavailable outcomes or provide precision when the record has little information. The registry's non-comparative designation is also important: the study was not presented as a formal head-to-head test. The record does not report the statistical plan needed to determine how outcomes were intended to be analyzed.
The appropriate interpretation is narrow: this registry record describes the study design, early closure, and a limited set of safety counts, while leaving the primary efficacy result unavailable. It does not support a claim that either treatment works or does not work, nor a claim that one is better or safer than the other.
13. Why This Trial Matters Statistically
Statistically, NIMBLe illustrates the gap between a well-defined clinical question and the evidence a trial ultimately produces. The study targeted advanced solid tumors with POLE or POLD1 mutations detected in blood and randomized participants between nivolumab alone and a nivolumab–ipilimumab combination. Yet screening found the relevant mutations less often than expected, and accrual was closed early. The registry records only 4 patients accrued and does not contain an estimable primary outcome.
This matters because statistical design depends on the availability of information about the population, endpoint, and outcomes. If the intended population is difficult to identify through screening, accrual can be limited, leaving estimates imprecise or not estimable. The registry does not report the study's sample-size justification or how the low prevalence affected the planned analysis, so those details cannot be assessed. The reported reason for closure is a practical constraint, not evidence about treatment effect.
The record also demonstrates why randomization should not be mistaken for a comparative conclusion. Allocation was randomized, but the official title describes a non-comparative study, and no arm-level primary results are posted. Without an outcome estimate and uncertainty measure, a randomized design does not yield a usable comparison by itself.
Finally, the safety counts show how descriptive observations differ from comparative inference. Counts can be reported even when a study is too limited to characterize risk precisely. Their interpretation depends on event definitions, follow-up, attribution, and the population analyzed, several of which are not reported. The key statistical lesson is to distinguish documented design and event counts from estimates that the record cannot support.
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 most consequential limitation is that the primary endpoint has no reported arm-level values and is identified as missing or not estimable for a binary comparison. No efficacy estimate or precision measure can therefore be interpreted. The record does not explain the specific reason for non-estimability, beyond documenting early accrual closure following screening review.
Accrual was closed prematurely on August 2, 2019, because the prevalence of relevant mutations was lower than expected, and the registry states that 4 patients were accrued. This limits the amount of information available for both efficacy and safety interpretation. The registry does not report a sample-size justification, target enrollment, or any revised analysis plan following early closure. Those details cannot be inferred from the phase designation or the planning assumptions.
The primary outcome description lists CR, PR, SD, or PD, but does not clarify how those categories were used to define the participant count. The record also does not report the analysis population, assessment schedule, handling of participants without evaluable outcomes, missing-data procedures, or sensitivity analyses. A detailed statistical analysis plan is not reported in the registry record.
The study was open-label, with no masking. The registry does not describe any safeguards for assessment or analysis that might address the absence of masking. The official title calls the trial non-comparative, and there is no estimable comparison posted. It would be inappropriate to treat the arm-level safety counts as a reliable head-to-head evaluation.
For safety, event attribution, timing, follow-up duration, and detailed serious adverse-event terms are not reported. For secondary outcomes, the record names the measures but provides no analyses. Together, these limitations mean that the record supports a description of study conduct and posted counts, not a complete assessment of efficacy, comparative safety, or the planned statistical approach.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Nivolumab Ipilimumab in Patients With hyperMutated Cancers Detected in Blood (NIMBLe) — NCT03461952.
- Registry record last updated 2024-02-22; status last verified 2024-02.
- Results first posted 2024-02-22.
- 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
NCT03461952 was a phase II randomized, open-label, parallel, non-comparative study of nivolumab alone or with ipilimumab in advanced solid tumors with blood-detected POLE or POLD1 mutations. The sponsor was the Canadian Cancer Trials Group. The study started on 2019-03-11, and the recorded primary completion date was 2021-06-01; its status is terminated.
The registry says accrual closed prematurely on August 2, 2019, after screening found relevant mutations less often than expected, with 4 patients accrued. The primary endpoint was objective response rate by RECIST 1.1 over 36 months. Its values are not reported or estimable for a binary comparison. The record lists 5 secondary outcomes but no analyses. Posted safety counts include 0 serious adverse events in each arm, other adverse events in 2 participants per arm, and deaths recorded as 1 and 0, respectively. These results are descriptive and do not support an efficacy or comparative treatment conclusion.