Open-label, Single Arm Trial of BI 695502 in Patients With Previously Untreated Metastatic Colorectal Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Open-label, Single Arm Trial of BI 695502 in Patients With Previously Untreated Metastatic Colorectal Cancer (NCT02776683), a na phase 3 study. The primary outcome is Percentage of Patients With Treatment-Emergent Adverse Events (TEAEs) in the Specified Categories Selected for Primary Endpoint Assessment.

2. Clinical Question

The central question is descriptive: what safety experience was recorded for patients receiving the study regimen, and what efficacy outcomes were designated for evaluation? Because the study has no comparison group, its results cannot isolate the effect of BI 695502 or distinguish treatment-related outcomes from the underlying course of the disease and other care. Open-label conduct also means treatment was not masked, although the registry does not report how that feature affected assessment.

3. Trial Design

This was an open-label, single-arm phase IIIb trial in patients with previously untreated metastatic colorectal cancer. The registry describes it as multicenter and multinational. Enrollment was recorded as 123. The trial status is completed. No random allocation or masking was used; allocation is listed as not applicable, and the model is listed as a single group.

The treatment objective was to evaluate the safety and tolerability of BI 695502 with mFOLFOX6 and, when applicable, as maintenance therapy. The registry lists BI 695502 and Avastin as drug interventions, but it does not describe a separate comparison arm. The design should therefore be read as a study of outcomes observed in a treated group, not as a head-to-head treatment comparison.

The primary endpoint concerned selected treatment-emergent adverse events. The registry defines the treated set as patients who signed informed consent and received at least one dose of trial medication. The primary safety observation period extended from baseline through the residual effect period after treatment, with a stated maximum overall duration. Efficacy outcomes listed in the record include centrally assessed response and progression measures, as well as overall survival.

The registry record does not report a statistical analysis plan, sample-size justification, interim-analysis plan, or efficacy analyses. It also does not describe detailed treatment schedules, eligibility criteria, or how treatment decisions were made. Those features should not be inferred from the trial title or the stated objectives.

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: All patients were screened for eligibility to participate in the trial. Patients attended specialist sites which would then ensure that they (the patient) met all strictly implemented inclusion/exclusion criteria. Patients were not to be randomised to trial treatment if any one of the specific entry criteria were violated.

Pre-switch Period

Reasons for not completing this period:

Post-switch Period

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Treated set (TS): The TS contained all patients who signed informed consent and who received at least one dose of trial medication.

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.

Population: Treated set (TS): The TS contained all participants who signed informed consent and who received at least one dose of trial medication.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Patients With Treatment-Emergent Adverse Events (TEAEs) in the Specified Categories Selected for Primary Endpoint Assessment, measured over From baseline up to 18 weeks after the last dose of trial medication prior to the Switch Visit. Maximum duration of up to 32 treatment cycles + safety follow up (up to 82 weeks overall)..

The primary endpoint was the percentage of patients with treatment-emergent adverse events in selected categories. The categories included infusion reactions; arterial or venous thromboembolic events; gastrointestinal perforations; hypertension; proteinuria; pulmonary hemorrhage; all hemorrhages; wound-healing complications, abscesses, or fistulas; posterior reversible encephalopathy syndrome; and ovarian failure. The endpoint description says that adverse events with onset between treatment start and the end of the residual effect period were considered.

A percentage endpoint expresses how common at least one event in the specified set was in the defined analysis population during the stated observation period. It can summarize safety concerns across several categories, but it does not by itself describe event severity, duration, recurrence, or causal attribution. The endpoint also differs from a count of all event episodes: the stated measure is the percentage of patients meeting the selected-event definition.

The registry specifies a 95% confidence interval calculated with the Wilson score method. A confidence interval conveys uncertainty around an estimated percentage; it is not a guarantee that the true value lies within the interval. The registry record describes the endpoint and method but does not report the primary endpoint estimate or its confidence interval.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical interpretation starts with the question the design can answer. A single-group study observes outcomes in patients who receive the study regimen. Without a concurrent comparator, it can describe what happened in that group, but it does not estimate a treatment contrast against another assigned group. Differences from an external expectation would require a suitable comparison and assumptions that are not described in the registry record. The study's open-label status means treatment assignment was not masked; the record does not state whether or how outcome assessment was blinded.

What does a percentage endpoint represent?

For a patient-level safety endpoint, the percentage represents the proportion of the defined analysis population who met the event definition during the specified observation period. The endpoint description groups selected adverse-event categories into a safety measure. This makes the result a summary of whether a patient experienced a qualifying event, rather than a measure of the total number of event episodes. A patient with more than one qualifying event may still contribute to a patient-level percentage as an affected patient; the registry description does not provide further counting rules.

The meaning of the percentage depends on its population and time window. The primary endpoint uses the treated set, defined as those who signed informed consent and received at least one dose of trial medication. The observation period starts at baseline and extends through the residual effect period after the last dose, with the record giving a maximum overall duration. A percentage based on this definition is not automatically comparable with a percentage from a study using a different population, event definition, or follow-up period.

Why report a confidence interval?

A confidence interval communicates sampling uncertainty around an estimate. The registry specifies a 95% confidence interval and the Wilson score method for the primary safety endpoint. The Wilson method is a way to form an interval for a proportion; it is useful because proportion estimates can have uncertainty that is not well represented by a simple symmetric interval, particularly near the boundaries of the possible range. This is general statistical context, not a reported result from the trial.

The confidence level describes the behavior of the interval procedure under repeated sampling assumptions. It does not mean that the interval gives the probability that this particular estimate is correct. Since the registry record does not report the primary endpoint percentage or its interval, the precision of that endpoint cannot be assessed from the record.

How should the adverse-event counts be read?

The adverse-event figures list patients at risk, patients affected under serious and other adverse-event measures, and deaths. These are separate descriptive summaries. A count of affected patients is not an event rate per unit of follow-up, and a death count does not identify cause or timing unless those details are separately reported. The record does not provide enough information to infer those details. The serious-event terms are reported with percentages, but the record does not supply a full breakdown of event episodes or the time distribution of events.

Rare or uncommon events can be difficult to estimate precisely in a study of this size, and the observed frequency can vary with case definitions, surveillance, and follow-up. This is a general point about safety data, not a claim about any particular event in this trial. A listed percentage should not be read as a precise estimate of risk without its denominator, follow-up context, and uncertainty interval.

What can the efficacy outcomes establish?

The registry lists progression-free survival and time to progression as centrally assessed outcomes, along with objective response, duration of response, and overall survival. In general, time-to-event outcomes concern the time until a defined event occurs; their analysis depends on definitions of the event, censoring, and follow-up. Censoring occurs when an individual's event time is not observed during the available follow-up, so the analysis uses the information available up to that point. The registry record does not report the analysis definitions, censoring rules, estimates, or uncertainty for these outcomes.

Objective response is listed as the proportion of patients with complete or partial response in the study summary. A response proportion is a different kind of measure from a time-to-event outcome: it does not by itself describe how long a response lasts. Duration of response is listed separately, as are progression-free survival, time to progression, and overall survival. Their distinct definitions matter, and one outcome should not be substituted for another.

What do planning calculations mean?

The registry record describes planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. For time-to-event illustrations, the reported planning quantity is the total number of events; for other illustrations, it is participants per arm. These are planning quantities, not observed trial outcomes and not evidence of observed power. The record does not report that this single-arm trial used those assumptions to plan its enrollment, and the illustrations do not establish that a valid two-arm comparison was conducted.

In general, a sample-size calculation estimates the amount of information needed under stated assumptions. If the assumed effect differs from the effect that would occur in a future study, the resulting planning target may not deliver the intended precision or power. A calculation based on an observed effect can be especially uncertain because the observed estimate itself is subject to sampling variability. The registry record does not report the assumptions or calculations used to justify the trial's enrollment.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary safety endpoint and several efficacy outcomes, including progression-free survival, objective response, overall survival, duration of response, and time to progression. It does not report a statistical analysis plan specifying which analyses were confirmatory, how multiple outcomes were prioritized, or whether any adjustment for multiple testing was planned. The record also does not report interim analyses or stopping rules. No such procedures should be inferred from the list of outcomes alone.

Multiplicity matters when several hypotheses or outcomes are tested: the chance of at least one apparently notable result can increase as the number of tests increases. This is a general statistical concern, not evidence that the trial used or omitted any particular adjustment. Because the registry reports no analyses, there are no reported efficacy tests to interpret for multiplicity in this record.

The record defines the treated set for the primary safety analysis, but it does not describe analysis populations for the listed efficacy outcomes. It does not report missing-data methods, censoring rules, handling of incomplete assessments, or sensitivity analyses. These omissions limit assessment of how the listed outcomes would be calculated and how missing observations might affect them. The registry record also does not provide enough detail to determine how patients who discontinued treatment or lacked follow-up were handled.

Accordingly, the named endpoints communicate planned areas of evaluation, not a complete statistical analysis strategy. Missing information about analysis populations, multiplicity, and missing data should be treated as unreported rather than filled in by assumptions.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pulmonary embolism (4.9%), sepsis (1.6%), febrile neutropenia (1.6%), deep vein thrombosis (1.6%), ileus (1.6%).

The registry lists 123 patients at risk for the adverse-event summaries. It reports 33 affected under the serious adverse-event measure, 118 under the other-adverse-event measure, and 41 deaths. The record does not provide a cause-of-death breakdown or enough detail to interpret deaths as treatment-related. These counts describe what was recorded; they do not establish causation or compare the regimen with another treatment.

The listed serious adverse-event terms include pulmonary embolism at 4.9%; sepsis, febrile neutropenia, deep vein thrombosis, and ileus at 1.6% each; and bacterial sepsis at 0.8%. These percentages should be read as the registry's reported term-level figures. The record does not give event-specific counts or additional precision measures for these terms, so the percentages should not be treated as exact or as a complete safety profile.

The primary safety endpoint is broader than the listed serious-event terms: it covers selected categories of treatment-emergent events and specifies a defined follow-up period. Its estimate and 95% confidence interval are not reported in the registry record. The adverse-event summaries therefore do not replace the missing primary endpoint result.

12. What the Result Does, and Does Not, Mean

The record establishes that this completed study was designed to evaluate safety and tolerability of BI 695502 with mFOLFOX6, including maintenance therapy when applicable, in previously untreated metastatic colorectal cancer. It identifies a single-group, open-label design and describes the selected adverse events that formed the primary safety endpoint. The safety summaries report affected-patient counts and selected serious-event terms, while the record's stated efficacy outcomes show which clinical outcomes were designated for evaluation.

These registry entries do not establish that the regimen caused the recorded events, nor do they show how those events compare with another treatment. A single-arm study has no concurrent control group to support that contrast. The counts also do not, by themselves, describe event severity beyond the serious-event classification, timing, duration, recurrence, or attribution.

The record does not report the primary endpoint percentage or its confidence interval. It also does not report analyses or results for the listed efficacy outcomes. Therefore, no conclusion about the size or precision of the primary safety estimate, or about the observed efficacy outcomes, can be drawn from the registry record. Listing an outcome as an objective is not the same as reporting its result.

For readers, the appropriate interpretation is limited to the study's design, endpoint definitions, and recorded safety summaries. Any broader comparative or causal conclusion would require information not reported in the registry record, including outcome estimates, analysis details, and an appropriate basis for comparison.

13. Why This Trial Matters Statistically

This trial is statistically informative as an example of the distinction between a study objective and a reported result. The record gives a detailed primary safety endpoint definition, a treated-set description, an observation period, and the stated confidence-interval method. Those details clarify what the investigators intended to measure. However, the endpoint estimate and interval are not reported, so the definition alone cannot show the observed frequency or its precision.

The single-group design also illustrates why descriptive safety data and comparative evidence answer different questions. Counts in a treated group can document events observed during treatment and follow-up. Without a concurrent control group, they do not estimate the difference attributable to the regimen. The same limitation applies to efficacy outcomes: an outcome name in the registry does not provide an effect estimate, and the record lists no analyses.

The record's adverse-event summaries provide concrete descriptive information, including counts and selected term-level percentages. They also show why clear denominators, time windows, event definitions, and uncertainty measures matter. The primary endpoint specifies several of these elements, but its numerical result is absent. A reader should not substitute a different safety summary for that missing endpoint result.

More broadly, registry interpretation requires separating what is defined, what is counted, and what is statistically estimated. This record supports statements about the trial's design and recorded safety figures. It does not support a comparative efficacy conclusion, a causal safety claim, or an assessment of the precision of the unreported primary endpoint.

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 main trial-specific limitation is the absence of a control group. The single-arm design can describe outcomes in treated patients but cannot directly estimate how outcomes would differ under another treatment. Open-label conduct is also recorded; the registry does not describe any masking of outcome assessment or explain how knowledge of treatment may have affected assessment.

The registry record does not report the primary endpoint percentage or its confidence interval, despite describing the endpoint and Wilson score method. It also lists efficacy outcomes without reporting their results or analyses. As a result, the record does not permit evaluation of the magnitude or precision of the primary safety estimate, or interpretation of efficacy estimates.

Further analysis details are not reported in the registry record. These include the statistical analysis plan, sample-size justification, interim-analysis procedures, missing-data methods, and analysis populations for efficacy outcomes. The record also does not provide detailed event-level information for the safety terms, such as timing, duration, recurrence, or attribution. The listed death count has no reported cause breakdown in the registry record.

Finally, the stated planning illustration assumptions should not be mistaken for the trial's own design justification or for evidence of observed power. The record does not establish that the trial was planned as a two-arm comparison, and the single-group design does not supply a randomized treatment contrast.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Open-label, Single Arm Trial of BI 695502 in Patients With Previously Untreated Metastatic Colorectal Cancer — NCT02776683.
• Registry record last updated 2019-11-21; status last verified 2019-10.
• Results first posted 2019-11-21.
• 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.

18. Record Summary

The registry describes NCT02776683 as a completed, open-label, single-arm phase IIIb trial in previously untreated metastatic colorectal cancer. Enrollment was 123. The study evaluated BI 695502 with mFOLFOX6 and maintenance therapy when applicable. No randomized comparator or masking is reported.

The primary endpoint was the percentage of treated patients with selected treatment-emergent adverse events, with a stated 95% Wilson score confidence interval. The endpoint estimate and interval are not reported in the registry record. Adverse-event summaries list 123 at risk, 33 affected under the serious-event measure, 118 under the other-event measure, and 41 deaths. Selected serious-event terms are reported with percentages.

The record lists efficacy outcomes including progression-free survival, objective response, overall survival, duration of response, and time to progression, but reports no analyses. The design supports descriptive reading of the recorded safety information, not a comparative treatment-effect conclusion.