A Study to Find the Best Dose of BI 905711 in Combination With Chemotherapy and to Test Whether This Dose Helps People With Advanced Gastrointestinal Cancers: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Find the Best Dose of BI 905711 in Combination With Chemotherapy and to Test Whether This Dose Helps People With Advanced Gastrointestinal Cancers (NCT05087992), a randomized phase 1 study. The primary outcome is Number of Patients With Dose Limiting Toxicity (DLT) During MTD Evaluation.

2. Clinical Question

The study asked whether BI 905711 could be combined with chemotherapy, with or without bevacizumab, in people with advanced gastrointestinal cancers. Its first part focused on participants with colorectal cancer receiving BI 905711, FOLFIRI and bevacizumab. The dose-evaluation question was whether dose-limiting toxicities occurred during the defined evaluation period at the tested BI 905711 doses.

The brief description also states that a second part included people with colorectal or pancreatic cancer and tested whether tumours shrank. Some participants in that part also received bevacizumab. The registry record does not report enough posted outcome results to assess tumour response from that part. The primary result concerns dose-limiting toxicity, not tumour shrinkage.

3. Trial Design

The registry describes a phase Ia/Ib, open-label, multicentre study with dose escalation followed by expansion cohorts. Its recorded phase category is Phase 1. Allocation is recorded as randomized, masking as none, and the model as sequential. These labels describe features registered for the study; the record does not explain how randomization was implemented or how sequential dose decisions were governed.

The first part was designed to evaluate tolerability and identify the highest BI 905711 dose participants could tolerate, in combination with FOLFIRI and bevacizumab for colorectal cancer. The primary endpoint captures dose-limiting toxicity during the maximum tolerated dose evaluation period. In the posted primary results, the 0.6 mg/kg BI 905711 group has 6 participants and 0 participants with a dose-limiting toxicity; the 1.2 mg/kg group has 3 participants and 2 participants with a dose-limiting toxicity.

The brief description says the second part included colorectal and pancreatic cancer cohorts and assessed tumour shrinkage, with some participants receiving bevacizumab. However, the available arm descriptions and primary results do not provide a complete account of enrollment or results for those expansion cohorts. The registry record does not report the detailed dose-escalation decision rules, cohort allocation process, or stopping rules.

The study is marked completed. Its recorded enrollment is 13, with a start date of 2021-11-24 and a primary completion date of 2023-10-23. These registry-level details do not establish how many participants contributed to each outcome: the primary endpoint population is specifically defined for dose evaluation, and safety results use their own at-risk counts. Treatment was open-label, so masking of participants or investigators was not used according to the 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.

Before assignment: All subjects were screened for eligibility prior to participation in the trial. Subjects attended a specialist site which ensured that they (the subjects) strictly met all inclusion and none of the exclusion criteria. Subjects were not to be allocated to a treatment group if any of the entry criteria were violated.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Maximum tolerated dose evaluation set (MTDS): This included all patients in the TS who were not replaced for the MTD determination. The MTDS was used for the primary analyses of dose-limiting toxicities (DLTs) and MTD determination.

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): This included all subjects who were dispensed study medication and were documented to have been treated with at least one dose of BI 905711 or chemotherapy. This TS was used for both safety and efficacy analyses.

6. Primary Endpoint

The registry lists the primary outcome as Number of Patients With Dose Limiting Toxicity (DLT) During MTD Evaluation, measured over From cycle 1 Day 1 until the day before cycle 3 Day 1 (2 14-day treatment cycles), or end of the REP (30 days + 5 days) in case of discontinuation before start of cycle 3, up to 35 days..

The primary endpoint is the number of participants with dose-limiting toxicity during maximum tolerated dose evaluation. It is a count of participants experiencing an event, rather than a measure of tumour response, survival, or the total number of toxic events. The specified period begins on cycle 1 Day 1 and continues until the day before cycle 3 Day 1, covering 2 14-day treatment cycles. If treatment was discontinued before cycle 3, the stated follow-up was through the end of the REP, described as 30 days + 5 days, up to 35 days.

This kind of endpoint is used in early dose evaluation because the occurrence of dose-limiting toxicity can inform whether a dose is tolerable enough to consider in subsequent study stages. It is a safety and dose-selection measure, not by itself evidence about whether treatment reduces tumours or improves survival. The registry defines the maximum tolerated dose evaluation set as treated patients in the dose-escalation part who were not replaced for maximum tolerated dose determination. It says this set was used for the primary analyses of dose-limiting toxicity and maximum tolerated dose determination.

The posted counts are 0 of 6 participants in the 0.6 mg/kg group and 2 of 3 in the 1.2 mg/kg group. These small group counts describe what was observed in the stated evaluation set. The registry record does not report a confidence interval, inferential test, or an adjusted estimate for this endpoint, so the precision of any comparison cannot be judged from an interval or model-based result.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a count endpoint represent? The primary outcome is a count of participants with at least one dose-limiting toxicity during a prespecified evaluation window. For this type of outcome, each participant is classified according to whether the event occurred within that window. The reported counts therefore describe participants with an event and the participants included in the endpoint group; they are not counts of the total number of toxicity episodes. The registry record does not provide a more detailed operational definition of dose-limiting toxicity.

Why is the evaluation window important? The endpoint timeframe specifies when an event contributes to the dose-evaluation result. Events outside that stated window need not contribute to this primary endpoint, even if they are recorded elsewhere as adverse events. The record gives a period from cycle 1 Day 1 to the day before cycle 3 Day 1, or, for specified discontinuations, through the end of the REP up to 35 days. Interpreting the count requires keeping that window and the endpoint population in view.

What does a confidence interval add? General statistical explanation: a confidence interval is a range of values compatible with the observed data under a stated statistical procedure. It conveys uncertainty in an estimate, while a single count or estimate does not show that uncertainty on its own. The registry record does not report a confidence interval for the primary dose-limiting toxicity results. Without one, readers cannot use the posted primary results to assess an interval-based range for the event rate or the difference between dose groups.

Why are small groups hard to compare? General statistical explanation: when groups contain few participants, each participant's outcome can have a substantial influence on the observed group result. Estimates may be imprecise, and the data may be compatible with a wide range of underlying event frequencies. This is relevant to the reported groups of 6 and 3 participants. The counts remain useful descriptions of what was observed, but they should not be mistaken for precise estimates of what would occur in a broader population.

What does a sequential design imply? General statistical explanation: in a sequential dose-escalation design, decisions about subsequent dosing can depend on information gathered at earlier stages. Such a design is different from a fixed comparison of treatments in a large confirmatory trial. The registry records a sequential model and describes dose escalation followed by expansion, but does not report the escalation algorithm, decision thresholds, or statistical operating characteristics. Those design details cannot be reconstructed from the endpoint counts.

How should randomization and masking be understood? The registry records randomized allocation and no masking. Randomization can be used to assign participants to groups in a way intended to limit systematic differences, but the record does not state how assignment occurred, whether it applied across all parts of the trial, or how the treatment groups were formed. Open-label status means treatment masking was not used as recorded. The record does not describe any procedures to reduce assessment bias for outcomes.

What do the planning calculations mean? General statistical explanation: sample-size planning uses assumptions about the effect worth detecting, variability or event frequency, allocation, and the desired error rates. The record's stated planning approach uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the target effect is described as a fraction of the observed effect. Those illustrations are not the trial's original design justification and are not observed power. For time-to-event outcomes, the record says planning quantities refer to total events; for other outcomes, they refer to participants per arm. The binary comparison is indicated as missing or not estimable, so there is no usable planning result for that comparison in the record.

Are adjusted or model-based results available? The registry record does not report the statistical model, covariate adjustment, estimation procedure, hypothesis-testing method, or handling of repeated outcomes. It also does not report a p-value or confidence interval for the primary result. The posted endpoint counts can be described directly, but additional inferential claims would require information not reported in the registry record.

Why distinguish outcome types? General statistical explanation: a count of participants with an event is a binary outcome summarized as the number of participants with the event, whereas a time-to-event outcome incorporates the timing of an event and often accounts for participants whose follow-up ends before the event. Tumour size change is a different type of measurement again. The secondary-outcome list includes these different outcome families, so one statistical method would not necessarily be suitable for all of them. The record does not report which methods were used for the secondary outcomes.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("0.6 mg/kg BI 905711 + FOLFIRI + Bevacizumab", "1.2 mg/kg BI 905711 + FOLFIRI + Bevacizumab")
n      <- c(6, 3)
events <- c(0, 2)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  0.6 mg/kg BI 905711 + FOLFIRI + Bevacizumab  6  0  0.0  0.0  45.9  0.0  39.0  0.4  41.0
ARM  1.2 mg/kg BI 905711 + FOLFIRI + Bevacizumab  3  2  66.7  9.4  99.2  20.8  93.9  19.4  93.2

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a range of secondary outcomes, including drug concentration measures, progression-free survival, change in target lesion size, duration of objective response and disease control. Examining multiple outcomes can create a multiplicity issue: general statistical explanation, the more comparisons or hypotheses considered, the greater the chance that at least one result appears notable by chance if each is interpreted separately. Methods for controlling or describing that issue depend on the analysis plan. A multiplicity strategy is not reported in the registry record.

The record lists 16 outcome measures and 12 posted secondary outcomes, but does not provide inferential analyses for the primary endpoint in the registry record. The statistical analysis plan, including any hierarchy of endpoints, prespecified subgroup analyses, or approach to repeated testing, is not reported in the registry record. It is therefore not possible to determine whether any reported secondary results were treated as confirmatory or exploratory.

Missing-data handling is also not reported in the registry record. The primary outcome population is described as treated patients in the dose-escalation part who were not replaced for maximum tolerated dose determination, but the record does not state how missing assessments, incomplete follow-up, or unevaluable participants were handled beyond that population definition. The record likewise does not report an imputation method or sensitivity analysis.

These omissions matter because conclusions can depend on which participants and assessments enter an analysis. The primary endpoint counts should be read as results in the defined maximum tolerated dose evaluation set, not automatically as results for every enrolled participant. The registry record does not report a complete accounting of participants excluded from that set or reasons for exclusion.

12. 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: diarrhoea (33.3%), enterocolitis (33.3%), biliary obstruction (33.3%), drug-induced liver injury (33.3%), leukopenia (11.1%).

The safety record reports serious adverse events, other adverse events and deaths separately, with a group-specific number at risk for each category. For the 0.6 mg/kg BI 905711 plus FOLFIRI and bevacizumab group, 4 of 9 participants at risk had a serious adverse event, 9 of 9 had an other adverse event, and 3 deaths are listed. For the 1.2 mg/kg group, the corresponding figures are 2 of 3, 3 of 3 and 2 deaths. In the FOLFIRI plus bevacizumab group, the figures are 0 of 1, 1 of 1 and 0 deaths.

These are descriptive safety counts. The record does not establish from these figures whether an event or death was caused by study treatment, and deaths should not be assumed to have been treatment-related. The groups also have very different at-risk counts, so their raw counts do not support a simple comparative conclusion. Serious adverse events are not the same measure as dose-limiting toxicity, which was the primary endpoint and had a specified evaluation window.

The listed serious adverse event terms include diarrhoea, enterocolitis, biliary obstruction and drug-induced liver injury, each reported as 33.3%; leukopenia and neutropenia are each reported as 11.1%. These percentages are reproduced as recorded. The registry record does not provide enough detail to interpret them as causal risks or to determine the event-specific denominators.

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

The posted primary result describes dose-limiting toxicity in the maximum tolerated dose evaluation set: 0 participants among 6 in the 0.6 mg/kg group and 2 among 3 in the 1.2 mg/kg group had a reported event during the defined evaluation period. This is evidence about the observed event counts in those groups, under the study's specific regimen and endpoint definition.

It does not establish a precise event rate for a broader population. The groups are small, and no confidence interval or formal comparison is reported for the primary endpoint. Nor does the count alone identify which dose should be used in another study: that would require the dose-selection rules and the full evaluation context, which are not reported in the registry record.

The result also does not answer whether BI 905711 makes tumours shrink, prolongs progression-free survival, or improves any other clinical outcome. Those questions are distinct from short-window dose-limiting toxicity. Although the study description identifies tumour response as an aim of the expansion part and lists relevant secondary outcomes, numerical results for those outcomes are not reported in the registry record for this analysis.

The registry's safety figures offer additional descriptive information, but they should not be interpreted as proof that a treatment caused an adverse event or death. The safety groups have differing at-risk counts, and the record does not provide a causal assessment in the information used. The findings therefore describe observed events rather than establishing treatment benefit or harm in a comparative sense.

Finally, the record's randomized allocation label does not by itself demonstrate that the primary dose groups were directly randomized against one another, or that the FOLFIRI plus bevacizumab group supplies a valid control for the primary endpoint. The record does not describe allocation procedures or report a comparative primary analysis. The evidence should be read as an early-phase, dose-evaluation result, not as a definitive estimate of comparative effectiveness.

14. Why This Trial Matters Statistically

This trial is statistically notable because it combines sequential dose escalation with a primary endpoint focused on dose-limiting toxicity. In early-phase research, the immediate statistical task differs from estimating a treatment effect in a large comparative trial: the priority is to describe tolerability within a defined evaluation window and population. A count endpoint is direct and interpretable, but its meaning depends on the event definition, follow-up period, and rules for deciding who is evaluable.

The reported dose groups are small, making uncertainty central to interpretation. A count such as 0 events among 6 participants or 2 among 3 is informative about those observations, but it does not provide a precise estimate of the frequency that might occur elsewhere. This is why confidence intervals and transparent analysis methods matter, even when a study's main purpose is dose evaluation rather than a confirmatory comparison.

The record also illustrates the importance of keeping endpoints distinct. Dose-limiting toxicity, serious adverse events, tumour response and time-to-event outcomes describe different questions and are not interchangeable. A safety count cannot stand in for tumour response, and a tumour-response outcome cannot establish tolerability during the dose-evaluation period.

Finally, the study's recorded randomized allocation should be interpreted alongside its sequential model, open-label status and missing analysis details. These features shape what can be inferred from the results. Clear reporting of allocation, dose-decision rules, analysis populations, missing-data handling and uncertainty would help readers distinguish descriptive findings from comparative evidence.

15. 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 primary endpoint results concern small groups: the registry reports 6 participants in the 0.6 mg/kg group and 3 in the 1.2 mg/kg group. With counts this limited, the observed results may be imprecise and should not be treated as stable estimates for a wider population. The registry record does not report confidence intervals or a formal test for the primary endpoint.

The result is specific to the maximum tolerated dose evaluation set and its stated timeframe. The record defines that set as treated patients in dose escalation who were not replaced for maximum tolerated dose determination, but does not provide a full accounting of replacements or other exclusions. It also does not report the detailed definition of dose-limiting toxicity or the rules used to make dose-escalation decisions.

Although randomized allocation is recorded, the registry does not describe how randomization was carried out or its relation to dose-escalation and expansion cohorts. The study was open-label, and the record does not report masking of outcome assessment. A primary comparison against the FOLFIRI plus bevacizumab group is not reported.

The available primary results do not answer the study's tumour-response question. The record lists relevant secondary outcomes, but numerical results for those outcomes are not included in the registry record for this analysis. Details of multiplicity control, statistical methods, missing-data handling, and sensitivity analyses are not reported in the registry record.

The study is marked completed, but completion status does not fill these reporting gaps. The posted results support a description of the recorded endpoint and safety counts; they do not establish the magnitude or precision of comparative treatment effects, nor do they support conclusions about efficacy.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study to Find the Best Dose of BI 905711 in Combination With Chemotherapy and to Test Whether This Dose Helps People With Advanced Gastrointestinal Cancers — NCT05087992.
• Registry record last updated 2025-02-19; status last verified 2025-02.
• Results first posted 2025-02-19.
• 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.

19. Record Summary

NCT05087992 is a completed phase Ia/Ib, open-label, multicentre study of BI 905711 with chemotherapy in advanced gastrointestinal cancers. The registry categorizes it as Phase 1, records randomized allocation, no masking and a sequential model, and lists enrollment of 13. The study began on 2021-11-24 and reached primary completion on 2023-10-23.

The primary endpoint was the number of participants with dose-limiting toxicity during maximum tolerated dose evaluation. The posted results are 0 of 6 participants at 0.6 mg/kg and 2 of 3 at 1.2 mg/kg, in combination with FOLFIRI and bevacizumab. The registry also lists safety counts for these groups and a FOLFIRI plus bevacizumab group. Numerical results for the tumour-response and other listed secondary outcomes, and details of inferential methods, are not reported in the registry record.