Phase I Trial of BI 836845 for Various Solid Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase I Trial of BI 836845 for Various Solid Cancer (NCT01403974), a randomized phase 1 study. The primary outcome is Part 1: Number of Patients With Dose Limiting Toxicities (DLTs) During the Maximum Tolerated Dose (MTD) Evaluation Period.

2. Clinical Question

This is a dose-finding and safety question, not a direct test of BI 836845 against another treatment. The trial description says its purpose was to determine a maximum tolerated dose or, if an MTD was not reached, a relevant biological dose. It also describes safety, pharmacokinetic and pharmacodynamic objectives. The posted primary results are counts of participants with DLTs during the first course, separated by dose cohort. They do not, by themselves, establish which dose was selected as the MTD or RBD. That selection and its supporting rationale are not reported in the registry record.

The registry lists an enrollment of 61, while the primary endpoint population is limited to Part 1 and is defined by treatment exposure and consent. These are not interchangeable populations. The posted results also include Part 2 groups, but those groups are outside the specified primary endpoint population. The clinical question should therefore be read in the context of the dose-escalation cohort counts, rather than as a comparison across all enrolled participants or as an efficacy comparison.

3. Trial Design

The study is registered as a phase I, open-label trial with randomized allocation and a sequential model. Its title and description identify it as a dose-escalation study of BI 836845 in advanced solid cancers, with repeated administration for participants showing clinical benefit. The registry reports the study as completed, with start and primary-completion dates of 2011-07-19 and 2015-07-08. These design labels do not fully explain how allocation operated in practice; that detail is not reported in the registry record.

Part 1 consisted of escalating dose cohorts. Participants received BI 836845 by intravenous infusion on Day 1, 8 and 15 of each 3-week course, continuing until disease progression or undue toxicities according to the arm descriptions. The listed Part 1 cohorts cover doses from 10 mg through 1800 mg. Part 2 included an Ewing's sarcoma family of tumours group and a group with biopsiable solid tumours. Both Part 2 descriptions specify BI 836845 1000 mg, given weekly in 21-day courses. The posted primary endpoint, however, is explicitly restricted to Part 1.

There is no listed placebo or active-control arm. The dose cohorts are not equivalent to a conventional two-arm randomized comparison: each represents a different dose level within an escalation study. The registry's randomized-allocation label should therefore be reported as a design classification, without assuming details of random assignment that are not described. Masking is listed as none, so the study was open label.

The overall enrollment is recorded as 61. Cohort sizes in the Part 1 primary results are small, and the registry presents participant counts by dose. The protocol's escalation rules, stopping boundaries, dose-selection algorithm, sample-size justification and any interim-review procedures are not reported in the registry record. The study also planned safety and drug-level assessments, including pharmacokinetic and pharmacodynamic objectives; specific results for those assessments are not included in the registry record described for the primary endpoint.

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 to participate in trial. Subjects attended specialist sites to ensure that they (the subjects) met all implemented inclusion/exclusion criteria. Subjects were to be entered to trial drug if any of the specific entry criteria was violated.

Only the first 6 of 16 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Treated set (TS) restricted to the part 1 (dose escalation) of the trial: All enrolled patients who gave informed consent and who were documented to have taken at least one dose of investigational treatment in the dose escalation part of the trial.

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: All enrolled patients who gave informed consent and who were documented to have taken at least one dose of investigational treatment.

Only the first 6 of 17 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Part 1: Number of Patients With Dose Limiting Toxicities (DLTs) During the Maximum Tolerated Dose (MTD) Evaluation Period, measured over During the first course of treatment, up to 21 days.

The primary endpoint counts participants with a dose-limiting toxicity during the first treatment course, with a timeframe of up to 21 days. It applies to the Part 1 dose-escalation population. A participant was counted if a drug-related adverse event met the registry's DLT definition. The listed criteria include specified severe neutropenia, febrile neutropenia, thrombocytopenia, liver-enzyme elevation, non-hematologic toxicity, infusion reaction, nausea or vomiting, skin toxicity, hyperglycemia, electrolyte abnormality, delayed recovery, sustained fatigue or asthenia, and other events judged to qualify by the investigator.

In dose-escalation research, a DLT endpoint is used to describe acute toxicity over a defined early treatment period. The count can inform decisions about whether a dose level merits further evaluation, alongside other clinical and biological evidence. It is not a complete measure of all safety concerns: events outside the defined window, events that do not meet the DLT criteria, and later toxicities may not be captured by this endpoint.

The posted results give a DLT count and participant count for each Part 1 dose cohort. They record zero DLTs for the listed cohorts except the 450 mg cohort, which records one; the 800 mg cohort records zero. These are descriptive counts. The registry record does not report a confidence interval, a formal comparison between doses, or a stated MTD or RBD decision with its supporting analysis. The endpoint is therefore best read as an early-course toxicity summary rather than as a complete account of safety or a dose-selection conclusion.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a participant count mean?

The endpoint is a count of participants who met the DLT definition within a specified period. It is a binary outcome at the participant level: each person either met the endpoint definition during the window or did not. The reported count is accompanied by the cohort size, so readers can see the amount of information behind each dose-specific result. This is different from counting every adverse event, because a participant may experience more than one event and the primary endpoint counts participants meeting the DLT criteria.

General statistical explanation: for a binary endpoint, an observed count is often used to estimate the proportion of a population with the event. Such an estimate is uncertain, especially when the cohort is small. A zero observed count does not prove that the underlying event probability is zero; it means that no qualifying event was recorded in that cohort under the endpoint definition and observation window. The registry does not report uncertainty intervals for these counts, so the precision of each dose-specific estimate cannot be quantified from the posted primary results.

Why do dose cohorts need separate interpretation?

The Part 1 structure is sequential dose escalation, rather than a set of concurrent treatment-versus-control groups. Each dose cohort has its own participant count and DLT count. Reading those counts separately preserves the link between dose level and observed toxicity. Combining the cohorts would answer a different question and would conceal the differences in dose and cohort size. A simple visual pattern across doses is not, by itself, a formal dose-response analysis.

General statistical explanation: sequential studies often use accumulating safety information to guide decisions about later cohorts. That makes cohort order and the rules for moving between doses important to interpretation. The registry does not report the escalation algorithm, stopping rules, or a model used to characterize dose-related risk. It is therefore not possible to determine from the posted endpoint counts how the investigators applied the protocol's dose-selection process.

What can be said about comparisons?

The record identifies randomized allocation, but it does not describe a separate placebo or standard-treatment comparator for the primary endpoint. The listed primary results are dose cohorts, and no between-group test, effect estimate or confidence interval is reported. The posted analysis status indicates that a two-arm binary comparison is not available or estimable. Accordingly, the counts should be described without a p-value or comparative risk estimate.

General statistical explanation: a confidence interval expresses the range of values reasonably compatible with the observed data under a statistical model. It communicates precision as well as the direction or scale of an estimate. A p-value addresses compatibility with a specified null hypothesis, not the size or practical importance of an effect. Neither measure is reported for the primary dose-specific counts in the registry record. Their absence should not be replaced by an informal claim that doses are statistically similar or different.

How should the planning illustrations be read?

The sample-size assumptions describe planning illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is expressed as a stated fraction of the observed effect. For non-time-to-event outcomes, the illustrations describe participants per arm for 80% or 90% power. These calculations are not the design justification for this phase I dose-escalation trial, and the registry record does not report a trial-specific sample-size justification based on them.

General statistical explanation: a planning calculation estimates the sample size that would be needed under stated assumptions for a future comparison. It does not describe the probability that the completed study detected an effect, and it is not observed power. The assumptions of equal allocation and a two-arm comparison also do not match the primary endpoint's presentation as multiple sequential dose cohorts without a separate control arm. The illustrations therefore should not be interpreted as an analysis of the dose-escalation design.

Why is the endpoint window important?

The DLT outcome concerns the first course, up to 21 days. This defined window makes the endpoint specific and helps distinguish acute dose-limiting events from broader safety follow-up. It also limits what the outcome can represent. Events occurring later, or events not satisfying the stated DLT criteria, are not necessarily reflected in the primary count. The registry describes broader safety assessment as a study objective, but the primary endpoint is narrower than the complete safety profile.

General statistical explanation: an endpoint definition and observation window determine which events can contribute to a result. Changing either can change the count, even if the participants and treatment remain the same. Readers should interpret the outcome using its stated criteria and period, rather than treating it as a count of all adverse events throughout participation. The registry record does not report a sensitivity analysis using alternative DLT definitions or observation windows.

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("Part 1: BI 836845 10 mg", "Part 1: BI 836845 20 mg", "Part 1: BI 836845 40 mg", "Part 1: BI 836845 60 mg", "Part 1: BI 836845 90 mg", "Part 1: BI 836845 135 mg")
n      <- c(3, 3, 3, 3, 3, 3)
events <- c(0, 0, 0, 0, 0, 0)
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  Part 1: BI 836845 10 mg  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Part 1: BI 836845 20 mg  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Part 1: BI 836845 40 mg  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Part 1: BI 836845 60 mg  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Part 1: BI 836845 90 mg  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Part 1: BI 836845 135 mg  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary DLT endpoint and nine posted secondary outcomes. The secondary outcomes concern tumour response and disease control, progression-free survival for the biopsiable-tumour group, and pharmacokinetic measures. A multiplicity adjustment strategy for testing multiple outcomes is not reported in the registry record. The available primary results are counts by dose cohort, with no reported p-values, so there is no basis for describing statistical significance across multiple dose comparisons. No formal hypothesis-testing plan is reported.

Missing-data methods are not reported in the registry record. The primary endpoint population is described as participants in Part 1 who consented and took at least one dose. The record does not explain how participants with incomplete observation, early discontinuation, or missing DLT assessment were handled. It also does not report an imputation method or sensitivity analysis for missing endpoint data. The population description in the registry is incomplete, so further details about its application cannot be established from that text.

Analysis populations can differ across outcomes. The primary endpoint is restricted to dose escalation, whereas Part 2 contains separate tumour groups, and the progression-free survival outcome is specified for the biopsiable-tumour group. The registry does not provide enough detail to reconstruct all analysis sets, exclusions, or participant flow for each outcome. The overall enrollment of 61 should not be taken as the denominator for the primary DLT endpoint, whose posted results are organized by Part 1 cohort.

These gaps matter because the interpretation of counts depends on who was included, whether endpoint status was known for all included participants, and how the analysis handled incomplete observations. Without those details, the safest reading is to report the posted counts and cohort sizes as recorded, while avoiding assumptions about unreported participants or analysis procedures.

12. Safety Results

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

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: hepatic function abnormal (33.3%), pneumonia (33.3%), urinary tract infection (33.3%), dyspnoea (33.3%), pneumonia aspiration (33.3%).

The serious adverse event summary gives, for each listed group, the number at risk and the number affected. In Part 1, the affected counts vary by dose cohort; the posted data include serious-event counts of 0, 1, 2 or 3 in cohorts with the listed group sizes. In Part 2, the Ewing's sarcoma family of tumours group has no participant recorded as affected, while the biopsiable-tumour group records four affected participants. These are serious adverse event summaries, not DLT counts, and they use a broader safety concept than the primary endpoint.

The registry lists hepatic function abnormal, pneumonia, urinary tract infection, dyspnoea, pneumonia aspiration and febrile neutropenia among the leading serious-event terms, each reported as 33.3%. The record does not give the context needed to interpret those percentages as treatment-specific risks or to determine whether they refer to participants, events, or another reporting unit. No deaths value is posted for the listed groups, so death counts are not reported in the registry record.

Serious adverse events should not automatically be treated as drug-related, and the registry counts do not establish causation. Conversely, a participant can have serious adverse events without meeting the primary DLT definition. The DLT endpoint is limited to specified drug-related events during the first course; the broader serious-event summary has a different scope. The registry record does not provide enough detail to reconcile individual serious events with DLT classifications or to assess their timing and attribution.

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

The primary result is a descriptive account of DLT counts during the first treatment course in the Part 1 dose-escalation cohorts. Most listed cohorts record zero participants with a DLT; the 450 mg cohort records one. The cohort sizes are small, and no uncertainty intervals or formal dose comparison are reported. This supports a factual summary of what was observed in the registry, but not a precise estimate of the underlying DLT risk at each dose.

The finding does not establish that a dose is safe, that the risk is absent at a dose with zero recorded events, or that a dose with a recorded event is less suitable than another dose. A small cohort can miss an event that could occur in a broader population. The result also does not describe the complete safety experience over all treatment courses or all follow-up, because the primary endpoint is restricted to the first course and its specified DLT criteria.

The results do not show whether BI 836845 improves tumour outcomes. Although the registry lists secondary tumour-response and disease-control outcomes, their result values are not reported in the registry record for this analysis. No comparison with a control treatment is listed for the primary endpoint. The registry's randomized-allocation label does not turn dose cohorts into an interpretable treatment-versus-control comparison without details of the allocation and an appropriate comparative analysis.

The trial description states that determining an MTD or RBD was an objective. The posted DLT counts do not identify the final MTD or RBD selection, and the selection rationale is not reported in the registry record. That conclusion would require the study's decision rules and possibly other safety or biological evidence. The result should therefore be understood as a narrow early safety summary, not a final dose recommendation, efficacy conclusion, or complete assessment of benefit and harm.

14. Why This Trial Matters Statistically

This study illustrates why dose-escalation results need to be read differently from conventional randomized treatment comparisons. The primary data are counts in sequential dose cohorts, not an estimated contrast between a treatment and a control. A count can be clinically informative for describing what was observed, while still carrying substantial uncertainty when cohort sizes are small. A zero count is particularly easy to overread as proof of no risk; statistically, it only describes the observations within the cohort and endpoint window.

The registry classification includes randomized allocation, while the posted treatment groups are dose levels in a sequential escalation study. That distinction highlights the importance of understanding the actual allocation structure before applying standard comparative methods. Without a stated comparator, analysis plan or escalation algorithm, readers should avoid assigning an inferential meaning that the registry does not support.

The endpoint also shows how definitions shape safety conclusions. DLT is not synonymous with every adverse event or every serious adverse event. Its specified criteria and first-course window create a focused dose-finding measure, while broader safety summaries address other aspects of tolerability. Keeping those outcomes distinct prevents a count from being interpreted beyond its definition.

Finally, planning calculations and completed-study evidence answer different questions. Illustrations based on an assumed two-arm design are not a substitute for the rationale and operating rules of a phase I escalation trial. This record emphasizes the value of transparent reporting of cohort progression, missing-data handling, dose-selection decisions and uncertainty around observed event counts. Those details are necessary to move from a descriptive safety pattern to a reproducible statistical conclusion.

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 registry record does not report a DLT confidence interval, a dose-trend model, a formal comparison between cohorts, or a statistical test for the primary endpoint. The small cohort sizes limit how precisely the observed counts can characterize risk. Since the study is sequential, cohorts may also reflect different points in the escalation process; the record does not describe the rules used to select or proceed between dose levels.

The primary endpoint is restricted to the first course, up to 21 days, and captures only events meeting its DLT definition. It cannot represent all adverse events, later toxicity, or the full safety experience over repeated courses. Although serious adverse event counts are listed separately, details on timing, attribution and the relation between those events and DLTs are not reported in the registry record. Death counts are also not reported.

The registry lists an overall enrollment of 61, but the primary results concern Part 1 dose cohorts. The record does not give a complete participant flow or fully explain missing endpoint observations and analysis-set exclusions. The description of the primary endpoint population is incomplete. A trial-specific sample-size justification, multiplicity strategy, and detailed statistical analysis plan are not reported.

Finally, the record states that MTD or RBD determination was an objective but does not report the final dose-selection conclusion or its rationale among the registry entries. Secondary outcomes are listed, but their results are not available in the described results. These limitations prevent the DLT counts from supporting broader claims about comparative safety, efficacy, or the preferred dose.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Phase I Trial of BI 836845 for Various Solid Cancer — NCT01403974.
• Registry record last updated 2025-07-25; status last verified 2025-07.
• Results first posted 2025-07-25.
• 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

NCT01403974 is a completed phase I, open-label study of BI 836845 in advanced solid cancers. The registry records randomized allocation and a sequential model. Part 1 escalated intravenous dose cohorts from 10 mg through 1800 mg, with weekly administration on Day 1, 8 and 15 of each 3-week course. Part 2 included Ewing's sarcoma family of tumours and biopsiable-tumour groups treated at 1000 mg weekly.

The primary endpoint counts participants with DLTs during the first course, up to 21 days, in the Part 1 dose-escalation population. The posted counts are zero in the listed cohorts except for one participant in the 450 mg cohort. No formal dose comparison, uncertainty interval, MTD or RBD conclusion, or supporting selection analysis is reported in the registry record. The results should be read as a dose-specific early safety description, not as evidence of comparative efficacy or a complete safety assessment.