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 A Study to Find the Best Dose of BI 1387446 Alone or in Combination With Ezabenlimab (BI 754091) in Patients With Different Types of Advanced or Metastatic Cancer (Solid Tumors) (NCT04147234), a non-randomized phase 1 study. The primary outcome is Maximum Tolerated Dose (MTD) Based on Number of Dose-limiting Toxicities (DLTs).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT04147234 |
| Phase | Phase 1 |
| Condition | Neoplasms |
| Allocation | Non-Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 42 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Boehringer Ingelheim |
| Results posted | Yes |
2. Clinical Question
Population
Adults with advanced or metastatic solid tumors whose previous treatment was not successful, as described in the registry record. The record lists Neoplasms as the condition. Further eligibility details, including any limits on tumor type or prior treatment, are not reported in the registry record.
Intervention
BI 1387446 was injected into a tumor at dose levels of 50 μg, 100 μg, 200 μg, or 400 μg when given alone. Combination arms used BI 1387446 at 50 μg, 100 μg, or 200 μg with ezabenlimab 240 mg given intravenously once every 3 weeks. BI 1387446 was administered intratumorally, with injection guidance depending on whether the tumor was visible or required imaging.
Comparator
There was no randomized control group. The design was non-randomized and parallel, with arms receiving different dose levels or the combination regimen. Comparisons between these arms are not like-for-like randomized comparisons, and differences could reflect factors other than dose or treatment combination.
Primary question
The primary question was which BI 1387446 dose met the registry’s maximum tolerated dose definition in each arm, based on dose-limiting toxicities during the evaluation period. This was a dose-finding question, not a confirmatory comparison of tumor response or survival. The registry record does not report the estimated maximum tolerated dose or the dose-limiting toxicity results.
3. Trial Design
This was a phase I, first-in-human study of BI 1387446 in solid tumors. The registry lists the trial as completed, with enrollment of 42. It records a start date of 2020-08-03 and a primary completion date of 2022-11-15. The design was non-randomized, unmasked, and parallel. Participants were not assigned to treatment by random allocation, and the record does not describe a blinded comparison.
The study examined BI 1387446 alone at 50 μg, 100 μg, 200 μg, and 400 μg, and in combination with ezabenlimab at BI 1387446 doses of 50 μg, 100 μg, and 200 μg. BI 1387446 was injected into tumors, using visual inspection for visible skin tumors or imaging guidance. The injection volume depended on tumor diameter. Ezabenlimab was given into a vein at 240 mg once every 3 weeks in the combination arms. The study description says BI 1387446 was given weekly at the beginning and then every 3 weeks.
The registry describes the study purpose as finding the highest BI 1387446 dose participants could tolerate, alone or with ezabenlimab. Participants could remain in the study for up to 2 years and 8 months if they benefited from treatment and could tolerate it. Regular visits included checks of health and recording unwanted effects. The registry record does not report randomization procedures, a control regimen, or further operational details of dose escalation.
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 participants were screened for eligibility prior to participation in the trial. Participants attended a specialist site which ensured that they (the participants) strictly met all inclusion and none of the exclusion criteria. Participants were not to be allocated to a treatment group if any of the entry criteria were violated.
Only the first 6 of 7 groups are shown.
Overall Study
| Stage | Arm A: BI 1387446 50 μg | Arm A: BI 1387446 100 μg | Arm A: BI 1387446 200 μg | Arm A: BI 1387446 400 μg | Arm B: BI 1387446 50 μg / Ezabenlimab 240 mg | Arm B: BI 1387446 100 μg / Ezabenlimab 240 mg |
|---|---|---|---|---|---|---|
| STARTED | 7 | 8 | 7 | 4 | 6 | 3 |
| COMPLETED | 0 | 0 | 1 | 0 | 0 | 0 |
| NOT COMPLETED | 7 | 8 | 6 | 4 | 6 | 3 |
Reasons for not completing this period:
| Reason | Arm A: BI 1387446 50 μg | Arm A: BI 1387446 100 μg | Arm A: BI 1387446 200 μg | Arm A: BI 1387446 400 μg | Arm B: BI 1387446 50 μg / Ezabenlimab 240 mg | Arm B: BI 1387446 100 μg / Ezabenlimab 240 mg |
|---|---|---|---|---|---|---|
| Clinical deterioration | 0 | 0 | 1 | 0 | 0 | 0 |
| Withdrawal by Subject | 0 | 1 | 1 | 0 | 0 | 0 |
| Clinical disease progression | 1 | 3 | 1 | 1 | 0 | 1 |
| Objective disease progression | 6 | 4 | 3 | 3 | 5 | 2 |
| Adverse Event | 0 | 0 | 0 | 0 | 1 | 0 |
Analysis population for the primary outcome, as defined in the registry: The MTD set includes all treated participants evaluable per the clinical trial protocol. This covers those completing required BI 1387446 (and ezabenlimab in Arm B) doses in Cycle 1 and undergoing the Echo/MUGA scan at Cycle 2 Visit 1. Patients stopping due to dose-limiting toxicities are still evaluable. Crossover participants from Arm A to B are included only if evaluable in Arm A. Replaced participants during MTD evaluation are excluded.
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 patient set includes all patients who were documented to have received at least one dose of any study medication. Cross-over patients are shown in both Arm A (reflecting their initial treatment) and Arm B (reflecting their first treatment post-crossover).
Only the first 6 of 8 groups are shown.
| Characteristic | Unit | Arm A: BI 1387446 50 μg | Arm A: BI 1387446 100 μg | Arm A: BI 1387446 200 μg | Arm A: BI 1387446 400 μg | Arm B: BI 1387446 50 μg / Ezabenlimab 240 mg | Arm B: BI 1387446 100 μg / Ezabenlimab 240 mg |
|---|---|---|---|---|---|---|---|
| Participants analysed | 7 | 8 | 7 | 4 | 7 | 5 | |
| Age, Customized: Arm A | Years | 62.0 (SD 16.6) | 55.3 (SD 10.1) | 65.0 (SD 7.2) | 55.3 (SD 24.4) | — | — |
| Age, Customized: Arm B | Years | — | — | — | — | 48.6 (SD 14.3) | 57.0 (SD 18.6) |
| Sex/Gender, Customized: Arm A- male | Participants | 5 | 4 | 3 | 3 | — | — |
| Sex/Gender, Customized: Arm A- female | Participants | 2 | 4 | 4 | 1 | — | — |
| Sex/Gender, Customized: Arm B- male | Participants | — | — | — | — | 4 | 4 |
| Sex/Gender, Customized: Arm B- female | Participants | — | — | — | — | 3 | 1 |
| Race/Ethnicity, Customized: Arm A- Hispanic or Latino | Participants | 0 | 0 | 0 | 0 | — | — |
| Race/Ethnicity, Customized: Arm A- Not Hispanic or Latino | Participants | 7 | 8 | 7 | 4 | — | — |
| Race/Ethnicity, Customized: Arm B- Hispanic or Latino | Participants | — | — | — | — | 1 | 0 |
| Race/Ethnicity, Customized: Arm B- Not Hispanic or Latino | Participants | — | — | — | — | 6 | 5 |
| Race/Ethnicity, Customized: Arm A - Asian | Participants | 0 | 0 | 1 | 0 | — | — |
| Race/Ethnicity, Customized: Arm A- Native Hawaiian or Other Pacific Islander | Participants | 1 | 0 | 0 | 0 | — | — |
| Race/Ethnicity, Customized: Arm A - White | Participants | 6 | 8 | 6 | 4 | — | — |
| Race/Ethnicity, Customized: Arm B - Asian | Participants | — | — | — | — | 0 | 0 |
| Race/Ethnicity, Customized: Arm B- Native Hawaiian or Other Pacific Islander | Participants | — | — | — | — | 0 | 0 |
| Race/Ethnicity, Customized: Arm B - White | Participants | — | — | — | — | 7 | 5 |
6. Primary Endpoint
The registry lists the primary outcome as Maximum Tolerated Dose (MTD) Based on Number of Dose-limiting Toxicities (DLTs), measured over From first administration of BI 1387446 until to end of treatment cycle 1 (up to 3 weeks)..
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | μg (microgram) |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was the maximum tolerated dose (MTD), based on the number of dose-limiting toxicities (DLTs). It was expressed in μg and assessed separately in each arm. The registry defines the MTD as the highest dose expected to have less than 25% risk that the true DLT rate is at least 33% during the MTD evaluation period.
The endpoint is suited to an early dose-finding study because it links dose selection to observed toxicities rather than treating tumor response as the main objective. Its definition uses a probability about the underlying DLT rate, not simply a rule that a particular observed count automatically determines the MTD. The registry says estimation was based on posterior probabilities of DLT incidence in toxicity categories among evaluable participants. It does not report the model details, prior assumptions, category definitions, or estimated probabilities.
The evaluation period ran from the first administration of any trial medication to the start of the second treatment cycle. More specifically, it ended at the second ezabenlimab administration or the fourth BI 1387446 administration, whichever occurred first; if that dose was not given, it ended 90 days after the last administration. Although the endpoint definition is described, the MTD result and the DLT counts are not reported in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry establishes that the study addressed dose finding for BI 1387446, both as a single agent and with ezabenlimab. It records several dose levels and gives a toxicity-based definition of the MTD. It does not post the estimated MTD, DLT counts, or posterior probabilities that would show how the definition was applied. Therefore, the posted information does not identify which dose, if any, met the MTD criterion in either treatment setting.
The non-randomized, parallel design matters when considering differences among dose groups. Participants were not randomly allocated to a common control comparison, so arm-level differences cannot be interpreted as randomized treatment effects. The registry does not provide enough information about how participants entered dose groups or about baseline characteristics to assess comparability between them. The arm structure is useful for describing the study’s dose-finding framework, but it does not by itself establish that a higher dose caused a change in toxicity or response.
The registry lists secondary outcomes involving objective response under RECIST 1.1 and itRECIST, and changes in lesion size using different versions of CTP. However, it does not report results for those outcomes in the record represented by these registry entries. No response estimates, confidence intervals, or analysis results are listed. The study’s primary dose-finding objective should therefore remain distinct from any claims about antitumor activity.
Safety figures are posted by arm for participants at risk and affected by serious adverse events and other adverse events, along with deaths. These are descriptive counts. They do not establish that a treatment caused an event, and the registry record does not provide event timing, exposure-adjusted rates, or a comparative safety analysis. The serious-event terms and percentages listed in the record also need to be read as reported, without assuming an unstated denominator or relationship to a particular dose group.
In short, the registry supports a description of the trial’s design, dose levels, endpoint definition, and recorded safety counts. It does not support a conclusion about the selected MTD, the magnitude or precision of response, or comparative effectiveness. The absence of posted primary and secondary efficacy results limits what can be inferred from the record.
9. Statistical Methods Explained
The primary endpoint uses a model-based dose-finding concept. Rather than defining the MTD as a dose with a particular observed toxicity count, the registry defines it using the estimated probability that the true DLT rate exceeds a threshold. The stated criterion is less than 25% risk that the true DLT rate is at least 33%. In general, a posterior probability expresses uncertainty about an unknown quantity after combining observed data with assumptions in a statistical model. The registry says posterior probabilities were used, but the specific model and its assumptions are not reported in the registry record.
This distinction matters because observed events are only one source of information about an underlying event rate. In general, a model can represent uncertainty in that rate and support a decision across dose levels. The registry does not say how information was shared across dose levels, how toxicity categories were constructed, or how many DLTs occurred. Without those details and the resulting estimates, the MTD rule can be described but not independently applied to identify a dose.
The MTD evaluation period is tied to treatment administration rather than a single universal calendar duration. The registry defines it as the time from the first trial medication to the start of the second cycle, operationalized as the second ezabenlimab administration or the fourth BI 1387446 administration, whichever occurred first. If the relevant administration did not happen, the period ended 90 days after the last administration. The primary outcome description also characterizes the period as extending from the first BI 1387446 administration until the end of cycle one, up to 3 weeks. The record does not explain how any differences in wording were handled in the analysis.
In general, time-window definitions affect which events can count toward an endpoint. Events outside an evaluation window may not contribute to that endpoint, even if they are recorded elsewhere as adverse events. The registry describes a specific DLT evaluation period, but the full DLT definition and event adjudication procedures are not reported in the registry record. Readers should not assume that every serious adverse event listed in the safety record was also a DLT.
The trial was non-randomized and unmasked. In general, randomization helps balance measured and unmeasured participant characteristics across treatment groups, while masking can reduce some risks of biased assessment or reporting. Neither feature is listed for this study. These design characteristics do not make descriptive safety or dose-finding information unusable, but they limit interpretation of between-arm contrasts. The registry does not report a planned inferential comparison among the arms.
The record lists 42 as enrollment, but it does not provide participant flow by arm or a complete analysis population for the primary endpoint. It describes the MTD set as treated participants evaluable under the protocol, including those completing required cycle-one doses and undergoing an Echo/MUGA scan at the cycle-two visit; the available population description ends before completing its account of participants stopping because of DLTs. The full criteria are not reported in the registry record. This makes it difficult to determine from the record who contributed to each endpoint analysis.
There are no reported primary endpoint estimates or analysis results to interpret for precision. In general, a confidence interval describes the uncertainty around an estimated quantity under a specified method. No confidence interval for the MTD or for a clinical outcome is reported in the registry record. The record also does not report a statistical analysis plan, interim analysis procedures, or a sample-size justification. Any sample-size planning illustration is a planning exercise based on assumptions, not observed power and not evidence that the trial had a particular probability of detecting an effect.
For the secondary outcomes, the registry names response criteria and lesion-size measures, including injected, target, and non-injected lesions. In general, outcome definitions determine what is counted and how an observed change is interpreted. Different response criteria or lesion groupings can answer distinct questions, so their results should not be treated as interchangeable. The record does not report the analysis populations, estimators, missing-data methods, or results for these outcomes. It records no analyses. These omissions prevent a detailed assessment of the statistical methods actually used for the posted outcomes.
More generally, over-dispersion occurs when the variability in observed counts exceeds what a simple probability model expects. Accounting for it can matter when estimating event rates or planning studies, because an overly restrictive model may understate uncertainty. The registry record does not report an over-dispersion assessment or a count model for the primary endpoint; no such method should be inferred from the posterior-probability description alone.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists secondary outcomes for objective response under RECIST 1.1 and itRECIST, plus best percentage changes in injected, target, and non-injected lesion sizes under different CTP versions. It records 8 outcome measures and no analyses. No multiplicity adjustment, hierarchy among outcomes, or prespecified strategy for interpreting multiple results is reported in the registry record. In general, testing multiple outcomes can increase the chance of seeing an apparently notable result by chance; an analysis plan can specify how that issue is handled. The record does not permit assessment of such a plan.
Missing-data handling is also not reported in the registry record. This includes how missing tumor assessments, incomplete follow-up, or discontinuation before the end of an assessment period would be handled for the secondary outcomes. For the primary MTD endpoint, the record describes an evaluable population and a treatment-cycle assessment window, but the population description is incomplete and no accounting of unevaluable participants is posted.
The analysis populations are not fully characterized. The MTD set is described as including treated participants evaluable according to the protocol, with specified treatment and scan requirements. The record also indicates that participants stopping because of DLTs remain relevant to evaluation, but the available wording does not complete that rule. Full analysis-population definitions and participant disposition by endpoint are not reported in the registry record.
These gaps do not establish that the study lacked internal procedures. They mean only that those procedures and the associated analysis details are not reported in the registry record. Without results, analysis populations, and missing-data rules, the listed outcomes cannot be evaluated for completeness or sensitivity to alternative assumptions.
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 |
|---|---|---|---|
| BI 1387446 50 ug | 2 | 7 | 28.6% |
| BI 1387446 100 ug | 4 | 8 | 50.0% |
| BI 1387446 200 ug | 3 | 7 | 42.9% |
| BI 1387446 400 ug | 0 | 4 | 0.0% |
| BI 1387446 50 ug / Ezabenlimab 240 mg | 5 | 7 | 71.4% |
| BI 1387446 100 ug / Ezabenlimab 240 mg | 2 | 5 | 40.0% |
| BI 1387446 200 ug / Ezabenlimab 240 mg | 2 | 4 | 50.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| BI 1387446 50 ug | 1 of 7 |
| BI 1387446 100 ug | 2 of 8 |
| BI 1387446 200 ug | 2 of 7 |
| BI 1387446 400 ug | 1 of 4 |
| BI 1387446 50 ug / Ezabenlimab 240 mg | 2 of 7 |
| BI 1387446 100 ug / Ezabenlimab 240 mg | 1 of 5 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| BI 1387446 50 ug | 6 of 7 |
| BI 1387446 100 ug | 8 of 8 |
| BI 1387446 200 ug | 7 of 7 |
| BI 1387446 400 ug | 4 of 4 |
| BI 1387446 50 ug / Ezabenlimab 240 mg | 7 of 7 |
| BI 1387446 100 ug / Ezabenlimab 240 mg | 5 of 5 |
Most frequent serious adverse event terms
| Event term | Organ system | BI 1387446 50 ug | BI 1387446 100 ug | BI 1387446 200 ug | BI 1387446 400 ug | BI 1387446 50 ug / Ezabenlimab 240 mg | BI 1387446 100 ug / Ezabenlimab 240 mg |
|---|---|---|---|---|---|---|---|
| Malignant neoplasm progression | Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 0 of 7 | 2 of 8 | 2 of 7 | 0 of 4 | 3 of 7 | 1 of 5 |
| Tumour haemorrhage | Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 1 of 7 | 0 of 8 | 1 of 7 | 0 of 4 | 0 of 7 | 0 of 5 |
| Intestinal obstruction | Gastrointestinal disorders | 0 of 7 | 1 of 8 | 0 of 7 | 0 of 4 | 0 of 7 | 0 of 5 |
| Liver disorder | Hepatobiliary disorders | 0 of 7 | 0 of 8 | 1 of 7 | 0 of 4 | 0 of 7 | 0 of 5 |
| Pneumonia | Infections and infestations | 1 of 7 | 0 of 8 | 0 of 7 | 0 of 4 | 0 of 7 | 0 of 5 |
| Injury | Injury, poisoning and procedural complications | 0 of 7 | 0 of 8 | 0 of 7 | 0 of 4 | 1 of 7 | 0 of 5 |
| Hypercalcaemia | Metabolism and nutrition disorders | 0 of 7 | 1 of 8 | 0 of 7 | 0 of 4 | 0 of 7 | 0 of 5 |
| Hypokalaemia | Metabolism and nutrition disorders | 0 of 7 | 1 of 8 | 0 of 7 | 0 of 4 | 0 of 7 | 0 of 5 |
| Hypovolaemia | Metabolism and nutrition disorders | 0 of 7 | 0 of 8 | 0 of 7 | 0 of 4 | 1 of 7 | 0 of 5 |
| Carotid artery stenosis | Nervous system disorders | 0 of 7 | 0 of 8 | 0 of 7 | 0 of 4 | 1 of 7 | 0 of 5 |
Most frequent non-serious adverse event terms
| Event term | Organ system | BI 1387446 50 ug | BI 1387446 100 ug | BI 1387446 200 ug | BI 1387446 400 ug | BI 1387446 50 ug / Ezabenlimab 240 mg | BI 1387446 100 ug / Ezabenlimab 240 mg |
|---|---|---|---|---|---|---|---|
| Fatigue | General disorders | 2 of 7 | 4 of 8 | 1 of 7 | 0 of 4 | 2 of 7 | 1 of 5 |
| Anaemia | Blood and lymphatic system disorders | 0 of 7 | 3 of 8 | 1 of 7 | 1 of 4 | 2 of 7 | 2 of 5 |
| Constipation | Gastrointestinal disorders | 0 of 7 | 2 of 8 | 2 of 7 | 1 of 4 | 1 of 7 | 2 of 5 |
| Asthenia | General disorders | 1 of 7 | 1 of 8 | 0 of 7 | 0 of 4 | 2 of 7 | 2 of 5 |
| Nausea | Gastrointestinal disorders | 0 of 7 | 3 of 8 | 0 of 7 | 0 of 4 | 0 of 7 | 2 of 5 |
| Vomiting | Gastrointestinal disorders | 0 of 7 | 2 of 8 | 1 of 7 | 0 of 4 | 1 of 7 | 1 of 5 |
| Injection site pain | General disorders | 0 of 7 | 2 of 8 | 2 of 7 | 1 of 4 | 0 of 7 | 0 of 5 |
| Aspartate aminotransferase increased | Investigations | 0 of 7 | 1 of 8 | 2 of 7 | 1 of 4 | 1 of 7 | 0 of 5 |
| Arthralgia | Musculoskeletal and connective tissue disorders | 1 of 7 | 2 of 8 | 0 of 7 | 0 of 4 | 1 of 7 | 1 of 5 |
| Pyrexia | General disorders | 0 of 7 | 2 of 8 | 2 of 7 | 0 of 4 | 0 of 7 | 0 of 5 |
Serious events with the highest arm-level rate: malignant neoplasm progression (42.9%), splenic vein thrombosis (25.0%), pneumonia (25.0%), sepsis (25.0%), respiratory failure (25.0%).
The registry lists serious adverse events and other adverse events by treatment group, with counts of participants at risk and affected, as well as deaths. For BI 1387446 50 μg alone, serious adverse events affected 2 of 7 participants at risk; other adverse events affected 6 of 7, and 1 death was listed. For 100 μg alone, the corresponding counts were 4 of 8 for serious events and 8 of 8 for other events, with 2 deaths. For 200 μg alone, they were 3 of 7 and 7 of 7, with 2 deaths. For 400 μg alone, they were 0 of 4 and 4 of 4, with 1 death.
In the combination groups, BI 1387446 50 μg with ezabenlimab 240 mg had 5 of 7 participants affected by serious events and 7 of 7 by other events, with 2 deaths. The 100 μg combination group had 2 of 5 and 5 of 5, with 1 death. The 200 μg combination group had 2 of 4 and 4 of 4, with 1 death. The registry also lists malignant neoplasm progression at 42.9%, and splenic vein thrombosis, pneumonia, sepsis, respiratory failure, and deep vein thrombosis at 25.0% each among its top serious-event terms.
These figures are descriptive, not estimates of comparative risk. The record does not establish that events or deaths were caused by study treatment, and the term percentages should not be assigned to a particular arm or denominator without further information.
12. What the Result Does, and Does Not, Mean
This trial’s statistical purpose was to guide dose selection for BI 1387446, both alone and in combination with ezabenlimab, using dose-limiting toxicities during a defined evaluation period. That objective is different from testing whether a treatment improves outcomes relative to a control. The registry lists a specific MTD decision rule, but does not report the estimated MTD, the DLT counts, or the posterior probabilities used to reach a dose decision.
The record also does not report results for its secondary tumor-response and lesion-size outcomes. Consequently, it cannot establish the size or precision of any observed response, nor whether response differed across arms. The listed safety counts describe events recorded in each group, but do not by themselves establish treatment causation or a reliable comparison between regimens.
In general, a dose-finding result can inform which dose is selected for further study, but it does not by itself establish clinical benefit. Likewise, an observed event count in a small group is not a complete characterization of a treatment’s safety profile. Interpretation depends on the endpoint definitions, analysis population, follow-up, and uncertainty, several of which are not fully reported in the registry record.
The non-randomized design further limits causal comparisons between groups. Different arms may not be comparable, and the registry does not report baseline characteristics or a control arm to support adjusted comparisons. The posted information therefore supports describing the study design, endpoint criteria, and adverse-event counts, but not drawing conclusions about comparative efficacy or establishing a treatment effect.
The key distinction is between the question the trial was designed to address and what the registry record reports as results. It describes a first-in-human dose-finding study and its MTD framework, yet does not report the primary endpoint result. The record should therefore be read as an account of the study structure and available safety summaries, not as a complete quantitative report of dose selection or antitumor outcomes.
13. Why This Trial Matters Statistically
Early dose-finding studies require careful separation of dose selection from later questions about clinical benefit. This trial’s primary endpoint makes that distinction explicit: the MTD was defined through a DLT-rate criterion and posterior probability, rather than through a tumor-response target. That framing helps readers recognize what the primary analysis was intended to answer.
The study also illustrates why endpoint definitions and evaluation windows matter. The DLT assessment was tied to treatment-cycle timing, including specified administrations and a fallback period when a scheduled dose was not given. Such rules determine which toxicity observations inform dose selection. Without the reported DLT results and full analysis details, however, readers cannot judge how the rule affected the dose decision.
The non-randomized design is also important statistically. Dose groups in a dose-finding study are not necessarily designed as equivalent comparison groups. Their descriptive results should not be read as if they arose from a randomized head-to-head trial. The same caution applies to the combination arms, which differ in regimen as well as in the presence of ezabenlimab.
Finally, the record underscores the value of complete reporting. A clearly defined endpoint is useful, but interpretation also depends on the estimate, its uncertainty, participant flow, analysis populations, and methods for incomplete observations. Those items are not fully reported in the registry record. The distinction between a prespecified statistical framework and posted outcome evidence is central to sound reading of an early-phase trial.
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 limitation is that the registry record does not report the primary endpoint result. It provides the MTD definition and describes a posterior-probability approach, but no selected dose, DLT count, or probability estimate is available. The record also does not provide results for the listed response and lesion-size outcomes, so efficacy cannot be assessed from these registry entries.
The study was non-randomized and unmasked. There was no like-for-like randomized control comparison, and the registry does not provide baseline characteristics that would help assess differences among arms. Consequently, observed differences among dose groups, if considered, would not establish causal effects of dose or combination therapy.
Some analysis details are incomplete. The MTD-set description is truncated, and the registry does not report full participant flow, analysis populations for each endpoint, missing-data methods, multiplicity procedures, or an analysis plan. It does not report whether or how interim analyses or sample-size justification were handled. The listed enrollment does not resolve the number evaluable for each endpoint or group.
The safety record gives participant counts by group and selected serious-event terms, but does not report exposure-adjusted rates, event timing, or a comparative analysis. The record also does not establish that listed adverse events or deaths were caused by treatment. These limitations are specific to the information available for this trial and restrict interpretation of both the dose-finding objective and the descriptive safety data.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: A Study to Find the Best Dose of BI 1387446 Alone or in Combination With Ezabenlimab (BI 754091) in Patients With Different Types of Advanced or Metastatic Cancer (Solid Tumors) — NCT04147234.
- Registry record last updated 2025-08-19; status last verified 2025-08.
- Results first posted 2025-06-12.
- 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
NCT04147234 is a completed phase I, first-in-human, non-randomized, unmasked, parallel study of BI 1387446 alone and with ezabenlimab in advanced or metastatic solid tumors. The registry lists enrollment of 42, a start date of 2020-08-03, and primary completion on 2022-11-15. BI 1387446 was administered intratumorally across several dose levels, with ezabenlimab given intravenously in the combination arms.
The primary endpoint was the maximum tolerated dose based on dose-limiting toxicities, using a posterior-probability criterion during the MTD evaluation period. The registry record does not report the MTD result or secondary outcome results. It does list adverse-event counts by arm and selected serious-event terms. The absence of a randomized control comparison, incomplete analysis reporting, and missing endpoint results constrain interpretation.