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 of Bemcentinib for the Treatment of COVID-19 in Hospitalised Patients (NCT04890509), a randomized phase 2 study. The primary outcome is Time to Sustained Clinical Improvement of at Least 2 Points.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT04890509 |
| Phase | Phase 2 |
| Condition | COVID-19 |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 115 |
| Arms in the primary outcome | 2 |
| Lead sponsor | BerGenBio ASA |
| Results posted | Yes |
2. Clinical Question
Population
Participants hospitalized with COVID-19 who were randomized and met the protocol’s inclusion and exclusion criteria. The primary outcome analysis included participants evaluable for that outcome.
Intervention
Bemcentinib added to standard of care. Bemcentinib was to be administered for up to 15 days or until hospital discharge, whichever came sooner. Standard of care followed local guidelines.
Comparator
Standard of care alone, administered according to local guidelines in place during the study.
Primary question
How long did it take participants to reach sustained clinical improvement of at least 2 points on the ordinal scale, live discharge, or fitness for discharge, whichever occurred first by Day 29?
The trial’s primary objective was to evaluate bemcentinib as an add-on to standard of care in hospitalized participants with COVID-19. Its primary measure was time from randomization to a qualifying improvement or discharge-related event. The result is reported as a median for each treatment group, with an 80% confidence interval.
This outcome combines clinical improvement on an ordinal scale with live discharge or fitness for discharge. It therefore captures the timing of the first qualifying event under the endpoint definition, rather than reporting separate times for improvement and discharge. The registry record does not report a separate comparative estimate for the primary outcome.
3. Trial Design
NCT04890509 is a completed, multicentre, randomized, phase 2 study sponsored by BerGenBio ASA. It used a parallel-group design with no masking. The registry lists an enrollment of 115 and records the study start as 2020-10-20 and primary completion as 2021-05-25.
Participants were assigned to bemcentinib plus standard of care or standard of care alone. In the combination arm, bemcentinib was planned for up to 15 days or until hospital discharge, whichever came sooner. Standard of care was administered according to local guidelines; those guidelines could reflect local practice at the time of treatment. The registry record does not give further detail about the randomization procedure, allocation concealment, treatment adherence, or how standard-of-care choices varied among participants.
The primary endpoint was assessed from randomization through Day 29. The registry defines the primary analysis population as the intention-to-treat analysis set of randomized participants who met protocol inclusion and exclusion criteria, while also specifying that the reported N for this outcome means participants evaluable for it. The posted primary outcome counts are 54 for bemcentinib plus standard of care and 53 for standard of care alone.
Twenty secondary outcomes are posted, covering clinical status, oxygen use, oxygen-free days, viral load, ventilation, and other measures. The registry record lists no separate analysis records. It does not report the study’s planned sample-size justification, detailed statistical analysis plan, or whether interim analyses were planned or conducted.
4. Participant Flow and Accounting
The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.
Overall Study
| Stage | Bemcentinib + Standard of Care (SoC) | Standard of Care |
|---|---|---|
| STARTED | 58 | 57 |
| COMPLETED | 54 | 53 |
| NOT COMPLETED | 4 | 4 |
Reasons for not completing this period:
| Reason | Bemcentinib + Standard of Care (SoC) | Standard of Care |
|---|---|---|
| Death | 4 | 4 |
Analysis population for the primary outcome, as defined in the registry: Intention to treat (ITT) analysis set included all participants who were randomized and matched the inclusion/exclusion criteria of the protocol. Here, 'N' (overall number of participants analyzed) is defined as participants who were evaluable for this outcome measure.
5. Baseline Characteristics
The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.
| Characteristic | Unit | Bemcentinib + Standard of Care (SoC) | Standard of Care | Total |
|---|---|---|---|---|
| Participants analysed | 58 | 57 | 115 | |
| Age, Continuous | years | 54.4 (SD 12.97) | 51.5 (SD 15.48) | 53.0 (SD 14.28) |
| Sex: Female, Male: Female | Participants | 17 | 22 | 39 |
| Sex: Female, Male: Male | Participants | 41 | 35 | 76 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 0 | 0 | 0 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 57 | 56 | 113 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 1 | 1 | 2 |
6. Primary Endpoint
The registry lists the primary outcome as Time to Sustained Clinical Improvement of at Least 2 Points, measured over From randomization up to Day 29.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | 80% Confidence Interval |
| Unit | days |
| Endpoint classification | Time-to-Event Analysis |
The primary endpoint measures the time from randomization to sustained clinical improvement of at least 2 points on a 9-point category ordinal scale, live discharge, or fitness for discharge, whichever occurred first by Day 29. Sustained improvement is defined as improvement without subsequent worsening. The scale ranges from 0, uninfected with no clinical or virological evidence of infection, through progressively more severe clinical states, to 8, death.
Time-to-event outcomes are useful when both whether an event occurs and when it occurs matter. Reporting a median describes the time by which half of the evaluable participants have reached the endpoint, under the endpoint’s analysis conventions. It does not by itself describe the full pattern of event times, or establish how the groups compare over the full follow-up period.
The endpoint’s discharge criteria are included alongside scale-based improvement. As a result, the measure represents a combined clinical milestone rather than a measure of any one component in isolation. The registry record does not report separate primary results for scale improvement, live discharge, and fitness for discharge, nor does it provide a hazard ratio or another comparative estimate for this endpoint.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The median time to sustained clinical improvement was 7 days in both groups. The 80% confidence interval was 6 to 8 days for bemcentinib plus standard of care and 6 to 9 days for standard of care alone. These are group-specific summaries; the registry record does not report a comparative estimate or a hypothesis-test result for the primary endpoint. The medians alone cannot establish that the groups are equivalent, or show whether a difference in the timing of events exists elsewhere in the distributions.
An 80% confidence interval describes uncertainty around an estimate under the analysis method used. It is not a range containing a fixed proportion of individual participants’ event times. The intervals overlap, but overlap alone is not a formal test of the difference between groups. The confidence level is also less than the commonly used 95% level, so the interval should be read at the level actually reported rather than as a 95% confidence interval.
The primary outcome analysis counts participants evaluable for the endpoint: 54 in the bemcentinib-plus-standard-of-care group and 53 in the standard-of-care group. The registry’s definition of the intended analysis population refers to randomized participants who met protocol criteria, but it does not explain why other enrolled or randomized participants were not evaluable for this measure, or how any such cases were handled.
Some secondary outcomes have unadjusted recalculations from posted summary statistics. For duration of oxygen use, the mean was 53.33% of days with bemcentinib plus standard of care and 56.38% with standard of care. The recalculated mean difference, expressed as standard of care minus bemcentinib plus standard of care, was 3.05 percentage points, with an interval from -8.138 to 14.24 and an unadjusted p value of 0.593. For oxygen-free days, the corresponding means were 46.67% and 43.62%; the recalculated difference was -3.05 percentage points, with an interval from -14.24 to 8.138 and an unadjusted p value of 0.593. Both intervals include zero, and the unadjusted results do not establish a clear difference.
For new ventilation use, the posted counts were 1 of 58 participants in the bemcentinib-plus-standard-of-care group and 6 of 57 in the standard-of-care group. The recalculated risk difference, defined as the standard-of-care risk minus the bemcentinib-plus-standard-of-care risk, was 0.088, with an interval from 0.002 to 0.174 and an unadjusted p value of 0.048. This is a secondary, unadjusted comparison; it should not be treated as confirmatory evidence, particularly because the record does not report a multiplicity plan. These secondary calculations do not supply the missing comparative estimate for the primary endpoint.
9. Statistical Methods Explained
What does the median describe?
A median is a measure of the central timing of an outcome. For this trial, each group’s median is the reported time to the endpoint as defined in the registry. The median does not show the distribution of all event times, and identical medians do not imply that every aspect of the groups’ outcomes was identical. A comparative analysis may use information beyond the group-specific medians; no such primary comparative estimate is reported in the registry record.
How should the 80% confidence intervals be read?
A confidence interval expresses uncertainty around an estimated quantity under a statistical procedure. The primary endpoint’s intervals are 80% confidence intervals, and should not be described as 95% intervals. The interval width reflects uncertainty in the group-specific median estimate, not the spread of individual participants’ outcomes. Comparing the two group-specific intervals informally is not a substitute for a direct estimate of the between-group contrast. The registry record does not report a confidence interval for that contrast.
What does “sustained” improvement add?
The endpoint requires improvement without subsequent worsening. This distinguishes the endpoint from a measure that would count a temporary improvement even if deterioration followed. The event is also defined to include live discharge or being fit for discharge, whichever qualifying event occurs first by Day 29. These rules make the endpoint clinically interpretable as a milestone, but they also mean it is a composite of several ways to qualify. The record does not provide component-specific primary results, so the contribution of each route to the reported median cannot be separated.
What do the secondary continuous-outcome recalculations show?
The posted oxygen-use outcomes are means with standard deviations. Their recalculated comparisons are simple, unadjusted comparisons from the posted summary statistics. A mean difference describes the difference in group averages in the stated direction; the interval indicates uncertainty around that estimated difference. For oxygen use, the reported interval runs from -8.138 to 14.24, and for oxygen-free days it runs from -14.24 to 8.138. Both include zero. The unadjusted p value for each is 0.593. These recalculations do not account for possible adjustment variables or other features of the sponsor’s analysis, and the registry record lists no corresponding reported analysis.
How should the binary comparison be interpreted?
New ventilation use is reported as a count of participants with the outcome. A risk difference compares the proportion with the outcome in one group with the proportion in the other. The recalculated estimate is oriented as standard care minus bemcentinib plus standard of care; its positive value corresponds to the larger posted risk in the standard-of-care group. The interval runs from 0.002 to 0.174, and the unadjusted p value is 0.048. This calculation is based on the posted counts and group sizes, is not adjusted for other factors, and is one of several secondary comparisons. It should be interpreted with caution rather than as a stand-alone confirmatory result.
What do the planning illustrations mean?
The sample-size illustrations are based on observed secondary effects, not on the trial’s own design justification. Their stated assumptions are a two-sided significance level of 0.05, equal allocation, and a normal approximation; the assumed effect is a stated fraction of the observed effect. For new ventilation use, planning for 100% of the observed effect gives 116 participants per arm for 80% power and 154 participants per arm for 90% power. Planning for 75% gives 171 participants per arm for 80% power and 229 participants per arm for 90% power. Planning for 50% gives 305 participants per arm for 80% power and 408 participants per arm for 90% power.
For the oxygen-use and oxygen-free-days recalculations, planning for 100% of the observed effect gives 1,581 participants per arm for 80% power and 2,117 participants per arm for 90% power. Planning for 75% gives 2,811 participants per arm for 80% power and 3,763 participants per arm for 90% power. Planning for 50% gives 6,323 participants per arm for 80% power and 8,465 participants per arm for 90% power. These are conditional planning illustrations; they are not observed power, and they do not show what the trial was designed to detect. The registry record does not report the trial’s own sample-size assumptions or justification.
In general, power is the chance that a study’s analysis will detect an effect under specified assumptions. A planning calculation depends on the assumed effect, variability, statistical procedure, and other design choices. If the observed effect used as the planning basis is uncertain, the resulting illustration is uncertain as well. These calculations should therefore be read as hypothetical planning scenarios, not as evidence about the reliability of the trial’s observed results.
10. Multiplicity, Missing Data and Analysis Populations
The registry record reports one primary outcome and 20 secondary outcomes. Multiple outcome comparisons create a multiplicity concern: as more hypotheses are examined, isolated nominal findings become less persuasive unless the analysis plan accounts for that testing. The registry record does not report a multiplicity adjustment, outcome hierarchy, gatekeeping procedure, or other strategy for controlling the interpretation of multiple comparisons. The unadjusted secondary p values should therefore be treated as exploratory rather than confirmatory.
Missing-data handling is not reported in the registry record. The primary outcome definition specifies an intention-to-treat analysis set of randomized participants who met the protocol’s inclusion and exclusion criteria, and clarifies that the displayed N means participants evaluable for the outcome. The record does not explain the difference between that population definition and outcome evaluability, describe missing observations, or state how incomplete follow-up or censoring was handled. No assumption about those procedures can be made from the posted information.
The registry record also does not report whether interim analyses were planned or conducted, whether the primary analysis was adjusted for baseline characteristics, or the full statistical analysis plan. It lists no separate analysis records. These omissions limit assessment of prespecification and the relationship between the posted summary statistics and any sponsor analysis.
For these reasons, the reported group summaries can be described directly, but the degree to which the analyses addressed missingness, repeated testing, and other design features cannot be assessed from the registry record. The sample-size illustrations are separate planning scenarios based on observed secondary effects; they do not supply the missing design justification or establish the analysis plan used for this trial.
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 |
|---|---|---|---|
| Bemcentinib + Standard of Care (SoC) | 6 | 58 | 10.3% |
| Standard of Care | 4 | 57 | 7.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Bemcentinib + Standard of Care (SoC) | 4 of 58 |
| Standard of Care | 4 of 57 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Bemcentinib + Standard of Care (SoC) | 18 of 58 |
| Standard of Care | 22 of 57 |
Most frequent serious adverse event terms
| Event term | Organ system | Bemcentinib + Standard of Care (SoC) | Standard of Care |
|---|---|---|---|
| Pneumonia | Infections and infestations | 2 of 58 | 0 of 57 |
| Sepsis | Infections and infestations | 1 of 58 | 1 of 57 |
| Septic shock | Infections and infestations | 0 of 58 | 2 of 57 |
| Cardiac arrest | Cardiac disorders | 1 of 58 | 0 of 57 |
| Myocardial infarction | Cardiac disorders | 1 of 58 | 0 of 57 |
| Death | General disorders | 1 of 58 | 0 of 57 |
| Multiple organ dysfunction syndrome | General disorders | 1 of 58 | 0 of 57 |
| Aspergillus infection | Infections and infestations | 1 of 58 | 0 of 57 |
| Candida infection | Infections and infestations | 1 of 58 | 0 of 57 |
| Acute respiratory distress syndrome | Respiratory, thoracic and mediastinal disorders | 0 of 58 | 1 of 57 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Bemcentinib + Standard of Care (SoC) | Standard of Care |
|---|---|---|---|
| Diarrhoea | Gastrointestinal disorders | 4 of 58 | 2 of 57 |
| Hyperglycaemia | Metabolism and nutrition disorders | 2 of 58 | 3 of 57 |
| Constipation | Gastrointestinal disorders | 3 of 58 | 1 of 57 |
| Hypertension | Vascular disorders | 1 of 58 | 3 of 57 |
| Anaemia | Blood and lymphatic system disorders | 2 of 58 | 1 of 57 |
| Thrombocytosis | Blood and lymphatic system disorders | 3 of 58 | 0 of 57 |
| Hypercoagulation | Blood and lymphatic system disorders | 0 of 58 | 2 of 57 |
| Leukocytosis | Blood and lymphatic system disorders | 1 of 58 | 1 of 57 |
| Tachycardia | Cardiac disorders | 1 of 58 | 1 of 57 |
| Pneumonia bacterial | Infections and infestations | 0 of 58 | 2 of 57 |
Serious events with the highest arm-level rate: septic shock (3.5%), pneumonia (3.4%), sepsis (1.8%), acute respiratory distress syndrome (1.8%), cardiac arrest (1.7%).
Serious adverse events were reported for 6 of 58 participants at risk in the bemcentinib-plus-standard-of-care group and 4 of 57 at risk in the standard-of-care group. Other adverse events were reported for 18 of 58 and 22 of 57, respectively. Deaths were reported for 4 participants in each group. These are descriptive counts; the registry record does not provide a comparative safety analysis or establish the cause of each death.
The registry lists septic shock at 3.5%, pneumonia at 3.4%, sepsis at 1.8%, acute respiratory distress syndrome at 1.8%, cardiac arrest at 1.7%, and myocardial infarction at 1.7% among the top serious adverse event terms. The record does not provide further detail about event timing, severity grading, relatedness to treatment, or whether participants could experience more than one event. The figures should therefore be read as reported safety summaries, not as adjusted comparisons between treatment groups.
12. What the Result Does, and Does Not, Mean
The primary result describes the observed median time to the defined sustained-improvement or discharge-related endpoint in each group. Both medians were 7 days, with 80% confidence intervals of 6 to 8 days for bemcentinib plus standard of care and 6 to 9 days for standard of care alone. This is a descriptive summary of the evaluable participants and does not establish that the treatments are equivalent or that no difference exists.
The registry record does not report a primary between-group effect estimate, confidence interval for the contrast, or p value. Consequently, the displayed medians and their group-specific intervals do not provide a complete inferential comparison. The endpoint also combines ordinal-scale improvement with live discharge and fitness for discharge, so it does not isolate any one of those components.
The unadjusted secondary recalculations provide additional descriptive information but do not replace the missing primary comparison. Their interpretation is limited by the number and variety of secondary outcomes and the absence of a reported multiplicity strategy. A nominal result for one secondary measure should not be taken as conclusive evidence, especially when the registry record does not state how that outcome was prioritized in the analysis plan.
The trial’s open-label design, locally determined standard of care, and unreported details about missing data and analysis methods also limit how fully the posted results can be evaluated. These limitations do not erase the reported observations; they define what can and cannot be concluded from them. The registry record supports describing the group summaries, but does not support a definitive comparative claim about the primary endpoint.
13. Why This Trial Matters Statistically
This trial illustrates why the endpoint and estimand matter as much as the numerical summary. The primary outcome is a time-to-event measure that combines sustained ordinal-scale improvement with discharge-related criteria. A median gives a concise description of timing, but by itself does not show the full distribution or quantify a treatment contrast. The record’s group-specific medians can therefore be reported without overstating what they establish.
It also demonstrates the importance of distinguishing a group summary from a comparative analysis. The posted primary results include medians and 80% confidence intervals, but no comparative estimate. Secondary recalculations can describe selected contrasts, yet they are not substitutes for the prespecified primary analysis. Their unadjusted status and multiplicity context matter when deciding how much inferential weight to place on an isolated finding.
For clinical researchers, the record highlights the value of transparent reporting of analysis populations, missing-data handling, confidence levels, multiplicity procedures, and sample-size rationale. The registry defines the intended primary analysis population while also reporting outcome-evaluable counts, but does not explain the handling of participants who are not evaluable. It also provides planning illustrations based on observed secondary effects without reporting the study’s own sample-size justification. Each distinction affects the strength and scope of statistical interpretation.
These are general lessons in reading trial registries: clear endpoint definitions and group summaries are useful, but comparative conclusions require the relevant analysis details and uncertainty estimates.
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 study was open label, with no masking recorded. Awareness of treatment assignment can affect care decisions and the assessment or recording of outcomes, although the registry record does not describe whether or how such effects occurred in this trial. Standard of care followed local guidelines, which means the comparator was not described as a single uniform regimen. The record does not detail the specific care participants received.
The primary endpoint combines sustained improvement on an ordinal scale with live discharge or fitness for discharge. This captures a broad milestone but does not show which component drove the group medians. The registry record reports group-specific medians and 80% confidence intervals, but no comparative primary estimate, hazard ratio, or p value. It also does not report the method used to account for censoring or incomplete follow-up.
The primary outcome Ns are described as participants evaluable for the measure, while the intended analysis population is defined in terms of randomized participants meeting protocol criteria. The registry record does not explain exclusions from the evaluable set, missing-data handling, or whether analyses followed the intended population definition in practice.
Other limitations include the absence of a reported multiplicity plan despite 20 posted secondary outcomes, no separate analysis records, and no reported interim-analysis information or full statistical analysis plan. The sample-size illustrations use observed secondary effects and are not the trial’s design justification. These gaps constrain interpretation of both the primary result and the secondary comparisons.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: A Study of Bemcentinib for the Treatment of COVID-19 in Hospitalised Patients — NCT04890509.
- Registry record last updated 2024-10-16; status last verified 2024-10.
- Results first posted 2024-10-16.
- 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
NCT04890509 is a completed, multicentre, randomized, parallel-group, phase 2 study of bemcentinib added to standard of care for hospitalized participants with COVID-19. The comparator was standard of care alone, with local guidelines determining care. The registry lists enrollment of 115, a start date of 2020-10-20, and primary completion on 2021-05-25.
The primary endpoint was time to sustained clinical improvement of at least 2 points on a 9-point ordinal scale, live discharge, or fitness for discharge, whichever occurred first by Day 29. The median was 7 days in both groups. The registry record reports 80% confidence intervals but no comparative primary estimate. It lists 20 secondary outcomes and provides safety summaries. Multiplicity handling, missing-data procedures, and the trial’s sample-size justification are not reported in the registry record.