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 LY3127804 in Participants With COVID-19 (NCT04342897), a randomized phase 2 study. The primary outcome is Ventilator Free Days.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT04342897 |
| Phase | Phase 2 |
| Condition | COVID-19, Pneumonia |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 95 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Eli Lilly and Company |
| Results posted | Yes |
2. Clinical Question
Population
Participants hospitalized with pneumonia and presumed or confirmed COVID-19 were eligible for the trial. The primary endpoint population was all randomized participants who received at least 1 dose of study drug. The registry record reports enrollment of 95, with 47 participants in the LY3127804 group and 48 in the placebo group for the posted primary outcome. Further eligibility criteria and details about the enrolled participants are not reported in the registry record.
Intervention
Participants assigned to LY3127804 received 20 mg/kg by intravenous infusion on Days 1 and 15. The study was randomized and double-blind. The registry record does not provide additional detail about how the blind was maintained or whether it was assessed.
Comparator
The comparison group received placebo at 20 mg/kg by intravenous infusion on Days 1 and 15. The trial used a parallel-group design, so participants were assigned to one of the two groups rather than receiving both interventions in sequence.
Primary question
The primary question concerned ventilator-free days from Day 1 through Day 28. The registry defines these as days breathing without assistance, provided an unassisted period lasted at least 24 consecutive hours and the participant did not die within 28 days from the first dose. The reported median was 28 days in both groups. The registry does not report a primary-endpoint comparison estimate, confidence interval, p-value, or sponsor-reported hazard ratio.
3. Trial Design
The registry describes a randomized, double-blind, placebo-controlled, parallel-group Phase 2 trial sponsored by Eli Lilly and Company. Its official title identifies participants hospitalized with pneumonia and presumed or confirmed COVID-19. Enrollment was 95. The trial began on 2020-04-20, and primary completion was recorded as 2020-10-12. Its status is terminated; the reason for termination is not reported in the registry record.
Participants were assigned to LY3127804 or placebo. Both were administered by intravenous infusion at 20 mg/kg on Days 1 and 15. The two groups are represented in the posted primary outcome by 47 participants assigned to LY3127804 and 48 assigned to placebo. These outcome denominators should be read as the groups represented in that result, not as evidence about any unreported exclusions or missing outcomes.
The primary endpoint was ventilator-free days from Day 1 through Day 28. The registry defines a qualifying day as one on which a participant breathed without assistance, with the unassisted period lasting at least 24 consecutive hours; a participant also had to remain alive within 28 days from the first dose for days to count under the definition. The registry lists secondary outcomes concerning ordinal-scale scores, complete response, death, hospitalization length, serious adverse events, and any treatment-emergent adverse event. It does not report the detailed protocol, allocation procedure, concealment method, or analysis plan.
The posted primary result gives medians and full ranges rather than a treatment-effect estimate with uncertainty. No primary inferential comparison is reported. Although the registry status is terminated, the reason and any implications for conduct or analysis are not reported in the registry 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.
Overall Study
| Stage | LY3127804 | Placebo |
|---|---|---|
| STARTED | 47 | 48 |
| Received at Least One Dose of Study Drug | 47 | 48 |
| COMPLETED | 33 | 36 |
| NOT COMPLETED | 14 | 12 |
Reasons for not completing this period:
| Reason | LY3127804 | Placebo |
|---|---|---|
| Death | 7 | 4 |
| Lost to Follow-up | 6 | 6 |
| Withdrawal by Subject | 1 | 2 |
Analysis population for the primary outcome, as defined in the registry: All randomized participants who received at least 1 dose of study drug.
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: All randomized participants who received at least 1 dose of study drug.
| Characteristic | Unit | LY3127804 | Placebo | Total |
|---|---|---|---|---|
| Participants analysed | 47 | 48 | 95 | |
| Age, Continuous | Years | 54.3 (SD 14.13) | 55.5 (SD 14.40) | 54.9 (SD 14.20) |
| Sex: Female, Male: Female | Participants | 18 | 17 | 35 |
| Sex: Female, Male: Male | Participants | 29 | 31 | 60 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 23 | 20 | 43 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 24 | 28 | 52 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Asian | Participants | 2 | 2 | 4 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 1 | 1 |
| Race (NIH/OMB): Black or African American | Participants | 8 | 7 | 15 |
| Race (NIH/OMB): White | Participants | 36 | 38 | 74 |
| Race (NIH/OMB): More than one race | Participants | 1 | 0 | 1 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Region of Enrollment: United States | Participants | 47 | 48 | 95 |
| Body Mass Index (BMI) | kilogram per square meter (kg/m^2) | 33.74 (SD 6.963) | 33.70 (SD 7.715) | 33.72 (SD 7.314) |
6. Primary Endpoint
The registry lists the primary outcome as Ventilator Free Days, measured over Day 1 through Day 28.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Full Range |
| Unit | days |
| Endpoint classification | Time-to-Event Analysis |
Ventilator-free days measure time spent breathing without assistance over the specified period, subject to the endpoint’s qualifying rules. In this trial, the period was Day 1 through Day 28. A day counted only when the participant breathed without assistance for at least 24 consecutive hours, and the definition excluded participants who died within 28 days from the first study-drug dose. The measure therefore combines information about respiratory support and survival under a single endpoint definition.
The posted summaries are medians with full ranges. The median describes the middle of the observed distribution, while the full range gives its lowest and highest reported values. For LY3127804, the median was 28 days and the full range was 12 to 28 days. For placebo, the median was 28 days and the full range was 4 to 28 days. Both groups therefore share the same reported median, while the lower ends of their ranges differ.
As a general statistical point, a median can describe a skewed distribution without assuming that outcomes are symmetrically distributed. It does not, by itself, show how much uncertainty there is around a difference between groups. The registry does not report a confidence interval or inferential test for the primary endpoint, so the posted medians and ranges do not establish the precision of a between-group effect. The registry record also does not report a primary-endpoint hazard ratio.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The primary outcome is reported as a median of 28 ventilator-free days in each group. The full range was 12 to 28 days for LY3127804 and 4 to 28 days for placebo. These descriptive results show that the middle value was the same in the two groups, while the lower observed values differed. The ranges alone do not indicate how observations were distributed between their endpoints, and they are not confidence intervals.
No primary comparison estimate, confidence interval, p-value, or sponsor-reported hazard ratio is reported in the registry record. Consequently, the posted summary does not quantify the uncertainty around a difference between the groups or support a formal conclusion about the primary treatment contrast. A shared median should not be read as proof that the underlying outcome distributions are identical; likewise, different range minima do not establish a reliable group difference. A range records the extremes of the observed values and can be affected by individual observations.
The registry also posts two secondary binary outcomes with unadjusted recalculated comparisons. For complete response, the reported percentages were 78.7% for LY3127804 and 87.5% for placebo. The recalculated unadjusted p-value is 0.253, and the corresponding confidence interval for the risk difference spans the no-difference value. For deaths between Day 1 and Day 28, the secondary outcome reports 7 participants in the LY3127804 group and 4 in the placebo group. Its recalculated unadjusted p-value is 0.318, and its confidence interval also spans the no-difference value. These are unadjusted re-calculations from posted summaries, not sponsor-reported analyses; the registry lists no reported analyses for these outcomes.
Such intervals indicate that the posted data are compatible with a range of possible group contrasts, including no difference. They do not prove equivalence, and the p-values do not measure the probability that a treatment effect is true. These comparisons should be interpreted cautiously because they are secondary, unadjusted, and based on the posted group summaries. The registry does not report a plan for multiplicity handling or the rationale for the trial’s size.
Safety summaries list serious adverse events in 6 participants receiving LY3127804 and 2 receiving placebo. They list deaths of 8 and 6 participants, respectively. The death counts in that safety summary do not match the secondary outcome counts for deaths between Day 1 and Day 28. The registry record does not explain the difference, so the two sets of counts should not be treated as interchangeable or reconciled by assumption.
9. Statistical Methods Explained
The registry reports a median and full range for the primary ventilator-free-days endpoint. A median is a measure of the center of a distribution: it identifies the middle observation or middle pair when values are ordered. A full range describes the lowest and highest values observed. These summaries answer different descriptive questions. Neither a median nor a range supplies a standard error or confidence interval for the difference between treatment groups.
Both groups had a reported median of 28 days. Their full ranges were 12 to 28 days for LY3127804 and 4 to 28 days for placebo. The endpoint is bounded by its Day 1 through Day 28 observation period, and the reported upper endpoint is 28 in both groups. When many values may lie near an endpoint’s boundary, a median can conceal differences in the rest of the distribution. That is a general statistical consideration, not evidence that a particular distributional pattern occurred in this trial.
The registry record does not report a sponsor analysis for the primary endpoint, a confidence interval for a group contrast, or a sponsor-reported hazard ratio. A hazard ratio is a relative comparison of event rates over time under a time-to-event analysis; it is not the same quantity as a difference between medians. The record does not provide a hazard ratio to interpret for this trial. The registry’s posted primary summaries therefore remain descriptive.
For the secondary binary outcomes, the posted results include group percentages or counts. The recalculated comparisons use unadjusted risk differences, with confidence intervals and p-values derived from those posted summaries. “Unadjusted” means these recalculations do not account for other variables or for testing multiple outcomes. They are not sponsor-reported estimates. The record lists no reported analyses for these secondary outcomes. As general statistical guidance, a confidence interval describes a range of values compatible with the data and the analysis assumptions; a wide interval signals limited precision. An interval that spans the no-difference value does not establish that the groups are equivalent.
The complete-response result was 78.7% in the LY3127804 group and 87.5% in the placebo group. Its recalculated unadjusted p-value is 0.253. The death outcome from Day 1 through Day 28 reports 7 participants in the LY3127804 group and 4 in the placebo group; its recalculated unadjusted p-value is 0.318. The corresponding recalculated risk-difference intervals for both outcomes span the no-difference value. These results do not, on their own, settle whether a treatment effect exists. The registry does not report adjustment for baseline factors, the method for handling missing secondary outcomes, or a statistical analysis plan against which these re-calculations could be checked.
The sample-size illustrations are planning calculations based on observed effects, not observed power and not the trial’s original design justification. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The illustrations ask how many participants per arm would be needed to detect specified fractions of the observed effect. For complete response, planning for 100% of the observed effect gives 285 participants per arm for 80% power and 381 participants per arm for 90% power; planning for 75% gives 481 and 643, respectively; planning for 50% gives 1,021 and 1,366, respectively. For the death outcome, planning for 100% of the observed effect gives 374 participants per arm for 80% power and 500 participants per arm for 90% power; planning for 75% gives 624 and 834, respectively; planning for 50% gives 1,309 and 1,752, respectively.
These planning values are illustrations conditional on the observed effects and stated assumptions. They do not show the probability that this trial detected an effect, and they should not be mistaken for the trial’s planned enrollment target. The registry record does not report a sample-size justification. The record also does not report interim analyses, stopping boundaries, or a complete statistical analysis plan. More generally, over-dispersion means that outcome variation exceeds what a simple model assumes; when that occurs, a model that ignores it can understate uncertainty. The registry record does not report a model for the primary endpoint or information that would allow an assessment of over-dispersion, so this general issue should not be used to infer a problem in the trial’s analysis.
10. Multiplicity, Missing Data and Analysis Populations
The registry record does not report a statistical analysis plan describing how multiplicity was handled. This matters because the trial lists several secondary outcomes, and separate testing across multiple outcomes can increase the chance of finding an apparently notable result even when no true difference exists. A prespecified hierarchy or adjustment can clarify which outcomes are confirmatory and how their error rates are controlled. No such approach is reported in the registry record, so the secondary comparisons should not be treated as multiplicity-adjusted evidence.
The recalculated secondary p-values and confidence intervals are unadjusted. They are simple re-calculations from posted group summaries, not sponsor-reported analyses. The record lists no reported analyses for these outcomes. Their interpretation is therefore conditional on the summaries and calculation approach, and should not be confused with a complete trial analysis.
Missing-data procedures are not reported in the registry record. The primary population is described as all randomized participants who received at least 1 dose of study drug, which specifies a treated, post-randomization population rather than all randomized participants irrespective of treatment received. The record does not report how many participants, if any, were excluded from this population, how missing endpoint values were treated, or whether sensitivity analyses were conducted. It also does not report a separate intention-to-treat analysis.
The registry does not report the sample-size justification, interim analyses, stopping rules, or full statistical analysis plan. Although trial status is terminated, the reason for termination and whether any planned analyses were affected are not reported in the registry record. These absences make it difficult to assess whether analyses or outcome handling followed a prespecified plan. They do not establish that any particular procedure was or was not used; they mean that the record does not document it.
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 |
|---|---|---|---|
| LY3127804 | 6 | 47 | 12.8% |
| Placebo | 2 | 48 | 4.2% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| LY3127804 | 8 of 47 |
| Placebo | 6 of 48 |
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 |
|---|---|
| LY3127804 | 30 of 47 |
| Placebo | 31 of 48 |
Most frequent serious adverse event terms
| Event term | Organ system | LY3127804 | Placebo |
|---|---|---|---|
| Atrial fibrillation | Cardiac disorders | 2 of 47 | 1 of 48 |
| Pneumothorax | Respiratory, thoracic and mediastinal disorders | 1 of 47 | 1 of 48 |
| Cardiac arrest | Cardiac disorders | 1 of 47 | 0 of 48 |
| Duodenal ulcer | Gastrointestinal disorders | 1 of 47 | 0 of 48 |
| Intestinal ischaemia | Gastrointestinal disorders | 1 of 47 | 0 of 48 |
| Pneumonia | Infections and infestations | 1 of 47 | 0 of 48 |
| Pneumonia bacterial | Infections and infestations | 1 of 47 | 0 of 48 |
| Sepsis | Infections and infestations | 1 of 47 | 0 of 48 |
| Urinary tract infection | Infections and infestations | 1 of 47 | 0 of 48 |
| Traumatic haemothorax | Injury, poisoning and procedural complications | 1 of 47 | 0 of 48 |
Most frequent non-serious adverse event terms
| Event term | Organ system | LY3127804 | Placebo |
|---|---|---|---|
| Constipation | Gastrointestinal disorders | 6 of 47 | 7 of 48 |
| Anxiety | Psychiatric disorders | 4 of 47 | 6 of 48 |
| Respiratory failure | Respiratory, thoracic and mediastinal disorders | 4 of 47 | 5 of 48 |
| Hypokalaemia | Metabolism and nutrition disorders | 5 of 47 | 3 of 48 |
| Insomnia | Psychiatric disorders | 5 of 47 | 3 of 48 |
| Acute kidney injury | Renal and urinary disorders | 3 of 47 | 4 of 48 |
| Hypertension | Vascular disorders | 4 of 47 | 3 of 48 |
| Leukocytosis | Blood and lymphatic system disorders | 5 of 47 | 1 of 48 |
| Diarrhoea | Gastrointestinal disorders | 3 of 47 | 3 of 48 |
| Hyperglycaemia | Metabolism and nutrition disorders | 3 of 47 | 3 of 48 |
Serious events with the highest arm-level rate: atrial fibrillation (4.3%), cardiac arrest (2.1%), duodenal ulcer (2.1%), intestinal ischaemia (2.1%), pneumonia (2.1%).
The safety summary reports serious adverse events in 6 participants in the LY3127804 group and 2 in the placebo group. It reports any treatment-emergent adverse events in 30 participants and 31 participants, respectively. These are counts from the registry’s safety summary; they do not, by themselves, describe the seriousness, duration, or clinical consequences of each event.
The registry lists several serious adverse-event terms, including atrial fibrillation at 4.3%, and cardiac arrest, duodenal ulcer, intestinal ischaemia, pneumonia, and pneumonia bacterial at 2.1% each. The listed terms do not provide a complete clinical account of each participant’s events, and the registry record does not report further detail needed to interpret individual cases.
The safety summary reports deaths of 8 participants with LY3127804 and 6 with placebo. Those counts differ from the secondary outcome counts for deaths between Day 1 and Day 28, which are 7 and 4. The registry record does not explain the difference in timeframe, definition, or analysis population, so the figures should be kept distinct. Counts alone cannot establish whether an intervention caused an event, and this summary should not be read as a causal comparison.
12. What the Result Does, and Does Not, Mean
The posted primary summaries establish that the median number of ventilator-free days was 28 in each group. They also show full ranges of 12 to 28 days for LY3127804 and 4 to 28 days for placebo. These are descriptive findings for the endpoint as defined in the registry. They do not quantify a treatment effect with uncertainty, because the registry does not report a primary comparison estimate, confidence interval, p-value, or sponsor-reported hazard ratio.
A matching median does not show that all outcomes were the same between groups or that the underlying distributions were equivalent. A range does not describe how frequently values occurred across the interval. Conversely, the different minimum values do not by themselves show a dependable difference. To assess the precision of a group contrast, readers would need an inferential estimate and its uncertainty, which are not reported for the primary endpoint.
The secondary results provide additional descriptions, but their recalculated comparisons are unadjusted and the registry does not report a multiplicity plan. The available evidence is therefore not a basis for a definitive claim about the secondary outcomes as a group. The safety counts are also descriptive. The differing death counts across the secondary and safety summaries are unexplained in the registry record.
These limitations do not prove that there was no effect, nor do the posted summaries establish that there was an effect. They describe what can be concluded from the record: a median of 28 days in both groups, ranges as posted, selected secondary and safety counts, and no reported primary inferential comparison. The record does not report enough about missing data, analysis procedures, or trial termination to resolve those uncertainties.
13. Why This Trial Matters Statistically
This trial illustrates why a reported endpoint summary and a treatment-effect analysis are not interchangeable. The primary outcome is a median with a full range in each group. Those summaries help describe observed values, but they do not show the estimated between-group contrast or its precision. Because the registry does not report a primary inferential comparison, readers cannot use the posted primary result to judge how compatible the data are with a range of possible treatment effects.
The endpoint also combines respiratory support and survival through its definition. A qualifying unassisted period had to last at least 24 consecutive hours, and a participant had to remain alive within 28 days from the first dose for days to count. Understanding the definition is essential before interpreting the median as a simple count of days without ventilation.
The secondary outcomes show a second statistical lesson: unadjusted comparisons across multiple outcomes carry less weight than a prespecified, fully reported analysis. The registry lists multiple secondary measures, provides no reported analyses for them, and does not report a multiplicity plan. Recalculations from posted summaries can aid description, but they cannot substitute for the sponsor’s full analysis or establish a confirmatory finding.
Finally, registry records can contain summaries that are not readily reconciled. Here, the secondary death outcome and safety summary report different death counts, with no explanation in the record. Careful interpretation preserves the distinction rather than silently combining or choosing between them. Together, these features make the trial a useful example of how endpoint definitions, uncertainty, multiplicity, analysis populations, and transparent reporting shape what can be learned from clinical-trial results.
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 principal limitation for interpreting the primary result is that the registry reports medians and full ranges without an inferential comparison. There is no reported confidence interval, p-value, or sponsor-reported hazard ratio for ventilator-free days. The descriptive summaries therefore do not establish the precision of a group contrast.
The registry record does not report the reason the trial was terminated. It also does not report a detailed protocol, a sample-size justification, interim analyses, stopping rules, or a full statistical analysis plan. Without those details, the relationship between the posted summaries and the planned analysis cannot be assessed from the registry record.
The described primary population includes randomized participants who received at least 1 dose of study drug. The record does not report how missing outcomes were handled, whether any participants were excluded from the primary analysis, or whether an analysis of all randomized participants was performed. This limits assessment of the effect of post-randomization exclusions or missingness.
The secondary recalculations are unadjusted and do not account for testing multiple outcomes. The registry does not report a multiplicity strategy or sponsor-reported analyses for those outcomes. In addition, death counts differ between the secondary outcome and the safety summary, and the reason is not reported. The record also lists six secondary outcomes while giving an overall outcome-measure count of 7; it does not explain that distinction. These reporting gaps should be considered when reading the results, but they do not identify what the missing details would have shown.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: A Study of LY3127804 in Participants With COVID-19 — NCT04342897.
- Registry record last updated 2021-08-12; status last verified 2021-08.
- Results first posted 2021-08-12.
- 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
NCT04342897 is a randomized, double-blind, placebo-controlled, parallel-group Phase 2 trial of LY3127804 in hospitalized participants with pneumonia and presumed or confirmed COVID-19. The recorded enrollment is 95, and the status is terminated. Participants received LY3127804 or placebo by intravenous infusion at 20 mg/kg on Days 1 and 15.
The primary endpoint was ventilator-free days from Day 1 through Day 28. The reported median was 28 days in both groups; the full ranges were 12 to 28 days for LY3127804 and 4 to 28 days for placebo. The registry does not report an inferential primary comparison. Secondary and safety summaries are also posted, but the registry does not report a multiplicity plan, missing-data procedures, or the reason for termination. Death counts differ between the secondary outcome and safety summary, without an explanation in the registry record.