Etoposide in Patients With COVID-19 Infection: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Etoposide in Patients With COVID-19 Infection (NCT04356690), a randomized phase 2 study. The primary outcome is Improvement in Pulmonary Status by Two Categories on an 8 Point Ordinal Scale of Respiratory Function.

2. Clinical Question

The study was designed as a randomized phase II evaluation of safety and efficacy in patients with COVID-19 infection. Its primary outcome focused on a clinically ordered measure of respiratory status rather than a single physiological measurement. The scale ranges from death and intensive forms of respiratory or other organ support to discharge without supplemental oxygen. A change of two categories was the specified improvement threshold.

The registry describes a 3:1 allocation plan, with etoposide compared with standard care. The posted arm counts are six in the etoposide group and two in the control group. With only eight participants, the results are descriptive and provide limited information about the between-group difference. The record does not report the eligibility criteria in sufficient detail to establish how broadly these findings might apply to other patients or settings.

3. Trial Design

This was a phase II, single-center, randomized, open-label, parallel-group study sponsored by Boston Medical Center. The registry lists COVID-19 as the condition and etoposide as the study drug. It describes randomization with a 3:1 allocation ratio and standard of care as the comparator. No placebo was used, and masking was listed as none.

The planned etoposide regimen was intravenous etoposide at 150 mg/m2 on Days 1 and 4 for eligible patients. The brief description allowed subsequent doses if the investigator and treating physician believed that a participant had clinical benefit but later showed recurrent clinical deterioration. The control group was to receive standard of care. Arm descriptions specify participants on ventilation in both groups.

Actual enrollment was eight participants: six assigned to etoposide and two to control. The registry lists the study as terminated. The recorded study start was 2020-05-08, and primary completion was 2022-07-01. The registry record does not explain why the trial was terminated or report whether the planned enrollment was reached.

The primary outcome was improvement in pulmonary status by two categories on an 8-point ordinal scale, assessed from baseline through hospital discharge or death. Secondary outcomes included overall survival, length of hospitalization, duration of ventilation after treatment, and changes in blood ferritin, C-reactive protein, D-dimer, white blood cell count, and platelet count. The registry lists eight secondary outcomes and nine outcome measures in total. It records no analyses for the posted results. Details of sequence generation, allocation concealment, the intended statistical model, and the planned sample-size justification 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

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.

6. Primary Endpoint

The registry lists the primary outcome as Improvement in Pulmonary Status by Two Categories on an 8 Point Ordinal Scale of Respiratory Function, measured over baseline, through hospital discharge or death.

The primary endpoint counts participants whose pulmonary status improved by two categories on an 8-point ordinal scale of respiratory function. The scale orders health states from death and substantial organ support through different levels of respiratory support to discharge without supplemental oxygen. The specified observation period begins at baseline and continues through hospital discharge or death.

An ordinal scale represents ordered categories: one category reflects a more favorable state than another, but the distance between adjacent categories is not necessarily identical. Defining a two-category improvement turns the scale into a binary outcome for the primary endpoint: each participant either meets the threshold or does not. This makes the endpoint focused on a clinically interpretable degree of change, while discarding information about the size and direction of changes that do not meet that threshold.

Such an endpoint can capture changes across levels of respiratory support and disposition that a single measurement may miss. It also requires clear and consistent assignment of scale categories and timing. The registry gives the scale definitions and the overall observation period, but does not report how measurements were collected, who assigned categories, or how deaths and other incomplete observations were handled in the analysis. The posted results give counts by arm, but do not provide an estimable statistical comparison for this binary endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary because it classifies each participant according to whether pulmonary status improved by the specified two-category threshold. The data posted for each arm are participant counts: the number meeting the endpoint and the total number in the arm. Such data can be presented descriptively, but a comparison between arms requires an estimate and an account of uncertainty. For this trial, the registry indicates that a two-arm comparison is missing or not estimable, so an inferential result should not be inferred from the counts alone.

What does a confidence interval communicate?

General explanation: a confidence interval expresses the uncertainty around an estimate under a specified statistical procedure. A narrow interval indicates greater precision than a wide interval, while an interval spanning a broad range of plausible effects signals limited precision. A confidence interval is not a guarantee that the true effect lies within its particular limits. No confidence interval for the primary comparison is reported in the registry record, so the precision of a comparative effect cannot be assessed from an interval.

Why does the sample size matter?

General explanation: when few participants contribute to an outcome, each event can materially change the observed group summaries. This makes estimates unstable and limits the ability to distinguish a true group difference from random variation. Small groups also make it difficult to assess whether participants were balanced on important characteristics. Here, the trial enrolled eight participants, with six in the etoposide group and two in control. The resulting counts are informative as a record of what was observed, but they do not support a precise estimate of a treatment effect.

What can randomization and open-label assignment tell us?

General explanation: randomization is designed to reduce systematic differences in treatment assignment, including differences in factors that may not be measured. It does not ensure balance in a small sample. An open-label design means treatment assignment is known to participants and clinicians. That knowledge can affect care and outcome assessment, depending on the outcome and the procedures used. The registry lists random allocation and no masking, but details of sequence generation, allocation concealment, and outcome-assessment safeguards are not reported in the registry record.

How should an ordinal scale be understood?

General explanation: an ordinal scale places outcomes in a meaningful order without assuming that the gap between each neighboring category is equal. The pulmonary-status scale ranges from death and levels of respiratory or organ support to discharge states. A threshold of improvement by two categories defines a readily interpretable event, but reduces the richer ordered scale to a yes-or-no classification. The registry reports the threshold and scale, but does not describe the statistical method intended for analyzing the full ordinal information, if any.

Why can over-dispersion matter?

General explanation: over-dispersion occurs when observed variation is greater than a statistical model assumes. It is particularly relevant to some count-data models, where underestimating variation can make uncertainty appear too small. The primary outcome in this study is reported as a binary participant count, not a count rate with an exposure period. No over-dispersion assessment or model is reported in the registry record, and it would not be appropriate to assume that a count-model analysis was used.

What do the planning assumptions mean?

General explanation: a sample-size illustration can show how many participants might be needed under stated assumptions about the effect to detect, significance level, allocation, and statistical approximation. Such an illustration is a planning aid; it is not a measure of the study's observed power and does not show that the trial was designed with those assumptions. The registry describes an illustration based on the observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the effect expressed as a fraction of the observed effect. The record does not give the numerical planning results in the results, and these assumptions do not establish the trial's own design justification. The registry's stated allocation plan was 3:1, unlike the equal allocation assumption used for the illustration.

Finally, the registry records zero analyses for the posted results. It does not report a statistical analysis plan specifying estimands, adjustment variables, confidence-interval procedures, or how multiple outcomes were handled. Consequently, methods beyond the recorded counts and stated design assumptions should not be attributed to the study.

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("Cohort 1 - Etoposide", "Cohort 1 - Control")
n      <- c(6, 2)
events <- c(2, 1)
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  Cohort 1 - Etoposide  6  2  33.3  4.3  77.7  9.7  70.0  9.9  71.0
ARM  Cohort 1 - Control  2  1  50.0  1.3  98.7  9.5  90.5  9.4  90.6

11. Multiplicity, Missing Data and Analysis Populations

The registry lists eight secondary outcomes, alongside the primary outcome. These include survival, hospitalization and ventilation duration, and changes in several laboratory measures. Examining multiple outcomes creates a multiplicity issue: when many statistical comparisons are made, the chance of at least one apparently unusual result can increase. General approaches include specifying a primary comparison in advance, controlling an overall error rate, or treating secondary findings as exploratory. The record does not report a multiplicity strategy, a hierarchy among secondary outcomes, or inferential analyses for them.

The registry records zero analyses for the posted results. It reports the primary participant counts, but marks a two-arm binary comparison as missing or not estimable. It does not provide an analysis population definition, such as whether the primary analysis was intended to include everyone randomized or only participants with evaluable outcome data. It also does not report the number of participants with missing primary outcome assessments or how missing outcomes were handled.

Missing data can affect both the number of observations contributing to an estimate and the interpretation of that estimate. General explanation: complete-case analysis, imputation, and other missing-data approaches rely on different assumptions; the best interpretation depends on why data are missing and on the planned analysis. No missing-data assumptions or methods are reported in the registry record. The absence of these details means that the posted counts cannot be used to reconstruct an analysis population or determine whether missingness could have influenced the result.

The record also does not report interim analyses, stopping rules, or whether the termination affected outcome collection or analysis. No conclusion about these procedures can be drawn from the status alone.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: febrile neutropenia (16.7%).

The registry records serious adverse events affecting one of six participants in the etoposide group and none of two participants in control. Other adverse events affected six of six participants in the etoposide group and one of two participants in control. Deaths were recorded for three participants in the etoposide group and one in control. The registry also lists febrile neutropenia at 16.7% among the leading serious adverse event terms.

These are event summaries, not an adjusted comparison of safety between treatments. The small groups mean each participant represents a substantial part of the recorded total. The record does not report event timing, severity grades beyond the serious-event classification, attribution to study treatment, or a statistical comparison. The listed term percentage should not be treated as a rate over person-time; the registry does not state such a denominator or follow-up adjustment. Nor do the death counts establish that treatment caused or prevented deaths. The summaries describe what was recorded in each group, with considerable uncertainty about comparative risk.

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

The results establish that the registry contains primary outcome counts for eight participants and safety summaries for the two randomized groups. Two participants in the etoposide group and one in control met the recorded primary endpoint. The registry does not report an estimable statistical comparison for that binary outcome. This means the posted results do not provide a precise estimate of the difference between groups or its uncertainty.

The findings do not establish that etoposide improves pulmonary status, nor do they establish that it does not. A small number of observed outcomes cannot resolve the treatment question reliably, particularly without an analysis estimate and uncertainty interval. The same caution applies to safety: the adverse-event and death counts are descriptive and cannot by themselves identify a causal effect.

The trial's randomized design is relevant, but randomization does not remove uncertainty when enrollment is this limited. Open-label treatment assignment and the lack of reported analysis details also constrain interpretation. The registry record does not report baseline characteristics, the reason for termination, missing-data handling, the intended statistical plan, or a sample-size justification. These gaps prevent a full assessment of comparability, bias, and the rationale for the study's size or analysis.

The appropriate interpretation is therefore narrow: these are the outcomes recorded for the participants in this terminated study, not a definitive comparative conclusion. Results from a larger, fully analyzed trial could differ, but the registry record alone cannot establish what such a study would find. No clinical decision should be based on these counts alone.

14. Why This Trial Matters Statistically

This trial illustrates a central issue in clinical research: a randomized design does not guarantee an informative estimate if very few participants contribute data. The registry reports eight enrolled participants and an outcome expressed as a binary threshold on an ordinal respiratory scale. The arm counts are easy to describe, but the record says the statistical comparison is missing or not estimable. Without an estimate and a measure of uncertainty, readers cannot judge the range of comparative effects compatible with the posted results.

It also highlights the distinction between a trial's design description and its results. The record describes randomization with a 3:1 allocation plan, while its sample-size illustration uses equal allocation. That illustration is explicitly a planning exercise based on the observed effect, not a record of the trial's own design justification. It should not be confused with evidence that the trial had adequate power or with a measure of observed power.

The open-label design and the use of an outcome involving respiratory support and discharge make transparent outcome-assessment procedures important. Yet the registry does not report those details, nor does it provide an analysis plan, an account of missing data, or a multiplicity strategy for the secondary outcomes. These omissions limit reproducibility and the ability to judge analytic choices.

For biostatisticians and clinical researchers, the record is a reminder to distinguish descriptive counts from comparative evidence and to make the planned analysis, outcome definitions, missing-data approach, and stopping procedures explicit. In a small study, uncertainty and reporting completeness are central to interpretation, not secondary technical details.

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 most immediate limitation is the small enrollment: eight participants, divided between six assigned to etoposide and two to control. Estimates based on such small groups are highly sensitive to individual outcomes. The registry does not provide a comparative estimate or confidence interval for the primary endpoint and states that a two-arm binary comparison is missing or not estimable.

The trial was open-label, with no masking. The primary endpoint includes clinical status categories related to respiratory support and discharge, so knowledge of assignment could potentially influence care or assessment. The registry record does not report blinded outcome assessment or other procedures to reduce this risk. It also does not report baseline characteristics, limiting assessment of group comparability.

The study is listed as terminated, but the reason is not reported in the registry record. The record does not state whether the planned enrollment was reached, whether outcome collection was complete, or whether termination affected follow-up. Missing outcome data and their handling are not described. Although the outcome timeframe runs through hospital discharge or death, the record does not explain the analysis treatment of death for the improvement endpoint.

Details of random-sequence generation, allocation concealment, eligibility criteria, analysis populations, statistical models, multiplicity handling, and sample-size justification are not reported in the registry record. There are no posted analyses for the secondary outcomes. These limitations prevent a comprehensive assessment of precision, bias, and the extent to which the recorded counts answer the study question.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Etoposide in Patients With COVID-19 Infection — NCT04356690.
• Registry record last updated 2023-06-15; status last verified 2023-05.
• Results first posted 2023-06-15.
• 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

The registry describes a randomized, open-label, parallel-group phase II study of etoposide versus standard of care in patients with COVID-19 infection on ventilation. Boston Medical Center is listed as sponsor. The study started on 2020-05-08, primary completion was 2022-07-01, and status is terminated.

Eight participants were enrolled: six in the etoposide arm and two in control. The primary outcome was improvement in pulmonary status by two categories on an 8-point ordinal respiratory-function scale through hospital discharge or death. The posted counts are two and one, respectively, but a two-arm comparison is not reported as estimable. Safety summaries include serious adverse events, other adverse events, and deaths, but no comparative safety analysis is reported. The record lists eight secondary outcomes and zero analyses.

The reason for termination, detailed statistical plan, missing-data methods, and sample-size justification are not reported in the registry record.