COVID-19 Outpatient Pragmatic Platform Study (COPPS) - Acebilustat Sub-Protocol: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for COVID-19 Outpatient Pragmatic Platform Study (COPPS) - Acebilustat Sub-Protocol (NCT04662060), a randomized phase 2 study. The primary outcome is For Clinical Domain: Time-to-sustained-resolution.

2. Clinical Question

3. Trial Design

COPPS was a pragmatic, randomized, double-masked, parallel-group Phase 2 platform trial. This acebilustat sub-protocol compared acebilustat with matching placebo in adults with COVID-19 who were not yet sick enough to require hospitalization. Participants were assigned to receive their allocated treatment for 28 days. The registry record lists 120 participants enrolled, with 60 in each arm.

The wider platform was designed to evaluate investigational therapeutics across a viral domain and a clinical domain. Its stated objectives included assessing viral shedding and clinical outcomes. The platform description says that individual treatments were covered by separate sub-protocols; the results concern acebilustat. The record lists six secondary outcomes, including viral shedding, viral cessation, symptom resolution, COVID-19-related hospitalizations, emergency department visits or death, and antibody development.

The study is marked completed. Its registry dates list a start of 2021-04-23 and primary completion of 2022-05-03. The primary outcome was time to sustained symptom resolution over 28 days, analyzed in the intent-to-treat analysis set. The sponsor-reported comparison used a Cox proportional hazards model. The registry record does not report the original design assumptions for this sub-protocol, the randomization procedure in further detail, or the planned approach to interim analyses.

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

Analysis population for the primary outcome, as defined in the registry: Intent-to-treat analysis set

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: Intent-to-treat population (all randomized participants)

6. Primary Endpoint

The registry lists the primary outcome as For Clinical Domain: Time-to-sustained-resolution, measured over 28 days.

The primary endpoint measures the time from randomization to sustained symptom resolution during a 28-day observation period. Resolution was defined as the first day on which no symptoms were self-reported on that day or any succeeding day through and including day 28. Sense of taste or smell was excluded from the definition; fatigue and cough counted as resolved when reported as mild or none. The analysis population was the intent-to-treat analysis set.

This is a time-to-event endpoint: the event is sustained symptom resolution, and the outcome includes when that event occurred, rather than only whether it occurred. In general, such endpoints can retain information about the timing of an event and can account for participants whose event time is not observed during follow-up. The registry record does not report how any censoring or missing symptom reports were handled.

The endpoint is useful for describing the course of symptoms over a defined period, but its interpretation depends on the operational definition. Here, resolution requires continued symptom-free reports through day 28, while taste and smell are excluded and fatigue and cough have specific thresholds. The reported medians and hazard ratio should therefore be read as results for this registry-defined outcome, not as a measure of every aspect of recovery.

7. Results Posted in the Registry

• Placebo vs Acebilustat: hazard ratio 0.60 (95% CI 0.34–1.04), reported p = 0.070 (as reported in the registry).

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Time to First Resolution

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor (Linear mixed-effects regression model); it was not recalculated.
• Result:
  • Placebo vs Acebilustat: hazard ratio 0.62 (95% CI 0.38–1.01), reported p = 0.050 (as reported in the registry).
• Interpretation:
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Placebo vs Acebilustat and the control. The interval gives the range of effects that remain compatible with the data.

Time to Full Resolution

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor; it was not recalculated.
• Result:
  • Placebo vs Acebilustat: hazard ratio 0.59 (95% CI 0.34–1.01) (as reported in the registry).
• Interpretation:
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Placebo vs Acebilustat and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary results cover several types of outcomes, including viral measures, symptom resolution, COVID-19-related acute-care events or death, and antibody development. Numerical comparisons are available for time to first resolution and time to full resolution; values for the other listed secondary outcomes are not reported in the registry record. Across the two symptom-resolution outcomes with posted estimates, the confidence intervals reach or include 1. These secondary findings provide context for the primary result but do not remove the uncertainty in it.

The secondary outcomes address related but distinct questions. First resolution and full resolution are not identical to sustained resolution as defined for the primary endpoint. Viral shedding and viral cessation concern the viral domain, while hospitalization, emergency department visits, or death concern clinical events. Because the outcomes measure different aspects of illness, one result should not be treated as a substitute for another or as a direct confirmation of the primary endpoint.

The registry record does not report a multiplicity adjustment or a hierarchy controlling interpretation across the several outcomes. In general, testing multiple outcomes creates more opportunities for apparently notable results to arise by chance. Unadjusted secondary results therefore carry less weight as confirmatory evidence than a prespecified primary analysis, especially when methods or p-values are not reported for some outcomes. The posted secondary findings are best read as exploratory context unless the record reports a confirmatory strategy; such a strategy is not reported in the registry record.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• For the Viral Domain: Change in Viral Shedding
• Time to Viral Cessation
• Time to First Resolution
• Time to Full Resolution
• Number of Participants With SARS-CoV-2 Related Hospitalizations, Emergency Department (ED) Visits, or Death in Outpatients With COVID-19 Disease.
• Number of Participants That Developed Antibodies to SARS-CoV-2

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is a time-to-event measure: the event is sustained symptom resolution, and the measured time begins at randomization. The registry describes a 28-day assessment window and defines resolution using symptom reports on the first qualifying day and every succeeding day through day 28. In general, time-to-event methods are used when both event occurrence and event timing matter. They can also accommodate incomplete follow-up through censoring, when the analysis method and censoring rules are specified. The registry record does not report the censoring rules used for this analysis.

What does the hazard ratio describe?

The sponsor used a Cox proportional hazards model for the primary comparison. In general, a Cox model estimates a relative hazard: the instantaneous rate of the event in one group compared with the other, conditional on the model. The posted comparison is labeled placebo versus acebilustat, and the reported hazard ratio is 0.60. A hazard ratio is not a direct comparison of medians, nor is it the chance of resolution by a particular day. It is interpreted in relation to the event definition, comparison order, and model assumptions.

The proportional-hazards assumption means, in general, that the relative hazard is treated as stable over time. The registry record does not report diagnostics or other information showing how that assumption was assessed. This matters because if relative hazards change over follow-up, a single hazard ratio may not describe that changing pattern fully. The posted record gives the model and estimate, but not additional model diagnostics.

Why use a confidence interval?

A confidence interval expresses statistical uncertainty around an estimated effect. In general, a 95% confidence interval is produced by a procedure that, under its assumptions, would contain the true parameter in 95% of repeated samples. It is not a probability statement about the particular interval after it has been calculated. For the primary estimate, the interval runs from 0.34 to 1.04. Because it includes 1, the data are compatible with the no-difference value as well as with other values in the interval. The width also reflects that the estimate is not highly precise.

How should the p-value be understood?

The reported p-value is 0.07. In general, a p-value measures how unusual data at least as incompatible with a null hypothesis would be, if that hypothesis and the model assumptions were correct. It does not give the probability that the null hypothesis is true, and it does not measure clinical importance. The p-value should be read together with the point estimate, confidence interval, endpoint definition, and analysis plan. The registry record does not report whether the p-value was adjusted for testing multiple outcomes.

What are the secondary endpoint models?

The time-to-first-resolution comparison is reported with a hazard ratio of 0.62, a 95% confidence interval from 0.38 to 1.01, and p = 0.05. Its listed method is a linear mixed-effects regression model. The time-to-full-resolution result is reported with a hazard ratio of 0.59 and a 95% confidence interval from 0.34 to 1.01; a method and p-value are not reported in the registry record. The record does not explain the model specification or the relationship between the listed method and the hazard-ratio estimand for the first secondary result. These method labels and estimates should be described as posted, without inferring details that the record does not provide.

What does the intent-to-treat population mean?

The primary result is identified as an intent-to-treat analysis. In general, this approach analyzes participants according to their randomized assignment, helping preserve the comparison created by randomization. The record does not provide further details about exclusions, the handling of missing outcome data, or whether every randomized participant contributed to each analysis. The primary outcome summaries list 60 participants in each arm, while the adverse-event groups list 59 at risk in each arm; the registry record does not explain that difference.

Statistical methods do not remove the need to examine how outcomes were defined, recorded, and analyzed. For this trial, interpretation is limited by the finite observation period, incomplete primary distribution summaries, and absent details on missing-data handling and model diagnostics. Those omissions do not change the reported estimate, but they leave some aspects of its context not reported in the registry record.

11. Confidence Intervals and Precision

The primary hazard-ratio estimate is 0.60, with a 95% confidence interval from 0.34 to 1.04. The interval crosses 1, the no-difference value for a hazard ratio. Thus, the posted data are compatible with no relative-hazard difference as well as with values on either side of that reference. The point estimate alone is not enough to convey this uncertainty.

In general, confidence intervals show the precision of an estimate: narrower intervals indicate less statistical uncertainty, while wider intervals indicate more. They do not account for every source of uncertainty, such as limitations in outcome measurement or model assumptions. The interval's upper bound is 1.04, and the reported p-value is 0.07. These results should be interpreted together rather than treating the point estimate as definitive.

The primary arm summaries are also incomplete. Placebo has a reported median of 26.0 days and lower quartile of 17.5 days. The acebilustat median is not reported in the registry record, although its lower quartile is 22.0 days. The upper quartile for placebo is also not reported in the registry record. Consequently, readers cannot compare complete median and interquartile-range summaries for the primary outcome using the posted values.

The interval describes precision under the analysis model; it does not guarantee that the estimate will reproduce in another study or population. The registry record does not report additional confidence intervals for the arm-specific primary medians, nor does it provide model diagnostics that would help assess assumptions underlying the hazard-ratio interval.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are illustrations based on the observed effect, not the trial's original design justification. The assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect targeted is the stated fraction of the observed effect. For time-to-event outcomes, the figures are total events across both arms, calculated using the Schoenfeld approach; they are not participants per arm. The primary-outcome planning display gives total-event targets for 80% and 90% power under each stated effect scenario.

In plain terms, the 80% power column gives the total number of events that the illustration estimates would be needed for an 80% chance of detecting the assumed effect under the stated conditions. The 90% column represents a higher target chance of detection and therefore has a larger event requirement. These are planning targets, not descriptions of how many events occurred in this trial. This is not observed power.

The rows describe planning for 100%, 75%, and 50% of the observed effect. A smaller assumed effect is harder to distinguish from random variation, so it requires more events to reach the same target power. For time-to-event planning, the event count is not interchangeable with enrollment: the number of participants needed depends in part on how many experience the event during the observation period. The registry record does not report the original sample-size assumptions for the acebilustat sub-protocol, so these illustrations should not be read as the trial's prospective design calculation.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

d is the total number of events needed (Schoenfeld's formula) with 1:1 allocation and proportional hazards.

Worked calculation for this trial

1. Hazard ratio HR = 0.600, so ln(HR) = -0.5108.
2. d = 4 × (1.96 + 0.842)² / [ln(0.600)]² = 120.3, rounded up to 121 events in total for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
hr <- 0.6                              # observed hazard ratio
events_needed <- function(power, frac = 1) ceiling(round(4 * (z_a + qnorm(power))^2 / log(hr^frac)^2, 6))   # Schoenfeld, 1:1 allocation
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: events = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(events_needed(0.80, f), big.mark = ","), format(events_needed(0.90, f), big.mark = ",")))

# Simulation: given d total events under proportional hazards, events in the treatment arm are Binomial(d, hr / (1 + hr)).
# The score (log-rank type) statistic is (D1 - d / 2) / sqrt(d / 4).
set.seed(4662060)
d    <- events_needed(0.80)
reps <- 20000
D1 <- rbinom(reps, d, hr / (1 + hr))
z  <- (D1 - d / 2) / sqrt(d / 4)
cat(sprintf("Simulated power with %s events: %.3f (%d replicates)\n", format(d, big.mark = ","), mean(abs(z) > z_a), reps))

Output from running the code

100% of observed effect: events = 121 (80% power), 162 (90% power)
 75% of observed effect: events = 214 (80% power), 287 (90% power)
 50% of observed effect: events = 482 (80% power), 645 (90% power)
Simulated power with 121 events: 0.783 (20000 replicates)

The simulation uses the fact that, with d events in total and proportional hazards, the number of events in the treatment arm is binomial. It is an approximation that conditions on the total number of events, and it should land close to the target power.

14. Analyses Reported by the Sponsor

The primary result is the sponsor-reported Cox proportional hazards analysis: hazard ratio 0.60 for placebo versus acebilustat, with a 95% confidence interval from 0.34 to 1.04 and p = 0.07. The registry record also lists this analysis as the reported result. No discrepancy is flagged between the posted result and its accompanying comparison summary.

For this time-to-event outcome, the comparison is presented using the sponsor's reported hazard ratio rather than an independent recalculation from summary statistics. An independent recalculation of a hazard ratio cannot be recovered simply from the two arm medians or quartiles; it would require participant-level event-time information or additional model output. The registry record does not report those data.

The sponsor's estimate is therefore the relevant posted comparative statistic, while the arm-specific summaries offer descriptive context. Those summaries are incomplete for the primary outcome: the acebilustat median and placebo upper quartile are not reported in the registry record. The estimate should be considered with its confidence interval and the model assumptions, rather than treating either a single summary value or the p-value as a complete account of the result.

15. Multiplicity, Missing Data and Analysis Populations

The study record lists one primary clinical endpoint and multiple secondary outcomes. It does not report whether the statistical plan adjusted for multiple comparisons, specified a testing hierarchy, or controlled a family-wise error rate across the outcomes. Without that information, the degree to which secondary findings were intended to be confirmatory is not reported in the registry record.

In general, multiplicity matters when several hypotheses are tested: the chance of at least one nominally unusual result can increase as more tests are considered. An adjustment or prespecified hierarchy can help control that issue, but neither should be assumed when it is not described. The p-values attached to posted outcomes should therefore be interpreted in the context of the complete outcome set and the absence of a reported multiplicity plan.

The primary analysis population is identified as the intent-to-treat analysis set. The registry record does not report detailed eligibility for that set, departures from randomized assignment, or the number of participants contributing outcome data at each time point. It also does not report how missing daily symptom reports were handled, whether any imputation was used, or how censoring was defined for time-to-event analyses.

These omissions matter because sustained resolution depends on a sequence of self-reported symptom assessments through day 28. In general, missing observations can affect whether an event can be classified and when its time is assigned. The impact depends on the amount and pattern of missingness and the analysis rules. Neither the missing-data pattern nor those rules are reported in the registry record, so their influence cannot be assessed from the posted information.

The record also does not report interim-analysis procedures, stopping rules, or details of the original statistical analysis plan. These should not be inferred from the platform description or from the final reported result. The available information supports a description of the posted analyses, but not a full audit of their prespecification or missing-data assumptions.

16. 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 5% reporting threshold.

Most frequent non-serious adverse event terms

For serious adverse events, the registry lists 0 affected participants among 59 at risk in the placebo group and 0 among 59 at risk in the acebilustat group. No deaths are listed in either group. These are the posted counts; they describe the recorded events in the listed at-risk groups and do not establish that no other safety events occurred.

For other adverse events, 3 participants are listed as affected among 59 at risk in the placebo group, and 1 among 59 at risk in the acebilustat group. The registry record does not provide serious adverse event terms, as the serious-event terms list is empty. It also does not explain why the adverse-event at-risk counts are 59 per group while the primary outcome summaries list 60 participants per arm.

In general, adverse-event counts are descriptive and depend on how events were defined, collected, and coded. The registry record does not provide event-level detail or exposure-adjusted rates. These counts should not be interpreted as a complete comparison of safety or as proof that an event cannot occur.

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

The posted primary analysis estimates a hazard ratio of 0.60 for placebo versus acebilustat, with a 95% confidence interval from 0.34 to 1.04 and p = 0.07. The interval includes 1, so the result does not provide a clear statistical distinction from the no-difference value under the reported model. The estimate and interval describe uncertainty in the comparison; they do not establish a definitive effect.

The endpoint is narrowly defined: sustained symptom resolution according to daily self-reports over 28 days, with taste and smell excluded and specified thresholds for fatigue and cough. It is not a measure of every possible clinical outcome. The record also lists viral and other clinical outcomes, but most of their numerical results are not reported in the registry record.

The arm-level primary summaries are incomplete because the acebilustat median and placebo upper quartile are not reported in the registry record. The hazard ratio, rather than a complete comparison of medians and interquartile ranges, is the posted comparative result. The registry record does not provide participant-level data, missing-data details, or model diagnostics that would permit a fuller independent assessment.

These results apply to the enrolled outpatient population and the specific study period and outcome definition. They do not establish what would happen in populations not represented in the record, over a different follow-up period, or for outcomes not numerically posted. The statistical result should be understood as an estimate with uncertainty, not as medical advice or a guarantee about an individual participant's course.

18. Why This Trial Matters Statistically

This trial illustrates several important features of interpreting pragmatic platform research. COPPS was designed to evaluate multiple investigational therapies within a broader platform, while individual treatments were described in separate sub-protocols. The acebilustat comparison therefore needs to be interpreted as its own randomized, placebo-controlled comparison, within the platform context and for the stated outpatient population.

The primary endpoint demonstrates why outcome definitions matter. Sustained resolution required symptom-free reports through day 28, with explicit exceptions for taste and smell and thresholds for fatigue and cough. That operational definition determines what event the time-to-event model analyzes. A different definition of resolution could represent a different clinical question.

The reported hazard ratio is accompanied by a confidence interval that includes the no-difference value. This makes the precision information central to interpretation: a point estimate alone can sound more conclusive than the underlying uncertainty supports. The record also contains incomplete primary arm summaries, which limits descriptive comparison and underscores the value of complete reporting alongside model-based results.

Finally, the trial record lists multiple secondary outcomes but does not report a multiplicity strategy or detailed missing-data procedures. Those gaps matter statistically because they affect how much confidence readers can place in individual secondary findings and how fully the analyses can be evaluated. The case illustrates that a clear endpoint definition, complete outcome reporting, and transparent analysis details are all part of interpreting a treatment comparison.

19. 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 posted primary arm summaries are incomplete: the acebilustat median time to sustained resolution is not reported in the registry record, and the placebo upper quartile is not reported. This makes it difficult to describe both groups' full time distributions from the posted summaries. The sponsor-reported hazard ratio remains the comparative estimate, but the missing descriptive values constrain interpretation of the arm-level pattern.

The registry record does not report how missing symptom assessments or censoring were handled, how the proportional-hazards assumption was assessed, or whether sensitivity analyses were conducted. These details are relevant to a time-to-event endpoint built from daily self-reports. The record also does not report the original sample-size justification or interim-analysis procedures, so the planning illustrations should not be confused with the trial's own prospective design rationale.

Several secondary outcomes are listed, but numerical results are posted for only some of them. A multiplicity adjustment or outcome hierarchy is not reported in the registry record. The method listed for time to first resolution is a linear mixed-effects regression model, while its comparison is expressed as a hazard ratio; the record does not explain that relationship. For time to full resolution, the analysis method and p-value are not reported.

Finally, the adverse-event at-risk counts are 59 per group, whereas the primary outcome summaries list 60 per group, and the record does not explain the difference. The listed serious adverse-event counts are low, but event-level detail is not reported. These reporting limits should be kept distinct from the study's numerical findings: they restrict how fully the results can be evaluated, but they do not change the values posted.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: COVID-19 Outpatient Pragmatic Platform Study (COPPS) - Acebilustat Sub-Protocol — NCT04662060.
• Registry record last updated 2023-06-13; status last verified 2023-06.
• Results first posted 2023-06-13.
• 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.

23. Record Summary

The registry record describes a completed, randomized, double-masked, parallel-group Phase 2 acebilustat sub-protocol within COPPS. It enrolled 120 outpatients with COVID-19, with 60 assigned to acebilustat for 28 days and 60 to matching placebo for 28 days.

The primary endpoint was time to sustained symptom resolution over 28 days in the intent-to-treat analysis set. The sponsor-reported placebo-versus-acebilustat hazard ratio was 0.60, with a 95% confidence interval from 0.34 to 1.04 and p = 0.07. The primary arm summaries are incomplete, and the registry record does not report detailed missing-data or multiplicity procedures.

The record lists secondary viral and clinical outcomes, with numerical comparisons posted for time to first resolution and time to full resolution. Serious adverse events and deaths were listed as 0 in both at-risk groups; other adverse events were listed for 3 placebo participants and 1 acebilustat participant. These findings should be read with the stated endpoint definitions and reporting limitations.