Evaluation of Protease Inhibition for COVID-19 in Standard-Risk Patients (EPIC-SR).: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluation of Protease Inhibition for COVID-19 in Standard-Risk Patients (EPIC-SR). (NCT05011513), a randomized phase 2, phase 3 study. The primary outcome is Time to Sustained Alleviation of Overall COVID-19 Signs and Symptoms Through Day 28.

2. Clinical Question

The primary question was about the timing of symptom alleviation, rather than simply whether a participant had alleviated symptoms by a single assessment date. The registry describes the primary hypothesis as testing whether there was a difference between groups in time to sustained alleviation through Day 28. Sustained alleviation required symptoms to meet the specified severity criteria for four consecutive days.

The trial also registered outcomes concerning safety, hospitalization or death, medical visits, severe symptoms, and symptom resolution. Those outcomes provide other perspectives on the course of illness and participant experience, but they do not change the primary question. Their interpretation depends on their own definitions, analysis populations, and statistical precision.

The registry record identifies the study as randomized, quadruple-masked, and parallel-arm. It reports 1,440 enrolled participants overall. The primary result lists 654 participants in the intervention group and 634 in the placebo group; the primary population definition is restricted to randomized participants who received at least one dose. The reason for the difference between overall enrollment and the group counts used for this outcome is not reported in the registry record.

3. Trial Design

EPIC-SR was a Phase 2/3, randomized, quadruple-masked, parallel-arm study comparing oral PF-07321332 with ritonavir against oral placebo. The registry identifies COVID-19 as the condition under study and describes the target group as nonhospitalized symptomatic adults at low risk of progressing to severe illness. The study enrolled 1,440 participants.

The registry lists the study as terminated. It records a start date of 2021-08-25 and a primary completion date of 2022-07-25. The reason for termination, the timing of the decision, and its consequences for conduct or interpretation are not reported in the registry record. A termination status alone does not establish why a study ended or whether its analyses were affected.

The primary endpoint was time to sustained alleviation of overall COVID-19 signs and symptoms from Day 1 through Day 28. The primary analysis population was modified intent-to-treat, defined as all randomized participants who received at least one dose of study intervention. Participants were analyzed according to the intervention to which they were randomized. The primary-result groups include 654 participants assigned to Nirmatrelvir 300 mg + Ritonavir 100 mg and 634 assigned to placebo.

The registry lists secondary outcomes addressing treatment-emergent adverse events, serious adverse events, hospitalization or death, death through Week 24, medical visits, hospital and intensive care duration, severe symptoms, and time to resolution or alleviation of symptoms. It lists 14 outcome measures and 8 analyses. The detailed protocol, the planned statistical analysis document, and a full account of outcome-specific analysis populations are not reported in the registry record. The allocation ratio, recruitment procedures, and reasons for termination are also 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.

Before assignment: A total of 1440 participants signed informed consent form. Out of which, 1296 subjects were randomized and assigned to study drug. 1288 participants were treated and remaining 8 participants were not treated.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified intent-to-treat (mITT1) population included all participants randomly assigned to study intervention, who received at least 1 dose of study intervention. Participants were analyzed according to the study intervention they were randomized.

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: Safety analysis set included all participants who received at least 1 dose of study intervention. Participants were analyzed according to the intervention they actually received. A randomized but not treated participant was excluded from the safety analyses.

6. Primary Endpoint

The registry lists the primary outcome as Time to Sustained Alleviation of Overall COVID-19 Signs and Symptoms Through Day 28, measured over From Day 1 to Day 28.

The primary endpoint measured the time from the start of study intervention or placebo on Day 1 until sustained alleviation of all targeted COVID-19 signs and symptoms, with follow-up through Day 28. The registry reports a median time in days and a 95% confidence interval for each group. The observed medians were 12.0 days for Nirmatrelvir 300 mg + Ritonavir 100 mg, with a 95% confidence interval of 11.0 to 13.0, and 13.0 days for placebo, with a 95% confidence interval of 12.0 to 14.0.

Alleviation was defined relative to each participant’s baseline symptom severity. Symptoms scored moderate or severe at enrollment had to be scored mild or absent; symptoms scored mild or absent at enrollment had to be absent. The qualifying status had to persist for four consecutive days. Missing baseline severity was considered mild. The endpoint was consolidated across overall COVID-19 signs and symptoms rather than reporting separate times for each symptom.

Time-to-event endpoints are useful when the timing of an event matters and participants may reach it at different points during follow-up. A median gives the time by which half of the analyzed group has experienced the defined event, subject to the analysis method and follow-up. It does not describe every participant’s experience or, on its own, establish that groups differ. A group comparison and its uncertainty are also relevant. The registry reports a log-rank comparison for this endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a time-to-event outcome. Each participant’s event was the first qualifying period of sustained symptom alleviation, defined by the registry as four consecutive days meeting the symptom-severity criteria. The time scale began on Day 1, the start of intervention or placebo, and the outcome was followed through Day 28. The registry reports group medians and 95% confidence intervals, and identifies the log-rank test as the method for comparing groups.

What does the median time describe?

As a general statistical concept, the median time is the point in the event-time distribution at which half of the group has experienced the defined event, taking account of the analysis approach. It summarizes the timing of an event across a group rather than the duration for every individual. A median can be useful when times are skewed or when some participants have not experienced the event by the end of observation. The registry’s posted medians are 12.0 days for the intervention group and 13.0 days for placebo. Those values alone do not quantify the uncertainty of the difference between groups.

What does the confidence interval contribute?

As a general statistical concept, a confidence interval expresses the uncertainty of an estimate under the model and sampling assumptions used. The registry reports 95% confidence intervals for each group median: 11.0 to 13.0 days for the intervention group and 12.0 to 14.0 days for placebo. These are intervals around separate group estimates, not an interval for the between-group contrast. The registry does not report a confidence interval for a primary hazard ratio or another primary comparative effect estimate. Consequently, the separate median intervals should not be mistaken for a direct interval around the treatment comparison.

Why use a log-rank test?

As a general statistical concept, the log-rank test compares event-time distributions between groups while using the timing of observed events and accounting for participants whose event times are not fully observed. Its result addresses evidence about a difference in the groups’ time-to-event experience under the test’s assumptions. The registry identifies a log-rank comparison and reports its p-value. A p-value is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The record does not report a hazard ratio for this primary analysis, so the log-rank result cannot be paired with a registry-reported primary hazard-ratio estimate.

What is the analysis population?

The primary population was modified intent-to-treat: participants had to be randomized and receive at least one dose. They were analyzed according to their randomized intervention. This preserves the randomized group assignment for those included, but it is narrower than a population including every randomized participant regardless of whether treatment was received. The record does not report a separate full intent-to-treat primary analysis or explain the relationship between this outcome population and all 1,440 enrolled participants.

How should secondary binary outcomes be read?

Several secondary outcomes are percentages of participants with an event or symptom status. A percentage is a group-level description of the observed proportion meeting that endpoint’s definition. The registry also posts logistic-regression odds ratios for two secondary outcomes and simple recalculations based on posted summary statistics. An odds ratio compares odds, not probabilities or percentage-point differences. A risk difference instead compares proportions on an absolute scale. These measures answer related but distinct questions and are not numerically interchangeable. The unadjusted recalculations do not replace the sponsor’s reported analyses.

As a general statistical concept, over-dispersion arises when observed variation exceeds what a chosen statistical model assumes. It can make standard errors and confidence intervals too small if ignored. The posted primary endpoint is time-to-event, not a count endpoint, and the registry does not report an over-dispersion assessment for it. No conclusion about over-dispersion or a specific correction is warranted from the registry record.

What do the sample-size illustrations mean?

The sample-size illustrations are planning calculations based on stated fractions of observed effects, two-sided significance level 0.05, equal allocation, and a normal approximation. They are not the trial’s documented design justification, and they are not observed power. The registry record does not report the trial’s original sample-size assumptions or justification. Planning based on an observed effect can be sensitive to the estimate used, especially when that estimate is uncertain; smaller target effects require larger planned samples under the stated illustration. Time-to-event illustrations use total event counts, while the other illustrations express participants per arm. Those planning quantities should not be interpreted as evidence about what the completed study was powered to detect.

10. Analyses Reported by the Sponsor

The registry reports a log-rank test for the primary time-to-sustained-alleviation outcome, alongside the group medians and their confidence intervals. It does not post a primary hazard ratio or an interval for a comparative effect. The sponsor’s own primary comparison therefore supplies the reported test result, while the medians describe each group separately.

For two secondary outcomes, the registry posts sponsor-reported logistic-regression odds ratios with 95% confidence intervals. For severe signs and symptoms, the reported odds ratio is 0.819, with a 95% confidence interval of 0.618 to 1.084. For progression to worsening status, the reported odds ratio is 0.802, with a 95% confidence interval of 0.613 to 1.050. Both intervals include the value corresponding to no odds difference.

The separate recalculations for those outcomes are simple, unadjusted risk-difference estimates derived from posted group summaries. They are on a different scale from the sponsor’s adjusted logistic-regression odds ratios. Differences in estimate or uncertainty are not evidence of an error when the statistical measures and adjustment approaches differ. The registry does not provide enough detail to identify every covariate or modeling choice in the sponsor’s logistic regressions.

The discrepancy flag is false, so the record does not indicate a discrepancy requiring a recalculation to take precedence. The sponsor’s reported analyses remain the relevant reported inferential results; recalculations are supplementary descriptions and should not be presented as replacements.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 14 outcome measures and 8 analyses, including a primary endpoint and multiple secondary outcomes. It does not report a multiplicity-control procedure, a testing hierarchy for secondary outcomes, or a statistical analysis plan describing how the family of comparisons was handled. Without that information, the degree to which secondary tests were intended to support confirmatory claims is not reported in the registry record.

As a general statistical concept, multiplicity arises when several hypotheses are tested. If each is interpreted without accounting for the others, the chance of at least one false-positive finding can rise. Adjustment procedures or a prespecified hierarchy can help control that risk, but the registry record does not identify such a procedure for these results. This is why secondary outcomes are most responsibly treated as supportive or exploratory unless the analysis plan establishes otherwise.

Missing-data reporting is limited. The primary endpoint description states that missing symptom severity at baseline was considered mild. This is a specific rule for baseline severity classification; it does not describe how missing follow-up symptom assessments, withdrawals, or incomplete event-time information were handled. The record does not report the amount or pattern of missing data, sensitivity analyses, or the assumptions used for missing observations across the outcomes.

The primary analysis population was modified intent-to-treat, consisting of randomized participants who received at least one dose and analyzed by randomized intervention. That definition excludes randomized participants who received no dose. The registry does not report a comparison with a full intent-to-treat analysis or a detailed accounting of exclusions by reason. Overall enrollment was 1,440, while the primary outcome groups list 654 and 634 participants. The reasons those counts do not encompass all enrolled participants are not reported in the registry record.

The record also does not report interim analyses, stopping rules, or whether the termination status was related to efficacy, safety, recruitment, or another reason. Their absence from the registry record should not be taken to mean that a particular procedure did or did not occur. These gaps constrain assessment of prespecification, missing-data sensitivity, and the full inferential context of the posted results.

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 1% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: covid-19 pneumonia (1.3%), pneumonia (0.3%), covid-19 (0.2%), sepsis (0.2%), creatinine renal clearance decreased (0.2%).

The posted safety counts include serious adverse events among 654 participants at risk in the Nirmatrelvir 300 mg + Ritonavir 100 mg group and 634 at risk in the placebo group. Serious adverse events were reported for 8 participants in the intervention group and 13 in the placebo group. Other adverse events were reported for 120 and 71 participants, respectively. The registry records 0 deaths in the intervention group and 1 in the placebo group in this safety summary.

These are observed counts, not adjusted comparisons. They do not by themselves establish that an intervention caused or prevented an event, and the counts should be interpreted in light of exposure, follow-up, event definitions, and the broader safety analysis. The registry lists COVID-19 pneumonia as the most frequent serious-event term, at 1.3%, followed by pneumonia at 0.3%; other listed terms are each reported at 0.2%.

The registry does not report a comparative statistical analysis for these safety counts in the posted results. A simple contrast between counts does not account for possible differences in observation time or explain the clinical context of each event. The available figures support a descriptive account of events recorded in each group, not a causal conclusion or a complete safety assessment.

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

The primary results describe time to a specifically defined event: four consecutive days meeting symptom-alleviation criteria, assessed from Day 1 through Day 28. The observed median was 12.0 days in the Nirmatrelvir 300 mg + Ritonavir 100 mg group and 13.0 days in the placebo group. The separate confidence intervals indicate uncertainty around each group median, and the reported log-rank comparison does not establish a clear difference in time-to-event experience.

This finding does not mean the groups had identical symptom courses. A study can produce estimates that differ while leaving uncertainty about the size and direction of the underlying difference. Nor does the result demonstrate that every participant experienced the group median or that an outcome for an individual can be predicted from the group summary. The endpoint combines several symptoms using a defined baseline-relative rule, so it is also not a direct description of every symptom considered separately.

The secondary outcomes offer additional descriptions, but they concern different events and are subject to their own uncertainty. The registry does not report a multiplicity procedure, detailed missing-data analyses, or a complete statistical analysis plan. Those limitations matter when assessing secondary findings and the strength of any broader interpretation.

The posted results do not establish a treatment recommendation. They describe one randomized study in the registered population, using the outcome definitions and analysis information available in the registry. They do not establish what would happen in other populations, under other care conditions, or for outcomes beyond those reported. The reason the study was terminated is not reported in the registry record, so termination should not be interpreted as evidence for a particular efficacy or safety conclusion.

14. Why This Trial Matters Statistically

EPIC-SR is statistically informative because its primary outcome captures when sustained symptom alleviation occurred, not only whether an event was observed at the end of follow-up. That distinction affects how the result should be interpreted: group medians summarize timing, while the log-rank test addresses evidence about the event-time distributions. Neither element alone conveys the full range of participant experiences.

The registry also illustrates why outcome definition and analysis population matter. Sustained alleviation depended on baseline symptom severity and four consecutive qualifying days. The primary population included randomized participants who received at least one dose. Those choices define precisely which event and participants the primary result describes; they are not interchangeable with a simpler symptom status or with all enrolled participants.

For clinical researchers, the record is a useful reminder to separate the primary result from secondary patterns. Secondary outcomes cover a broad range of questions, and their estimates can point in a similar direction without constituting independent confirmation. Without a reported multiplicity plan, missing-data strategy, or full statistical analysis plan, the extent of confirmatory support from those outcomes remains uncertain.

The record also demonstrates the distinction between sponsor analyses and simple re-calculations. Logistic-regression odds ratios and unadjusted risk differences are different statistical quantities. Reading them as though they were interchangeable can misstate the evidence. The primary endpoint’s reported test, its group medians, and its confidence intervals likewise answer related but different questions.

Finally, the termination status and the difference between overall enrollment and primary analysis counts highlight the importance of transparent accounting. The registry does not explain those issues, so their implications cannot be determined from the posted record alone.

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.

Several limitations are specific to the information posted for this trial. The study is listed as terminated, but the reason for termination and its timing relative to the analyses are not reported in the registry record. Without that context, termination cannot be linked to efficacy, safety, recruitment, or another cause.

The primary result provides group medians and confidence intervals, together with a log-rank comparison, but no primary comparative effect estimate such as a hazard ratio or its confidence interval. This limits direct assessment of the size and precision of the between-group time-to-event contrast using a single effect measure. The two separate median intervals are not a substitute for an interval around that contrast.

The primary analysis population is modified intent-to-treat, not all enrolled participants. Although the record reports 1,440 enrolled overall, the primary groups include 654 and 634 participants. The reasons for the difference, and the disposition of participants not represented in those outcome groups, are not reported in the registry record. The record also does not report an outcome-specific missing-data summary or sensitivity analyses.

The available secondary outcomes involve different definitions and analysis approaches. The registry does not report a multiplicity adjustment, a hierarchy of secondary hypotheses, or the full statistical analysis plan. Sponsor-reported logistic-regression estimates are posted for two secondary outcomes, but the registry does not fully detail the model choices. Recalculated unadjusted comparisons are descriptive supplements and do not reproduce those adjusted analyses.

Finally, details such as the reason for the study’s low-risk population designation, recruitment procedures, and original sample-size justification are not reported in the registry record. The sample-size illustrations based on observed effects do not fill that gap, because they are planning examples rather than evidence about the original design or observed power.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Evaluation of Protease Inhibition for COVID-19 in Standard-Risk Patients (EPIC-SR). — NCT05011513.
• Registry record last updated 2023-08-14; status last verified 2023-07.
• Results first posted 2023-08-14.
• 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

EPIC-SR (NCT05011513) was a randomized, quadruple-masked, parallel-arm Phase 2/3 study of oral PF-07321332 with ritonavir versus oral placebo in nonhospitalized symptomatic adults with COVID-19 who were described as at low risk of progressing to severe illness. The registry reports 1,440 enrolled participants and lists the study as terminated.

The primary endpoint was time to sustained alleviation of overall COVID-19 signs and symptoms through Day 28. In the modified intent-to-treat population, the reported median was 12.0 days for Nirmatrelvir 300 mg + Ritonavir 100 mg and 13.0 days for placebo. The registry reports a log-rank comparison; the posted results do not establish a clear between-group difference.

Secondary outcomes cover symptom severity and progression, death, healthcare use, and safety. Their interpretation is limited by the absence of a reported multiplicity plan and detailed missing-data information. The reason for termination and the original sample-size justification are not reported in the registry record.