Camostat Efficacy vs. Placebo for Outpatient Treatment of COVID-19 (CAMELOT): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Camostat Efficacy vs. Placebo for Outpatient Treatment of COVID-19 (CAMELOT) (NCT04583592), a randomized phase 2 study. The primary outcome is Disease Progression at Day 28.

2. Clinical Question

The trial is registered as NCT04583592 and is titled CAMELOT. It addresses a short-term clinical outcome in an outpatient population at risk of severe illness. The outcome combines hospitalization, including an emergency room visit, with death from any cause; it does not separately estimate each component in the posted primary result.

3. Trial Design

The registry record describes CAMELOT as a Phase 2, randomized, double-blind, placebo-controlled, parallel-group trial. Its masking designation is quadruple. The study was sponsored by Sagent Pharmaceuticals Inc. The registry lists enrollment of 295 and a completed status. The recorded start date is 2020-11-09, and the primary completion date is 2021-03-31.

Participants were ambulatory patients with confirmed COVID-19 and at least one risk factor for severe illness. One group received camostat mesilate for 14 days alongside standard of care treatment; the comparator group received placebo for 14 days alongside standard of care treatment. Random allocation and a parallel-group design support a direct comparison of the assigned groups, while masking is intended to limit the influence of treatment knowledge on trial conduct or assessment. The registry record does not report further detail about who was masked or how masking was maintained.

The primary outcome was disease progression at Day 28, defined as hospitalization, including an emergency room visit, or death due to any cause within 28 days of randomization. The posted results give counts for the two groups and identify the Intent-to-Treat population as all randomized subjects. The registry lists 16 outcome measures and one reported analysis. Listed secondary outcomes include survival rate, time to fever resolution, resolution of viral shedding, adverse events and serious adverse events, discontinuation, and laboratory measures including platelet count and potassium level. A complete account of how each outcome was assessed is not reported in the registry record.

The registry record does not report the trial's original sample-size assumptions, its detailed allocation procedures, or whether any interim analyses were planned or performed. Those details should not be inferred from the posted outcome data or from the separate planning illustrations.

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: The Intent-to-Treat population includes all subjects who 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: The Intent-to-Treat population includes all subjects who were randomized. \[Note: Subjects may have more than one risk factor upon entry into study.\]

6. Primary Endpoint

The registry lists the primary outcome as Disease Progression at Day 28, measured over 28 days.

The primary endpoint, disease progression at Day 28, is a binary outcome: each participant either met the definition or did not. The registry defines an event as requiring hospitalization, including an emergency room visit, or dying due to any cause within 28 days of randomization. The reported group values are counts of participants with an event.

A combined endpoint can capture more than one clinically important route to worsening while keeping the primary analysis focused on a single yes-or-no outcome. It also means the combined result does not reveal whether any group difference came from hospitalization, emergency room visits, or deaths. The posted primary result does not report those components separately.

Using a fixed follow-up window makes the outcome interpretable as progression during the first 28 days after randomization. In general, binary endpoints are often summarized by comparing the proportion with an event in each group. That comparison should be interpreted alongside its uncertainty: counts alone do not show how precisely the underlying group difference is estimated.

The registry identifies the primary analysis population as the Intent-to-Treat population, comprising all randomized subjects. The record does not report additional endpoint-adjudication details or how participants with incomplete follow-up were handled.

7. Results Posted in the Registry

• Camostat Mesilate: 5.2% (10/194) versus 5.9% (6/101) on control; risk difference -0.8 percentage points (95% CI -6.3 to 4.8), risk ratio 0.87 (0.32–2.32); unadjusted two-proportion z-test p = 0.777.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary because the endpoint records whether each participant met the disease-progression definition by Day 28. The posted group data give event counts and group sizes. For a binary outcome, a group event proportion describes the frequency of participants meeting the definition relative to the group size. Comparing those proportions gives an interpretable contrast, but the contrast remains subject to sampling uncertainty.

What does the risk difference express?

The re-calculation uses a risk difference, comparing the event proportion in the camostat group with that in the placebo group. A risk difference is an absolute contrast: zero represents equal event proportions, a negative value in this comparison points toward a lower observed proportion in the camostat group, and a positive value points toward a higher one. The estimate is -0.008. Its confidence interval runs from -0.064 to 0.048, crossing zero. That crossing means the interval includes no difference as well as possible differences in either direction. It does not prove that there is no effect.

Risk differences are one useful way to summarize binary outcomes, but they are not the only possible measure. The registry analysis reports a chi-squared method, while the re-calculation reports the risk difference and also includes relative measures. Different measures answer related but distinct questions: an absolute contrast describes the difference in event proportions, while a relative measure compares them in relation to one another. No single measure removes the need to consider uncertainty and the study context.

What does the chi-squared test assess?

In general, a chi-squared test assesses whether observed categorical counts are compatible with a null hypothesis of no association between group and outcome. In this trial, the posted analysis lists a chi-squared method for the two treatment groups and reports p = 0.787. The p-value describes how unusual results at least as discrepant as those observed would be under the test's null model and assumptions. It is not the probability that the null hypothesis is true, and it does not quantify clinical importance.

The re-calculation is an unadjusted analysis of posted summary statistics and gives p = 0.777. It is not a replacement for an analysis conducted by the sponsor using the full trial data and its prespecified procedures. The registry record does not report enough detail to determine all analysis choices behind the sponsor's result, including any covariate adjustment. The sponsor's analysis is therefore the reported trial analysis, while the re-calculation offers a separate summary-statistic check.

Why does the confidence interval matter alongside the p-value?

In general, a confidence interval describes a range of effect estimates compatible with the data and the method used to construct it. Its width conveys precision: a wider interval permits a broader range of underlying effects, while a narrower interval is more constraining. Here, the risk-difference interval includes both negative and positive values. The p-value and interval tell a consistent broad story: the posted data do not sharply distinguish a difference from no difference, and the estimate is uncertain.

A confidence interval is not a guarantee that the true effect lies within its endpoints, nor is it a range of outcomes for individual participants. Its interpretation relies on the statistical procedure and assumptions used to calculate it. The registry record does not report the confidence-interval method for the sponsor's chi-squared analysis; the numerical interval belongs to the unadjusted re-calculation.

What does a binary-count analysis leave out?

Turning the endpoint into an event count makes the main comparison straightforward, but it does not show when an event occurred within the follow-up window or how severe the individual courses were. The composite endpoint also combines hospitalization, emergency room visits, and death due to any cause. The posted primary result does not separate those components, and a count-based analysis does not describe outcomes beyond the defined 28-day period.

General statistical caution also applies when event counts are limited: estimates may remain imprecise even when group sizes are larger. This is one reason to interpret the effect estimate, interval, endpoint definition, and p-value together rather than treating a single test result as decisive. The registry record identifies the primary population as all randomized subjects but does not report missing-data procedures, additional analysis populations, or details of any adjusted model. Those matters therefore cannot be assumed.

10. Confidence Intervals and Precision

The re-calculated risk difference is -0.008, with a confidence interval from -0.064 to 0.048. The interval includes zero and extends to values on both sides of it. In practical statistical terms, the posted summary data do not pin down the direction of the underlying difference with precision. They are compatible with a lower event proportion in the camostat group, no difference, or a higher event proportion.

In general, a confidence interval reflects both the observed estimate and the amount of information available under the calculation method. Its width is a useful guide to precision, not a measure of whether an effect is important. A wide interval can include effects that might matter in either direction, even when the point estimate lies near zero. The posted event counts are 10 in the camostat group and 6 in the placebo group, so the primary comparison is based on relatively few observed events.

The interval is from an unadjusted re-calculation using posted summary statistics, not a confidence interval reported for the sponsor's chi-squared test. The sponsor reports p = 0.787, while the re-calculation gives p = 0.777. The registry record does not report a sponsor confidence interval for the primary effect. The numerical interval should therefore be read as a description of the re-calculated risk difference rather than as a full substitute for the sponsor's analysis or its underlying records.

Precision is also specific to the chosen endpoint and follow-up. The estimate concerns the combined Day 28 outcome of hospitalization, including an emergency room visit, or death due to any cause. It does not separately quantify precision for each component or for outcomes after that period.

11. 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 calculation. They use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Each row asks what sample size would be needed to detect a stated fraction of the observed effect under those assumptions. The figures are participants per arm, not total events.

For planning at 100% of the observed effect, the illustration gives 13,315 participants per arm for 80% power and 17,824 participants per arm for 90% power. For planning at 75% of the observed effect, it gives 24,066 participants per arm for 80% power and 32,216 participants per arm for 90% power. For planning at 50% of the observed effect, it gives 55,034 participants per arm for 80% power and 73,674 participants per arm for 90% power.

In plain language, 80% power means that, if the assumed effect and calculation assumptions were true, the design would have an 80% chance of detecting an effect using the stated significance rule. The 90% power column expresses the same idea with a higher chance of detection under those assumptions. These are planning illustrations, not observed power.

The pattern follows a general principle: smaller effects are harder to distinguish from random variation, so detecting them requires more participants. The trial's own original sample-size assumptions and justification are not reported in the registry record. The illustration should not be mistaken for evidence about the trial's intended design or for a recommendation about a future study.

12. 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

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.059; treatment proportion p₁ = 0.052; average p̄ = 0.055.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.3237 = 0.6345.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.3237 = 0.2724.
4. n = (0.6345 + 0.2724)² / (-0.0079)² = 13,314.5, rounded up to 13,315 participants per arm 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
p0 <- 0.0594059405940594                          # control proportion
p1 <- 0.0515463917525773                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(4583592)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 13,315 (80% power), 17,824 (90% power)
 75% of observed effect: n per arm = 24,066 (80% power), 32,216 (90% power)
 50% of observed effect: n per arm = 55,034 (80% power), 73,674 (90% power)
power.prop.test: 13314.5 
Simulated power at n = 13,315 per arm: 0.798 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Analyses Reported by the Sponsor

The registry reports a sponsor analysis using a chi-squared method, with p = 0.787 for the primary comparison. The unadjusted re-calculation from posted summary statistics gives p = 0.777 and estimates the risk difference as -0.008, with a confidence interval from -0.064 to 0.048. The registry record does not report a sponsor estimate or confidence interval for the risk difference.

The re-calculation is a simple check using the posted counts and group sizes. It is not the sponsor's analysis and may not reproduce every detail of the trial's statistical procedures. The sponsor's reported chi-squared result is the trial analysis to cite when describing the registry's reported test. The two p-values are close in their broad interpretation: neither indicates a clear difference under the test used. They are not identical, and the re-calculation should not be presented as though it were the sponsor's result.

The registry record does not report a complete statistical analysis plan or enough information to reconstruct every analytic decision. Consequently, interpretation should distinguish the sponsor's reported result from the separate unadjusted calculation, rather than treating them as interchangeable estimates from one procedure.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists 16 outcome measures, and the posted record includes 15 secondary outcomes. The named secondary outcomes cover survival, fever resolution, viral shedding, adverse events, discontinuation, and laboratory measures including platelet count and potassium level. Considering multiple outcomes creates a multiplicity issue: when many comparisons are examined, the chance of at least one apparently unusual result can increase. In general, prespecified priorities and appropriate multiplicity procedures help readers interpret a family of tests.

The registry record does not report whether any multiplicity adjustment was planned or used for secondary outcomes, nor does the posted primary result describe an adjustment across outcomes. No conclusion about such procedures should be inferred from the number of outcomes alone. The sponsor's reported primary chi-squared p-value is the one posted analysis identified for the comparison; the record lists one reported analysis.

The primary population is described as the Intent-to-Treat population, comprising all randomized subjects. The registry record does not report detailed missing-data handling, whether all randomized participants contributed to every outcome, or whether additional analysis populations were used. It also does not report how participants with incomplete follow-up were treated in the primary endpoint analysis.

The original sample-size assumptions, justification, and any interim analyses are not reported in the registry record. The planning illustrations based on the observed effect do not establish what the trial was designed to detect. Without the original plan and detailed analysis methods, the registry results cannot answer whether particular choices were prespecified or how unreported missingness or interim monitoring may have been handled.

These omissions limit the level of methodological detail that can be assessed from the registry entry. They do not establish that a procedure was absent; they mean that the relevant procedure is not reported in the registry record.

15. 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: pneumonia (2.6%), fatigue (2.0%), oxygen saturation decreased (1.0%), acute respiratory distress syndrome (1.0%), dyspnea (1.0%).

The posted serious adverse event figures give 10 affected participants among 195 at risk in the camostat group and 5 among 99 at risk in the placebo group. These are counts and denominators for the safety information, not the primary outcome analysis. The registry record also reports 1 death in the camostat group and 0 in the placebo group. Those figures do not by themselves establish that an event or death was caused by study treatment.

The listed serious adverse event terms include Pneumonia at 2.6%, Fatigue at 2.0%, Oxygen saturation decreased at 1.0%, Acute Respiratory Distress Syndrome at 1.0%, Dyspnea at 1.0%, and Hypoxia at 1.0%. The posted percentages are descriptive entries; the registry record does not provide further context in the registry record to interpret them as treatment-related risks or to establish how the terms were distributed between groups.

Safety counts should be read with their stated at-risk groups and definitions. They are not directly interchangeable with the randomized group sizes used for the primary endpoint. The registry record does not report further details about attribution, duration, or the assessment process for these events. These figures describe what was posted and do not establish a causal safety comparison.

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

The primary result is an uncertain comparison of Day 28 disease progression between randomized outpatient groups. The event counts were 10 in the camostat group and 6 in the placebo group. The unadjusted risk-difference estimate points toward a lower observed event proportion in the camostat group, but its confidence interval crosses zero and permits values in either direction. The sponsor's chi-squared analysis reports p = 0.787; the separate unadjusted calculation reports p = 0.777.

These results do not provide clear statistical evidence of a difference in the primary endpoint. They also do not demonstrate that the treatments are equivalent or that meaningful differences are ruled out. Failure to show a clear difference is not the same as proving that there is no effect. The width and direction of the re-calculated interval are important: the data leave substantial uncertainty about the size and direction of the underlying contrast.

The result is specific to the registry-defined population, treatment period, comparator, and combined endpoint. Participants were ambulatory patients with confirmed COVID-19 and at least one risk factor for severe illness. The intervention and placebo were given for 14 days alongside standard of care treatment, and the primary outcome covered 28 days after randomization. The result does not separately answer whether the groups differed for hospitalization, emergency room visits, or death, because those components are combined in the posted primary measure.

The findings do not establish effects in other patient groups, with different treatment schedules, or over longer follow-up. Nor do they support a causal interpretation of the posted safety counts. The registry record provides a limited view of the statistical plan and missing-data procedures, so conclusions should remain tied to the results and definitions that are reported.

17. Why This Trial Matters Statistically

This trial illustrates why an observed difference and a statistically clear difference are not the same thing. The primary event counts differ between the groups, but the re-calculated risk-difference interval includes no difference and values in either direction. Looking only at which group has the larger count would ignore the group sizes and the uncertainty around the comparison.

It also shows why the endpoint definition matters. Disease progression combines hospitalization, including an emergency room visit, with death due to any cause during a specified follow-up window. That provides one primary binary outcome, but the combined result cannot reveal which component accounts for the observed pattern. A statistical summary is only as specific as the outcome it summarizes.

The distinction between the sponsor's analysis and an unadjusted re-calculation is important for transparent interpretation. The posted sponsor analysis uses a chi-squared method and reports p = 0.787. The separate calculation supplies another way to describe the summary data, including an estimated risk difference and its uncertainty. Keeping those analyses distinct avoids implying that the re-calculation reproduces every detail of the trial analysis.

Finally, planning illustrations based on an observed effect are not the trial's original design rationale and are not observed power. They demonstrate the general relationship between the assumed effect size and the sample size needed under stated assumptions. The registry record's missing details about original design assumptions, multiplicity, and missing-data handling also show why statistical interpretation should distinguish what is reported from what remains unknown.

18. 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 primary comparison is based on 10 events in the camostat group and 6 in the placebo group. The re-calculated confidence interval crosses zero and covers effects in both directions, limiting precision about the underlying difference. The registry record reports the primary endpoint as a composite of hospitalization, including an emergency room visit, or death due to any cause. It does not provide separate primary results for those components.

The population is defined as ambulatory patients with confirmed COVID-19 and at least one risk factor for severe illness. This limits how broadly the result can be interpreted beyond that study population. The treatment period was 14 days, and the primary outcome window was 28 days after randomization; the posted primary result does not describe outcomes beyond that window.

The registry record does not report the original sample-size justification, detailed statistical analysis plan, missing-data procedures, multiplicity adjustment, or interim analysis plans. It identifies the Intent-to-Treat population for the primary outcome but does not report further detail about incomplete follow-up or other analysis populations. These gaps restrict assessment of how the reported result relates to the full planned analysis.

Safety information is also limited to posted counts, denominators, deaths, and listed serious adverse event terms and percentages. The registry record does not provide further attribution details in the registry record, so the counts should not be interpreted as establishing causation. The completed status and enrollment do not resolve these limitations in the information available for interpretation.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Camostat Efficacy vs. Placebo for Outpatient Treatment of COVID-19 (CAMELOT) — NCT04583592.
• Registry record last updated 2022-01-18; status last verified 2022-01.
• Results first posted 2022-01-18.
• 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.

22. Record Summary

NCT04583592, CAMELOT, is a completed Phase 2 randomized, double-blind, placebo-controlled, parallel-group trial sponsored by Sagent Pharmaceuticals Inc. It enrolled 295 ambulatory patients with confirmed COVID-19 and at least one risk factor for severe illness. Camostat mesilate or placebo was given for 14 days alongside standard of care treatment.

The primary endpoint was disease progression by Day 28, defined as hospitalization, including an emergency room visit, or death due to any cause. The posted counts were 10 in the camostat group and 6 in the placebo group. The sponsor's chi-squared analysis reports p = 0.787. The unadjusted re-calculation estimates a risk difference of -0.008, with a confidence interval from -0.064 to 0.048 and p = 0.777. The interval includes no difference and values in both directions, so the primary comparison remains uncertain.