A Study to Assess the Safety, Tolerability, Pharmacokinetics and Efficacy of EDP-305 in Subjects With Non-Alcoholic Steatohepatitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Assess the Safety, Tolerability, Pharmacokinetics and Efficacy of EDP-305 in Subjects With Non-Alcoholic Steatohepatitis (NCT03421431), a randomized phase 2 study. The primary outcome is Mean Change From Baseline (Average) in Alanine Aminotransferase (ALT) at Week 12.

2. Clinical Question

3. Trial Design

This was a randomized, triple-masked, parallel-group Phase 2 study sponsored by Enanta Pharmaceuticals, Inc. The registry record lists an enrollment of 134 and a completed study. The recorded start date was 2018-04-25, and the primary completion date was 2019-06-14. The study evaluated safety, tolerability, pharmacokinetics, and efficacy of EDP-305 in people with non-alcoholic steatohepatitis.

Participants were assigned to placebo, EDP-305 1 mg, or EDP-305 2.5 mg. The intervention descriptions state that participants in each group took two tablets orally once a day for 12 weeks. The parallel design means each participant was assigned to one group rather than receiving each treatment in sequence. Randomization is intended to create groups that are comparable on average, although chance differences can remain, especially in a trial of this size.

The registry labels masking as triple. The identities of the three masked parties are not reported in the registry record. The original allocation ratio, detailed randomization method, and original sample-size justification are also not reported in the registry record. The record identifies a Phase 2 dose-ranging study, but the information reported does not establish the planned statistical contrasts or whether any other planned analyses were conducted.

The primary outcome was mean change from baseline in ALT at Week 12. Baseline was the average of screening and Day 1 measurements, using the non-missing measurement if either value was missing. The efficacy population was defined as participants who received at least one dose; for each outcome, the analyzed count represents participants evaluable for that measure. Therefore, the outcome-specific analyzed counts need not match the enrollment count. Secondary outcomes included liver fat assessed by MRI-PDFF, APRI, and several lipid and other laboratory measures.

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: The Safety or Efficacy Population consisted of all participants who received at least one dose of the study treatment. Out of 134 randomized participants, 2 participants (1 participant each in the EDP-305 2.5 mg and placebo groups) were randomized but not dosed because the participants withdrew and were not included in the Safety and Efficacy Populations.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Efficacy Population: included all participants who received at least one dose of the study treatment. Here "Overall Number of Participants Analyzed" = participants evaluable for this outcome measure. Only those participants with valid measurements for the outcome measure at the specified time point were reported.

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: Demographics were summarized for the Safety Population, which included all the randomized participants who received at least one dose of the study treatment. The numbers of participants included in the Efficacy Population were same as the Safety Population.

6. Primary Endpoint

The registry lists the primary outcome as Mean Change From Baseline (Average) in Alanine Aminotransferase (ALT) at Week 12, measured over Baseline and Week 12.

The primary endpoint is mean change from baseline in alanine aminotransferase (ALT) at Week 12, measured in units per liter (U/L). ALT is a blood-test measure used in the trial as a laboratory indicator relevant to liver assessment. The registry describes blood samples collected at specific time points for ALT evaluation.

Baseline was defined as the average of the screening and Day 1 values. If one of those measurements was missing, the non-missing value was used. The endpoint’s change score was defined as the Week 12 mean minus the baseline mean. A negative change therefore denotes a lower mean ALT at Week 12 than at baseline; the endpoint itself does not say why the value changed or what that change means for a participant’s health.

The registry reports the mean and standard deviation for each group. The efficacy population included participants who received at least one dose of study treatment, while the number analyzed for this outcome is the number evaluable for ALT at the specified time point. That outcome-specific analysis count is important when interpreting precision and comparing results across measures.

ALT is one laboratory measure, not a direct account of every aspect of NASH. The trial also assessed other outcomes, including MRI-PDFF liver fat and APRI. Statistical differences in a laboratory endpoint should be interpreted as differences in the measured endpoint under the analysis used, not as proof of a broader clinical effect.

7. Results Posted in the Registry

• EDP-305 1 mg: mean -23.76 versus -13.85 on control; difference -9.91 (95% CI -21.16 to 1.34), standardized effect d = -0.39; z-test p = 0.084.
• EDP-305 2.5 mg: mean -26.14 versus -13.85 on control; difference -12.29 (95% CI -25.54 to 0.96), standardized effect d = -0.42; z-test p = 0.069.

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.

Mean Change From Baseline in Percentage of Fat in the Liver as Assessed by Magnetic Resonance Imaging - Proton Density Fat Fraction (MRI-PDFF) at Week 12

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • EDP-305 1 mg: mean -3.76 versus -2.84 on control; difference -0.92 (95% CI -3.33 to 1.50), standardized effect d = -0.18; z-test p = 0.457.
  • EDP-305 2.5 mg: mean -6.43 versus -2.84 on control; difference -3.59 (95% CI -6.41 to -0.76), standardized effect d = -0.54; z-test p = 0.013.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between EDP-305 1 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for EDP-305 2.5 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Mean Change From Baseline in Aspartate Aminotransferase to Platelet Ratio Index (APRI) at Week 12

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • EDP-305 1 mg: mean -0.11 versus -0.18 on control; difference 0.07 (95% CI -0.12 to 0.27), standardized effect d = 0.21; z-test p = 0.458.
  • EDP-305 2.5 mg: mean -0.19 versus -0.18 on control; difference -0.01 (95% CI -0.22 to 0.19), standardized effect d = -0.04; z-test p = 0.892.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between EDP-305 1 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between EDP-305 2.5 mg and the control. The interval gives the range of effects that remain compatible with the data.

Mean Change From Baseline in Triglycerides (TG) at Week 12

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • EDP-305 1 mg: mean -0.37 versus -0.15 on control; difference -0.21 (95% CI -0.80 to 0.37), standardized effect d = -0.15; z-test p = 0.471.
  • EDP-305 2.5 mg: mean 0.13 versus -0.15 on control; difference 0.28 (95% CI -0.23 to 0.79), standardized effect d = 0.27; z-test p = 0.285.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between EDP-305 1 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between EDP-305 2.5 mg and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover several types of measurement, including liver fat by MRI-PDFF, APRI, triglycerides, cholesterol measures, and other laboratory indicators. The listed secondary outcomes include MRI-PDFF, APRI, triglycerides, total cholesterol, HDL-C, LDL-C, adiponectin, and ApoA-1. Numerical results are reported in the registry information for MRI-PDFF, APRI, and triglycerides; numerical results for the other listed measures are not reported in the registry record in the information summarized for these outcomes.

Across the reported secondary results, the clearest contrast is for liver fat in the 2.5 mg group relative to placebo. Its recalculated unadjusted mean difference is -3.586 percentage points, with a 95% confidence interval from -6.408 to -0.764, and the sponsor’s ANCOVA estimate is 4.717 in the opposite contrast order, with a 95% confidence interval from 1.980 to 7.455. The corresponding 1 mg comparison has an interval that includes zero in both the recalculation and sponsor analysis. For APRI and triglycerides, the sponsor-reported comparisons with placebo have confidence intervals that include zero. These outcomes therefore do not show the same pattern of a clearly separated adjusted interval as the higher-dose MRI-PDFF comparison.

The ALT result and the secondary pattern are related as different measurements collected in the same trial, not as interchangeable evidence. ALT is the primary laboratory endpoint; MRI-PDFF assesses liver fat, while APRI and triglycerides represent other measures. A secondary finding can add context, but it does not replace the primary analysis or prove that changes across endpoints share a common mechanism.

There are many secondary outcomes and analyses in the registry record. The reported multiplicity approach is not reported in the registry record. When many outcomes or contrasts are tested, some results may appear notable by chance even if there is no underlying effect. Unless the testing plan and adjustment are known, unadjusted secondary p-values carry less weight than a prespecified primary analysis. The pattern is useful for generating a fuller statistical picture, but it should be treated cautiously rather than as a collection of independent confirmations.

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

• Mean Change From Baseline in Percentage of Fat in the Liver as Assessed by Magnetic Resonance Imaging - Proton Density Fat Fraction (MRI-PDFF) at Week 12
• Mean Change From Baseline in Aspartate Aminotransferase to Platelet Ratio Index (APRI) at Week 12
• Mean Change From Baseline in Triglycerides (TG) at Week 12
• Mean Change From Baseline in Total Cholesterol at Week 12
• Mean Change From Baseline in High-Density Lipoprotein Cholesterol (HDL-C) at Week 12
• Mean Change From Baseline in Low-Density Lipoprotein Cholesterol (LDL-C) at Week 12
• Mean Change From Baseline in Adiponectin at Week 12
• Mean Change From Baseline in Apolipoproteins A1 (ApoA-1) at Week 12

9. Statistical Interpretation

10. Statistical Methods Explained

Continuous outcomes such as ALT are commonly described by a mean and a measure of spread. The mean gives the average observed change within a group; the standard deviation describes how much individual measurements vary around that average. A standard deviation is not the same as the uncertainty in the group mean or in a contrast between groups. It describes participant-level variability, while a confidence interval for a contrast describes uncertainty in the estimated difference.

What does a change score represent?

For ALT, the registry defines change as the Week 12 mean minus the baseline mean. Baseline uses the average of screening and Day 1 values, with the single non-missing value used if one of those measurements is absent. A negative group mean change means that the average Week 12 value is below the average baseline value. The difference between groups then asks whether those average changes differ. A within-group decline is not by itself evidence that the treatment group changed differently from placebo, because the placebo group also had a negative mean change.

What is the recalculated comparison?

The unadjusted recalculation compares the posted group summary statistics without covariate adjustment. It gives an estimated mean difference, a confidence interval, and an unadjusted p-value. The recalculated ALT comparisons use the observed means and standard deviations and are not the same analysis as the registry’s ANCOVA. The values are useful as a transparent summary-statistics comparison, but they do not reproduce the sponsor’s adjusted model. Because the registry’s reported ANCOVA results differ from these recalculations, the sponsor’s own analysis takes precedence.

What does ANCOVA add?

ANCOVA, or analysis of covariance, estimates group differences while adjusting for covariates in the model. In a randomized trial, adjustment can account for specified baseline or other characteristics and may improve precision. The registry identifies ANCOVA and reports least-squares mean differences, but the covariates, model details, and missing-data handling for this analysis are not reported in the registry record. Without those details, the adjusted estimate cannot be independently reconstructed from the information. The least-squares mean difference is a model-based contrast, not simply the subtraction of the two displayed group means.

How should the p-value be read?

A p-value describes how compatible the observed data are with a specified null model, under the assumptions of the analysis. A smaller p-value is not a measure of effect size, clinical importance, or the probability that the null hypothesis is true. The recalculated p-values are explicitly unadjusted. The registry’s ANCOVA p-values belong to the sponsor’s reported analysis. The original statistical testing plan and any multiplicity adjustment are not reported in the registry record, so a single p-value should not be interpreted as resolving all uncertainty about the trial’s results.

Why do confidence intervals matter?

A confidence interval gives a range of estimates compatible with the observed data under the model and its assumptions. Its width reflects precision: wider intervals indicate less precision, while narrower intervals indicate greater precision. For an estimated mean difference, an interval that includes zero does not rule out no difference under that analysis. An interval that excludes zero indicates that zero is outside the interval, but this does not establish that the size of the effect matters clinically. Interpretation also depends on the analysis plan, the outcome’s meaning, the quality of measurement, and whether multiple comparisons were considered.

Why can variability affect precision?

General statistical principle: greater variability among participants makes it harder to estimate a group mean or contrast precisely when the analyzed sample size is held constant. Standard deviations describe that variability. Unequal analyzed counts can also affect the uncertainty of a contrast. Over-dispersion is a general concern when a model assumes less variability than the data contain; it can make standard errors too small and intervals too narrow if not handled appropriately. The registry information listed for this analysis does not establish that over-dispersion occurred or that it was relevant to the continuous ALT endpoint.

What does the analyzed population mean?

The registry defines the efficacy population as participants who received at least one dose of study treatment. For each outcome, the reported analyzed count is the number evaluable for that specific measure and time point. Consequently, the number analyzed can vary by outcome and can be smaller than the overall enrollment. The rules for missing measurements beyond the baseline convention, and the details of any imputation or sensitivity analysis, are not reported in the registry record. This matters because analyses based only on participants with valid observed measurements can represent a different subset than all randomized participants.

How do design and analysis relate?

Randomization supports comparisons between assigned groups, while masking can reduce the influence of expectations on care, measurement, or reporting. Neither feature removes all uncertainty or guarantees that the analyzed groups are identical. The study used a parallel-group design and placebo comparator, but the allocation ratio and specific randomization procedures are not reported in the registry record. The posted analysis counts are outcome-specific, so precision should be assessed from the participants analyzed for each outcome rather than from enrollment alone.

11. Confidence Intervals and Precision

The recalculated ALT comparisons have broad confidence intervals relative to their point estimates. For EDP-305 1 mg versus placebo, the estimated mean difference was -9.91 U/L, with a 95% confidence interval from -21.16 to 1.343. For EDP-305 2.5 mg versus placebo, it was -12.29 U/L, with a 95% confidence interval from -25.54 to 0.960. Each interval includes zero, so the unadjusted summary-statistics comparisons do not precisely distinguish the estimated difference from no difference.

The interval bounds also show that the data remain compatible with effects of different magnitudes. The point estimate is only one value within a range of plausible values under the analysis. Readers should not treat the point estimate as the exact effect or interpret a confidence interval as containing every possible true value. A 95% confidence interval is a procedure-based uncertainty summary: in repeated use under its assumptions, intervals constructed in this way would cover the true parameter at the stated rate.

The sponsor’s ANCOVA intervals are distinct estimates. The placebo versus 2.5 mg interval runs from 0.029 to 24.891 U/L, while the placebo versus 1 mg interval runs from -5.754 to 18.268 U/L. The adjusted interval for the higher dose is narrowly separated from zero, whereas the interval for the lower dose includes zero. Because the adjusted estimates differ from the recalculations, these intervals should be interpreted as results from different analyses, with the sponsor’s analysis taking precedence.

Precision is limited by the outcome-specific analyzed counts, variability, and the width of the resulting intervals. The registry lists 20 placebo participants evaluable for the primary outcome, 48 for EDP-305 1 mg, and 38 for EDP-305 2.5 mg. These counts are not the same as the number enrolled in the trial. The registry does not report a clinically important difference for ALT, so statistical precision cannot be translated into a judgment about whether the range is clinically meaningful.

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 participants per arm, not total events. They illustrate how many participants per arm would be required under specified planning assumptions for a two-group comparison. The assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set as a stated fraction of the observed effect. These are planning illustrations based on the observed data, not the trial’s own original design calculations; the original design assumptions are not reported in the registry record.

The 80% power column means the planned design would have an 80% chance of rejecting the null hypothesis under the assumed effect and model, if those assumptions hold. The 90% power column represents a higher target probability under the same kind of assumptions. Neither column reports the probability that the observed trial result is correct. These figures are not observed power.

The rows describe planning for 100%, 75%, or 50% of the observed effect. A smaller effect is harder to distinguish from random variation, so detecting a smaller effect generally requires more participants. The planning numbers can therefore rise substantially when the target is reduced from the full observed effect to a fraction of it. Such figures should not be read as evidence that the observed effect will recur, or as a recommendation for a future trial’s size. They are conditional illustrations; an appropriate future design would require a justified target effect, variance assumptions, a defined analysis plan, and consideration of multiplicity and attrition. The registry record does not report the original sample-size justification, so the posted planning calculations cannot be used to infer why the completed study enrolled its participants.

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

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -9.910; pooled standard deviation σ = 25.662.
2. n = 2 × (1.96 + 0.842)² × 25.662² / -9.910² = 105.3, rounded up to 106 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
delta <- -9.91                      # observed difference in means
sigma <- 25.6621091370845                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3421431)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 106 (80% power), 141 (90% power)
 75% of observed effect: n per arm = 188 (80% power), 251 (90% power)
 50% of observed effect: n per arm = 422 (80% power), 564 (90% power)
power.t.test: 106.2 
Simulated power at n = 106 per arm: 0.802 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-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.

14. Analyses Reported by the Sponsor

The registry’s sponsor-reported primary analyses use ANCOVA and report least-squares mean differences. For placebo compared with EDP-305 2.5 mg, the estimate was 12.460 U/L, with a 95% confidence interval from 0.029 to 24.891. For placebo compared with EDP-305 1 mg, the estimate was 6.257 U/L, with a 95% confidence interval from -5.754 to 18.268. The registry also reports an ANCOVA comparison between the two EDP-305 dose groups.

The recalculations are unadjusted comparisons based on posted summary statistics. Their contrast is expressed as dose group minus placebo, while the sponsor’s reported placebo-versus-dose contrast is ordered in the opposite direction. This reverses the sign convention, but the numerical estimates and uncertainty also differ because the sponsor used ANCOVA rather than the simple unadjusted calculation.

The sponsor’s own analysis takes precedence when describing the trial’s reported inferential results. The registry identifies the method but does not report the covariates and full model details needed to reproduce it. The difference between the adjusted and unadjusted estimates therefore cannot be attributed to a specific modeling choice from the registry record alone. These analyses should be distinguished rather than blended or treated as interchangeable. The sponsor’s reported adjusted result is the primary reference for the trial’s stated analysis, while the recalculation is a supplementary summary-statistics check.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates several outcomes, treatment contrasts, time points, or analysis variants. Each additional statistical test creates another opportunity for a result to appear notable by chance. This does not make every individual result invalid, but it changes how a set of p-values should be interpreted. A prespecified primary endpoint and a defined hierarchy or adjustment can help control the interpretation of multiple tests.

The registry lists a primary ALT outcome, a range of secondary outcomes, and many analyses. However, the planned multiplicity strategy, including whether any adjustment was applied, is not reported in the registry record. The registry information also does not establish which analyses were prespecified, whether any interim analyses occurred, or how the reported outcomes fit into a testing hierarchy. Those details should not be inferred from the reported estimates or p-values.

Missing-data methods are also not fully described. For the primary endpoint, baseline is the average of screening and Day 1 values, using the available value when one is missing. That is a specific baseline rule, not a complete description of how missing Week 12 outcomes or other missing measurements were handled. Beyond the outcome-specific evaluability definition, the imputation approach, sensitivity analyses, and assumptions about missingness are not reported in the registry record.

The efficacy population included participants who received at least one dose, and each outcome’s analyzed count represents participants evaluable for that measure at the specified time. The registry does not provide enough information to determine how those counts relate to all randomized participants, whether exclusions occurred after randomization, or how missing results might influence estimates. Analyses restricted to evaluable participants may be vulnerable to bias if the reason measurements are missing is related to treatment or outcome.

These reporting gaps limit independent assessment of the planned statistical framework. They do not establish that the analysis was inappropriate; they mean that conclusions about multiplicity control, missing-data assumptions, or unreported analyses cannot be drawn from the registry record.

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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: cerebrovascular accident (4.2%), asthma (1.8%).

The serious adverse event figures are reported for participants at risk in each group. The registry lists 24 at risk in the placebo group, 55 in the EDP-305 1 mg group, and 53 in the EDP-305 2.5 mg group. One participant in placebo and one participant in the 1 mg group were reported with a serious adverse event; none were reported in the 2.5 mg group. The registry lists zero deaths in each group.

The serious adverse event denominators differ from the primary efficacy outcome counts. These are distinct reporting populations and should not be treated as if they were the same analysis set. The registry also lists cerebrovascular accident at 4.2% and asthma at 1.8% among the top serious adverse event terms. Those percentages are reported terms, not figures that can be assigned to a particular treatment arm based on the information.

Counts in a study of this size describe what was recorded during the trial; they do not establish whether a treatment caused an event or provide a precise estimate of uncommon risks. The registry information does not give event narratives or a causal assessment for the listed serious events. Safety comparisons should therefore be read descriptively, alongside the reported denominators and the limited follow-up duration.

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

The results describe changes in measured outcomes over the trial period and differences between randomized groups under the analyses reported. Mean ALT declined from baseline in all three groups. The sponsor’s ANCOVA analysis estimated a placebo-versus-EDP-305 2.5 mg difference with a confidence interval just above zero, while the corresponding adjusted interval for EDP-305 1 mg included zero. The unadjusted recalculations have intervals that include zero for both dose comparisons, illustrating that the analytic method affects the estimated contrasts and their uncertainty.

These results do not establish that EDP-305 does or does not work. A statistical contrast in ALT is not itself a direct measure of how participants feel, function, or experience a health outcome. The registry record does not report a threshold for a clinically important ALT difference, so the clinical meaning of the estimated change cannot be determined from these statistics alone. The trial also evaluated other outcomes, but multiple measurements should not be treated as interchangeable evidence.

The registry describes a Phase 2 study with 12 weeks of treatment. The reported results therefore concern a limited period and a particular set of measured outcomes. They do not establish longer-term effects, nor do they answer questions not measured in this study. The registry record does not report all details needed to assess the original analysis plan, missing-data strategy, or multiplicity handling.

Confidence intervals are essential to interpretation. The point estimates give the center of the observed comparisons, but the intervals express uncertainty around them. The sponsor’s interval for the 2.5 mg comparison is narrowly separated from zero, while the recalculated interval includes zero. This should prompt attention to the distinction between adjusted and unadjusted analyses rather than selecting whichever estimate appears more favorable. For the trial’s reported analysis, the sponsor’s ANCOVA results take precedence.

Safety findings are also descriptive. The recorded serious adverse events and deaths are useful parts of the trial record, but event counts from a study with these denominators cannot settle questions about rare events or causality. A balanced reading keeps the endpoint, analysis method, precision, reporting limitations, and duration in view without converting statistical results into medical advice.

18. Why This Trial Matters Statistically

This trial illustrates why a reported group mean and a treatment comparison answer different questions. Each group’s mean ALT change was negative, but the treatment question concerns the difference in change between an EDP-305 group and placebo. The placebo group’s change provides context for interpreting the changes observed in the dose groups. Looking only at within-group movement would not isolate a treatment contrast.

It also demonstrates the importance of distinguishing adjusted from unadjusted analyses. The sponsor reported ANCOVA estimates, while the recalculations compare summary statistics without adjustment. Their estimates and intervals differ. Reporting both can help readers understand the sensitivity of conclusions to the analysis, but it does not make the two methods equivalent. The registry’s adjusted results are the sponsor’s own analysis and take precedence for the trial’s reported inference.

Precision matters as much as the point estimate. The ALT recalculation intervals are broad and include zero, while the sponsor’s interval for the higher dose is close to the zero boundary. This makes it especially important not to reduce the interpretation to a binary significant-or-not label. The size and range of effects compatible with the data, the outcome’s clinical meaning, and the planned analysis all matter.

The trial also highlights the challenge of interpreting multiple endpoints. Liver fat, ALT, APRI, lipids, and other measures describe different aspects of the study’s assessment. Without a reported multiplicity strategy, secondary results should be treated as less definitive than the prespecified primary analysis. Finally, outcome-specific analyzed counts and incomplete reporting of missing-data methods matter for understanding who contributed to each estimate and how robust the conclusions are.

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 primary outcome analyses include different numbers of evaluable participants by group: 20 for placebo, 48 for EDP-305 1 mg, and 38 for EDP-305 2.5 mg. These counts are smaller than the overall enrollment and are specific to the endpoint. The registry defines analyzed participants as those with valid measurements at the specified time point, but does not report the full reasons that outcome data were unavailable or the missing-data methods used beyond the baseline convention.

The sponsor’s ANCOVA estimates cannot be reconstructed from the listed information because the covariates and full model specification are not reported in the registry record. The recalculated comparisons are unadjusted and differ from the sponsor’s reported estimates. This makes it important to keep the analysis types separate and to give precedence to the sponsor’s own analysis for the reported trial inference.

The registry does not report the original sample-size justification, allocation ratio, full randomization method, or the planned multiplicity approach. It also does not report the details needed to identify any interim analyses or distinguish all prespecified from exploratory analyses. With several outcomes and comparisons, the absence of this information constrains interpretation of secondary p-values.

The primary endpoint is a laboratory measure observed at Week 12. The registry does not give a clinically important difference for ALT, and these results do not establish longer-term outcomes. The serious adverse event counts are descriptive, with denominators that differ from the primary efficacy counts; they do not allow a reliable assessment of uncommon harms or causal attribution. These are limitations of what can be concluded from the reported record, not evidence that unreported procedures were or were not undertaken.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Assess the Safety, Tolerability, Pharmacokinetics and Efficacy of EDP-305 in Subjects With Non-Alcoholic Steatohepatitis — NCT03421431.
• Registry record last updated 2021-09-16; status last verified 2021-08.
• Results first posted 2021-09-16.
• 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

NCT03421431 is a completed, randomized, triple-masked, parallel-group Phase 2 study sponsored by Enanta Pharmaceuticals, Inc. It evaluated EDP-305 1 mg and 2.5 mg against placebo in participants with non-alcoholic steatohepatitis. Each group took two tablets orally once a day for 12 weeks. The registry lists enrollment of 134.

The primary outcome was mean change from baseline in ALT at Week 12. The sponsor’s ANCOVA analysis reported a placebo-versus-2.5 mg estimate with a confidence interval narrowly above zero; the placebo-versus-1 mg interval included zero. Simple unadjusted recalculations differ, so the sponsor’s own analysis takes precedence. The registry also reports secondary endpoint and safety information, with outcome-specific analyzed counts. Details of the original sample-size justification, multiplicity strategy, and full missing-data methods are not reported in the registry record.