PRV-015 in Gluten-free Diet Non-responsive Celiac Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for PRV-015 in Gluten-free Diet Non-responsive Celiac Disease (NCT04424927), a randomized phase 2 study. The primary outcome is Absolute Change From Baseline in Celiac Disease Patient-Reported Outcome Abdominal Symptoms Domain Score Through Week 24.

2. Clinical Question

3. Trial Design

This was a Phase 2b, randomized, double-blind, placebo-controlled, parallel-group study of PRV-015 in adults with non-responsive celiac disease following a gluten-free diet. The registry record lists quadruple masking. Participants were assigned to placebo or one of three PRV-015 dose groups: 100 mg, 300 mg, and 600 mg. Both PRV-015 and placebo are described as sterile solutions for subcutaneous administration. The study was designed to evaluate efficacy and safety, with PRV-015 used as an adjunct to the diet.

The registry lists enrollment of 388 and a completed status. The study start date is 2020-08-24 and the primary completion date is 2024-07-30. These registry entries describe the study as recorded; the enrollment figure should not be treated as the analysis population for every outcome. For the primary outcome, the registry specifies a modified intent-to-treat population that included randomized participants who received at least one dose of double-blind treatment, and it reports only participants with data at baseline and through Week 24.

The posted primary-outcome arm counts are 56 for placebo, 55 for PRV-015 100 mg, 56 for PRV-015 300 mg, and 52 for PRV-015 600 mg. These are the counts attached to the reported endpoint results, not a statement about the number enrolled or the population used for every other outcome. Additional details about the randomization procedure, allocation concealment, study sites, treatment duration beyond the endpoint timeframe, and any protocol changes are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: A total of 388 participants entered the single-blind placebo run-in period and among them, 1 participant never received treatment and was not started after signing the ICF, 9 participants discontinued during run-in or were not dosed, and 27 participants were considered run-in failures. Another 126 participants were considered randomization failures. A total of 226 participants were enrolled and randomized in the study.

Single-blind Period (4 Weeks)

Reasons for not completing this period:

Double-blind Period (24 Weeks)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The modified intent-to-treat (mITT) analysis set included all randomized participants who received at least 1 dose of double-blind treatment. Only participants with data collected at baseline and up to Week 24 are 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: Single-blind period: Participants treated with single-blind placebo in run-in period are included. Double-blind period: The intent-to-treat analysis set included all randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Absolute Change From Baseline in Celiac Disease Patient-Reported Outcome Abdominal Symptoms Domain Score Through Week 24, measured over Baseline (average of Day -7 to Day -1) up to Week 24.

The primary endpoint is the absolute change from baseline in the Celiac Disease Patient-Reported Outcome abdominal symptoms domain score through Week 24. Participants recorded symptoms daily in an eDiary. The questionnaire included nine symptoms, while this domain used abdominal cramping, abdominal pain, bloating, and gas. Each symptom was rated on an 11-point scale from 0, not experiencing the symptom, to 10, the worst possible symptom experience. The abdominal symptoms domain total ranges from 0 to 40, and higher scores indicate worse symptoms.

Baseline was defined as the average of daily abdominal symptoms domain scores during the last week of the placebo run-in period. The stated timeframe runs from baseline, described as the average of Day -7 to Day -1, up to Week 24. An absolute change expresses how the score at follow-up differs from its baseline level; a negative change therefore represents a lower score than baseline. That direction alone does not establish why the score changed or whether a between-group difference is meaningful.

A daily patient-reported symptom score can capture a participant’s experience over time, rather than relying only on a single assessment. The domain focuses on four abdominal symptoms relevant to the study’s stated question. General statistical consideration: repeated symptom measurements can support an estimate of average change, while the uncertainty around that estimate helps show how precisely the data distinguish possible values. The registry record does not report a rationale for selecting this endpoint, a threshold for a clinically important difference, or a separate interpretation rule.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The posted primary outcome is a continuous measure: a score that can take a range of numerical values. The registry describes the arm summaries as least-squares means and gives 95% confidence intervals. It reports three comparisons against placebo, each using a mixed model for repeated measures (MMRM). The reported contrast is an LS mean difference. Detailed specifications of the fitted model, including its covariates and covariance structure, are not reported in the registry record.

What does a least-squares mean represent?

General explanation: a least-squares mean is a model-based estimate of a group’s average outcome. Unlike a simple raw average, it may reflect adjustments included in the statistical model. Here, the registry labels the group values as least-squares means, but does not list which adjustments were made. The values should therefore be read as the model-based means reported for the endpoint, without assuming particular covariates or adjustment details.

Why use a repeated-measures model?

General explanation: when an outcome is measured at multiple times, observations from the same participant are related rather than independent. An MMRM can model the pattern of repeated outcomes and compare groups over time while accounting for that within-participant relationship. Its estimates depend on the model specification and the available observations. The registry names MMRM as the comparison method, but does not provide enough detail to reconstruct the model or evaluate all its assumptions.

The primary endpoint is described as change from baseline through Week 24, with a baseline average based on daily scores from the last week of the placebo run-in period. The report also specifies the modified intent-to-treat analysis set and limits the displayed primary-outcome data to participants with data collected at baseline and up to Week 24. Those criteria help define who contributes to the posted results, but do not explain how observations that are missing at some visits were handled by the model.

How should the confidence intervals be read?

General explanation: a confidence interval gives a range of values compatible with the estimate under the analysis and its assumptions. A narrower interval indicates greater precision than a wider interval, though precision is not the same as clinical importance. The posted 95% intervals for all three between-group contrasts include zero. Thus, each interval includes no difference as well as nonzero possibilities. The intervals should not be read as a probability that a particular treatment effect is true.

What does the p-value add?

General explanation: a p-value describes how unusual data at least as inconsistent with a specified null model would be if that model were true, subject to the analysis assumptions. It is not the probability that the null hypothesis is true, nor a measure of effect size or importance. The registry reports p-values of 0.854, 0.655, and 0.871 for the three comparisons. Read alongside the estimates and intervals, they do not provide evidence of a clear difference under conventional testing interpretations. The registry record does not report whether or how these p-values were adjusted for multiple comparisons.

Why does multiplicity matter?

General explanation: when several hypotheses are tested, the chance of obtaining at least one apparently notable result can increase if each is judged separately. The primary outcome has three dose-versus-placebo comparisons. The record lists 11 outcome measures and 12 analyses, and identifies 10 posted secondary outcomes. It does not report a testing hierarchy, family-wise error procedure, or other multiplicity strategy. The three p-values should therefore be understood as the posted individual comparison results, without assuming a particular family-level error control.

The registry also gives planning assumptions for a sample-size illustration based on an observed effect: a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. Such planning values are illustrations rather than observed power, and do not establish the study’s original design rationale. The sample-size justification and any interim analyses are not reported in the registry record.

10. Analyses Reported by the Sponsor

The posted comparison results are the registry’s MMRM analyses: each PRV-015 dose is contrasted with placebo using an LS mean difference, a 95% confidence interval, and a p-value. The registry record does not report a separate recalculated comparison for this continuous endpoint. It also does not flag a discrepancy between a recalculation and the posted result. Accordingly, the sponsor-reported contrasts are the available inferential results to interpret, rather than treating an independent recalculation as a replacement.

The three reported estimates are 0.04 for PRV-015 100 mg, 0.11 for PRV-015 300 mg, and 0.04 for PRV-015 600 mg, with confidence intervals that all include zero. Their p-values are 0.854, 0.655, and 0.871, respectively. These figures are model-based comparisons, not simple arithmetic differences calculated from the displayed group means. The registry does not provide the model details needed to reproduce the MMRM estimates from the arm summaries.

General statistical consideration: a recalculation from summary statistics, when available, is generally unadjusted and may not match a model-based analysis that uses repeated measurements or other adjustments. Here, no such recalculated comparison is reported in the registry record, so there is no numerical recalculation to compare with the sponsor analysis. The sponsor’s reported analysis should be described on its own terms, including its uncertainty and the limits of the available methodological detail.

11. Multiplicity, Missing Data and Analysis Populations

The primary outcome has three reported dose-versus-placebo comparisons. The registry reports a p-value and 95% confidence interval for each, but does not report whether the comparisons were treated as a family, whether an adjustment was used, or whether a particular comparison was designated as primary among the dose groups. The record lists 11 outcome measures, 12 analyses, and 10 posted secondary outcomes. It does not describe a testing hierarchy or explain how conclusions across outcomes were controlled for multiplicity. The individual p-values should not be interpreted as if a family-level error-control procedure had been documented.

The primary-outcome analysis population is described as the modified intent-to-treat set, including randomized participants who received at least one dose of double-blind treatment. Only participants with data collected at baseline and up to Week 24 are reported. The displayed arm counts are 56 for placebo, 55 for PRV-015 100 mg, 56 for PRV-015 300 mg, and 52 for PRV-015 600 mg. The registry does not report a full account of missing observations, reasons for missingness, or the assumptions and procedures used to handle missing repeated measurements. It also does not report sensitivity analyses assessing how alternative missing-data assumptions might affect the estimates.

The record names MMRM but does not provide the model specification or detailed analysis plan. Sample-size justification, interim analyses, and a prespecified clinically important difference are not reported in the registry record. The planning illustration uses an observed effect, two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. That illustration is not observed power and does not substitute for the trial’s original design rationale. These omissions limit how fully an independent reader can assess the planned analysis, missing-data robustness, and strength of conclusions across multiple outcomes.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: presyncope (1.9%), pelvic fracture (1.8%), gluten sensitivity (0.3%), intussusception (0.3%).

The registry separates adverse-event summaries into a single-blind placebo group and double-blind treatment groups. In the single-blind placebo group, 2 serious adverse events were reported among 387 at risk, and no other adverse events were reported among 387 at risk. No deaths were reported in that group. The registry lists the double-blind placebo group as 1 serious adverse event among 57 at risk and 22 other adverse events among 57 at risk, with no deaths.

For double-blind PRV-015 100 mg, the serious adverse event count was 0 among 56 at risk and the other adverse event count was 18 among 56 at risk; no deaths were reported. For PRV-015 300 mg, the corresponding counts were 0 serious adverse events among 57 at risk and 22 other adverse events among 57 at risk; no deaths were reported. For PRV-015 600 mg, the counts were 1 serious adverse event among 54 at risk and 19 other adverse events among 54 at risk; no deaths were reported.

The registry lists Presyncope at 1.9%, Pelvic Fracture at 1.8%, Gluten Sensitivity at 0.3%, and Intussusception at 0.3% among the top serious adverse event terms. It does not map those terms to treatment groups in the posted summary. These counts describe reported events in the listed groups; they do not by themselves establish that treatment caused an event or allow a full comparative safety assessment. Definitions, exposure-adjusted rates, event timing, and detailed attribution are not reported in the registry record.

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

The results describe changes in a patient-reported abdominal symptom score from baseline through Week 24 in the stated analysis population. All four posted least-squares mean changes are negative: -1.32 for placebo, -1.28 for PRV-015 100 mg, -1.21 for PRV-015 300 mg, and -1.28 for PRV-015 600 mg. Because higher scores indicate worse symptoms, the negative values represent lower estimated scores than baseline within each group. The between-group contrasts, rather than these within-group changes alone, are the relevant results for comparing PRV-015 with placebo.

The three contrasts are small in the units of the reported scale, and their 95% confidence intervals include zero. The estimates and intervals therefore do not establish a clear difference between any posted dose group and placebo on this endpoint. They also do not establish equivalence or rule out all potentially meaningful differences. The registry record does not provide a clinically important difference threshold against which the estimates could be judged.

These data do not answer every question about the study. The primary result concerns the specified abdominal symptoms domain, analysis population, and Week 24 timeframe. Secondary outcomes include other patient-reported measures, intraepithelial lymphocyte density, and safety assessments, but their results are not part of the primary endpoint comparisons. A result for this score cannot stand in for all those outcomes, nor does it establish a broader effect beyond the recorded measure and follow-up.

General statistical explanation: a non-small p-value is not proof that groups are identical, just as a numerical difference is not by itself proof of an important effect. Interpretation depends on the estimate, its uncertainty, the endpoint’s meaning, and the analysis choices. The absence of reported multiplicity and missing-data details also matters when assessing the full body of findings. These results should be read as the posted statistical evidence for this endpoint, not as medical advice or a stand-alone treatment recommendation.

14. Why This Trial Matters Statistically

This trial raises a useful statistical question: how should multiple dose groups be compared with a shared placebo group when the primary outcome is a repeated patient-reported symptom score? The registry reports a model-based change estimate for each group and an MMRM contrast for each PRV-015 dose versus placebo. Reading both the group changes and the contrasts helps distinguish improvement from baseline within groups from evidence about differences between groups.

The endpoint also illustrates the importance of defining what a score measures. The abdominal symptoms domain aggregates four symptoms, each recorded daily, into a score with a stated range and direction. A numerical change has meaning only in relation to that scale, its baseline definition, and its follow-up timeframe. Here, the registry specifies a baseline average from the last week of the placebo run-in period and outcomes through Week 24, allowing readers to understand the reported change without assuming that every symptom or outcome is represented by the domain.

The posted contrast intervals include zero, making precision central to interpretation. Statistical nonsignificance is not equivalent to evidence of no difference, and a claim of equivalence would require additional design features not reported in the record. The absence of a stated clinically important difference threshold further limits judgment about the practical meaning of the estimates.

Finally, the trial is a reminder that a named method does not reveal every analytical choice. The registry identifies MMRM and the modified intent-to-treat population, but does not report all model details, missing-data procedures, or multiplicity handling. Those omissions affect reproducibility and the breadth of conclusions that can be drawn from the posted endpoint results. Clear reporting of these details helps readers assess not only the estimates, but also the assumptions supporting them.

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 trial-specific reporting gaps constrain interpretation. The registry identifies MMRM as the comparison method but does not describe the model specification, such as which covariates were used or how the repeated measurements were modeled. The analysis population is defined, and only participants with baseline and through-Week-24 data are reported, but procedures for handling missing measurements and sensitivity analyses are not reported in the registry record.

Three dose-versus-placebo comparisons are reported for the primary endpoint. A multiplicity strategy, testing hierarchy, or adjustment is not reported. The record also does not state a clinically important difference threshold for the abdominal symptoms score. Without those details, the statistical evidence cannot be translated into a prespecified judgment about clinical importance or interpreted as an equivalence assessment.

The primary-outcome estimates apply to the specified modified intent-to-treat population and the available endpoint data, not automatically to every participant enrolled or to every secondary outcome. The registry’s enrollment figure is 388, while the primary-outcome arm counts are 56, 55, 56, and 52. These figures refer to different recorded quantities and should not be conflated. The registry does not explain the difference between them in the endpoint record.

For safety, the registry reports counts of serious and other adverse events in groups with different at-risk counts, including both single-blind and double-blind groups. The summary does not provide event timing, exposure-adjusted rates, detailed attribution, or a mapping of the listed top serious-event terms to arms. The reported counts alone cannot establish causality or support a complete comparison of risks.

Finally, the sample-size illustration is based on an observed effect and stated planning assumptions, not the original design justification. The original sample-size rationale, interim analyses, detailed protocol decisions, and any changes to the planned analysis are not reported in the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: PRV-015 in Gluten-free Diet Non-responsive Celiac Disease — NCT04424927.
• Registry record last updated 2025-10-20; status last verified 2025-10.
• Results first posted 2025-10-20.
• 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

NCT04424927 is a completed Phase 2b randomized, double-blind, placebo-controlled, parallel-group study of PRV-015 as an adjunct to a gluten-free diet in adults with non-responsive celiac disease. The registry lists quadruple masking and enrollment of 388. The study start date is 2020-08-24, and the primary completion date is 2024-07-30.

The primary endpoint was absolute change from baseline in the Celiac Disease Patient-Reported Outcome abdominal symptoms domain score through Week 24. The posted MMRM comparisons of PRV-015 100 mg, 300 mg, and 600 mg versus placebo each have 95% confidence intervals that include zero. The registry also reports adverse-event counts for single-blind placebo and double-blind groups. Further details about multiplicity, missing-data handling, the model specification, and the original sample-size justification are not reported in the registry record.