Study of Glycerol Phenylbutyrate & Sodium Phenylbutyrate in Phenylbutyrate Naïve Patients With Urea Cycle Disorders (UCDs): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Glycerol Phenylbutyrate & Sodium Phenylbutyrate in Phenylbutyrate Naïve Patients With Urea Cycle Disorders (UCDs) (NCT03335488), a randomized phase 4 study. The primary outcome is Rate of Treatment Success (Percentage of Participants Defined as Treatment Success at Week 4) During the Initial Treatment Period.

2. Clinical Question

3. Trial Design

This was a randomized, controlled, open-label, parallel-arm phase 4 study of RAVICTI and NaPBA in patients with urea cycle disorders who were not currently or previously chronically treated with phenylacetic acid prodrugs. The registry records enrollment of 16 and lists the study as completed. The recorded start date is 2018-02-20, and the primary completion date is 2022-07-05.

The design included a baseline period and an initial treatment period, followed by a RAVICTI-only transition period, maintenance period, and safety extension period. In the initial period, one arm received RAVICTI and the other received NaPBA. The NaPBA arm then transitioned to RAVICTI; the RAVICTI arm continued RAVICTI. The registry describes the RAVICTI oral liquid as having a maximum total daily dose of 17.5 mL, with dosing based on disease and treatment status at entry. The brief summary says the study would run for approximately 25 weeks.

Open-label means treatment assignment was not masked, as the registry records masking as none. This can matter when outcomes involve clinical assessment or decisions, although the primary endpoint also incorporates measured biochemical criteria. The parallel-arm structure means the initial-treatment groups were compared during the same study period rather than each participant serving as their own comparator. Random allocation can help balance participant characteristics across treatment groups on average, but a small study can still have chance imbalances.

The posted primary outcome concerns treatment success at Week 4 during the initial treatment period. The registry also lists outcomes involving drug discontinuation, fasting plasma ammonia change, ammonia area under the curve, and peak ammonia concentration. The registry record does not report the rationale for the planned enrollment, details of allocation concealment, or a fuller account of the randomization procedure.

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: In the Initial Treatment Period (approximately 28 days), eligible participants were randomized to one of two treatment arms: RAVICTI or NaPBA. For all Initial Treatment Period NaPBA participants and RAVICTI participants who were treatment failures, a Transition Period (7 days ± 2 days) with RAVICTI followed. All participants received RAVICTI in the Maintenance Period (8 Weeks) and the Safety Extension Period (12 Weeks).

Initial Treatment Period

Transition Period

Maintenance Period

Reasons for not completing this period:

Safety Extension Period

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified Intent-to-Treat Population: all participants from the Safety population with no major eligibility violations and participants who had ammonia data post-randomization.

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.

6. Primary Endpoint

The registry lists the primary outcome as Rate of Treatment Success (Percentage of Participants Defined as Treatment Success at Week 4) During the Initial Treatment Period, measured over Week 4.

The primary endpoint was the percentage of participants classified as treatment successes at Week 4, at the end of the initial treatment period. A participant had to avoid an unprovoked hyperammonemic crisis on the assigned treatment and meet at least 2 of 3 biochemical criteria. These criteria concerned plasma ammonia measurements at pre-dose and after dose at 4 hours and 8 hours, glutamine at the zero-hour time point, and essential amino acids at zero hour. The registry defines normal results in relation to the upper limit of normal.

This composite combines a clinical event criterion with laboratory measures. Such an endpoint can represent several aspects of disease control in one classification, but it also means that participants can qualify through different combinations of the biochemical criteria, provided the crisis condition is met. It is not a direct measure of the magnitude of change in any one laboratory value. The individual criteria and secondary outcomes help distinguish those components.

For the posted modified intent-to-treat population, treatment success was 81.8% in the RAVICTI group and 80.0% in the NaPBA group. These are observed percentages in the analyzed groups, not estimates that establish equivalence. The registry record does not report a prespecified equivalence or non-inferiority margin. Interpretation should therefore focus on the observed estimates and the uncertainty in the reported comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry reports a Fisher exact test and an odds ratio for the primary binary outcome. Each participant was classified as a treatment success or not under the endpoint definition, and the analysis compared the resulting proportions between the randomized groups. The arm-level percentages were 81.8% for RAVICTI and 80.0% for NaPBA.

Why use an exact test?

General statistical explanation: Fisher’s exact test is commonly used for a comparison of two categorical groups when the counts are small. It calculates a probability for the observed arrangement of outcome classifications, and arrangements at least as incompatible with the null hypothesis are used to assess the evidence against that hypothesis. Unlike methods that rely on a large-sample approximation, the exact test is based on the possible allocations of the observed counts under its assumptions. The registry identifies the method as Fisher exact but does not report further details about its implementation.

The small group sizes make this choice understandable: the analyzed groups contained 11 and 5 participants. In a small sample, conventional large-sample methods can give unreliable approximations, especially when some cells contain very few observations. Exact methods avoid relying on that particular approximation, but they cannot create information that the study did not collect. They may yield wide confidence intervals and limited ability to distinguish between groups when observations are sparse.

What does the odds ratio describe?

The odds ratio compares the odds of treatment success between groups, with the direction depending on which group is placed in the numerator. The registry reports an odds ratio of 1.1 and a 95% confidence interval from 0.0 to 28.1. An odds ratio near 1 is a point estimate near equal odds, but that estimate must be considered alongside its uncertainty. The interval is broad and includes the null value of 1. It is compatible with a wide range of relative effects, so it does not provide a precise estimate of the comparison.

General statistical explanation: odds and probabilities are related but are not interchangeable. The endpoint percentages describe proportions of participants classified as successes, while the odds ratio expresses a ratio of odds. Readers should not interpret the odds ratio as a difference in percentages or as a direct measure of how many additional participants would succeed. No such difference is reported in the registry record, and none is needed to interpret the reported analysis.

How should the p-value be read?

The reported p-value is 1.000. General statistical explanation: a p-value describes how compatible the observed data are with a null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, and it does not measure the probability that the findings are due to chance. A large p-value does not establish that the groups are equivalent. In this study, the uncertainty interval is especially important because it shows how imprecisely the relative effect is estimated.

The analysis is labeled as superiority. For superiority testing, the null hypothesis is generally framed as no difference under the specified measure, with the alternative indicating a difference. The registry does not report the full statistical analysis plan or the precise direction and tail conventions of the test. It also does not report a non-inferiority or equivalence margin. Therefore, the posted superiority result should not be reinterpreted as a formal test of equivalence.

What does the sample-size illustration mean?

General statistical explanation: planning illustrations based on an observed effect are hypothetical calculations for future study planning. The stated assumptions are a two-sided significance level of 0.05, equal allocation, a normal approximation, and an effect to detect expressed as a fraction of the observed effect. They are not observed power and do not describe the design justification for this completed trial. The registry record does not report the trial’s sample-size justification. Because the observed effect comes from a small study, planning calculations anchored to it may be unstable; a planning illustration is not a guarantee that a future study would achieve the stated power.

For a binary endpoint, planning estimates depend on the assumed success rates, the target effect, and the analysis assumptions. Altering the assumed fraction of the observed effect changes the planning question. Such calculations are useful for illustrating how design choices affect required enrollment, but they should not be mistaken for outcomes observed in this study. No planning value should be read as a participant count enrolled or analyzed in the trial.

What can an unadjusted re-calculation add?

General statistical explanation: a simple unadjusted re-calculation from group-level counts can serve as a check on a reported summary, but it does not replace an analysis that uses participant-level data or the sponsor’s specified analysis. For this binary endpoint, the posted summary does not support an additional re-calculated two-arm estimate in the registry record. The reported analysis remains the Fisher exact result. The registry record reports no discrepancy flag.

More broadly, summary-level results constrain what can be checked. They do not reveal the distribution of baseline characteristics, individual measurements, missing observations, or the exact handling of all analysis cases. Those details can matter for adjusted analyses and for assessing whether the population included in the analysis matches the stated definition. The registry record does not report enough information to reconstruct those procedures beyond the population definition and posted test.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("RAVICTI", "NaPBA")
n      <- c(11, 5)
events <- c(9, 4)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  RAVICTI  11  9  81.8  48.2  97.7  52.3  94.9  51.6  94.5
ARM  NaPBA  5  4  80.0  28.4  99.5  37.6  96.4  35.9  95.7

11. Analyses Reported by the Sponsor

The registry’s own primary comparison is the Fisher exact analysis: an odds ratio of 1.1, a 95% confidence interval from 0.0 to 28.1, and p = 1.000. The reported treatment-success percentages are 81.8% for RAVICTI and 80.0% for NaPBA. These results should be read together: the point estimate is close to equal odds, but the confidence interval is broad and the comparison is imprecise.

No additional unadjusted two-arm re-calculation is available from the posted summary for this binary endpoint. Accordingly, there is no separate recalculated estimate to set beside the registry’s analysis. The sponsor’s reported analysis is the available inferential result for the primary comparison, and the posted summary does not identify a discrepancy.

General statistical explanation: when a sponsor reports an analysis based on the trial’s specified approach, an unadjusted check based only on summary values would answer a narrower question and could differ if its assumptions or included observations differ. Here, the registry record does not provide enough information to reconstruct alternative analyses. The reported result remains uncertain; the lack of a separate recalculation does not make the confidence interval narrower or establish equivalence.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and four secondary outcomes. The secondary outcomes concern drug discontinuation during initial treatment, change in fasting plasma ammonia, ammonia area under the curve, and peak ammonia concentration. Multiple outcomes can lead to multiple statistical comparisons. General statistical explanation: when many hypotheses are tested, the chance of at least one apparently notable result can rise unless the analysis plan accounts for multiplicity. The registry record does not report whether a multiplicity adjustment, hierarchical testing strategy, or other control of multiple comparisons was planned or used. The secondary results therefore require cautious interpretation as separate outcome analyses.

The registry record defines the primary analysis population as the modified intent-to-treat population: participants in the safety population with no major eligibility violations and ammonia data after randomization. This is not described as every randomized participant regardless of post-randomization data. The arm sizes for the primary endpoint were 11 for RAVICTI and 5 for NaPBA. The enrollment record gives 16, but the record does not describe in detail how every enrolled or randomized participant contributed to each outcome analysis.

The registry record does not report how missing primary endpoint components were handled, whether any observations were imputed, or how participants with incomplete biochemical assessments were classified. Those choices can affect a composite outcome, because the endpoint relies on crisis status and laboratory criteria. The record also does not give a detailed flow of exclusions or missing measurements for the primary analysis.

The registry record does not report interim analyses, a stopping rule, or a complete statistical analysis plan. It also does not report the sample-size justification. These are unknowns rather than evidence that a particular procedure was or was not used. Without those details, the posted result can be interpreted as the reported endpoint comparison, but not as a full audit of prespecification, multiplicity control, or missing-data sensitivity.

13. 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: pyrexia (20.0%), hyperammonaemic crisis (18.2%), neutropenia (9.1%), hyperammonaemia (9.1%), bone marrow failure (6.2%).

Safety results are reported across different study periods, so the groups should not be treated as if they were all concurrent randomized comparisons. During initial treatment, serious adverse events were reported for 2 of 11 participants receiving RAVICTI and 0 of 5 receiving NaPBA. In the transition period, the RAVICTI group had 1 serious adverse event among 5 participants at risk. For the combined maintenance and safety periods, the registry reports 3 among 16 participants at risk. No deaths are recorded in the listed groups.

These counts describe participants affected, not necessarily the total number of individual event episodes. The registry also lists serious adverse-event terms including pyrexia at 20.0%, hyperammonaemic crisis at 18.2%, neutropenia at 9.1%, hyperammonaemia at 9.1%, and bone marrow failure at 6.2%. The record does not provide enough context alongside these terms to treat them as directly comparable treatment-specific rates across the different periods.

General statistical explanation: safety counts in small groups are descriptive and can be unstable. A zero count in a group does not establish that the underlying risk is zero. The later period counts also reflect different exposure histories, including transition to RAVICTI, so they should not be interpreted as a randomized head-to-head safety comparison. These figures summarize reported events; they do not by themselves establish causality.

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

The primary result shows that treatment success, as defined by the registry, was common in both initial-treatment groups among participants included in the modified intent-to-treat analysis: 81.8% in the RAVICTI group and 80.0% in the NaPBA group. The sponsor’s reported odds ratio was 1.1. These are the observed results for a composite endpoint assessed at Week 4.

The result does not establish that the treatments are equivalent. The 95% confidence interval, from 0.0 to 28.1, is wide and includes the null value. That interval indicates substantial uncertainty about the relative odds. The reported p-value of 1.000 does not prove that the groups have the same outcomes, nor does it show that any difference is clinically unimportant. The registry record does not report an equivalence or non-inferiority margin.

The endpoint combines the absence of an unprovoked hyperammonemic crisis with biochemical criteria involving ammonia, glutamine, and essential amino acids. A treatment-success percentage does not show the individual laboratory results, the pattern of criteria met, or the distribution of measurements. The listed secondary outcomes address some related measures, but they are distinct outcomes and have their own uncertainties.

The study was open-label and included small analyzed groups. Randomization helps make a fair comparison in principle, but with few participants, chance variation can remain substantial. The precision of the primary comparison is limited, and the registry record does not provide the missing-data details or full analysis plan needed for a more complete assessment.

Finally, the primary comparison concerns the initial treatment period. Participants assigned to NaPBA transitioned to RAVICTI afterward, and later periods are not a continued randomized comparison of the original treatments. The result should therefore be understood as a Week 4 comparison in this study population, not as a direct comparative estimate for all later treatment periods or for every patient with a urea cycle disorder.

15. Why This Trial Matters Statistically

This trial illustrates an important distinction between a point estimate and the precision of that estimate. The treatment-success percentages were close: 81.8% with RAVICTI and 80.0% with NaPBA. Yet the reported confidence interval for the odds ratio extends from 0.0 to 28.1. Looking only at the percentages could suggest a settled comparison; considering the interval makes clear that the estimate is uncertain.

The example also shows why a non-significant superiority analysis is not the same as evidence of equivalence. A study can fail to distinguish groups because the groups are similar, because information is limited, or for both reasons. The wide interval in this study leaves substantial uncertainty. Formal equivalence or non-inferiority conclusions would require an appropriate design and margin, neither of which is reported in the registry record.

Composite endpoints raise a related interpretive issue. They can combine a clinical event and several laboratory criteria into a single outcome, which may be useful for capturing a broader definition of control. But the overall success percentage does not reveal which criteria drove classification. Component outcomes and their missingness matter when readers assess what the composite represents.

Finally, the study’s sequence of treatment periods matters. The initial randomized comparison is not carried forward unchanged once the NaPBA group transitions to RAVICTI. Separating the randomized initial-period evidence from later descriptive safety observations helps prevent an inappropriate head-to-head interpretation of results that were collected under different treatment histories.

16. 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 analysis includes small groups: 11 participants in the RAVICTI arm and 5 in the NaPBA arm. The broad 95% confidence interval around the odds ratio reflects substantial uncertainty. With small groups, estimates can be sensitive to outcome classifications, and the study has limited ability to rule out a range of possible differences.

The open-label design means treatment assignment was not masked. The primary endpoint includes objective laboratory measurements, but it also depends on whether an unprovoked hyperammonemic crisis occurred and on the definition and classification of that event. The registry record does not report independent blinded adjudication of crises or other safeguards against assessment effects.

The endpoint is composite and permits success through different combinations of biochemical criteria, so the overall percentage does not identify which components contributed most. The registry record does not report detailed component-level results within the primary outcome. It also does not report how missing endpoint components were handled, whether sensitivity analyses were conducted, or the full statistical analysis plan.

The study was designed with later RAVICTI-only periods after initial treatment, including transition, maintenance, and safety extension periods. Consequently, those periods do not maintain the initial randomized comparison between RAVICTI and NaPBA. Safety observations across periods should be interpreted with their different treatment histories in mind.

The registry record does not report the sample-size justification, multiplicity procedures, interim analyses, or allocation concealment details. It also does not report a formal equivalence or non-inferiority margin. These limits do not invalidate the posted result, but they restrict how confidently it can be generalized or used to make claims beyond the initial Week 4 comparison.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Study of Glycerol Phenylbutyrate & Sodium Phenylbutyrate in Phenylbutyrate Naïve Patients With Urea Cycle Disorders (UCDs) — NCT03335488.
• Registry record last updated 2024-07-01; status last verified 2024-06.
• Results first posted 2023-07-10.
• 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.

20. Record Summary

NCT03335488 is a completed randomized, controlled, open-label, parallel-arm phase 4 study of RAVICTI and NaPBA in phenylbutyrate-naïve patients with urea cycle disorders. The registry records enrollment of 16, a start date of 2018-02-20, and primary completion on 2022-07-05. The primary endpoint was treatment success at Week 4 during initial treatment, defined by absence of an unprovoked hyperammonemic crisis and meeting at least 2 of 3 biochemical criteria.

The posted success percentages were 81.8% for RAVICTI and 80.0% for NaPBA. The registry reports a Fisher exact odds ratio of 1.1, a 95% confidence interval from 0.0 to 28.1, and p = 1.000. The interval is broad, so the comparative estimate is imprecise and does not establish equivalence. Later periods involved transition to RAVICTI and do not provide a continuing randomized comparison. The registry record does not report the full statistical plan or missing-data procedures.