Efficacy and Safety of Cobiprostone in Patients With Portal Hypertension: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of Cobiprostone in Patients With Portal Hypertension (NCT00737594), a randomized phase 2 study. The primary outcome is Change From Baseline in Hepatic Venous Pressure Gradient (HVPG) After Four (4) Weeks of Treatment.

2. Clinical Question

The question was framed around a change in a physiological measure, rather than a reported clinical event or patient-reported outcome. The posted primary endpoint is “Change From Baseline in Hepatic Venous Pressure Gradient (HVPG) After Four (4) Weeks of Treatment.” Although the study summary describes its purpose as evaluating efficacy and safety for lowering portal hypertension, the record contains no collected primary endpoint results. It therefore does not show the size or direction of any change in HVPG, nor whether the groups differed.

The two cobiprostone arms make dose-specific comparisons relevant, but the registry record does not state which comparisons were designated as primary statistical contrasts. It also does not report a planned analysis population or the number assigned to each arm. The clinical question can be identified from the intervention groups and endpoint, but its answer cannot be estimated from the posted results.

3. Trial Design

The registry describes a randomized, double-blind, parallel-group, placebo-controlled phase 2 study. It lists three arms: matching placebo TID, cobiprostone 12 mcg TID, and cobiprostone 18 mcg TID. The arm descriptions specify capsules taken three times daily. The registry lists enrollment as 9 and gives the overall study status as terminated.

In a parallel-group design, participants are assigned to one group for the study comparison rather than crossing between treatments. Random allocation is generally intended to reduce systematic differences between groups, while blinding is generally intended to limit the influence of treatment knowledge on participant experience, care, measurement, or assessment. These are general features of the design; the registry does not describe how allocation was generated or concealed, who was blinded, or how masking was maintained.

The registry lists a start date of 2008-07 and a primary completion date of 2009-02. It does not report a detailed recruitment timeline, the reason for termination, or how many participants completed treatment. Enrollment is not enough to determine group sizes or completion. The posted outcome record says that the study terminated early and data were not collected for the primary endpoint.

The study summary states that its purpose was to assess efficacy and safety of two cobiprostone dose levels against placebo for lowering portal hypertension. That statement describes the planned question, not a result. There is one primary outcome measure listed and no analyses reported. The record does not describe further design details such as stratification, treatment duration beyond the endpoint timeframe, adherence assessment, or a statistical analysis plan. Those details 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: 9 participants were recruited, but the study terminated before data were collected

Overall Study

Analysis population for the primary outcome, as defined in the registry: Study terminated early, data were not collected.

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 study terminated before data were collected

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Hepatic Venous Pressure Gradient (HVPG) After Four (4) Weeks of Treatment, measured over 4 weeks.

The primary endpoint was change from baseline in hepatic venous pressure gradient (HVPG) after four weeks of treatment. The registry gives the timeframe as 4 weeks but does not provide a unit, summary parameter, dispersion measure, or endpoint description beyond the title. Most importantly, it states that the study terminated early and data were not collected.

In general, HVPG is a measurement of pressure in the hepatic venous circulation used to characterize portal hypertension. A change-from-baseline endpoint compares a participant’s measurement after treatment with that participant’s starting measurement. This can describe whether the measure shifted during the specified period; comparing changes across randomized groups can then address whether the groups showed different patterns. These are general explanations of the endpoint, not findings from this trial.

The endpoint was chosen to address a hemodynamic question connected to the study’s stated purpose of lowering portal hypertension. However, the registry does not report the measurement method, the planned summary statistic, a clinically meaningful threshold, or how endpoint observations would be analyzed. No baseline or follow-up HVPG values, group-level changes, or treatment contrasts appear in the posted results. Thus the endpoint’s intended role is identifiable, but its observed behavior and precision are not.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Randomization, blinding, an endpoint, and a timeframe are described, but the registry record does not report a statistical analysis plan or any analyses. The following explanations describe general statistical ideas relevant to this design; they should not be read as a claim that the trial used a particular method.

What randomization contributes

In general, random assignment is used to make treatment groups comparable with respect to both measured and unmeasured characteristics, on average across repeated assignments. It does not guarantee that groups in a particular small study will have similar baseline characteristics. The registry identifies the study as randomized but does not state the allocation sequence method, concealment procedures, or group sizes. With no participant-level or group-level results posted, balance cannot be examined.

Why masking can matter

In general, double blinding is intended to reduce the influence of knowledge about treatment assignment on participant behavior and on decisions or assessments made during a study. The registry calls the study double-blinded and describes a matching placebo. It does not specify which roles were masked or report whether masking was successfully maintained. No claim about the effectiveness of masking can be made from the posted record.

What a change-from-baseline endpoint represents

In general, a change score is calculated by comparing a follow-up measurement with a baseline measurement for the same participant. A group summary can describe the average or another chosen feature of those changes. A treatment comparison might then contrast the change in one arm with the change in another. The registry names change in HVPG after four weeks as the primary endpoint, but does not report the chosen summary measure or comparison method. It supplies no observed values from which a change or contrast could be summarized.

Uncertainty and statistical tests

In general, an effect estimate describes the size and direction of a comparison, while a confidence interval describes the range of values reasonably compatible with the data and the analysis assumptions. A hypothesis test can summarize how inconsistent observed data are with a specified null hypothesis. These tools depend on having analyzable observations and a defined analysis. The registry reports no outcome analysis, estimate, confidence interval, or p-value. It is therefore not possible to explain a trial-specific inferential result or its precision.

Why over-dispersion can matter

In general, over-dispersion occurs when observed variability is greater than a statistical model expects. If a model assumes less variability than is actually present, standard errors and inferential conclusions may be unreliable. Whether this issue is relevant depends on the outcome and analysis model. The registry does not report an analysis model or outcome data, so there is no basis to assess over-dispersion in this trial.

Planning calculations are not trial results

Sample-size planning illustrations answer a hypothetical design question: how many participants per arm might be needed to detect a specified effect under stated assumptions. The assumptions listed for the planning illustration describe a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. Such illustrations are not observed power and are not evidence that a study had a particular chance of detecting an effect. Here, the registry reports no observed primary endpoint effect, so the posted record does not supply a trial result on which to base an effect-specific interpretation. Any planning display must be understood as an illustration under its stated assumptions, not as an analysis of participants in this study.

The registry does not report whether the primary endpoint was analyzed using a model that adjusted for baseline HVPG, whether the analysis used change scores, or how any departures from assumptions would have been handled. These are not interchangeable details: choices about summaries, covariates, missing observations, and variance assumptions can affect estimates and their uncertainty. Without a documented plan or posted analyses, the statistical method behind an endpoint estimate cannot be evaluated because no endpoint estimate is available.

10. Multiplicity, Missing Data and Analysis Populations

The study lists three arms and one primary outcome measure. A comparison of each cobiprostone dose with placebo would involve more than one treatment contrast, but the registry record does not state which contrasts were planned, whether both were confirmatory, or how multiplicity was to be addressed. It also does not report a hierarchy of hypotheses or a prespecified adjustment. The presence of multiple arms alone does not establish what inferential procedure was intended.

Missing data are central to interpreting this record, but the reported situation is more specific than a set of missing follow-up values in an otherwise completed analysis. The primary outcome population says the study terminated early and data were not collected. The registry does not explain the circumstances of termination, whether any measurements were taken for other purposes, or how many participants had any endpoint observation. A missing-data method is not reported.

The record does not name an intention-to-treat, per-protocol, or safety analysis population, and it reports zero analyses. It does not say how participants would have been handled if they withdrew, missed a visit, or had an unusable measurement. These analysis-population and missing-data decisions cannot be inferred from the randomized design or the enrollment count. In the absence of an endpoint dataset and a reported analysis plan, there is no treatment comparison to assess and no basis for judging whether missingness could have biased an estimate.

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

This record establishes that a randomized, double-blind, placebo-controlled phase 2 study was planned to compare two cobiprostone doses with matching placebo in participants with portal hypertension. It also identifies change from baseline in HVPG after four weeks as the primary endpoint. It does not establish what happened to HVPG in any group, because the registry says the study terminated early and data were not collected.

As a result, the posted record does not show that either dose lowered portal pressure, increased it, or had no effect. It also cannot show whether any difference between groups was large or small, precise or uncertain, or statistically compatible with chance. Those statements require observed outcome results and an analysis. No such results are reported.

The same caution applies to safety. The adverse-event group entries list zero at risk and zero affected for serious and other events, and deaths are not reported. In the context of the explicit statement that data were not collected, those entries should not be read as evidence that no events occurred or that the treatment was safe. They do not permit a comparison of event rates between groups.

More generally, absence of posted results is not evidence that the treatments had identical effects, and it is not evidence that one treatment was better than another. It means the registry record does not supply the observations needed to answer the intended question. The terminated status is known, while the reason for termination is not reported in the registry record. This distinction matters: the design and question can be described, but neither the trial’s efficacy findings nor a reason for its early termination can be inferred from this record.

12. Why This Trial Matters Statistically

This trial is statistically informative mainly as an example of the distinction between a planned comparison and an observed result. Its registry entry identifies randomization, blinding, three treatment groups, and a physiological primary endpoint. Yet the record also states that the study terminated early and that endpoint data were not collected. A clear research question and a defined endpoint do not, by themselves, produce an estimate of treatment effect.

The enrollment listing of 9 also illustrates why a total enrollment figure cannot stand in for an analysis. It does not provide arm-specific counts, follow-up completeness, or the number contributing to an endpoint. Even if group sizes were known, sample size alone would not establish an effect or its precision. The relevant evidence would be observed measurements and a transparent analysis, neither of which is reported.

The three-arm structure raises questions about dose-specific comparisons and multiplicity, but the registry does not report the planned contrasts or correction strategy. That absence prevents evaluation of how the inferential burden across comparisons would have been handled. It is a limit of the available record, not proof that no plan existed.

Finally, the safety listing shows why registry entries must be interpreted in context. Zero counts paired with zero participants at risk and an explicit statement that data were not collected cannot be treated like a completed safety assessment. The record supports a careful account of what was intended and what is missing; it does not support conclusions about the treatment’s effects.

13. 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 most consequential limitation is that the primary endpoint was not collected, according to the registry record. There are no posted baseline values, follow-up measurements, group summaries, or treatment contrasts. As a result, statistical precision, variability, and the direction or magnitude of any effect cannot be assessed.

The record lists enrollment as 9 but does not report arm-specific enrollment, allocation counts, treatment completion, or endpoint follow-up. The reason for early termination is not reported. It is therefore not possible to determine how the recorded enrollment relates to the number of participants exposed to treatment or to any intended analysis population.

Important design and analysis details are also absent from the registry record. These include the allocation procedure, the exact masking arrangements, a statistical analysis plan, the planned contrasts, multiplicity handling, and methods for addressing missing data. The endpoint entry does not report a unit, a summary parameter, or dispersion. These omissions limit assessment of how the planned question would have been answered even apart from the lack of collected endpoint data.

The adverse-event entries list zero participants at risk and zero affected in each group, while death counts are not reported. Given the record’s statement that data were not collected, these entries do not provide an evaluable comparison of safety. The record also does not explain whether other data were collected before termination. All interpretations must therefore remain limited to the registered design and the explicit absence of primary endpoint results.

14. Related Tutorials

15. Related Statistical Calculators

16. Sources

• ClinicalTrials.gov: Efficacy and Safety of Cobiprostone in Patients With Portal Hypertension — NCT00737594.
• Registry record last updated 2019-11-18; status last verified 2016-02.
• Results first posted 2016-03-16.
• 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.

17. Record Summary

The registry describes a phase 2 randomized, double-blind, parallel-group, placebo-controlled study of cobiprostone in participants with portal hypertension. The three listed arms are matching placebo TID, 12 mcg TID, and 18 mcg TID. Enrollment is listed as 9; the study status is terminated. The record gives a start date of 2008-07 and a primary completion date of 2009-02.

The primary endpoint was change from baseline in HVPG after four weeks of treatment. The registry states that the study terminated early and data were not collected. It lists one outcome measure and zero analyses. No primary endpoint estimates or treatment comparisons are reported. The reason for termination, group-specific participant counts, analysis populations, and statistical plan are not reported in the registry record.

The adverse-event entries do not provide an evaluable safety comparison in light of the statement that data were not collected. The record supports description of the intended design and endpoint, but not conclusions about efficacy or safety.