This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for A Study of Baricitinib (LY3009104) in Participants With Primary Biliary Cholangitis Who do Not Respond or Cannot Take UDCA (NCT03742973), a randomized phase 2 study. The primary outcome is Change From Baseline in Alkaline Phosphatase (ALP).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03742973 |
| Phase | Phase 2 |
| Condition | Primary Biliary Cholangitis |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 2 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Eli Lilly and Company |
| Results posted | Yes |
2. Clinical Question
Population
Participants with primary biliary cholangitis who had an inadequate response to ursodeoxycholic acid (UDCA) or were intolerant to it. The registry record reports enrollment of 2, while the primary-outcome population is described as zero participants reported in Cohort A because of personal-information protection based on enrollment futility.
Intervention
Baricitinib, an oral study drug. Cohort A is described as receiving 2 mg once a day for 12 weeks. Cohort B was planned to receive 4 mg orally; the registry says the study was terminated because of enrollment futility.
Comparator
Placebo, administered orally. Cohort A is described as receiving placebo once a day for 12 weeks. Placebo was also planned for Cohort B, which was not carried forward after the termination decision.
Primary question
The registered primary outcome was change from baseline in alkaline phosphatase (ALP), assessed from baseline to Week 12. The record does not report a numerical comparison of ALP change between baricitinib and placebo, so it cannot establish the direction or precision of a treatment contrast.
The study was described as a proof-of-concept evaluation of efficacy and safety. That description identifies its broad purpose, but does not substitute for a reported estimate of treatment effect. The registry lists additional outcomes concerning ALP response and total bilirubin, itch, and fatigue. It does not report their results in the information available for this record.
The record therefore identifies the clinical population, assigned treatments, comparator, and intended primary measure, but leaves the central comparative question unanswered. It does not report whether ALP changed differently between groups, how uncertain any difference was, or whether the other listed outcomes showed group differences.
3. Trial Design
This was a randomized, double-blind, placebo-controlled, parallel-group phase 2 study. Random allocation is the design feature intended to create groups that are comparable, on average, with respect to factors measured and unmeasured at the outset. Double masking is intended to reduce the influence of treatment knowledge on participant experience, care, and outcome assessment. The registry does not report the specific allocation procedure, masking arrangements, or who was masked.
The record describes Cohort A as including baricitinib at 2 mg orally once a day for 12 weeks and placebo orally once a day for 12 weeks. It also describes a planned Cohort B using baricitinib at 4 mg and placebo. The record says Cohort B was planned, but that the study was terminated following an enrollment-futility decision. It does not report Cohort B outcome data.
The study began on 2019-03-28, and the primary completion date is listed as 2019-09-26. Its current status is terminated. The registry reports enrollment of 2. Cohort A arm descriptions state that results are not reported because of protection of personal identifiable information based on enrollment futility. The primary outcome population is likewise described as zero participants reported for that reason. These statements do not provide a usable group-level outcome summary.
The registry identifies four outcome measures and zero analyses. It does not report an analysis population, the number allocated to each arm, or an estimate of the primary contrast. It also does not describe the method used to generate the random sequence, allocation concealment, adherence, or treatment discontinuation. Those details are not reported in the registry record, so the study design can be characterized only at the level of the registered labels and arm descriptions.
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: Zero participants reported in cohort A due to protection of personal identifiable information based on enrollment futility and cohort B is not reported due to early study termination based on enrollment futility.
Overall Study
| Stage | Baricitinib Cohort A | Placebo Cohort A |
|---|---|---|
| STARTED | 0 | 0 |
| COMPLETED | 0 | 0 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: Zero participants reported in cohort A due to protection of personal identifiable information based on enrollment futility.
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: Zero participants reported in cohort A due to protection of personal identifiable information based on enrollment futility.
| Characteristic | Unit | Placebo Cohort A | .Baricitinib Cohort A | Total |
|---|---|---|---|---|
| Participants analysed | 0 | 0 | 0 | |
| Age, Categorical: <=18 years | — | — | — | |
| Age, Categorical: Between 18 and 65 years | — | — | — | |
| Age, Categorical: >=65 years | — | — | — | |
| Sex: Female, Male: Female | — | — | — | |
| Sex: Female, Male: Male | — | — | — | |
| Ethnicity (NIH/OMB): Hispanic or Latino | — | — | — | |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | — | — | — | |
| Ethnicity (NIH/OMB): Unknown or Not Reported | — | — | — | |
| Race (NIH/OMB): American Indian or Alaska Native | — | — | — | |
| Race (NIH/OMB): Asian | — | — | — | |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | — | — | — | |
| Race (NIH/OMB): Black or African American | — | — | — | |
| Race (NIH/OMB): White | — | — | — | |
| Race (NIH/OMB): More than one race | — | — | — | |
| Race (NIH/OMB): Unknown or Not Reported | — | — | — | |
| Region of Enrollment: United States | — | — | — |
6. Primary Endpoint
The registry lists the primary outcome as Change From Baseline in Alkaline Phosphatase (ALP), measured over Baseline, Week 12.
| Element | Registry entry |
|---|---|
| Result type | N/A |
| Measure of spread | not applicable |
| Unit | not stated |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is change from baseline in alkaline phosphatase (ALP), with measurements designated at baseline and Week 12. A change endpoint compares a participant’s value at follow-up with that participant’s starting value. In a randomized trial, the principal comparison would ordinarily concern the distribution of those changes between the assigned groups; the registry record does not report such a comparison.
ALP is a laboratory measurement. As a trial endpoint, it gives a defined quantity that can be measured at specified time points and compared across treatment assignments. A baseline-to-follow-up change can describe movement during the study period, while a between-group contrast can help distinguish observed movement under one assignment from movement under the comparator. Neither within-group movement alone nor the endpoint label establishes a treatment effect.
The registry does not report the ALP unit, the summary measure, a measure of variability, a numerical result, or the analysis method. It also does not give a rationale for selecting ALP as the primary measure or state how the result was intended to relate to clinical outcomes. Those matters are not reported in the registry record.
The other registered measures include an ALP-and-bilirubin response definition, change in itch Numeric Rating Scale (NRS), and change in fatigue NRS. They do not fill the primary endpoint reporting gap: no results or analyses for those outcomes are reported in this record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record documents a randomized, double-blind, placebo-controlled study, but it does not provide a numerical result for the primary endpoint. There is no reported estimate of change from baseline in ALP by group, no reported contrast between baricitinib and placebo, and no uncertainty interval or test result. Consequently, the size, direction, and precision of a primary treatment comparison cannot be evaluated from the record.
The status is terminated, and the record attributes the decision to enrollment futility. That is a statement about the reason for ending the study, not a numerical finding about ALP or another outcome. The record does not report the decision rule used to define enrollment futility, any interim analysis, or whether the decision was based on a formal statistical assessment. No such details should be inferred from the termination status alone.
The registry lists enrollment of 2, while the Cohort A descriptions say results are not reported to protect personal identifiable information in view of enrollment futility. The primary-outcome population field states that zero participants are reported for the same reason. These entries describe what the record makes available, but do not permit reconstruction of group-level outcomes or a treatment contrast. The registry does not report how participants were allocated, and it would be inappropriate to infer arm sizes from the overall enrollment entry.
There are zero analyses listed. In practical terms, the record offers no sponsor-reported statistical analysis to interpret for the listed outcomes. It gives no point estimates, standard errors, confidence intervals, or p-values. The absence of these statistics is not evidence that the groups had the same outcomes, nor does it show that either assigned treatment had a particular effect.
The planned Cohort B does not provide additional evidence: the record says it was planned and that the study was terminated following enrollment futility. It reports no Cohort B observations. Likewise, the arm descriptions for Cohort A do not contain outcome summaries. The study’s proof-of-concept label describes its intended role, but the available registry information is insufficient to judge whether the primary endpoint supported that aim.
Overall, the record supports a description of the design, planned treatments, endpoint, and reason for termination. It does not support a quantitative conclusion about efficacy, safety, or comparative ALP change. A reader should distinguish the fact that the study ended early from any claim about what the treatment did; the former is reported, while the latter cannot be assessed from the posted results.
9. Statistical Methods Explained
Randomization assigns participants to treatment groups using a chance-based process. In general, its statistical purpose is to reduce systematic differences between groups at the point of assignment, so that a later contrast can be interpreted as a comparison of the assigned strategies, subject to the trial’s conduct and analysis. The registry identifies the study as randomized but does not report the sequence-generation method or allocation concealment. The details needed to assess how randomization was implemented are therefore not reported in the registry record.
Double masking means that treatment assignment is concealed from designated people involved in the trial. In general, masking can limit differences in expectations, behavior, co-interventions, and outcome assessment that might otherwise arise from knowing the assignment. The record gives the double-masked label but does not identify which people were masked or describe procedures for maintaining masking. The label alone does not establish how masking worked in practice.
A parallel-group design assigns participants to a study group and compares outcomes across groups. The registry lists baricitinib and placebo arms, with a Cohort A regimen and a planned Cohort B regimen. For Cohort A, the descriptions state oral daily treatment for 12 weeks; for Cohort B, the descriptions state that it was planned before the termination decision. The record does not report results for the planned cohort or an analysis comparing the Cohort A assignments.
What would a change endpoint represent?
Change from baseline represents the follow-up value relative to the starting value for the same participant. A group summary of change describes how those individual changes are distributed within an assigned group. A between-group comparison of change would address whether the observed changes differed by assigned treatment. The registry names ALP change from baseline as the primary endpoint and specifies baseline and Week 12, but supplies no group summaries, contrast, or measure of uncertainty. There is therefore no numerical estimate to interpret.
In general, a confidence interval gives a range of values compatible with an estimate under a statistical model and its assumptions. Its width conveys precision: wider intervals indicate less precision than narrower intervals. A confidence interval is not reported for the primary endpoint in this registry record. Without an estimate and interval, the magnitude and uncertainty of any treatment contrast cannot be assessed. A p-value is also not reported, and no significance conclusion can be drawn.
Why does precision matter?
Statistical precision depends on the amount and variability of information available, as well as on the analysis. In general, fewer observations or greater outcome variability can make estimates less precise. This is a general explanation, not a claim about the actual variability or precision in this study: the registry does not report an ALP dispersion measure or a numerical estimate. The information available does not support calculating either one.
The record includes a description of planning illustrations based on the observed effect, a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation; it says the effect to detect is a stated fraction of the observed effect. It also distinguishes illustrations expressed as participants per arm from time-to-event illustrations expressed as total events. These are planning assumptions, not evidence that the trial had a particular achieved power. The registry does not report a numerical observed effect or a completed sample-size justification for the trial. No observed power can be inferred from planning assumptions.
In general, over-dispersion occurs when outcome variation is greater than a model assumes. If it is ignored, standard errors can be too small and uncertainty can be understated. This is a general statistical consideration, not a feature reported for this trial. The registry does not report the distribution of ALP changes or a model that would allow an assessment of over-dispersion. No particular model or variance adjustment should be attributed to the investigators.
Registered outcome measures include a response definition requiring ALP <1.67 x the upper limit of normal, at least 15% decrease from baseline, and total bilirubin less than the upper limit of normal; change from baseline in itch NRS; and change from baseline in fatigue NRS. The record does not report estimates or analyses for these measures. Their inclusion in the registry is not evidence that a result was obtained or that a comparison was statistically supported.
Finally, the record reports zero analyses. This distinguishes the presence of registered outcomes from the presence of reported statistical findings. No analysis population, missing-data method, endpoint model, covariate adjustment, or multiplicity procedure is described. These are important methodological details because they determine how outcomes are summarized, how incomplete observations are handled, and how multiple questions are interpreted. In this case, those details are not reported in the registry record.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists a primary ALP change outcome and additional measures covering an ALP-and-bilirubin response, itch NRS change, and fatigue NRS change. Considering multiple outcomes can raise a multiplicity issue: in general, conducting several statistical tests can increase the chance of at least one apparently notable result when all null hypotheses are true. A prespecified testing strategy, adjustment, or hierarchy can address that issue, but the registry does not report any such plan. It also reports zero analyses, so there are no posted tests to interpret.
Missing-data handling is not reported in the registry record. The available outcome information does not state how missing Week 12 measurements, withdrawals, or incomplete assessments were treated. In general, the handling of missing observations matters because an analysis based only on observed values can differ from one that makes explicit assumptions about missing values. No missingness pattern or method can be inferred.
The analysis population is also not reported. The record does not specify whether the primary analysis was intended to include all randomized participants, participants who received treatment, or another defined group. The primary outcome description says zero participants are reported in Cohort A because of personal-information protection based on enrollment futility; that entry does not define an analysis set or permit a reconstruction of one.
Neither a statistical analysis plan nor rules for endpoint derivation, exclusions, protocol deviations, or interim analyses are reported in the registry record. These omissions limit assessment of prespecification and analytic choices. The record’s termination explanation should not be treated as evidence of an unreported interim test or a particular stopping boundary.
11. What the Result Does, and Does Not, Mean
The record establishes that this was a randomized, double-blind, placebo-controlled phase 2 study of baricitinib in participants with primary biliary cholangitis who had an inadequate response to UDCA or could not take it. It documents a primary endpoint of ALP change from baseline to Week 12 and identifies several secondary outcomes. It also reports that the study was terminated following an enrollment-futility decision.
Those details do not establish the size or direction of any treatment effect. The registry does not report a numerical ALP result, a baricitinib-versus-placebo estimate, or a confidence interval. It does not report statistical tests or results for the secondary outcomes. The record therefore cannot show how the groups compared or how precisely any difference was estimated.
Termination is not itself an efficacy result. The stated enrollment-futility reason does not reveal whether observed outcomes favored either group, whether any formal interim analysis occurred, or what criterion informed the decision. Such details are not reported in the registry record. Similarly, the trial’s proof-of-concept description indicates its stated study purpose but does not demonstrate that the intended question was answered.
The absence of reported adverse-event counts in the available arm entries should also be read cautiously. Those entries list zero participants at risk and zero affected for the displayed categories, while Cohort A outcome reporting is restricted for personal-information protection based on enrollment futility. These records do not provide a complete, interpretable safety comparison. No conclusion about comparative safety can be drawn from them.
Thus, the posted information is useful for understanding the registered design and the limits of the available report. It does not establish benefit, lack of benefit, or harm. A quantitative interpretation would require outcome estimates and their uncertainty, along with the analysis methods and populations; these are not reported in the registry record.
12. Why This Trial Matters Statistically
This study illustrates an important distinction between a registered research question and an answer supported by reported data. A detailed design label and a named primary endpoint do not, by themselves, produce an interpretable treatment comparison. The record identifies randomization, double masking, a placebo comparator, and an ALP endpoint, but does not report the estimate or uncertainty needed to evaluate the comparison.
Early termination for enrollment futility can leave a study with limited information. The registry reports that reason for termination, but gives no decision rule or statistical details. For readers, that makes it especially important not to treat the termination status as a surrogate for an efficacy finding. The record indicates why the study ended as recorded, not what the treatment effect was.
The record also demonstrates why outcome availability and analysis reporting are separate issues. It names primary and secondary outcomes and describes planned regimens, but reports zero analyses. The primary endpoint population entry and the privacy-based suppression of Cohort A results further restrict what can be learned from the registry. Without numerical outcome summaries, precision and potential group differences cannot be evaluated.
From a statistical reporting perspective, useful interpretation depends on an estimate, a measure of uncertainty, a defined analysis population, and information about missing data and multiplicity. None of these analytic details is reported. Their absence does not imply that no analyses occurred outside the registry record; it means that the record does not document them.
The study therefore matters as an example of how a registered design can remain visible even when outcome-level evidence is not interpretable from the public record. Keeping design details separate from unreported results helps prevent overstatement and clarifies exactly what further information would be needed for a statistical assessment.
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 main limitation for interpretation is the lack of reported numerical outcome results. The primary endpoint is named, but there is no group-level ALP change, treatment contrast, variability measure, confidence interval, or test result. Without these, neither effect magnitude nor precision can be assessed.
Enrollment is reported as 2, but the record does not give allocation by group. It states that Cohort A results are not reported because of personal-information protection based on enrollment futility, and describes the primary-outcome population as zero participants reported for that reason. These entries prevent reconstruction of outcome summaries and do not clarify how many participants contributed to any analysis.
The study was terminated following enrollment futility, and Cohort B is described as planned rather than completed. The registry does not report a futility definition, a stopping procedure, interim analyses, or outcome information for Cohort B. It is therefore not possible to assess whether the termination decision was tied to an outcome analysis or how it affected planned comparisons.
The registry reports zero analyses and does not describe analysis populations, missing-data methods, endpoint models, covariate adjustments, or multiplicity control. It also does not report implementation details for randomization or masking. The absent information limits assessment of both the conduct and statistical analysis of the study. These are limits of the registry record; they do not establish that the details or analyses do not exist elsewhere.
Finally, adverse-event entries show zero counts for the displayed Cohort A fields, alongside the record’s privacy-based restriction on outcome reporting. Those entries do not provide a sufficiently described basis for comparative safety interpretation. No efficacy or safety conclusion should be inferred from the lack of reportable numerical comparisons.
14. Related Tutorials
15. Related Statistical Calculators
16. Sources
- ClinicalTrials.gov: A Study of Baricitinib (LY3009104) in Participants With Primary Biliary Cholangitis Who do Not Respond or Cannot Take UDCA — NCT03742973.
- Registry record last updated 2020-10-14; status last verified 2019-10.
- Results first posted 2020-10-14.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
17. Record Summary
The registry record describes a randomized, double-blind, placebo-controlled, parallel-group phase 2 study of baricitinib in primary biliary cholangitis among participants with an inadequate response to UDCA or intolerance to it. The primary endpoint was change from baseline in ALP at Week 12. Cohort A descriptions list baricitinib at 2 mg orally once a day and placebo for 12 weeks; Cohort B, involving planned baricitinib at 4 mg and placebo, was not carried forward after the enrollment-futility termination decision.
The record lists enrollment of 2, a status of terminated, four outcome measures, and zero analyses. It does not report a numerical primary-outcome comparison, statistical uncertainty, or results for the secondary outcomes. Cohort A reporting is restricted for personal-information protection based on enrollment futility. The design and termination are documented, but the treatment effects and their precision cannot be assessed from the registry record.