Varlitinib in Combination With Capecitabine for Advanced or Metastatic Biliary Tract Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Varlitinib in Combination With Capecitabine for Advanced or Metastatic Biliary Tract Cancer (NCT03093870), a randomized phase 2, phase 3 study. The primary outcome is Objective Response Rate (ORR) - Part 1.

2. Clinical Question

The trial asks whether adding varlitinib to capecitabine changes objective response in patients with advanced or metastatic biliary tract cancer receiving second-line systemic therapy. Its comparison is varlitinib plus capecitabine versus placebo plus capecitabine, so capecitabine is included in both groups and the randomized contrast concerns the addition of varlitinib rather than capecitabine itself.

The primary endpoint is objective response rate (ORR) in Part 1. The registry defines a response as at least one visit assessment of complete response or partial response. This makes the question one about the proportion of participants meeting that response definition during the specified assessment period. The record describes the assessment period as extending until progression or, when progression was absent, the last evaluable assessment, up to 2 years. It also states that assessments are included regardless of whether a participant discontinued treatment or received subsequent therapy before progression.

The stated aim includes assessing safety and efficacy, but the registry record does not provide primary endpoint results that would answer the ORR comparison. The trial question can therefore be described from its design and endpoint definition, but the posted information does not establish the direction or precision of any observed difference.

3. Trial Design

This was a randomized, double-blind, parallel-group study with placebo control. The registry lists Phase 2 and Phase 3. Its official title identifies the comparison as varlitinib plus capecitabine versus placebo plus capecitabine in patients with advanced or metastatic biliary tract cancer receiving second-line systemic therapy. The trial is marked completed, with enrollment of 151. The listed start date is 2017-07-04 and the primary completion date is 2019-11-12.

The record distinguishes a safety lead-in from Part 1. The primary ORR endpoint is identified as a Part 1 outcome. The adverse-event information separately lists a varlitinib-and-capecitabine safety lead-in group, a varlitinib-and-capecitabine Part 1 group, and a placebo-and-capecitabine Part 1 group. The safety lead-in should not be treated as another randomized Part 1 comparison: the record identifies the randomized, placebo-controlled comparison for Part 1, while the lead-in is presented separately.

Randomization is intended, in general, to reduce systematic differences between groups at allocation, while masking is intended to limit the influence of treatment knowledge on care or assessment. The record identifies the study as double-blind, but further operational details about who was masked and how masking was maintained are not reported in the registry record. The allocation method, stratification factors, and details of treatment assignment are also not reported in the registry record.

The registry describes the primary endpoint’s observation window as continuing until progression or the last evaluable assessment in the absence of progression, up to 2 years. Assessments were included regardless of discontinuation or subsequent therapy before progression. Other outcome measures listed include overall survival, duration of response, disease control rate, tumor size, and laboratory safety measures. The record lists 15 outcome measures, but does not provide analysis estimates for them in the information. More detailed treatment schedules, eligibility criteria, and protocol procedures 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: Period 1: Safety lead-in phase had only 1 arm (Varlitinib and Capecitabine). Period 2: Part 1 phase had 2 arms (Varlitinib and Capecitabine vs Placebo and Capecitabine)

Overall Study

Reasons for not completing this period:

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 Objective Response Rate (ORR) - Part 1, measured over Data obtained up until progression, or until last evaluable assessment in the absence of progression, regardless of whether subjects discontinued treatment or received a subsequent therapy prior to progression, up to 2 years..

Objective response rate is a binary endpoint at the participant level: each participant either meets the registry’s response definition or does not. The record defines a response as at least one visit response of complete response or partial response. It says best overall RECIST response is based on overall visit responses after randomization and before RECIST progression or the last evaluable assessment when progression is absent. The listed response categories also include stable disease, progressive disease, and not evaluable.

ORR is used to describe how many participants meet a specified tumor-response criterion during a defined assessment period. In general, a response endpoint can provide a measure of observed tumor shrinkage under the study’s assessment rules. It does not, by itself, describe how long a response lasts or whether participants live longer; duration of response and overall survival are listed as separate outcomes in this trial.

The registry gives the ORR timeframe as assessments up to progression or the last evaluable assessment in the absence of progression, up to 2 years. It specifies that assessments remain part of the ORR evaluation even if treatment was discontinued or subsequent therapy was received before progression. The record does not report the number of responders, the number assessed, the ORR in either Part 1 group, a between-group estimate, or uncertainty around an estimate. Thus, the definition and observation window are available, but the posted primary endpoint result is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies ORR as a count of participants and defines the endpoint using complete or partial response. It does not report an inferential analysis for that outcome in the record. The distinctions below explain the statistical ideas relevant to this kind of endpoint in general; they should not be read as a description of methods used by the trial unless those methods are explicitly stated in the registry record.

What does a binary endpoint represent?

A binary endpoint classifies each participant into one of two analysis outcomes, such as meeting or not meeting a prespecified response definition. For a group, the observed response proportion is based on the number who meet the definition and the denominator used for analysis. The denominator matters: it may depend on the analysis population and on how participants without an evaluable assessment are treated. The registry gives the response definition and assessment window, but does not report the ORR counts, analysis denominator, or missing-assessment rules.

In general, an estimated proportion is a description of the observed participants, while an estimate of a treatment contrast compares the groups. A contrast might be expressed using a difference or another comparative measure, but the registry record does not report a comparative estimate for ORR. No such value should be inferred from the enrollment total or from the safety counts.

What does uncertainty add?

In general, a confidence interval gives a range of values compatible with the observed data under the statistical method and its assumptions. It helps show precision: a broad interval indicates that the data are compatible with a wider range of effects than a narrow interval. A confidence interval does not ensure that a particular value is true, and it does not remove concerns about bias, missing data, or analysis choices. No confidence interval for the primary outcome is reported in the registry record.

A p-value, in general, describes how unusual data at least as incompatible with a specified null hypothesis would be if that hypothesis and the test assumptions held. It is not the probability that the null hypothesis is true, and it does not measure clinical importance. The registry record does not report a primary ORR p-value or identify a primary test for the comparison.

Why do the response rules matter?

Response assessment depends on the criteria used, the timing and completeness of assessments, and the rule for selecting a participant’s best response. The registry says best overall RECIST response is based on overall visit responses after randomization and before progression or the last evaluable assessment when progression is absent. It lists complete response, partial response, stable disease, progressive disease, and not evaluable as categories. The ORR definition counts participants with at least one visit response of complete response or partial response. These rules establish what counts as an event, but do not reveal how many participants met the definition.

The registry states that ORR data were included regardless of treatment discontinuation or receipt of subsequent therapy before progression. That statement describes the stated assessment scope. It does not fully specify the statistical handling of intercurrent events, missing evaluations, or participants classified as not evaluable. Detailed analysis rules for those situations are not reported in the registry record.

How should planning illustrations be understood?

Sample-size calculations are planning tools, not trial outcomes. In general, a calculation depends on a target effect, significance level, allocation, outcome variability or event rate, and statistical method. A calculation based on an observed effect is especially conditional: it uses the observed estimate as an input and does not establish that the effect is reliable or that the trial had a particular probability of success. Such a calculation is not observed power. The planning assumptions identified for any illustration are a two-sided significance level of 0.05, equal allocation, and a normal approximation; the calculation is framed around detecting a stated fraction of an observed effect. The registry record does not report the original trial’s sample-size justification or design assumptions.

General concepts such as confidence intervals, p-values, and planning calculations help organize interpretation, but they cannot fill gaps in the reported results. For this trial, the registry defines the endpoint and design but does not report the response counts, treatment-effect estimate, uncertainty interval, or primary hypothesis-test result. Consequently, there is no primary statistical result available in the registry record to interpret.

10. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates more than one outcome, time point, group contrast, or hypothesis. In general, conducting many tests can increase the chance of obtaining at least one apparently notable result by chance. Prespecified prioritization or adjustment procedures can address this issue, but the registry record does not report a multiplicity strategy for this trial. It lists multiple outcomes, including ORR, overall survival, duration of response, disease control rate, tumor size, and laboratory safety measures. The record does not specify which secondary outcomes were confirmatory, whether any were adjusted for multiple testing, or how the primary and secondary analyses were organized.

The primary ORR description defines an observation period through progression or the last evaluable assessment without progression, up to 2 years, and states that assessments are included regardless of treatment discontinuation or subsequent therapy before progression. This does not fully specify the analysis population or how missing or unevaluable response data were handled. The number of participants with missing endpoint assessments, the reasons for missingness, and the statistical method for addressing missing data are not reported in the registry record.

The registry record also does not report whether the primary analysis included all randomized participants, only participants with evaluable assessments, or another population. It does not report interim analyses, stopping rules, or a statistical analysis plan. These omissions limit assessment of how analysis choices and multiple outcomes may affect interpretation. The stated endpoint definition remains informative, but it cannot replace the missing details about multiplicity, analysis populations, and incomplete outcome data.

11. 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: acute kidney injury (8.3%), bile duct obstruction (8.3%), enterocolitis (8.3%), blood bilirubin increased (7.9%), cholangitis (6.3%).

The registry lists safety information for three groups: varlitinib and capecitabine in the safety lead-in, varlitinib and capecitabine in Part 1, and placebo and capecitabine in Part 1. For serious adverse events, it reports 13 affected among 24 at risk in the safety lead-in, 25 among 64 at risk in the varlitinib Part 1 group, and 27 among 63 at risk in the placebo Part 1 group. The record also lists deaths of 2, 35, and 35 in those respective groups. These are counts, not estimates of a causal treatment effect.

The registry separately records other adverse-event counts: 24 among 24 at risk in the safety lead-in, 64 among 64 at risk in the varlitinib Part 1 group, and 59 among 63 at risk in the placebo Part 1 group. These categories should not be conflated with serious adverse events. The registry’s list of top serious adverse-event terms includes acute kidney injury, bile duct obstruction, and enterocolitis at 8.3%; blood bilirubin increased at 7.9%; cholangitis at 6.3%; and hyperbilirubinaemia at 6.2%. The group attribution and percentage denominators for these terms are not reported in the registry record.

Adverse-event counts describe reported events in the listed groups; they do not by themselves establish that treatment caused an event. The record does not provide sufficient detail about event definitions, severity grading, exposure time, or adjusted safety comparisons. The safety lead-in and randomized Part 1 groups also represent distinct study components, so their counts should be read in their stated group context.

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

The record establishes the trial’s design, clinical comparison, and primary endpoint definition. It describes a randomized, double-blind study of varlitinib plus capecitabine versus placebo plus capecitabine in advanced or metastatic biliary tract cancer, with ORR as the primary outcome in Part 1. It also defines the response categories and the assessment period. These details make clear what comparison the trial was designed to address.

However, the registry record does not report the primary ORR counts or a comparative estimate, confidence interval, or p-value. It therefore does not show how frequently participants in either Part 1 group met the response definition or how precisely the groups differed. The completed status and enrollment total are study descriptors, not evidence about the primary outcome. The listed safety counts and other outcomes likewise cannot stand in for the missing ORR result.

In general, a trial result is interpreted using both the size of an estimated effect and its uncertainty, alongside study design and limitations. A response endpoint is only one dimension of outcome, and the registry lists overall survival and duration of response separately. Their analysis results are not reported in the registry record. Nor does the available information establish the clinical importance of any effect, because no primary comparative result is posted.

Accordingly, the record supports a description of the research question and how ORR was defined, but not a conclusion about the observed efficacy comparison. It does not establish that either regimen was superior, equivalent, or ineffective. Those judgments would require outcome results and their uncertainty, together with the relevant analysis details. The trial’s stated aim to assess safety and efficacy should not be mistaken for a reported finding.

13. Why This Trial Matters Statistically

This trial is statistically informative as an example of why a clear study question and a complete results record are both necessary. Its randomized, double-blind, placebo-controlled design identifies a planned comparison in which both groups receive capecitabine and the investigational contrast is the addition of varlitinib. The record also distinguishes a safety lead-in from the randomized Part 1 groups, which matters when identifying which participants belong to the primary efficacy comparison.

The primary endpoint has a recognizable participant-level definition: complete or partial response on at least one visit, assessed using best overall RECIST response within a specified follow-up window. That definition helps readers understand what an ORR analysis would count. Yet interpretation depends on more than the endpoint label. Response counts, denominators, a comparative estimate, and uncertainty are needed to judge the observed result, and they are not reported in the registry record.

The distinction between endpoint definition and endpoint finding is central. A registry can describe what was intended to be measured without posting the data needed to quantify the comparison. Likewise, safety counts across a lead-in and randomized groups do not supply the missing efficacy analysis. The record’s multiple listed outcomes also make prespecification and multiplicity procedures relevant, but those methods are not reported.

For biostatistical readers, the case highlights the importance of analysis-population definitions, missing-data rules, and transparent uncertainty estimates. These determine how the available observations become an estimate and how cautiously it should be interpreted. Without those results and details, the registry supports a precise account of the design and endpoint, but not a quantitative conclusion about the primary question.

14. 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 statistical interpretation is that the registry record does not report the primary ORR result. It gives no Part 1 responder counts, analysis denominators, treatment-effect estimate, confidence interval, or p-value. Without these, neither the direction nor precision of the efficacy comparison can be assessed from the record.

The record describes the response observation window and states that assessments were included regardless of treatment discontinuation or subsequent therapy before progression. It does not report how missing or unevaluable assessments were handled, how the primary analysis population was defined, or the number of participants with incomplete endpoint information. It also does not report a statistical analysis plan, a multiplicity strategy, or interim-analysis procedures.

The safety information separates a safety lead-in from two Part 1 groups, but does not provide detailed context about exposure, event timing, severity grading, or causal assessment in the information. The listed serious adverse-event term percentages do not have reported group attribution or denominators. Consequently, these terms should not be treated as comparative rates.

Although the trial is marked completed and enrollment is reported as 151, those details do not address the primary outcome. The record lists multiple secondary outcomes but does not report their analysis estimates. Further eligibility and treatment-procedure details are also not reported in the registry record. These gaps restrict assessment of generalizability, analysis choices, and the robustness of any efficacy or safety interpretation.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Varlitinib in Combination With Capecitabine for Advanced or Metastatic Biliary Tract Cancer — NCT03093870.
• Registry record last updated 2021-08-03; status last verified 2021-06.
• Results first posted 2021-08-03.
• 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.

18. Record Summary

The registry describes a completed, randomized, double-blind, parallel-group study of varlitinib plus capecitabine versus placebo plus capecitabine as second-line systemic therapy for advanced or metastatic biliary tract cancer. It lists Phase 2 and Phase 3, enrollment of 151, a start date of 2017-07-04, and a primary completion date of 2019-11-12.

The primary Part 1 outcome is objective response rate, defined as the number of participants with at least one visit response of complete response or partial response. Assessment extends until progression or the last evaluable assessment without progression, up to 2 years, including assessments after treatment discontinuation or subsequent therapy before progression.

The record lists safety lead-in and Part 1 adverse-event counts, as well as secondary outcomes including overall survival and duration of response. Primary ORR counts and comparative analysis results, along with key details about missing data, analysis populations, and multiplicity, are not reported in the registry record.