A Study of SHR-1210 in Combination With Apatinib or Chemotherapy in Subjects With Advanced PLC or BTC: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of SHR-1210 in Combination With Apatinib or Chemotherapy in Subjects With Advanced PLC or BTC (NCT03092895), a non-randomized phase 2 study. The primary outcome is The Safety and Tolerability.

2. Clinical Question

3. Trial Design

This was a phase 2 study with a parallel-group model, non-randomized allocation, and no masking. It enrolled 157 participants with advanced PLC or BTC. The registry lists the study as completed. Its recorded start date is 2017-04-24, and its primary completion date is 2021-03-31. The sponsor was Jiangsu HengRui Medicine Co., Ltd.

The study evaluated SHR-1210 in combination with apatinib or chemotherapy. The treatment pathway depended on disease and treatment history: participants with advanced PLC who had failed or could not tolerate prior systemic therapy were to receive SHR-1210 plus apatinib; participants with advanced PLC or BTC who had not received prior systemic therapy were to receive SHR-1210 plus FOLFOX4 or GEMOX. The registry’s adverse-event groups distinguish an apatinib group, a hepatocellular carcinoma chemotherapy group, a cholangiocarcinoma chemotherapy group, and an “Others” group.

Because allocation was not randomized and masking was absent, the treatment groups do not constitute randomized comparisons. In particular, the groups differ in the clinical pathways described in the registry, including cancer type and prior-treatment status. This design can describe safety events observed among treated participants, but it cannot by itself separate treatment effects from those differences or from other sources of confounding.

The primary endpoint was safety and tolerability, expressed as the incidence of adverse events and serious adverse events assessed by NCI-CTCAE v4.03. The registry gives a timeframe of up to approximately 4 years and states that the safety endpoint was analyzed in the safety population, defined as participants who were treated. The registry record does not report a detailed statistical analysis plan, allocation procedures beyond non-randomized assignment, or a sample-size justification.

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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: The primary safety endpoint was analyzed in SS. The SS included all participants who were treated .

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 The Safety and Tolerability, measured over Up to approximately 4 years.

The primary endpoint addressed safety and tolerability by assessing the incidence of adverse events (AEs) and serious adverse events (SAEs). The registry specifies NCI-CTCAE v4.03 as the assessment standard and identifies the safety population as all participants who were treated. The stated timeframe is up to approximately 4 years.

In general, an adverse-event endpoint records unwanted medical events occurring during a study; a serious adverse event is a separate seriousness classification, not simply a synonym for an event that is severe in intensity. A standardized grading framework can support consistent recording, but the registry record does not provide the operational definitions, attribution rules, or detailed tabulation methods used for this study.

Safety endpoints are used to describe the types and frequency of events observed during treatment and follow-up. Counts help show how many participants in a group experienced an event, while event descriptions provide context about what was recorded. Such summaries do not, on their own, establish that treatment caused an event. Interpretation also depends on the population at risk, follow-up, how events were collected, and the clinical differences among groups.

The registry identifies safety and tolerability as the primary endpoint but does not give an inferential estimate, confidence interval, or hypothesis test for that endpoint. The posted adverse-event summaries report participant counts by group. They should therefore be read as descriptive safety information, not as a randomized comparison or a complete estimate of treatment-attributable risk.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The study’s key design feature for statistical interpretation is its non-randomized allocation. Randomization, in general, is a method for assigning participants by chance to treatment groups. Its purpose is to reduce systematic differences between groups, including differences that may not be measured. This study did not use randomized allocation, so the treatment groups may differ in ways related to prognosis or treatment choice. The registry describes treatment pathways linked to prior systemic therapy and cancer type, making a direct comparison especially difficult to interpret as a treatment effect.

What does the parallel-group design mean?

In a parallel-group study, participants receive one assigned treatment pathway during the study rather than crossing between study treatments as part of the design. Here, the registry lists an SHR-1210 plus apatinib arm and chemotherapy combinations involving SHR-1210 plus FOLFOX4 or GEMOX, with further group distinctions in the adverse-event record. The parallel model does not make the groups comparable by itself. Comparability depends on how participants entered each group and on their clinical characteristics.

How should open-label status be understood?

Open-label means masking was absent: treatment assignment was not concealed from participants and study personnel in the way it would be in a blinded design. In general, knowledge of treatment can influence reporting, assessment, or management of events, although the extent and direction of any such influence cannot be determined from this record. The registry identifies the study as unmasked but does not describe event-review procedures or whether any independent adjudication was used.

What do the safety counts represent?

The primary endpoint concerns the incidence of adverse events and serious adverse events in the safety population. The record defines that population as participants who were treated. In general, participant counts indicate how many people experienced at least one event in a category; they are distinct from event counts, since one person may experience more than one event. The posted group summaries report participants with events and participants at risk. They do not include a participant-level account of repeated events, event duration, or exposure time.

The registry specifies NCI-CTCAE v4.03 for assessment. Such a grading framework is intended to standardize descriptions of adverse events. However, the record does not report the detailed event coding, grade-specific summaries, attribution to treatment, or rules for deciding whether events were related to the study drugs. It also does not report a statistical comparison of the groups for safety.

What can uncertainty measures add?

In general, a confidence interval describes the range of estimates compatible with the data and a specified statistical procedure. It communicates precision as well as the estimated quantity. The registry record does not report confidence intervals for the primary safety endpoint. Without them, readers cannot assess the statistical precision of the group-specific event summaries using the record alone. A lack of an interval is not evidence that the groups are equivalent or that an event rate is known precisely.

In general, over-dispersion occurs when outcome variability is greater than a model assumes. It can make standard errors too small if ignored. The registry record does not report a model for the primary endpoint or state whether over-dispersion was assessed; therefore, it is not possible to determine whether that issue arose in the study’s analysis.

How should planning calculations be interpreted?

Any planning illustrations based on an observed effect are hypothetical calculations rather than evidence about the study’s original design. The stated assumptions for such illustrations include a two-sided significance level of 0.05, equal allocation, and a normal approximation; the effect to detect is expressed as a fraction of the observed effect. These assumptions do not establish that the trial used those methods to plan enrollment. The registry record does not report the original sample-size justification or a prospective power calculation. Planning illustrations are not observed power and do not substitute for an analysis of the study’s uncertainty.

For time-to-event outcomes, general methods often account for the number and timing of events, while binary-outcome methods typically concern participant counts in outcome categories. The registry lists several time-to-event secondary outcomes but does not report their results or the analysis procedures used. It also does not provide the detailed analysis plan needed to assess censoring, model assumptions, or handling of incomplete follow-up. Those elements are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple secondary outcomes: objective response rate, duration of response, disease control rate, time to progression, overall survival, time to response, and progression-free survival. These are distinct measures and may involve different outcome definitions and follow-up. When several outcomes are tested, the chance of at least one nominally significant finding can increase if no multiplicity strategy is used. In general, a prespecified hierarchy or adjustment can address this issue, but the record does not report whether any such approach was planned or applied.

The registry record does not report results for these secondary outcomes, nor does it provide analysis estimates or inferential tests for the primary safety endpoint. The statistical analysis plan, outcome-specific analysis populations, and any planned interim analyses are not reported in the registry record. No interim analysis should be assumed from the information available.

Missing-data handling is also not reported in the registry record. This matters because incomplete outcome assessment or loss to follow-up can affect interpretation, particularly for response and time-to-event outcomes. The record does not describe how missing assessments, withdrawals, censoring, or participants without evaluable outcomes were handled.

The primary safety analysis population is described as all treated participants. This identifies a population for safety reporting, but the record does not give further rules for inclusion, exclusions, or analysis of participants with incomplete safety follow-up. Without a reported plan for missing data and multiplicity, the registry record does not support a full assessment of the statistical reliability of unreported efficacy outcomes or of potential selection among analyses.

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: multiple organ dysfunction syndrome (33.3%), febrile neutropenia (33.3%), hepatic failure (14.3%), upper gastrointestinal haemorrhage (11.8%), disease progression (10.9%).

The serious adverse-event summaries give participants at risk and participants affected within each registry group. The counts are 14 of 28 in the SHR-1210 plus apatinib group, 20 of 34 in the hepatocellular carcinoma SHR-1210 plus FOLFOX4 group, 58 of 92 in the cholangiocarcinoma SHR-1210 plus FOLFOX4 or GEMOX group, and 2 of 3 in the “Others” group. For other adverse events, the corresponding affected counts are 28 of 28, 34 of 34, 92 of 92, and 3 of 3.

These figures are participant counts, not counts of every event. A participant may have experienced more than one event, and the summary does not provide event-level detail or follow-up time by group. The serious-event terms listed in the registry are accompanied by percentages, but the denominator and calculation are not described. The percentages therefore need that context before being interpreted as participant-level risks.

Because allocation was non-randomized and the groups follow different clinical pathways, these summaries should not be treated as controlled comparisons. The counts describe recorded events in the listed groups; they do not establish that treatment caused an event or that one combination is safer than another.

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

The registry record establishes that the study was a completed, non-randomized, open-label phase 2 study of SHR-1210 combinations in advanced PLC or BTC. It identifies safety and tolerability as the primary endpoint, specifies NCI-CTCAE v4.03 assessment, defines the safety population as treated participants, and posts group-specific counts of participants with serious and other adverse events. Those observations can be read descriptively as safety information from the study groups.

They do not establish comparative treatment effects. Participants were not randomized, and the registry describes distinct treatment pathways based on disease and prior systemic therapy. Consequently, group differences cannot be attributed to the treatment combinations alone. The posted summaries also do not provide adjusted comparisons, confidence intervals, or a detailed analysis plan that would help quantify uncertainty or account for differences between groups.

The record does not report results for its listed secondary outcomes, including response, progression, and survival measures. It therefore does not support conclusions about the comparative efficacy of the combinations. Similarly, adverse-event counts alone do not show whether individual events were caused by study treatment, how long participants were followed, or how many repeated events occurred.

In general, a study’s findings are most useful when considered together with its design, outcome definitions, analysis procedures, and uncertainty estimates. Here, several of those details—including the original sample-size justification, multiplicity approach, and missing-data methods—are not reported in the registry record. The appropriate interpretation is limited to the study characteristics and descriptive safety counts that are recorded, without extending them to causal or comparative claims.

13. Why This Trial Matters Statistically

This study illustrates why design information is essential when reading clinical trial statistics. Although the registry lists several treatment groups and records adverse events for each, allocation was non-randomized. The groups were also organized around different clinical circumstances, including cancer type and prior systemic therapy. Their counts are therefore not interchangeable estimates from otherwise comparable groups.

The distinction between a participant count and an event count is another important statistical consideration. The safety summary reports how many participants were recorded with serious and other adverse events, but it does not show the total number of individual events or repeated events per participant. Those quantities answer different questions and should not be conflated.

The record also demonstrates the difference between having an outcome listed and having its results reported. Several response and time-to-event outcomes are named, yet their estimates and analyses are not reported in the registry record. Without effect estimates and measures of precision, a list of intended outcomes cannot establish efficacy or the strength of evidence.

Finally, the study’s safety endpoint is clinically important as a descriptive measure, but its interpretation depends on population definition, assessment, follow-up, and analytical detail. The record identifies the treated safety population and assessment standard, while leaving several methods unreported. This is a useful reminder that trial registration details and posted results can support some questions—such as what was studied and what event counts were recorded—while leaving other questions unresolved.

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 registry record contains limited detail about how non-randomized treatment assignment was implemented. It does not report the criteria used to select among the treatment pathways beyond the summary description relating to cancer type and prior systemic therapy. It also does not provide baseline characteristics by group. Without those details, the degree of clinical comparability between groups cannot be assessed from the record.

The adverse-event summary does not provide event-level data, grade-specific counts, exposure-adjusted rates, or group-specific follow-up duration. It does not explain the denominator for the percentages attached to serious-event terms. The record also does not report how treatment relatedness was assessed or whether events were reviewed independently. These omissions limit interpretation of the safety summaries.

Although the registry lists secondary outcomes, it does not report their results or the detailed definitions and statistical methods used for them. The analysis plan, missing-data procedures, multiplicity strategy, and original sample-size justification are not reported in the registry record. Confidence intervals and inferential tests for the primary safety endpoint are also absent.

The enrollment figure and the group-specific safety counts describe the registered study, but they do not resolve these design and reporting limitations. In particular, the non-randomized structure means that observed differences between groups should not be treated as causal estimates. The record supports a bounded description of study conduct and posted safety counts, not a comprehensive evaluation of comparative benefit or risk.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of SHR-1210 in Combination With Apatinib or Chemotherapy in Subjects With Advanced PLC or BTC — NCT03092895.
• Registry record last updated 2026-01-29; status last verified 2022-04.
• Results first posted 2026-01-29.
• 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

NCT03092895 is a completed, open-label, non-randomized phase 2 study of SHR-1210 combined with apatinib or chemotherapy in participants with advanced PLC or BTC. The registry records 157 participants, a start date of 2017-04-24, and a primary completion date of 2021-03-31. The sponsor was Jiangsu HengRui Medicine Co., Ltd.

The primary endpoint was safety and tolerability, measured through adverse events and serious adverse events assessed by NCI-CTCAE v4.03 in participants who were treated. The registry posts group-specific participant counts for serious and other adverse events. It also lists response, progression, and survival outcomes, but their results are not reported in the registry record.

Because the study was non-randomized and its treatment pathways reflect different clinical groups, the safety summaries are descriptive rather than like-for-like comparisons. The record does not report a detailed statistical analysis plan, missing-data methods, multiplicity strategy, or original sample-size justification.