Study of BGB-10188 as Monotherapy, and in Combination With Zanubrutinib, and Tislelizumab: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of BGB-10188 as Monotherapy, and in Combination With Zanubrutinib, and Tislelizumab (NCT04282018), a non-randomized phase 1, phase 2 study. The primary outcome is Part E: Overall Response Rate (ORR) as Per Response Evaluation Criteria in Solid Tumors (RECIST) v.1.1.

2. Clinical Question

3. Trial Design

This was a Phase 1/2, parallel-group study, recorded as completed, with enrollment of 97. The study began on 2020-04-29, and primary completion was 2024-08-28. BeiGene is identified as the sponsor. Masking is listed as none. The overall allocation is recorded as non-randomized.

The study had distinct parts rather than a single treatment comparison. Part A evaluated BGB-10188 monotherapy through dose escalation in participants with relapsed or refractory chronic lymphocytic leukemia or small lymphocytic lymphoma, marginal zone lymphoma, follicular lymphoma, mantle cell lymphoma, or diffuse large B-cell lymphoma. Part B evaluated BGB-10188 with zanubrutinib in relapsed or refractory follicular lymphoma, mantle cell lymphoma, and diffuse large B-cell lymphoma. Part D evaluated BGB-10188 with tislelizumab in participants with unresectable locally advanced or metastatic solid tumors. Part E was a dose expansion in participants with platinum-resistant ovarian cancer.

The recorded dose regimens varied across these parts. Part A lists BGB-10188 doses of 60 mg, 120 mg, 240 mg, 360 mg, and 540 mg. Part B lists BGB-10188 240 mg with zanubrutinib 160 mg. Part D lists BGB-10188 doses from 20 mg through 540 mg with tislelizumab 200 mg. Part E lists BGB-10188 160 mg or 320 mg with tislelizumab 200 mg. These are different disease and dose cohorts, not interchangeable arms in a single controlled comparison.

There is a design-description inconsistency worth noting: the overall allocation field says non-randomized, while the Part E arm descriptions say participants were randomized between the dose regimens. The record does not explain this difference. The study purpose included determining a maximum tolerated dose and a recommended dose for expansion, as well as assessing safety and tolerability. The primary outcome available in the record is instead the Part E overall response rate.

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: A total of 97 participants were enrolled, out of which 96 were randomized and treated. One participant was enrolled but not treated. Part C was cancelled by the sponsor and not initiated.

Only the first 6 of 14 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analysis was performed on safety analysis set that included all participants in Part A who received at least one dose of BGB-10188.

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: Analysis was performed on participants from safety analysis set that included participants who received at least one dose of BGB10188 and/or zanubrutinib and/or tislelizumab.

Only the first 6 of 15 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Part E: Overall Response Rate (ORR) as Per Response Evaluation Criteria in Solid Tumors (RECIST) v.1.1, measured over Up to 10.0 months.

The primary endpoint is Part E overall response rate (ORR) under RECIST v.1.1. It is defined as the percentage of participants who had a complete response (CR) or partial response (PR), assessed up to 10.0 months. Under the registry definition, CR requires disappearance of target and non-target lesions and no new lesions; pathological lymph nodes must have a short axis less than 10 mm. PR requires at least a 30% decrease in the sum of target-lesion diameters relative to the baseline sum.

ORR is a binary response outcome summarized as a proportion of participants. It gives a direct description of how many participants met the prespecified response categories during the stated assessment period. It does not, by itself, describe how long a response lasts, the time until progression, or survival. Those are distinct outcomes and should not be inferred from a response percentage.

The registry specifies a 95% confidence interval for the endpoint and defines the efficacy analysis set as participants who received at least one dose of BGB-10188 and/or tislelizumab on or after Cycle 1 Day 1. However, neither the ORR estimate nor its confidence interval is reported in the registry record. The endpoint definition therefore clarifies what was intended to be measured, but it does not provide the observed Part E result.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

For a binary endpoint such as ORR, each participant is classified according to whether the response definition was met during the relevant assessment period. The group response estimate is the proportion classified as having CR or PR. The registry expresses the endpoint as a percentage of participants. A percentage is an intuitive summary, but it does not show how much uncertainty surrounds the estimate unless an interval and the underlying analysis population are also available.

What does the confidence interval contribute?

A confidence interval is a range of estimates compatible with the observed data under a specified statistical procedure. It communicates sampling uncertainty around the estimated response percentage; it is not a range containing a known fixed proportion with certainty, nor does it indicate the probability that a particular treatment effect is true. The registry identifies a 95% confidence interval as the dispersion measure for the primary endpoint, but the interval limits are not reported in the registry record. Without them, readers cannot assess precision or see which values remain compatible with the observed data.

For binary outcomes, the interval method matters, especially when the analysis group is small or the response proportion is near a boundary. Different procedures can produce different limits. The statistical procedure used for the Part E interval is not reported in the registry record. No method should be imputed from the endpoint label alone.

How should dose groups be interpreted?

Dose escalation is intended to examine different dose levels as part of evaluating tolerability and selecting doses for further study. Dose expansion examines selected regimens in a specified population. These aims differ from estimating a treatment effect against a concurrent control. In this study, the parts involved different tumor populations, drug combinations, and dose levels. A raw difference between group response percentages could reflect differences in population, eligibility, follow-up, or other features in addition to dose. Without a reported analysis that addresses those issues, such a contrast is descriptive rather than a clean treatment-effect estimate.

The registry’s overall allocation classification is non-randomized, while the Part E arm descriptions state that participants were randomized to the two dose regimens. Random allocation, when used and implemented as planned, can balance measured and unmeasured baseline factors on average. But the record’s inconsistent descriptions make it important not to assume a particular assignment procedure. The method of randomization, any allocation concealment, and the explanation for this discrepancy are not reported in the registry record.

What do the planning illustrations mean?

Planning calculations are not trial outcomes. The stated illustrations use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They frame the effect to detect as a stated fraction of the observed effect. Such scenarios can illustrate how a hypothetical future study’s required enrollment may vary with the target effect and assumptions. They do not describe the number of participants actually analyzed, establish the original protocol’s sample-size justification, or estimate the probability that this completed study detected an effect. Sample sizes in these calculations are planning illustrations, never observed power.

For a binary response, the planning result depends on the assumed response rates in the comparison groups as well as the desired power, allocation, and significance threshold. The assumptions and calculated values should therefore be interpreted as conditional on that setup, rather than as a feature of the observed trial. The registry does not report the protocol’s sample-size justification or the reasoning used to select the study enrollment. It also does not report a statistical analysis plan for the primary endpoint.

Why can over-dispersion matter?

In general, over-dispersion means that observed variability is greater than a statistical model expects. A model that fails to account for that extra variability can make uncertainty appear too small. This concept is more often relevant to modeled counts or clustered outcomes than to a simple, unadjusted binary proportion. The registry information does not identify a model for the primary ORR analysis, so there is no basis to state whether over-dispersion was assessed or whether it was relevant to the sponsor’s analysis.

The registry lists 28 outcome measures and 0 analyses. The count of outcomes indicates that the study record includes many measured endpoints; it does not reveal how many were tested, how they were prioritized, or which analyses were prespecified. A complete interpretation would require the endpoint-specific analysis methods, analysis population, handling of participants without an evaluable response, and any adjustment or hierarchy used for multiple comparisons. Those details are not reported in the registry record.

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("Part E: Dose Expansion- BGB-10188- 160 mg + Tislel", "Part E: Dose Expansion- BGB-10188- 320 mg + Tislel")
n      <- c(9, 5)
events <- c(0, 0)
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  Part E: Dose Expansion- BGB-10188- 160 mg + Tislel  9  0  0.0  0.0  33.6  -0.0  29.9  0.3  30.8
ARM  Part E: Dose Expansion- BGB-10188- 320 mg + Tislel  5  0  0.0  0.0  52.2  0.0  43.4  0.4  45.9

11. Multiplicity, Missing Data and Analysis Populations

The study record lists 28 outcome measures, including the Part E primary ORR and several secondary measures related to response and pharmacokinetics. With multiple outcomes, dose groups, and disease settings, testing many hypotheses can increase the chance of obtaining apparently notable findings by chance. Statistical approaches to this issue include prespecifying a primary endpoint, ordering tests, or adjusting significance thresholds. Which approach was used for this study is not reported in the registry record. The record also reports 0 analyses, so it does not document endpoint-specific analysis results or a multiplicity procedure.

The primary endpoint population is described as the efficacy analysis set, including participants who received at least one dose of BGB-10188 and/or tislelizumab on or after Cycle 1 Day 1. This identifies an inclusion criterion, but does not fully describe the analysis population. The number meeting that criterion, the number with response assessments, and the handling of participants who lacked an evaluable assessment are not reported in the registry record.

Missing data can affect a binary response estimate if participants without an assessment are excluded, counted as not responding, or handled using another method. These choices can yield different estimates, particularly when missingness is related to disease status, tolerability, or follow-up. The missing-data rule for the primary endpoint is not reported in the registry record. No imputation method or sensitivity analysis is described.

The record does not report interim analyses, stopping rules, or the protocol’s sample-size justification. It also does not explain how the overall non-randomized classification relates to the Part E descriptions stating random assignment. These omissions limit assessment of prespecification, the analysis population, and the interpretation of any comparison between dose regimens.

12. 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 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pyrexia (40.0%), alanine aminotransferase increased (30.0%), aspartate aminotransferase increased (30.0%), staphylococcal bacteraemia (25.0%), acute myeloid leukaemia (25.0%).

The registry records serious adverse events, other adverse events, and deaths separately for the listed treatment groups. The group entries give counts at risk and counts affected; these should be read as group-specific counts, not as comparative estimates adjusted for differences between groups. The number affected is not a measure of event severity beyond the category in which it is listed, and the entries do not provide the timing, duration, or outcome of each event.

The listed serious-event terms and frequencies are pyrexia, 40.0%; alanine aminotransferase increased, 30.0%; aspartate aminotransferase increased, 30.0%; Staphylococcal bacteraemia, 25.0%; acute myeloid leukaemia, 25.0%; and febrile neutropenia, 20.0%. The registry record does not specify in the registry record the denominator or calculation details behind the listed term percentages, so they should be read as registry-reported frequencies rather than independently interpreted rates.

Deaths are recorded in several groups, with the group-specific counts varying across the listed regimens. A death count in a group does not by itself establish its cause or attribution to study treatment. The record does not provide a controlled safety comparison, pooled risk estimate, or causal assessment in the information. Differences in group size, dose, combination, and disease context prevent these descriptive entries from serving as like-for-like comparisons.

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

The registry record identifies a clinically defined response outcome in a specific expansion population: ORR among participants with platinum-resistant ovarian cancer receiving BGB-10188 with tislelizumab. It also describes the RECIST v.1.1 response criteria and the efficacy analysis set. These details clarify the endpoint’s intended meaning and the kinds of response that count toward it.

However, the endpoint’s observed result is not reported in the registry record. The response percentage, confidence interval, and endpoint denominator are missing in the registry record. Therefore, the record cannot support a numerical conclusion about the frequency of response or the precision of that estimate. A treatment effect, whether favorable or unfavorable, cannot be inferred from an endpoint definition alone.

Nor does ORR alone describe response durability, time to progression, survival, or comparative safety. Those outcomes require their own estimates, definitions, follow-up information, and analysis. The record lists other study outcomes, but their results are not given in the registry record for this explanation. The safety entries are descriptive by group and do not establish that a regimen caused an event or changed risk relative to a control.

The broader study included several populations and regimens, and its overall allocation is recorded as non-randomized. Although the Part E descriptions say participants were randomized to dose regimens, the record does not reconcile that statement with the overall classification. Accordingly, the design should not be treated as a simple, definitive comparison between treatment strategies. The data described in the record set out the study’s aims and selected safety counts; they do not provide enough primary endpoint information to quantify the Part E response result.

14. Why This Trial Matters Statistically

This study illustrates why a registry description and an outcome result answer different questions. The record is informative about the study’s scope, treatment parts, population definitions, and RECIST-based primary endpoint. Those elements allow a reader to understand what the investigators intended to assess. But without the ORR estimate, confidence interval, and analysis denominator, the endpoint cannot be evaluated quantitatively.

The multi-part design also highlights the limits of reading across cohorts. Monotherapy escalation, a combination with zanubrutinib, and a combination with tislelizumab were studied in different disease contexts and dose groups. This structure supports dose and safety exploration, but it does not create a single common comparator. Group counts or unadjusted response contrasts should not be mistaken for randomized treatment effects, particularly given the difference between the overall allocation entry and the Part E randomization wording.

Statistically, the record emphasizes the importance of reporting uncertainty alongside an estimate. A percentage without a denominator or interval can be difficult to judge, while a stated confidence level without the interval limits gives no direct view of precision. Analysis-population rules and missing-data handling also matter because they determine which participants contribute to the endpoint and how unavailable assessments are treated.

Finally, planning calculations have a separate role from trial results. Assumption-based scenarios may help frame future study sizes, but they are not evidence that the current study had a particular probability of success. They do not replace the protocol’s rationale, a prespecified analysis, or outcome estimates. For this record, the key statistical lesson is that design descriptions and safety counts cannot fill the gap left by an unreported primary endpoint result.

15. 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 limitation for interpreting efficacy is the absence of the Part E ORR estimate, its 95% confidence interval, and the number of participants contributing to that analysis. The registry defines the endpoint and gives an assessment timeframe of up to 10.0 months, but it does not provide the actual response result. This prevents an assessment of both magnitude and precision.

The record covers distinct disease populations, treatment combinations, and dose levels. Results from one part or group would not automatically generalize to another, and the overall enrollment of 97 does not establish the size of the Part E analysis population. The record also has an unresolved design-description discrepancy: overall allocation is listed as non-randomized, while the Part E descriptions state that participants were randomized to dose regimens. Details that might clarify this, including the assignment method, are not reported in the registry record.

Additional endpoint-specific limitations include unreported analysis methods, missing-data rules, multiplicity handling, and interim-analysis procedures. The efficacy-set definition is available, but the record does not give the actual number included or describe how participants without evaluable response data were handled. The statistical procedure for calculating the confidence interval is also not reported.

The safety information is descriptive and organized by group. It does not give a comparative analysis or sufficient detail to interpret the cause, timing, and clinical course of the recorded events. The listed serious-event term percentages also lack denominator and calculation detail in the registry record. These limitations mean that neither response nor safety differences across the groups can be interpreted as controlled comparative effects from the registry information alone.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study of BGB-10188 as Monotherapy, and in Combination With Zanubrutinib, and Tislelizumab — NCT04282018.
• Registry record last updated 2026-02-20; status last verified 2026-02.
• Results first posted 2026-02-20.
• 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.

19. Record Summary

NCT04282018 is a completed Phase 1/2 study sponsored by BeiGene, with enrollment of 97. It evaluated BGB-10188 as monotherapy and in combinations with zanubrutinib or tislelizumab across mature B-cell malignancies and solid tumors. The study included dose escalation and expansion parts, with Part E focused on platinum-resistant ovarian cancer.

The primary endpoint was Part E ORR by RECIST v.1.1, defined as the percentage of participants with CR or PR and assessed up to 10.0 months. A 95% confidence interval was specified, and the efficacy analysis set was defined by receipt of at least one dose on or after Cycle 1 Day 1. The endpoint result and interval are not reported in the registry record. The record includes group-specific adverse-event counts and selected serious-event terms, but no controlled safety comparison. Its overall non-randomized classification conflicts with Part E descriptions that state random assignment.