Study Investigating the Safety and Efficacy of Blinatumomab in Combination With Pembrolizumab in Adults With Relapsed or Refractory Diffuse Large B Cell Lymphoma (DLBCL): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study Investigating the Safety and Efficacy of Blinatumomab in Combination With Pembrolizumab in Adults With Relapsed or Refractory Diffuse Large B Cell Lymphoma (DLBCL) (NCT03340766), a non-randomized phase 1 study. The primary outcome is Number of Participants With Dose Limiting Toxicities (DLTs).

2. Clinical Question

3. Trial Design

This was a Phase 1b, open-label, non-randomized, sequential study of blinatumomab combined with pembrolizumab. The registry lists an enrollment of 31 and a completed status. The recorded study start was 2018-03-16, and the primary completion date was 2020-11-06. These registry dates describe the study record; they do not by themselves indicate when any particular participant was enrolled or assessed.

The listed cohorts used different blinatumomab dose sequences with pembrolizumab: Cohort Ia used 9/28 µg/day, Cohort IIa used 9/28/56 µg/day, and Cohort IIIa used 9/28/112 µg/day. The registry describes blinatumomab as a continuous intravenous infusion for 8 weeks, followed by a 28-day treatment-free interval. Participants with stable disease or better may have received a second 28-day consolidation cycle. An expansion cohort was intended to test the maximum tolerated dose identified in Part 1. The cohort descriptions in the registry record are incomplete, so further details of dose administration are not reported in the registry record.

The design is consistent with a dose-finding objective: the primary objective was to determine the maximum tolerated dose of blinatumomab in combination with pembrolizumab. However, the registry record does not report the dose-escalation decision rules, how cohort transitions were governed, or the final dose selected. It also does not report a randomized comparator, masking, or a parallel control group. The listed total enrollment should not be treated as the size of each cohort or as the number included in the primary analysis.

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: Participants were enrolled into each dose cohort sequentially using a rolling 6 study design. A Dose Level Review Team (DLRT) reviewed the safety data to evaluate possible drug effects and dose-limiting toxicities (DLTs).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: DLT Analysis Set: Included participants who experienced a DLT, OR were removed from treatment for an adverse event/toxicity that was not a DLT or for other reasons and had been exposed to study treatment for ≥ 12 days OR completed the DLT evaluation 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.

Population: Full Analysis Set: Included all participants who received blinatumomab.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Dose Limiting Toxicities (DLTs), measured over The DLT evaluation period was 42 days from initiation of pembrolizumab treatment (Day 15 for Cohort Ia and Day 19 for Cohorts IIa and IIIa).

The primary endpoint is the number of participants with dose-limiting toxicities (DLTs), recorded as a count of participants. The registry defines DLTs as grade 3–5 adverse events during the DLT-evaluation period that the Investigator judged possibly, probably, or definitely related to study drug administration. Toxicities were graded using National Cancer Institute Common Terminology Criteria for Adverse Events Version 4.0. The description identifies grade 3 as severe or medically significant but not immediately life-threatening, grade 4 as life-threatening, and grade 5 as death.

The DLT-evaluation period was 42 days from initiation of pembrolizumab treatment: Day 15 for Cohort Ia and Day 19 for Cohorts IIa and IIIa. These cohort-specific starting points matter when interpreting which adverse events could count toward the primary endpoint. The DLT Analysis Set included participants who experienced a DLT, or who were removed from treatment for an adverse event or toxicity that was not a DLT or for other reasons after exposure to study treatment for at least 12 days, or who completed the DLT-evaluation period.

General statistical context: in early dose-finding research, a DLT count can help characterize tolerability at candidate dose levels and inform dose selection. The count is not itself a complete description of safety, and it does not establish a treatment effect on lymphoma. The registry record does not report the observed DLT count, cohort-specific DLT results, or the maximum tolerated dose reached.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

General statistical context: a clinical endpoint is the prespecified quantity used to address a study question. Here, the primary endpoint is a count of participants who experienced DLTs, while the study objective concerns dose selection. The count and the decision about a maximum tolerated dose are related but are not interchangeable: dose selection ordinarily depends on how toxicities are evaluated across dose levels and on decision rules. Those rules and the final selection are not reported in the registry record.

What does the DLT count represent?

The endpoint counts participants, rather than counting every adverse-event episode. A participant who experiences more than one qualifying event would still be a participant counted for this endpoint; the registry defines the endpoint as a number of participants with DLTs. The recorded definition also applies a severity threshold, a time window, and an investigator judgment about the relationship to study drug administration. Therefore, the endpoint is narrower than a count of all adverse events or all serious adverse events. Seriousness and severity are distinct concepts in adverse-event reporting, and the registry's DLT definition is based on grade 3–5 toxicity plus the specified attribution and timing criteria.

The DLT window begins from initiation of pembrolizumab, with different cycle-day starting points stated for Cohort Ia and Cohorts IIa and IIIa. This gives the event-counting period a cohort-specific anchor. General statistical context: a fixed evaluation window makes safety counts more comparable when participants have similar opportunities for observation during that window. However, participants who discontinue early or have different treatment exposure can still have different observation histories. The registry's DLT Analysis Set definition addresses some of this by including participants who experienced DLTs, those removed under specified circumstances after exposure for at least 12 days, and those who completed the DLT period. The actual analysis-set size and participant-level follow-up are not reported in the registry record.

Why is this a dose-finding endpoint?

General statistical context: early-phase dose-finding studies commonly use treatment-limiting toxicity to help decide which dose or dose sequence should be studied further. The aim is not simply to identify the dose with the lowest observed toxicity count, since counts depend on the number of participants assessed and the dose-escalation process. A dose decision may also rely on prespecified rules about acceptable toxicity and how evidence from each cohort affects the next cohort. The registry record does not report such rules or the dose-selection analysis. It would therefore be inappropriate to infer a maximum tolerated dose from the cohort labels or from the safety summaries alone.

The listed design is non-randomized and sequential, and masking is listed as none. General statistical context: sequential dose assignment means later treatment decisions can depend on earlier experience, while non-randomization means that cohort membership is not established by chance allocation. Such designs can be appropriate for a dose-finding objective, but cohort-level outcome differences are not automatically unbiased estimates of dose effects. In an open-label study, investigator and participant knowledge of treatment can also affect subjective assessments or reporting. The registry record does not describe additional procedures intended to address these issues.

How should response outcomes be understood?

The registry lists objective response rate and complete response rate outcomes, measured over the first 12 weeks and over the treatment period. It lists revised response Cheson criteria for these outcomes and additional Lugano Classification outcomes for Cohort IIIa. General statistical context: response endpoints classify participants according to defined disease-response criteria, then summarize the number or proportion meeting the category. The criteria and assessment period are important because they determine what counts as a response and when it is assessed. The registry record does not provide the response results, denominators, assessment schedules, or further details about adjudication.

A response rate is distinct from the primary DLT count. Response outcomes concern observed disease response under the specified criteria; the DLT endpoint concerns treatment toxicity within its defined window. Neither outcome alone establishes a causal treatment effect in a non-randomized study. Without a randomized comparator, reported response findings, if available, would be descriptive of participants treated in this study rather than a direct estimate of how outcomes differ from an alternative treatment.

What can be inferred from the posted analyses?

The registry lists 19 outcome measures and reports no analyses. Accordingly, there are no posted inferential results to interpret, such as confidence intervals, hypothesis tests, adjusted comparisons, or model estimates. General statistical context: a confidence interval expresses the statistical uncertainty around an estimate under the assumptions of the method used; it does not correct for biases caused by study design or incomplete reporting. A p-value, when reported, addresses compatibility with a specified statistical model and null hypothesis, not clinical importance or whether a treatment works. Neither type of result is available for the primary endpoint in the registry record.

A planning sample-size illustration based on an observed effect, if considered, is not a trial result and is not observed power. The registry's stated planning assumptions describe a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect framed as a fraction of the observed effect. These are planning assumptions, not a reported design justification for this non-randomized study. The trial's own sample-size justification is not reported in the registry record. No new sample-size calculation or power estimate can be derived from the reported outcomes because the primary observed effect and analysis results are not reported.

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("Cohort Ia: Blinatumomab 9/28 µg/Day + Pembrolizuma", "Cohort IIa: Blinatumomab 9/28/56 µg/Day + Pembroli", "Cohort IIIa: Blinatumomab 9/28/112 µg/Day + Pembro")
n      <- c(10, 6, 5)
events <- c(1, 0, 2)
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  Cohort Ia: Blinatumomab 9/28 µg/Day + Pembrolizuma  10  1  10.0  0.3  44.5  1.8  40.4  2.3  41.3
ARM  Cohort IIa: Blinatumomab 9/28/56 µg/Day + Pembroli  6  0  0.0  0.0  45.9  0.0  39.0  0.4  41.0
ARM  Cohort IIIa: Blinatumomab 9/28/112 µg/Day + Pembro  5  2  40.0  5.3  85.3  11.8  76.9  11.8  77.7

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 19 outcome measures, including several objective response and complete response outcomes across different assessment periods and response criteria. Cohort IIIa has additional response outcomes using the Lugano Classification. Multiple outcomes create a multiplicity issue when several statistical tests are performed: general statistical context, the more questions tested, the greater the chance that at least one result may appear notable by chance if each is interpreted without accounting for the full testing plan. Appropriate interpretation depends on the prespecified hierarchy, any adjustment strategy, and which outcomes are considered confirmatory or exploratory. These details are not reported in the registry record.

The record reports no analyses. It therefore does not report whether any multiplicity adjustment was used, whether outcomes were tested formally, or how missing outcome data were handled. The registry does not report analysis denominators or the number of participants with missing assessments for the listed response outcomes. The primary DLT Analysis Set is described, but its size and the actual application of its inclusion rules are not reported in the registry record.

General statistical context: missing observations can affect an estimate when participants with unobserved outcomes differ systematically from those with observed outcomes. The implications depend on the amount, timing, and reasons for missingness and on the analysis strategy. Those details are not reported in the registry record, so the direction or size of any potential missing-data effect cannot be assessed. Likewise, the DLT population definition should not be mistaken for a reported missing-data method; it defines who may enter the analysis set, not how every unobserved outcome was treated.

The registry also does not report a statistical analysis plan, prespecified hypothesis tests, or interim-analysis procedures. Their absence from the record does not establish that none existed; their details are not reported in the registry record. Without analysis methods and outcome-level denominators, the listed outcomes cannot be interpreted as complete estimates of efficacy or safety.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: aphasia (66.7%), disease progression (41.7%), neurotoxicity (33.3%), vomiting (22.2%), dysarthria (22.2%).

The posted safety summaries give serious adverse event counts by listed cohort: Cohort Ia reports 9 affected participants among 12 at risk, Cohort IIa reports 7 among 10, and Cohort IIIa reports 9 among 9. The summaries also report counts for other adverse events, alongside death counts: 6 in Cohort Ia, 7 in Cohort IIa, and 4 in Cohort IIIa. These are descriptive counts for the cohorts and should not be treated as randomized comparisons or as estimates of a treatment effect. The registry does not report event timing, duration, recurrence, exposure-adjusted rates, or details sufficient to connect each death to a specific adverse event or treatment attribution.

The listed serious-event terms include aphasia, disease progression, neurotoxicity, vomiting, dysarthria, and encephalopathy, with percentages reported for those terms. The denominator and calculation basis for those percentages are not reported in the registry record, so they should not be read as rates across all enrolled participants without further information. The cohort safety summaries are also distinct from the primary DLT endpoint: a serious adverse event is not automatically a DLT, because DLT status depends on the specified grade, evaluation window, and investigator assessment of relationship to study treatment.

General statistical context: small cohort counts can be imprecise and are sensitive to individual observations. The listed figures describe what was recorded in these cohorts; they do not establish comparative risk or causation.

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

The registry record describes an early-phase, non-randomized study designed to evaluate blinatumomab with pembrolizumab and determine a maximum tolerated blinatumomab dose. It identifies a specific DLT endpoint and evaluation window, and it lists response outcomes and cohort-level safety summaries. These details clarify what the study set out to assess and how the primary toxicity endpoint was defined.

The record does not report the observed DLT count, the maximum tolerated dose selected, or results for the listed response outcomes. It also reports no analyses. Consequently, it does not provide a quantitative basis for judging the primary dose-finding result or estimating response outcomes. The enrollment of 31 does not substitute for outcome denominators, and the registry does not show how many participants contributed to each primary or secondary outcome analysis.

The safety counts and selected serious-event terms are descriptive registry summaries. They do not establish that an event was caused by either drug, and the primary DLT definition requires a specific investigator attribution in addition to severity and timing. Nor do these data alone permit a controlled comparison of safety between dose cohorts. The study had no randomized comparator, and the registry record does not report exposure-adjusted analyses or the details needed to account for differences in observation.

General statistical context: a study result is interpreted in light of its design, endpoint definition, analysis population, uncertainty, and potential sources of bias. Here, the design supports a dose-finding purpose but does not provide the protections of randomized allocation for treatment comparisons. Missing outcome results and analysis details further limit interpretation. These limitations do not demonstrate that an unreported result is favorable or unfavorable; they mean that the result cannot be quantified from the registry record.

The record should therefore be read as documentation of the study design, registered objectives, outcome definitions, and posted descriptive safety information. It does not establish efficacy, comparative safety, or a final recommended dose.

14. Why This Trial Matters Statistically

This study is statistically notable for the way its primary question differs from the usual efficacy comparison. The stated objective is dose finding: determining the maximum tolerated dose of blinatumomab in combination with pembrolizumab. The registered primary endpoint is a count of participants with DLTs, defined using toxicity grade, a cohort-specific evaluation window, and investigator-assessed relation to study drug administration. That definition makes clear that a general adverse-event count cannot be substituted for the primary endpoint.

The sequential, non-randomized, open-label design also shapes what comparisons can support. Different blinatumomab dose sequences are listed across cohorts, but participants were not randomly allocated to a concurrent control. General statistical context: a dose-finding design can provide information about treatment experience at studied doses, while direct causal comparisons between doses require careful attention to cohort assignment, changing study experience, and other differences between groups. The registry record does not report the decision rules or final dose choice needed to interpret how the dose-finding objective was resolved.

The distinction between an objective and a reported result is especially important. The registry lists response outcomes and safety measures, but no analyses are reported, and the observed primary DLT count is not reported. The posted serious adverse event counts describe safety experience, but they do not supply the missing primary endpoint or establish comparative risk. Thus, the study record illustrates why endpoint definitions, analysis populations, and result availability must be considered together rather than treating enrollment or listed outcomes as evidence of a finding.

For clinical researchers, the central statistical lesson is that design determines the questions a study can answer. A non-randomized dose-finding study can be framed around tolerability and dose selection, but it cannot be interpreted as a randomized test of efficacy. Missing outcome estimates and analysis details leave the study's primary statistical conclusion unresolved in the registry record.

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 main limitation for interpreting outcomes is that the registry record does not report the observed primary DLT count or the maximum tolerated dose identified. It lists the study objective and endpoint definition but provides no primary-endpoint estimate, uncertainty interval, or analysis. The secondary response outcomes are listed without reported results, denominators, or analysis methods. Although 31 participants are listed as enrolled, the record does not report the number assessed for each outcome or the cohort-specific enrollment totals for the primary analysis.

The study is non-randomized, open label, and sequential, with no listed control arm. This limits direct comparison between dose cohorts and does not support a randomized estimate of treatment effect. The registry record does not report cohort assignment rules, baseline characteristics by cohort, dose-escalation decision rules, or how the final dose was selected. The cohort descriptions are incomplete, so additional details of administration and treatment exposure are not reported in the registry record.

Safety interpretation is also limited. The record gives cohort-level counts and selected serious-event terms but does not report timing, duration, recurrence, exposure-adjusted rates, or a denominator for the selected term percentages. The death counts are not accompanied by information establishing cause or treatment relationship. Serious adverse events cannot be equated with DLTs because the DLT definition uses a specific grade range, period, and attribution judgment.

The registry does not report a statistical analysis plan, missing-data methods, multiplicity procedures, interim-analysis procedures, or a sample-size justification. Their details are not reported in the registry record. Without these results and methods, precision, missingness, and potential selective interpretation cannot be fully evaluated.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study Investigating the Safety and Efficacy of Blinatumomab in Combination With Pembrolizumab in Adults With Relapsed or Refractory Diffuse Large B Cell Lymphoma (DLBCL) — NCT03340766.
• Registry record last updated 2024-10-17; status last verified 2024-07.
• Results first posted 2022-02-25.
• 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

NCT03340766 is a completed Phase 1b open-label, non-randomized, sequential study of blinatumomab with pembrolizumab in adults with relapsed or refractory DLBCL. The registry lists enrollment of 31, a study start of 2018-03-16, and primary completion on 2020-11-06. Cohorts Ia, IIa, and IIIa used different blinatumomab dose sequences with pembrolizumab; an expansion cohort was intended to test the maximum tolerated dose identified in Part 1.

The primary endpoint was the number of participants with DLTs during a 42-day evaluation period anchored to pembrolizumab initiation, with cohort-specific starting days. The record lists response outcomes and provides descriptive adverse-event summaries. It does not report the observed primary endpoint result, the selected maximum tolerated dose, response results, or statistical analyses. The sample-size justification, missing-data methods, multiplicity procedures, and dose-selection rules are not reported in the registry record.