Pembrolizumab and CXCR4 Antagonist BL-8040 in Treating Patients With Metastatic Pancreatic Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pembrolizumab and CXCR4 Antagonist BL-8040 in Treating Patients With Metastatic Pancreatic Cancer (NCT02907099), a na phase 2 study. The primary outcome is Overall Response Rate.

2. Clinical Question

3. Trial Design

This completed phase IIb pilot study was sponsored by M.D. Anderson Cancer Center. Its title describes pembrolizumab and CXCR4 antagonist BL-8040 in patients with metastatic pancreatic cancer. The registry lists an enrollment of 20 and identifies the study model as single group. Allocation is not applicable, and masking is none. These design features mean the registry does not describe random assignment or a blinded comparison group.

The treatment group is listed as pembrolizumab plus BL-8040. The registry describes BL-8040 as given subcutaneously on days 1–5 and 8–12 of cycle 1, and on days 1, 4, 8, and 11 of subsequent cycles. Beginning in cycle 2, pembrolizumab is given intravenously over about 30 minutes on day 1. Cycles repeat every 21 days for up to 1 year in the absence of disease progression or another condition, but the treatment description ends mid-sentence in the registry record. Further detail about that stopping condition is not reported in the registry record.

The study start date is 2016-12-01, the primary completion date is 2023-10-06, and the status is completed. The primary endpoint is overall response rate measured by complete or partial response according to RECIST 1.0, with a stated timeframe of 3 years. The registry also lists secondary outcomes concerning T-cell infiltration, circulating T-cells, and treatment-related adverse events assessed by CTCAE v4.0. The registry record does not describe the allocation of follow-up visits or the timing of each secondary assessment.

A single-group design can describe outcomes among participants receiving the study treatment, but it does not by itself separate treatment-associated changes from other explanations. Without a concurrent control group, the record cannot support a randomized comparison of outcomes between treatment strategies. Details of the protocol’s rationale for the design and the statistical plan 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.

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.

Population: Metastatic Pancreatic Cancer (mPC).

6. Primary Endpoint

The registry lists the primary outcome as Overall Response Rate, measured over 3 years.

Overall response rate is the primary endpoint. In this study, the registry defines a response as either a complete response or a partial response under RECIST 1.0 after treatment with BL-8040 and pembrolizumab. The endpoint is expressed as a count of participants, and the stated assessment timeframe is 3 years. The registry record does not report the observed count or the corresponding rate.

In general, an objective response endpoint classifies whether tumors meet prespecified criteria for substantial shrinkage or disappearance. A complete response and a partial response are distinct response categories, but both count toward the endpoint as defined in this record. Using a named response framework makes the outcome definition more explicit than an informal description of improvement. It also makes clear which categories are included in the reported response measure.

An overall response rate describes the proportion of an assessed group meeting the response definition; it does not describe every aspect of disease course or the duration of response. Interpretation depends on knowing which participants were assessed, how missing assessments were handled, and how response was determined. Those analysis details are not reported in the registry record. The record also does not provide an estimate of the response rate, its confidence interval, or a comparison with another group. The endpoint definition alone therefore explains what was intended to be counted, not what the study observed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a binary participant-level outcome: a participant is classified as having a complete or partial response, or not meeting either response category, according to RECIST 1.0. The registry describes the endpoint as a count of participants rather than a continuous measurement. In general, a binary endpoint is summarized by counting participants in each outcome category and estimating the proportion with the event of interest. The registry record does not report those primary outcome counts or an estimated response proportion.

What does a confidence interval add?

In general, an estimate from a study is subject to sampling uncertainty: another group drawn from the same underlying population could produce a different estimate. A confidence interval communicates the precision of an estimate under a specified statistical procedure. A narrower interval indicates greater precision under that procedure; a wider interval indicates less precision. It does not resolve limitations in the study design, remove bias, or prove that the estimated value is correct. No confidence interval for overall response rate is reported in the registry record, so its precision cannot be assessed.

What would a p-value address?

In general, a p-value is calculated under a statistical model and a specified null hypothesis. It describes how compatible the observed data are with that hypothesis under the model; it is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. A p-value also does not replace an effect estimate and its uncertainty. The registry record does not report a primary-endpoint p-value or identify a null hypothesis and testing procedure for that endpoint. Therefore, no inferential test result can be interpreted from the registry information.

Why does the design affect interpretation?

The study is listed as single group, with allocation not applicable and masking none. In general, a single-group study can estimate outcomes observed during treatment, but it lacks a concurrent group that would represent what might have happened under a different strategy. Randomization, when used, can help balance measured and unmeasured prognostic factors across comparison groups on average. No randomized comparison is described for this study. The response endpoint can therefore be interpreted as a descriptive outcome for the studied group if results are available, not as a randomized treatment contrast.

Masking can reduce the influence of knowledge about treatment assignment on assessments or other study procedures. The registry lists masking as none. That fact does not establish that response assessments were biased, but it is relevant context when considering an outcome that involves classification. Further information about review procedures, independent assessment, and adjudication is not reported in the registry record.

How should planning calculations be read?

In general, sample-size calculations are design tools: they explore how much information a proposed study might collect under stated assumptions. The registry’s planning-illustration assumptions describe a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect to detect expressed as a fraction of the observed effect. The record also characterizes the illustrations as computed from the observed effect, rather than as the trial’s own design. Such assumptions do not turn this single-group study into a two-arm randomized trial. They are not observed results and are not observed power.

A planning illustration depends on its assumptions, including the target effect and the statistical model. If the effect used for planning is uncertain, the resulting illustration is also conditional on that uncertainty. The registry record does not report a sample-size justification for the actual study or enough information to assess how its enrollment was selected. It does not report a prespecified target effect, a protocol-based power calculation, or an interim analysis plan. These details should not be inferred from a separate planning illustration.

Why can over-dispersion matter?

In general, over-dispersion occurs when observed variability is greater than a model assumes. A model that does not accommodate over-dispersion can understate uncertainty, making intervals too narrow or tests appear more informative than the data warrant. This issue can arise in some count and grouped-data analyses, but the registry’s primary endpoint is described as a participant count and no specific primary statistical model is reported. There is no basis in the registry record to say whether over-dispersion was assessed or relevant to the sponsor’s primary analysis.

The distinction between sponsor analyses and simple recalculations is also important in general. A recalculation from summary statistics may use a simpler, unadjusted method and need not reproduce a sponsor’s prespecified analysis, which may account for design or analysis details not available in a summary. The registry information does not report a primary sponsor analysis or a recalculated primary estimate. No numerical result should be inferred in its absence.

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("Pembrolizumab Plus BL-8040")
n      <- c(15)
events <- c(1)
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  Pembrolizumab Plus BL-8040  15  1  6.7  0.2  31.9  1.2  29.8  1.6  30.2

11. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures and two analyses, but it does not describe the statistical procedures used for those analyses. The primary endpoint is identified as overall response rate; secondary outcomes include T-cell infiltration, circulating T-cells, and treatment-related adverse events. The registry record does not state whether the analyses were prespecified, how the two analyses relate to the listed outcomes, or whether any adjustment for multiple comparisons was used.

In general, multiplicity arises when several hypotheses or outcomes are tested. Conducting multiple tests without an appropriate strategy can increase the chance of at least one apparently positive result under a collection of null hypotheses. Approaches to multiplicity can include prioritizing a primary endpoint, controlling an overall error rate, or treating secondary analyses as exploratory. The record identifies a primary endpoint, but does not report a multiplicity strategy. It is therefore not possible to determine how the reported analyses, if any, should be interpreted in relation to one another.

Missing outcome data can affect both the estimated response and its uncertainty. Interpretation depends on how many participants had an assessment, why data were missing, and how the analysis handled those cases. The registry record does not report missingness, the primary endpoint analysis population, or the method used to handle missing response assessments. Enrollment alone cannot establish the primary analysis denominator.

The registry record also does not report whether the primary analysis followed an intention-to-treat principle, used an evaluable-participant group, or applied another population definition. The relationship between the enrollment figure and the adverse-event at-risk counts is not explained. No assumptions about missing outcomes, exclusions, or analysis populations can be made from the listed information.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: dyspnea (13.3%), nausea (6.7%), vomiting (6.7%), hypertension (6.7%), alkaline phosphatase increased (6.7%).

The registry’s safety summary lists a group titled pembrolizumab plus BL-8040. It reports 15 participants at risk for serious adverse events and 4 affected, and 15 at risk for other adverse events and 5 affected. It also reports 2 deaths. These are the record’s stated counts; the registry information does not explain the relationship between the safety group and the enrolled population or provide a detailed account of follow-up for these summaries.

The listed serious-event terms include dyspnea at 13.3%, and nausea, vomiting, hypertension, alkaline phosphatase increased, and abdominal pain at 6.7% each. These percentages are presented as registry values. The record does not explain whether participants could contribute to more than one term, so the entries should not be added together or interpreted as mutually exclusive categories. The summary does not establish that every listed event was caused by study treatment.

In general, serious adverse events and other adverse events are distinct safety classifications, and the counts should be read using the definitions and collection rules in the study record. The registry lists treatment-related adverse events as a secondary outcome assessed by CTCAE v4.0, but detailed assessment methods, event duration, severity breakdowns, and attribution procedures are not reported. These safety data describe reported events; they do not establish a comparative safety profile because no concurrent control group is described.

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

The registry record establishes that a completed phase IIb pilot study evaluated pembrolizumab with CXCR4 antagonist BL-8040 in a single-group design. It specifies overall response rate as the primary endpoint and defines response as complete or partial response according to RECIST 1.0, with a stated timeframe of 3 years. These details make the clinical question and intended outcome identifiable.

The record does not report the observed primary response count or rate, a confidence interval, or a p-value. Consequently, the primary outcome’s magnitude and precision cannot be described from this record. Study completion does not itself establish that the combination produced a particular response result. Nor does the enrollment figure tell how many participants contributed to the primary endpoint.

Even if a response estimate were reported, the single-group design would constrain what it could establish. A result could describe outcomes among participants treated in the study, but without a concurrent randomized comparator it would not directly estimate how outcomes differ under another treatment strategy. Observed outcomes can reflect multiple influences, and the design does not provide a randomized contrast to separate them.

The registry also includes selected safety counts and event terms. Those entries are useful as descriptions of events recorded in the study, but the available information does not provide the detail needed to assess every aspect of safety, such as event timing, duration, attribution, or the population included in each summary. The event counts should not be combined or treated as evidence of a comparison against another treatment.

In general, a study’s endpoint definition, design, analysis population, and uncertainty all shape the conclusions that can be drawn. Here, the endpoint and broad design are described, while the central response result and several analysis details are not reported in the registry record. The appropriate interpretation is therefore limited: the registry identifies what the study set out to assess and provides selected safety information, but it does not permit a statistical assessment of the primary response outcome or a comparative treatment effect.

14. Why This Trial Matters Statistically

This study is statistically informative as an example of why a clearly named endpoint does not substitute for a reported result. The registry defines overall response rate using complete or partial response according to RECIST 1.0 and states a 3 years timeframe. Yet it does not report the response count, estimate, or uncertainty. Readers can identify the intended question but cannot evaluate the primary finding’s size or precision from the registry record.

The single-group design also illustrates the difference between describing outcomes and estimating a comparative effect. A study can record what happened among participants receiving a treatment combination, but a concurrent randomized comparison is needed to provide a direct randomized contrast with another strategy. The registry lists no such control group. This limitation concerns the design and remains relevant regardless of whether an observed outcome appears favorable or unfavorable.

The record further distinguishes enrollment from the group used in a specific analysis. It lists an enrollment of 20, while its safety summary lists 15 at risk for the relevant event counts. The registry does not explain how these populations relate, and the primary analysis denominator is not reported. This is a practical reminder that endpoint counts need an explicit analysis population and clear missing-data handling to be interpretable.

Finally, planning illustrations and observed trial results answer different questions. The record describes planning assumptions based on a fraction of an observed effect, but those assumptions do not define the actual study design or establish observed power. Keeping design calculations separate from observed estimates is essential when interpreting a pilot study. The registry’s incomplete outcome reporting makes that distinction especially important: assumptions cannot fill in an unreported primary 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 main trial-specific limitation in interpreting this registry record is that the primary endpoint result is not reported. Although overall response rate is defined as complete or partial response by RECIST 1.0, no response count, rate, confidence interval, or p-value is available. The primary analysis population and missing-response handling are also not reported, preventing assessment of the endpoint denominator and its uncertainty.

The single-group design has no concurrent control group in the registry description. It cannot support a randomized comparison of the combination with another treatment strategy. Masking is listed as none; the record does not provide further detail about response review or safeguards for outcome assessment.

The enrollment figure and safety at-risk counts are not reconciled in the registry record. The safety summary lists 15 at risk for serious events and 15 at risk for other events, while enrollment is 20. The primary analysis denominator is not stated. These quantities should not be treated as describing the same analysis population without supporting information.

The treatment description ends mid-sentence after stating that cycles continue in the absence of disease progression or another condition. The additional stopping detail is not reported in the registry record. The record also does not provide a statistical analysis plan, sample-size justification for the actual study, multiplicity strategy, interim analysis details, or an explanation of how the two listed analyses relate to the outcomes. These omissions limit evaluation of the conduct and statistical interpretation of the study beyond the basic design and endpoint definition.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Pembrolizumab and CXCR4 Antagonist BL-8040 in Treating Patients With Metastatic Pancreatic Cancer — NCT02907099.
• Registry record last updated 2025-02-19; status last verified 2024-11.
• Results first posted 2025-02-19.
• 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

NCT02907099 is a completed phase IIb pilot study sponsored by M.D. Anderson Cancer Center. It enrolled 20 participants and is listed as a single-group study with no masking. The treatment group received CXCR4 antagonist BL-8040 with pembrolizumab. The registry gives a study start date of 2016-12-01 and a primary completion date of 2023-10-06.

The primary endpoint is overall response rate, defined as complete or partial response according to RECIST 1.0 after treatment, with a stated timeframe of 3 years. The registry record does not report the primary response result or its statistical uncertainty. Safety information includes counts of serious and other adverse events and selected serious-event terms, but the relationship between the listed safety population and enrollment is not explained. The single-group design does not provide a concurrent randomized comparison.