Pembrolizumab + Poly-ICLC in MRP Colon Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pembrolizumab + Poly-ICLC in MRP Colon Cancer (NCT02834052), a non-randomized phase 1, phase 2 study. The primary outcome is Phase 1 and 2: Determine the Response Rate of Metastatic MRP Colon Cancer (That Has Progressed Following Two Lines of Therapy in the Metastatic Setting) to the Combination of Pembrolizumab and Poly-ICLC.

2. Clinical Question

The study also asks whether Poly-ICLC can be combined safely with pembrolizumab and how the combination relates to immune activation and tumor response. Its stated secondary outcomes include adverse events and dose-limiting toxicities, 20-week progression-free survival, and duration of response. These questions concern different aspects of treatment and should not be treated as interchangeable evidence: a response endpoint describes tumor changes under a defined assessment rule, while safety outcomes describe reported harms.

3. Trial Design

This is a completed Phase I/II trial registered as NCT02834052. The registry lists an enrollment of 42 and describes the allocation as nonrandomized, masking as none, and the model as single group. The study began on 2018-01-10 and had a primary completion date of 2022-07-29. Those registry entries describe the study record; they do not establish how many participants received each treatment or contributed to each endpoint.

The Phase 1 portion was described as a run-in to determine whether Poly-ICLC could be safely combined with standard-dose pembrolizumab. Pembrolizumab was specified at 200 mg intravenously every 3 weeks. Poly-ICLC was specified for intramuscular administration twice weekly at one of two dose levels, with the Phase 1 cohort descriptions identifying 1 mg and 2 mg. Phase 2 used the standard pembrolizumab dose with the maximum tolerated Poly-ICLC dose determined in Phase 1, described as either 1 mg or 2 mg. The record says up to 30 participants would be treated in Phase 2 and that participants may receive treatment for 1 year.

The dose cohorts are not randomized arms, and the registry explicitly cautions that they are not a like-for-like control comparison. Differences between cohorts could reflect more than dose, and an apparent contrast between them cannot be interpreted as a randomized treatment effect. The record does not report a detailed allocation procedure, masking, or a separate control treatment. It also does not report the rationale for the enrollment target or a statistical design justification. The completion status indicates that the registry marks the study completed; it does not supply the missing outcome results or analysis details.

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 safety population for this trial includes all participants who received at least one dose of the investigational product and excludes any participants who were screen fails.

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 Phase 1 and 2: Determine the Response Rate of Metastatic MRP Colon Cancer (That Has Progressed Following Two Lines of Therapy in the Metastatic Setting) to the Combination of Pembrolizumab and Poly-ICLC, measured over From baseline to disease progression (Expected 12-24 months).

The primary endpoint is the count of participants with a complete or partial response, used to determine the response rate in the study population. The registry defines complete response as disappearance of all target lesions. Partial response is defined as a decrease of at least 30% in the sum of the longest diameters of target lesions. The assessment is described as using RECIST v1.0 criteria and MRI. The stated timeframe is from baseline to disease progression, with an expected duration of 12-24 months.

Response criteria give researchers a prespecified way to classify tumor changes rather than relying on an informal impression. A count of participants meeting the complete- or partial-response definition can be expressed as a response rate when the relevant outcome population is known. The registry says that the analysis population is patients who received treatment, while the outcome population is determined within each dataset or cohort by the number with a qualifying response. The exact response count, denominator used for any reported rate, and response estimate are not reported in the registry record.

This endpoint answers a tumor-response question, not every question about treatment. It does not by itself describe how long a response lasted, whether disease remained controlled, or the frequency of adverse events. The registry lists duration of response and progression-free survival as separate secondary outcomes, reflecting those distinct questions. No primary endpoint result is reported in the record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

For a binary endpoint, each participant is classified according to whether the event of interest occurred. In this trial's primary outcome, the event is a complete or partial response under the stated RECIST v1.0 criteria. A response count is the number of participants classified as having that event. A response rate requires both that count and a clearly defined denominator. The registry describes the treated population as the analysis population, while the outcome population is determined within each dataset or cohort by the participants with evaluable response classifications. It does not report the actual response counts or rates.

In general, binary outcomes can be summarized using counts and proportions, with uncertainty expressed using a confidence interval. A confidence interval communicates how imprecisely an estimate is determined by the data and statistical method. It does not mean that the true value has a specified probability of falling in the particular interval, nor does it account automatically for design limitations, missing data, or bias. No response estimate or confidence interval is reported in the registry record, so the precision of the primary outcome cannot be evaluated.

RECIST criteria provide a rule-based classification for tumor response. As registered, complete response requires disappearance of all target lesions, and partial response requires a decrease of at least 30% in the sum of the longest diameters of target lesions. The response assessment is described as MRI-based. Applying a common criterion can make response classification more consistent, but interpretation still depends on the assessment process, timing, evaluability, and handling of missing scans. Those operational details are not reported in the registry record.

The study design matters when interpreting estimates. Randomization can help make comparison groups similar on average at baseline, while a nonrandomized design does not provide that protection. This trial is listed as nonrandomized and single group, and the registry notes that its arms are not a like-for-like control comparison. The dose cohorts can describe outcomes observed under their respective regimens, but they cannot establish that a difference between cohorts was caused by dose. The Phase 2 Poly-ICLC dose was selected based on the Phase 1 arm, which further means the cohorts have different roles in the study rather than serving as interchangeable randomized groups.

General statistical principle: when several outcomes are tested, the chance of at least one apparently notable finding can rise if each is assessed independently without accounting for multiplicity. Researchers may address this through a prespecified testing hierarchy or other adjustment, depending on the design and goals. The registry lists six outcome measures and identifies secondary questions about adverse events and dose-limiting toxicities, 20-week progression-free survival, and duration of response. It does not report a multiplicity plan, hypothesis-testing hierarchy, or adjustment method. The number of registered measures alone does not establish that all were tested inferentially.

Safety summaries require clear definitions of the event category and denominator. Serious adverse events and other adverse events are distinct categories in the registry's cohort entries. A count of affected participants is not the same as a count of individual events, because one participant may experience more than one event; the record's labels describe affected participants. General principle: event rates can be compared meaningfully only when populations, observation periods, ascertainment, and definitions are sufficiently clear. The record gives cohort-specific at-risk and affected counts, but does not provide all those details for interpreting rates.

Statistical analysis plans commonly specify the analysis population, endpoint rules, missing-data handling, and any adjustments before results are examined. Such details help readers understand which observations contribute to an estimate and how incomplete information is handled. The registry's primary endpoint description defines a treated analysis population and a cohort-based outcome population, but further analysis methods are not reported. The registry lists no analyses. It does not report a model, inferential procedure, confidence interval method, or missing-data approach.

Sample-size planning is also distinct from results observed in a trial. General principle: a planning calculation depends on assumptions about the effect to detect, variability or event frequency, allocation, and statistical error rates. It is not a measure of the evidence actually observed. The registry record does not report the trial's sample-size justification or a prospective statistical design calculation. Thus the enrollment entry should not be taken on its own as evidence that the study was adequately powered for its primary endpoint.

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("Phase 1: Poly-ICLC: 1mg", "Phase 1: Poly-ICLC 2mg", "Phase 2: Poly-ICLC: 2mg")
n      <- c(7, 7, 17)
events <- c(0, 0, 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  Phase 1: Poly-ICLC: 1mg  7  0  0.0  0.0  41.0  0.0  35.4  0.3  36.9
ARM  Phase 1: Poly-ICLC 2mg  7  0  0.0  0.0  41.0  0.0  35.4  0.3  36.9
ARM  Phase 2: Poly-ICLC: 2mg  17  1  5.9  0.1  28.7  1.0  27.0  1.4  27.3

11. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures, including the primary response-rate question and secondary outcomes concerning adverse events and dose-limiting toxicities, 20-week progression-free survival, and duration of response. It does not report how these outcomes were prioritized for statistical testing, whether a hierarchy was used, or whether any multiplicity adjustment was planned. In general, multiplicity matters when several statistical claims are considered together, because repeated testing can increase the chance of a false-positive finding. The number of listed measures does not show that all were tested or that any particular adjustment was used.

The record describes the analysis population as patients who received treatment. It further says that the outcome population is based on the number of patients within each dataset or cohort with complete or partial response under RECIST criteria. The number assessed for the primary outcome, the response counts, and the handling of patients without evaluable response data are not reported in the registry record. The missing-data method is not reported, including whether incomplete outcome data were excluded, imputed, or handled in another way.

The registry lists no analyses and does not report a statistical analysis plan, model, confidence interval method, or adjusted analysis. It also does not report a sample-size justification. These omissions limit the ability to assess whether outcome definitions and analytic decisions were specified in advance, how uncertainty was quantified, and how incomplete observations might affect the results. A completed study status does not resolve these reporting gaps.

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

Serious events with the highest arm-level rate: dyspnea (37.5%), abdominal distention (12.5%), small bowel obstruction (12.5%), death (12.5%), acute kidney injury (12.5%).

The registry gives cohort-specific counts for participants at risk and affected by serious adverse events, other adverse events, and deaths. For Phase 1: Poly-ICLC: 1mg, the serious adverse event entry is 7 affected among 8 at risk, the other adverse event entry is 0 affected among 8 at risk, and the deaths entry is 8. For Phase 1: Poly-ICLC: 2mg, the corresponding entries are 7 affected among 10 at risk, 0 affected among 10 at risk, and 10 deaths. For Phase 2: Poly-ICLC: 2mg, they are 17 affected among 24 at risk, 0 affected among 24 at risk, and 23 deaths.

The registry also lists serious-event terms and percentages: Dyspnea, 37.5%; Abdominal Distention, 12.5%; Small Bowel Obstruction, 12.5%; Death, 12.5%; Acute Kidney Injury, 12.5%; and Hyperkalemia, 12.5%. The denominator and calculation basis for these term percentages are not reported in the registry record. The counts should not be read as evidence of causation, and the record does not establish that deaths were caused by treatment. Timing and observation duration for these safety entries are not reported.

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

The record establishes that a completed Phase I/II study evaluated pembrolizumab with Poly-ICLC in metastatic mismatch repair-proficient colon cancer, with a primary question about response and additional safety and disease-control questions. It provides response criteria, a timeframe, cohort descriptions, and safety counts. It does not report the primary response count or rate, so it cannot establish from the posted results how often participants met the response definition or how precise an estimate would be.

The design also limits what can be inferred. The trial is nonrandomized and single group, and its dose cohorts are not a like-for-like control comparison. Observations in these cohorts cannot determine how outcomes would have differed with pembrolizumab alone or another treatment. Nor can differences between cohorts be attributed to Poly-ICLC dose on the basis of this design alone.

Safety counts describe recorded affected participants and deaths in the cohort entries, but they do not by themselves show that treatment caused an event. The record does not include enough detail to interpret the listed serious-event term percentages fully, and it does not report the timing or duration of the events. The listed counts should therefore be understood as descriptive entries rather than causal estimates.

These limitations do not mean that the study found no responses or that no safety events occurred. They mean that the registry record does not report the information needed to make those outcome claims. The response result, uncertainty around estimates, missing-data handling, and analysis methods are not reported. The study's status as completed does not substitute for those results. Any conclusion about efficacy, comparative benefit, or treatment-related risk would require information beyond the outcomes and methods described in this record.

14. Why This Trial Matters Statistically

This trial raises a statistical question that is common in early-phase clinical research: how to describe a treatment combination when dose selection and outcome evaluation occur in a nonrandomized design. Phase 1 was intended to determine whether Poly-ICLC could be combined safely with pembrolizumab and to identify a dose for Phase 2. Phase 2 then used the dose selected through Phase 1. That sequence is relevant because the cohorts serve different design purposes and should not be treated as randomized alternatives.

The primary outcome uses a defined tumor-response rule, which makes the intended endpoint more specific than a general judgment that tumors changed. The registry identifies RECIST v1.0 criteria, MRI assessment, and a complete- or partial-response definition. In statistical reporting, those definitions are only part of the evidence: readers also need response counts, denominators, and an account of missing or unevaluable assessments. Those results are not reported in the registry record.

The study also illustrates why endpoint results and statistical methods must be considered together. The registry lists multiple outcome measures, but does not report a testing hierarchy or multiplicity approach. It describes the treated analysis population and cohort-based outcome population, but not how missing endpoint data were handled. These details affect what an estimate represents and how much confidence readers can place in it.

Finally, descriptive safety counts and a completed-study status are not substitutes for an analysis of treatment effects. The nonrandomized, single-group design cannot provide a like-for-like control comparison, and the registry does not report the primary response result or analysis plan. The statistical significance of this trial record lies in recognizing what its design and reported information can support, and where its evidence remains incomplete.

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 interpretation is that the registry record does not report the response count or rate, despite identifying response as the primary outcome. It also does not report uncertainty estimates or the number of participants with evaluable response data. As a result, the central efficacy question cannot be answered from the posted results described in the record.

The nonrandomized, single-group design has no like-for-like control comparison. The Phase 1 and Phase 2 cohorts have different roles, and Phase 2 dose selection depended on Phase 1. Cohort outcomes cannot isolate the effect of Poly-ICLC dose or the combination relative to another treatment. The registry does not report baseline characteristics or enough cohort-level outcome information to assess comparability.

Important analysis details are not reported, including the statistical analysis plan, multiplicity approach, missing-data handling, methods for uncertainty estimation, and sample-size justification. Although the outcome definition specifies RECIST v1.0 and MRI, further operational details about assessment and evaluability are not reported. Safety entries also lack a reported observation period and denominator basis for the serious-event term percentages. These gaps constrain interpretation of both benefit and risk and should not be filled by assumption.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Pembrolizumab + Poly-ICLC in MRP Colon Cancer — NCT02834052.
• Registry record last updated 2024-06-07; status last verified 2024-06.
• Results first posted 2024-06-07.
• 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

NCT02834052 is a completed, nonrandomized Phase I/II single-group trial of pembrolizumab and Poly-ICLC in metastatic mismatch repair-proficient colon cancer. The registry lists an enrollment of 42. Phase 1 assessed combination dosing, and Phase 2 used a Poly-ICLC dose selected from Phase 1 with pembrolizumab at 200 mg intravenously every 3 weeks.

The primary outcome is the count of participants with complete or partial response under RECIST v1.0 criteria assessed by MRI, from baseline to disease progression, with an expected timeframe of 12-24 months. The record lists safety, 20-week progression-free survival, and duration of response among secondary outcomes. It does not report the primary response result, a statistical analysis plan, multiplicity procedures, or missing-data handling.