Combination Immunotherapy in Biochemically Recurrent Prostate Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Combination Immunotherapy in Biochemically Recurrent Prostate Cancer (NCT03315871), a non-randomized phase 2 study. The primary outcome is Percentage of Participants With a Toxicity.

2. Clinical Question

The trial concerns adults with biochemically recurrent prostate cancer: a rise in prostate-specific antigen after prior treatment such as radiation or surgery. The registry description notes androgen deprivation therapy and monitoring as approaches used for this setting, and also describes monitoring PSA over time. The research objective was to assess whether the study combination could induce an anti-tumor impact. The primary endpoint was a toxicity measure, so the posted results do not directly answer that objective with a reported tumor-response estimate.

These distinctions matter when reading the results. The study objective concerns anti-tumor impact, while the stated primary endpoint concerns toxicity. Safety counts are reported for named cohorts, but no statistical comparison between treatment strategies is reported. The available information therefore supports a description of the registered design and reported adverse-event counts, not a comparative conclusion about anti-tumor effects or relative safety.

3. Trial Design

This is a phase II, parallel, non-randomized, unmasked trial listed as completed, with an enrollment of 40. The registry lists a start date of 2018-03-20 and a primary completion date of 2024-06-04. The record describes three cohort or arm groupings, but these do not form a randomized comparison.

The first grouping is a safety lead-in combination cohort, marked closed December 2018, in which participants received PROSTVAC, CV301, and MSB0011359C (M7824). The second grouping concerns biochemical recurrence and describes surveillance followed by PROSTVAC and CV301, then the addition of MSB0011359C (M7824); it is marked closed. The third grouping describes biochemical recurrence with combination antigen direct immunotherapy, with surveillance as needed followed by PROSTVAC and CV301. The safety results use a combined label for Arm 2 and Arm 3, so those results do not separately identify the adverse-event counts for each of those treatment descriptions.

The registry summary describes screening with medical history, physical examination, blood and urine tests, imaging, and review of previously collected tissue. It also describes drug administration over at least 7 months and frequent study visits. Some participants were to have close monitoring with monthly PSA checks. These descriptions do not establish that every participant followed an identical schedule or received the same treatment sequence.

Enrollment size is not the same as the denominator for every endpoint. The adverse-event results give different numbers at risk across the listed groups. The registry record does not report a randomized allocation process, masking, a separate control group, or an inferential analysis plan. Because treatment assignment was non-randomized and the cohort descriptions differ, comparisons across groups can be affected by differences in eligibility, treatment timing, and participant composition that cannot be assessed from the posted information.

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.

Safety Lead-In

PROSTVAC + CV301 + M7824

Reasons for not completing this period:

PROSTVAC + CV301

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.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With a Toxicity, measured over At the end of therapy (7 months).

The primary endpoint is the percentage of participants with a toxicity assessed using the Common Terminology Criteria for Adverse Events, version 4.0, at the end of therapy, defined as 7 months. It is a binary event outcome at the participant level: each participant is classified according to whether a qualifying toxicity occurred within the stated assessment period. The registry describes a serious adverse event separately by its consequences, including death, a life-threatening experience, hospitalization, disruption of normal life functions, congenital anomaly or birth defect, or an important medical event requiring intervention to prevent such outcomes.

General statistical context: an event percentage describes how often an event was observed in a defined group. It is useful for summarizing toxicity because it relates the number of participants with an event to the number assessed or at risk. Its interpretation depends on a clear event definition, assessment period, and denominator. A confidence interval, if reported, would express uncertainty in the estimated event percentage; a narrow interval indicates greater precision than a wide interval, without resolving possible bias in how participants were selected or assessed.

The registry record does not report the primary endpoint percentage, its denominator, or a confidence interval. The adverse-event counts include serious and other adverse events, but those counts should not be treated as the missing primary endpoint result. The primary endpoint is toxicity more broadly, whereas the available safety breakdown distinguishes seriousness and does not supply a complete participant-level account of all toxicity at the specified endpoint time.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies the primary outcome as a percentage, which is a binary-outcome summary. It does not report the observed primary endpoint estimate or an analysis model. The explanations below describe general statistical concepts relevant to interpreting a percentage outcome and the listed safety counts; they do not imply that the trial used a particular unreported method.

What does an event percentage represent?

General statistical context: for a binary outcome, a participant either meets the event definition during the specified period or does not. An event percentage expresses the number meeting the definition relative to the group assessed. To interpret it, readers need the numerator, the denominator, the event definition, and the assessment window. A toxicity percentage at the end of therapy is not interchangeable with a count of serious events, a count of all events, or an event rate over person-time.

The registry states that the primary outcome is toxicity assessed by CTCAE v4.0 at the end of therapy, defined as 7 months. It does not report the percentage or denominator for that primary outcome. The adverse-event groups instead give numbers at risk and numbers affected for serious and other adverse events. Those entries provide some descriptive information, but they do not establish that their categories and follow-up exactly reproduce the primary endpoint definition.

What would an interval add?

General statistical context: a confidence interval describes the uncertainty around an estimated quantity under the assumptions of its statistical method. It communicates precision, not certainty that the estimate is unbiased or applicable to other populations. For an event percentage, an interval is useful alongside the numerator and denominator because a proportion based on limited observations can be imprecise. The registry record does not report a confidence interval for the primary endpoint or for the displayed safety counts.

A confidence interval also cannot repair design limitations. In a non-randomized study, differences between cohorts may reflect who entered each cohort, when treatment was given, or which regimen was received. An interval around a group-specific percentage would quantify sampling uncertainty under its assumptions, but would not by itself remove those sources of confounding.

Why does the non-randomized design matter?

General statistical context: random allocation aims to make treatment groups comparable on average, including for characteristics that were not measured. Non-randomized allocation does not have that protection. A direct contrast between groups can therefore mix treatment differences with baseline or clinical differences. Here, the registry labels the allocation non-randomized and explicitly warns that the arms are not a like-for-like control comparison. The arm descriptions also differ in surveillance and drug sequence, and the safety record combines Arm 2 and Arm 3 in one group.

These features make descriptive summaries more defensible than causal claims. The registry record does not report an adjusted comparison, covariate model, or a method for handling differences between cohorts. It would be inappropriate to infer that a difference in reported counts was caused by a treatment strategy.

What do the planning assumptions mean?

General statistical context: sample-size planning asks how many observations may be needed to detect a specified effect under stated assumptions. The planning illustrations associated with this analysis are described as using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They use a stated fraction of the observed effect. Such illustrations are planning exercises, not evidence that the trial had that design or that it achieved a particular probability of detecting an effect.

The registry describes a non-randomized trial with multiple cohort and treatment descriptions, so equal allocation between two arms is an illustrative assumption rather than a description of the trial allocation. The planning assumptions are explicitly not the trial's own design. Sample-size illustrations should not be read as observed power, as a retrospective test of whether the trial was adequately sized, or as a formal analysis of its results. The record does not report its sample-size justification.

What is not available for statistical evaluation?

The registry record does not report an analysis plan, an inferential analysis, a primary endpoint estimate, a confidence interval, or a p-value. It also does not report how missing observations were handled, the analysis populations, or whether outcome assessment was complete for all participants. Without those details, readers cannot determine whether the reported safety denominators represent all enrolled participants, all treated participants, or an endpoint-specific subset. The enrollment of 40 should not be substituted for a missing endpoint denominator.

General statistical context: over-dispersion is relevant when count data vary more than a simple model expects, often because observations are clustered or event rates differ across individuals. That issue would matter for a suitable count model if the data and analysis called for one. The registry record does not report a count model or an over-dispersion assessment, and the binary primary endpoint should not be reinterpreted as a count outcome. The appropriate statistical description is therefore limited to the definitions and counts actually 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("1/ Lead-In Cohort (Closed December 2018)/Arm 1/Com", "2/Biochemical Recurrence Cohort/Arm 1/Combination ", "2/Biochemical Recurrence Cohort/Arm2/Combo Therapy")
n      <- c(6, 8, 26)
events <- c(0, 1, 6)
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  1/ Lead-In Cohort (Closed December 2018)/Arm 1/Com  6  0  0.0  0.0  45.9  0.0  39.0  0.4  41.0
ARM  2/Biochemical Recurrence Cohort/Arm 1/Combination   8  1  12.5  0.3  52.7  2.2  47.1  2.8  48.2
ARM  2/Biochemical Recurrence Cohort/Arm2/Combo Therapy  26  6  23.1  9.0  43.6  11.0  42.1  11.1  42.3

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 5 outcome measures, including the primary toxicity endpoint and secondary outcomes concerning grade 3 and grade 4 adverse events and a 20% change in PSA doubling time among evaluable participants with biochemical recurrence. It reports 0 analyses. The statistical plan for handling multiple outcomes or repeated assessments is not reported in the registry record.

General statistical context: when investigators test or estimate many outcomes, the chance of at least one apparently notable finding can increase if no multiplicity strategy is used. Possible approaches include identifying a primary endpoint in advance, adjusting inference, or treating secondary outcomes as descriptive. The registry identifies a primary endpoint, but does not report a multiplicity adjustment, hierarchy of tests, or inferential results. It is therefore not possible to determine how multiplicity was addressed in the study analysis.

Missing-data procedures and analysis populations are also not reported in the registry record. The outcome description refers to evaluable participants for the PSA doubling-time measure, but does not define evaluability or state how participants without the required information were handled. The record does not specify whether the primary analysis included all enrolled participants, all treated participants, or a subset with an endpoint assessment.

These omissions affect interpretation of both denominators and uncertainty. A reported percentage can change depending on who is included and how missing outcome status is treated. The adverse-event counts identify participants at risk within each safety grouping, but do not explain whether anyone was excluded from those denominators or why. No assumptions about missing observations, intention-to-treat analysis, per-protocol analysis, or imputation should be attributed to this trial from the registry record alone.

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: cardiac disorders - other, specify: accelerated junctional rhythm (16.7%), cpk increased (12.5%), diarrhea (12.5%), infections and infestations - other, specify: covid (12.5%), muscle weakness lower limb (12.5%).

The safety record separates serious adverse events from other adverse events within three groupings. In the lead-in group, 6 participants were at risk for serious events and 1 was affected; 6 were at risk for other events and 6 were affected. In the biochemical recurrence Arm 1 group, the corresponding counts were 8 and 3 for serious events, and 8 and 7 for other events. In the group labeled Arm 2/Arm 3, the counts were 26 and 2 for serious events, and 26 and 25 for other events. Each grouping reports 0 deaths.

These are participant counts, not a comparison of treatment effects. The non-randomized design, differing cohort descriptions, and combined Arm 2/Arm 3 safety label prevent a like-for-like interpretation. The record does not give the number of individual events per participant, event onset or duration, or detailed severity distributions beyond the serious-versus-other categorization shown. It also does not establish from these figures alone whether each participant had one or multiple events.

The listed serious-event terms and percentages are descriptive registry entries. They do not necessarily represent the probability of an event for every participant in every cohort. The registry record does not report the denominators behind those term percentages. These counts should not be confused with the primary endpoint result, which is a broader toxicity percentage assessed at the end of therapy and is not reported in the registry record.

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

The record documents a completed phase II study of immunotherapy strategies in people with biochemically recurrent prostate cancer and reports adverse-event counts by cohort. It also gives a registered primary safety question: the percentage of participants with toxicity assessed at the end of therapy. That primary percentage is not reported. The registry record likewise does not report an anti-tumor outcome estimate, despite describing anti-tumor impact as the study objective.

The posted safety counts show that adverse events were recorded in each of the three listed groupings, with serious and other adverse events reported separately and 0 deaths in each group. The counts are useful as descriptive safety information for the named at-risk populations. They do not indicate comparative risk because treatment was not randomized, groups differed in treatment sequence, and one safety grouping combines Arm 2 and Arm 3. No statistical comparison or uncertainty estimate is reported.

General statistical context: a study result is most informative when the outcome definition, denominator, estimate, and uncertainty are all available. Without the primary endpoint percentage and its analysis population, the registered primary question cannot be quantitatively interpreted. Without a randomized comparison or suitable adjusted analysis, group differences cannot be attributed to treatment alone. Even a precise estimate would not resolve selection or confounding concerns inherent in the design.

Accordingly, the registry entry supports an account of design features and reported safety counts, not a conclusion that one regimen is superior, safer, or effective. The absence of an endpoint estimate in the registry record should not be mistaken for a finding of no toxicity or no anti-tumor activity. It means the relevant result is not reported in the registry record. The trial's findings should be interpreted only to the extent that the outcome and population are actually specified.

14. Why This Trial Matters Statistically

This trial illustrates why design labels and endpoint definitions are essential when interpreting clinical results. The registry identifies a phase II study, but its non-randomized allocation and differing cohort descriptions mean the groups are not interchangeable treatment arms. A descriptive count in one group cannot, on its own, quantify a treatment effect relative to another group.

The distinction between the primary toxicity endpoint and the trial's anti-tumor objective is also statistically important. A study objective describes the broader research aim; the primary endpoint specifies the registered measure intended to be assessed. Here, the primary measure is a toxicity percentage at the end of therapy, while the registry summary describes anti-tumor impact. The record does not report an estimate for either question, so readers should not infer an efficacy result from safety counts.

The safety denominators show why enrollment and endpoint denominators must not be treated as identical. The registry lists an enrollment of 40, while the safety groups report their own numbers at risk. These group-specific figures are the relevant denominators for understanding the posted adverse-event counts, though the record does not explain all denominator selection details.

Finally, statistical precision and study design answer different questions. General statistical context: intervals can describe uncertainty around estimates, but cannot remove systematic differences between non-randomized groups. Here, neither an interval nor an inferential analysis is reported. The record therefore offers limited but useful descriptive safety information while leaving major questions about the primary endpoint and comparative effects unresolved.

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.

Several limitations specific to this record constrain interpretation. Allocation was non-randomized, and the registry explicitly notes that the arms are not a like-for-like control comparison. Treatment descriptions also vary: some include surveillance before treatment, some use surveillance as needed, and the described drug combinations are not identical across cohorts. Consequently, differences in adverse-event counts cannot be isolated as treatment effects.

The safety results combine Arm 2 and Arm 3 in one grouping, so the counts do not distinguish those strategies. The record gives numbers at risk and affected for serious and other adverse events, but does not provide participant-level data, event timing, recurrence, duration, or a complete account of event severity. The listed serious-event term percentages have no stated denominators in the record.

The primary endpoint result is not reported in the registry record. There is no reported estimate, confidence interval, or inferential analysis, and the record lists 0 analyses. Missing-data methods, analysis populations, and the sample-size justification are not reported in the registry record. The record also does not explain whether the safety denominators include all enrolled participants or only those meeting a particular treatment or assessment criterion.

The enrollment is listed as 40, but that figure does not supply missing outcome denominators or establish the precision of any unreported endpoint. The completed status and primary completion date do not resolve the absence of results. These limitations mean the posted information should be treated as descriptive registry data, not as a complete statistical report of the trial's primary outcome or anti-tumor objective.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Combination Immunotherapy in Biochemically Recurrent Prostate Cancer — NCT03315871.
• Registry record last updated 2025-06-19; status last verified 2025-06.
• Results first posted 2025-06-19.
• 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

The registry lists NCT03315871 as a completed phase II, parallel, non-randomized, unmasked trial with enrollment of 40. The study concerns combination immunotherapy strategies for biochemically recurrent prostate cancer. The registered primary endpoint is the percentage of participants with toxicity assessed by CTCAE v4.0 at the end of therapy, defined as 7 months.

Three safety groupings report serious and other adverse-event counts, with 0 deaths in each. The registry also lists serious-event terms and percentages. The primary toxicity percentage, an anti-tumor outcome estimate, confidence intervals, p-values, and inferential analyses are not reported in the registry record. Because cohorts were not randomized and the safety results combine two arm labels in one grouping, the reported counts are descriptive and do not support a like-for-like treatment comparison.