Safety and Efficacy of Therapies for Metastatic Castration-resistant Prostate Cancer (mCRPC): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy of Therapies for Metastatic Castration-resistant Prostate Cancer (mCRPC) (NCT04631601), a randomized phase 1, phase 2 study. The primary outcome is Subprotocols A, B and C (Parts 1 and 2): Number of Participants Who Experienced a Dose-limiting Toxicity (DLT).

2. Clinical Question

3. Trial Design

NCT04631601 is a Phase 1 and Phase 2 master protocol evaluating investigational therapies in participants with mCRPC. The registry lists randomized allocation, no masking, and a sequential study model. Enrollment is listed as 55. The study started on 2021-01-15, had a primary completion date of 2023-10-23, and is marked terminated. The reason for termination is not reported in the registry record.

The protocol encompasses several treatment subprotocols and cohort types. Acapatamab was studied with enzalutamide, abiraterone, and AMG 404, as well as as monotherapy. AMG 404 was also studied as monotherapy. Dose exploration was intended to estimate a maximum tolerated dose or recommended phase 2 dose. Dose expansion was intended to confirm safety and tolerability and further evaluate efficacy. Some expansion cohorts were designated for participants in Asia, with safety, tolerability, and pharmacokinetics among their stated aims.

The primary endpoint is narrower than the full master protocol: it covers Subprotocols A, B, and C, Parts 1 and 2, and counts participants experiencing a DLT. The listed primary results include cohorts receiving acapatamab with enzalutamide, abiraterone, or AMG 404. The adverse-event results also include an AMG 404 monotherapy cohort and an acapatamab monotherapy cohort, but those are not among the stated primary endpoint subprotocols and parts.

The registry lists 35 outcome measures and 28 posted secondary outcomes. Secondary outcomes include objective response rate, CTC responses and conversions, PSA response and duration measures, and duration of response per RECIST 1.1. The record does not report further details of the randomization procedure, allocation concealment, or the operational sequence for adding or closing cohorts.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: A total of 55 participants were enrolled and 54 received trial treatment. For subprotocol A, no participants were enrolled in part 2.

Only the first 6 of 11 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: DLT Analysis Set: Included all participants who received at least one acapatamab dose and had an evaluable DLT endpoint (had a DLT or completed all planned doses without a DLT within 28-day DLT window in Cycle 1).

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: Safety Analysis Set: all participants who were enrolled and received at least one dose of acapatamab or AMG 404.

Only the first 6 of 12 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Subprotocols A, B and C (Parts 1 and 2): Number of Participants Who Experienced a Dose-limiting Toxicity (DLT), measured over Cycle 1: Day 1 to Day 28 (28-day cycle).

The primary endpoint is the number of participants who experienced a dose-limiting toxicity (DLT) in Subprotocols A, B, and C, Parts 1 and 2. Its timeframe is Cycle 1, Day 1 to Day 28, described as a 28-day cycle. A participant belongs to the DLT Analysis Set if they received at least one acapatamab dose and had an evaluable DLT endpoint: they either had a DLT or completed all planned doses without a DLT within the Cycle 1 window.

The registry defines DLTs as certain adverse events occurring within 28 days of the first AMG 160 dose and possibly related to treatment. The listed categories include Grade 5 toxicity, specified severe blood-related toxicities, febrile neutropenia, selected Grade 3 or higher non-hematologic toxicities, and specified liver-test abnormalities. Some criteria include duration, treatment delay, dose administration, or other qualifications; the registry refers readers to the protocol for the complete list. The primary endpoint is reported as a participant count, not as a rate or time-to-event measure.

General: DLT endpoints are used in early-phase dose exploration to describe treatment-limiting toxicity during a defined observation window. They can inform dose selection alongside other safety and efficacy evidence. A count alone does not describe the severity or duration of every event, and it does not provide a direct comparative estimate unless an appropriate comparison and analysis are reported.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What kind of outcome is reported?

The primary endpoint is a binary event summarized as a count: each participant either experienced a DLT within the defined window or did not. The registry posts the number with an event and the cohort size. A count is easy to interpret as an observed description, but it does not, by itself, quantify the uncertainty around the population event probability or estimate a treatment contrast.

General: A confidence interval describes a range of values compatible with the observed data under a statistical model. For an event proportion, it can communicate uncertainty that a raw count cannot. The registry record does not post confidence intervals for the primary results, so the uncertainty cannot be read from an interval estimate in the reported endpoint results.

Why are the cohort results not a two-arm comparison?

The endpoint results are organized across cohorts with differing combinations, doses, and protocol parts. A comparative analysis would need a defined contrast, an analysis population for that contrast, and a method for estimating the difference or another chosen effect measure. The registry record does not post such an estimate. It explicitly notes that values are missing or not estimable for a two-arm comparison of this binary endpoint. Therefore, a comparative p-value or effect estimate should not be inferred from the separate counts.

Although the registry lists randomized allocation, that designation does not turn every reported cohort count into a direct randomized comparison. The record does not report which cohorts formed a particular randomized contrast, how randomization was implemented, or the analysis associated with any such contrast. These details are not reported in the registry record.

How should a planning illustration be understood?

The sample-size assumptions describe a planning illustration computed from an observed effect, not the trial's own design. The stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. For non-time-to-event outcomes, the illustration is expressed as participants per arm for a specified power level. No planning values are included in the registry record for this analysis, so no sample-size result can be stated.

General: A sample-size calculation asks how many participants a hypothetical future study might need under chosen assumptions. It is not an observed result and is not evidence that the completed trial had a particular probability of detecting an effect. It should not be confused with observed power, nor treated as a record of the study's original design or justification. The original sample-size justification is not reported in the registry record.

What does over-dispersion mean in general?

General: Over-dispersion occurs when data vary more than a statistical model expects. In count outcomes, for example, an analysis that assumes a particular mean-variance relationship may understate uncertainty if the observations are more variable than that assumption allows. The primary endpoint is posted as participant counts by cohort, and the registry does not report a fitted count model or an over-dispersion assessment. It would therefore be inappropriate to claim that over-dispersion was present or addressed.

What can be inferred from no posted analysis?

The registry lists 0 analyses for the posted results. This means no analysis results of that type are available in the record for interpretation. It does not establish that investigators conducted no analyses outside the registry. The registry does not report an analysis plan, model specification, hypothesis tests, adjusted estimates, or a statistical decision rule for the primary endpoint.

General: Statistical inference depends on both the observed data and the method used to analyze them. A method can account for design features, covariates, or multiple comparisons when those are part of a prespecified plan. Without such details and posted estimates, interpretation should stay descriptive: report what was observed in each cohort, distinguish it from a comparative conclusion, and note that precision is not quantified by a posted interval.

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("Subprotocol A Part 1: Acapatamab 0.15 mg and Enzal", "Subprotocol A Part 1: Acapatamab 0.15 mg and Enzal", "Subprotocol A Part 1: Acapatamab 0.3 mg and Enzalu", "Subprotocol A Part 1: Acapatamab 0.3 mg and Enzalu", "Subprotocol B Part 1: Acapatamab 0.15 mg and Abira", "Subprotocol B Part 1: Acapatamab 0.3 mg and Abirat")
n      <- c(3, 3, 4, 4, 4, 5)
events <- c(1, 0, 2, 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  Subprotocol A Part 1: Acapatamab 0.15 mg and Enzal  3  1  33.3  0.8  90.6  6.1  79.2  6.8  80.6
ARM  Subprotocol A Part 1: Acapatamab 0.15 mg and Enzal  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Subprotocol A Part 1: Acapatamab 0.3 mg and Enzalu  4  2  50.0  6.8  93.2  15.0  85.0  14.7  85.3
ARM  Subprotocol A Part 1: Acapatamab 0.3 mg and Enzalu  4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2
ARM  Subprotocol B Part 1: Acapatamab 0.15 mg and Abira  4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2
ARM  Subprotocol B Part 1: Acapatamab 0.3 mg and Abirat  5  1  20.0  0.5  71.6  3.6  62.4  4.3  64.1

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 35 outcome measures and 28 posted secondary outcomes, spanning response, biomarker, and duration measures. The primary DLT outcome is reported for specified subprotocols and parts, while secondary outcomes have their own stated scopes. The registry record does not report a multiplicity strategy, such as a prespecified ordering of hypotheses or an adjustment for testing multiple outcomes. Without that information, nominal findings across many outcomes cannot be treated as if they arose from a single isolated test. No inferential secondary outcome results are detailed in the registry record for this analysis.

Missing-data handling is not reported in the registry record. The record does define the DLT Analysis Set: participants had to receive at least one acapatamab dose and have an evaluable endpoint, meaning a DLT occurred or all planned doses were completed without a DLT within the 28-day DLT window in Cycle 1. This definition describes who is included in the endpoint population, but does not explain how unevaluable participants, withdrawals, incomplete follow-up, or other missing observations were handled.

The primary count results and the adverse-event group listings have different stated scopes. The primary endpoint covers Subprotocols A, B, and C, Parts 1 and 2; the safety listings also include AMG 404 monotherapy and acapatamab monotherapy. Those listings should not be assumed to represent the primary endpoint analysis population. The record does not report further analysis-population definitions for all secondary outcomes.

General: Multiplicity refers to the increased chance of misleading positive findings when many hypotheses are examined. Missing-data methods can also affect estimates if outcome availability differs across participants. Since neither a multiplicity approach nor detailed missing-data methods are reported in the registry record, these aspects cannot be evaluated from the posted 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: cytokine release syndrome (75.0%), atrial fibrillation (33.3%), myocardial infarction (33.3%), covid-19 (33.3%), staphylococcal sepsis (33.3%).

The serious adverse-event listings report, for each group, the number at risk and the number affected by serious adverse events. For example, the acapatamab and AMG 404 Part 1 group lists 11 at risk and 9 affected, while the AMG 404 monotherapy group lists 10 at risk and 1 affected. The figures are cohort-specific descriptions; they are not comparative estimates adjusted for differences between groups or exposure.

The registry separately lists participants with other adverse events and deaths. A serious adverse event, an event meeting the primary DLT definition, and a death are distinct reporting categories. Their counts should not be substituted for one another. The posted group-level death counts include 6 in the acapatamab and AMG 404 Part 1 group and 0 in the acapatamab monotherapy group. The record does not establish from these counts whether a death was caused by treatment.

The listed serious-event terms include cytokine release syndrome at 75.0%; atrial fibrillation, myocardial infarction, COVID-19, staphylococcal sepsis, and fall at 33.3% each. The denominators underlying these term percentages are not reported in the registry record. Because cohorts are small and differ in composition, these percentages should be read as registry descriptions rather than stable estimates of event frequency in a broader population.

General: Safety counts describe observed events in the stated groups and follow-up context. They do not alone establish causality or compare risks across treatments. The registry record does not report enough detail to assess event timing, duration, exposure-adjusted incidence, or attribution for each listed event.

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

The primary results document DLT counts in the reported dose-exploration cohorts for acapatamab combinations with enzalutamide, abiraterone, and AMG 404. Some cohorts have no recorded DLTs, while others have one or more. The counts are useful as a descriptive record of what occurred in these groups during the defined Cycle 1 window.

They do not, on their own, establish that any combination or dose is safer or less safe than another. Cohort sizes are small, and the groups differ by treatment combination, dose, and protocol part. No comparative estimate, confidence interval, or inferential analysis is posted for the primary endpoint. The registry record also does not report the decision criteria used to select a dose or explain how the observed counts affected dose selection.

The primary endpoint is limited to DLTs as defined for the specified parts of the study. It does not summarize every adverse event, all treatment-related harms, long-term safety, or efficacy. The registry lists additional outcomes such as objective response rate, PSA response, CTC measures, and duration of response, but those outcomes are not represented by the primary DLT counts. Safety listings for monotherapy cohorts have a different scope from the primary endpoint.

General: An observed event count is not a guarantee about what would happen in another group of participants. A count can be affected by the cohort size, who entered the cohort, follow-up, and the event definition. Comparative conclusions generally require a clearly defined contrast and an analysis that accounts for the study design; those results are not reported in the registry record.

The study is marked terminated, but the reason is not reported in the registry record. That status does not, by itself, show why the study stopped or indicate the direction of its safety or efficacy findings. The posted data should therefore be read as limited descriptive evidence from a multi-part protocol, not as a definitive treatment comparison or a complete account of the study's conclusions.

14. Why This Trial Matters Statistically

This trial is statistically interesting because it uses a master-protocol structure to evaluate multiple investigational treatment approaches in the same disease area. The registry describes dose exploration followed by dose expansion, with cohorts involving acapatamab combinations, monotherapy, and an Asia expansion component. Such a structure creates several distinct questions: dose selection, safety characterization, pharmacokinetics, and preliminary efficacy. Results from one cohort do not automatically answer the others.

The primary outcome is a count of participants with a DLT, which makes the reported result concrete but also limits what can be concluded without an explicit analysis. The varying cohort sizes and treatment combinations make unadjusted comparisons especially difficult to interpret. Reporting the participant count and event count separately by cohort preserves that context rather than implying a single common estimate across the protocol.

The registry also illustrates why design labels need careful interpretation. Randomized allocation is listed, yet the posted endpoint results do not identify a two-arm contrast or provide an estimate for one. A design feature and a posted analysis are not the same thing: readers need the particular comparison, analysis population, and method to interpret a randomized treatment effect.

Finally, the record lists many outcome measures but does not report a multiplicity strategy, detailed missing-data methods, or statistical analyses for the primary results. These omissions constrain inference even when observed counts are available. Statistically, the central lesson is to distinguish descriptive cohort evidence from comparative evidence and to keep the endpoint's population and timeframe in view.

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 endpoint results are small cohort counts, with cohort sizes ranging from the listed group sizes rather than a common analysis population. The cohorts vary in treatment combination, dose, and study part. The registry record does not post confidence intervals or inferential estimates for the primary endpoint, and it identifies values as missing or not estimable for a two-arm comparison of this binary outcome.

The endpoint definition refers to events possibly related to treatment and gives examples of qualifying toxicities, while directing readers to the protocol for the complete DLT list. The complete protocol criteria are not reproduced in the registry record summarized for these results. The endpoint is also limited to the specified DLT window in Cycle 1; it does not capture all possible safety outcomes or later events.

There is no reported explanation for the terminated status. The registry record does not report detailed eligibility criteria, randomization procedures, allocation concealment, the statistical analysis plan, sample-size justification, interim analyses, multiplicity handling, or missing-data methods. It also does not report the denominators for the listed serious adverse-event term percentages.

The master protocol includes cohorts outside the primary endpoint scope, including monotherapy cohorts. Adverse-event listings from those groups cannot be treated as if they were part of the primary DLT analysis. The registry's sample-size assumptions describe a planning illustration based on an observed effect, not the original trial design; no numerical planning results are included in the registry record for this analysis.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Safety and Efficacy of Therapies for Metastatic Castration-resistant Prostate Cancer (mCRPC) — NCT04631601.
• Registry record last updated 2026-08-26; status last verified 2026-07.
• Results first posted 2026-08-26.
• 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

NCT04631601 is a randomized, unmasked Phase 1 and Phase 2 sequential master protocol in metastatic castration-resistant prostate cancer. It evaluates investigational therapies including acapatamab, enzalutamide, abiraterone, and AMG 404. Enrollment is listed as 55; the study started on 2021-01-15, reached primary completion on 2023-10-23, and is marked terminated.

The primary endpoint counts participants with a DLT during Cycle 1 for Subprotocols A, B, and C, Parts 1 and 2. Posted results are cohort-specific counts for acapatamab combinations with enzalutamide, abiraterone, and AMG 404. The registry lists 35 outcome measures and 28 posted secondary outcomes, with 0 analyses listed. It does not report a two-arm estimate for the primary binary endpoint, detailed statistical methods, or the reason for termination. The displayed counts are descriptive and do not establish comparative treatment effects.