Cabozantinib and Pembrolizumab for Advanced Metastatic Melanoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Cabozantinib and Pembrolizumab for Advanced Metastatic Melanoma (NCT03957551), a na phase 1, phase 2 study. The primary outcome is Phase I: Number of Participants With a Dose Limiting Toxicity (DLT) Using CTCAE, Version 4.03.

2. Clinical Question

The study asks whether cabozantinib combined with pembrolizumab can be assessed for safety and preliminary efficacy as front-line therapy for advanced metastatic melanoma. The registry describes a sequential Phase I and Phase II study, rather than a randomized comparison. Its Phase I primary outcome concerns dose-limiting toxicity (DLT), while listed secondary outcomes include Disease Control Rate, Progression-Free Survival, and Overall Survival at 2 Years.

The registry summary frames the study as an evaluation of safety and preliminary efficacy. However, the record does not report results for the listed disease-control or survival outcomes. It also does not provide a comparative estimate of the combination against another treatment. The available record therefore supports a description of the study question, design and posted safety counts, but not a conclusion about comparative efficacy.

3. Trial Design

The registry records a Phase I and Phase II, sequential study with an enrollment of 28 and a status of terminated. The recorded start date is 2019-08-27, and the primary completion date is 2024-06-10. The study is described as front-line therapy research in advanced metastatic melanoma. These registry fields describe the study record; they do not establish why enrollment ended or whether every intended study activity was completed.

Allocation is recorded as NA, masking as none, and the model as sequential. There is no randomized comparator arm in the record. Participants are described in dose cohorts, so observed outcomes across those cohorts should not be interpreted as treatment-versus-control effects. Differences between cohorts could reflect factors other than dose, and the record does not report an adjusted comparison that would account for such factors.

The intervention description specifies cabozantinib tablets at three dose levels: 40, 20 and 60 mg per day. Pembrolizumab is described as a fixed 200 mg intravenous infusion once every 3 weeks. The Phase I outcome population identifies cohorts at 40mg and 60 mg. The posted adverse-event groups include Phase I cohorts at 40mg and 60mg and a Phase II group at 40mg. The presence of a dose level in the regimen description does not itself establish that a corresponding cohort has results in the posted adverse-event information.

The primary completion date marks a registry milestone, not necessarily the date of final analysis or publication. The registry record does not report the reason for termination, the recruitment history, the full protocol, or whether termination affected the planned outcome assessments. The listed secondary outcomes are Disease Control Rate, Progression-Free Survival, and Overall Survival at 2 Years, but results for them 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:

Analysis population for the primary outcome, as defined in the registry: Number of participants enrolled to Phase I cohorts (6 at 40mg and 2 at 60 mg levels)

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 I: Number of Participants With a Dose Limiting Toxicity (DLT) Using CTCAE, Version 4.03, measured over Up to 21 days.

The Phase I primary endpoint is the number of participants with a dose-limiting toxicity, assessed using CTCAE, Version 4.03, over up to 21 days. A participant count is a way to summarize how many people in the defined cohort experienced an event meeting the study's DLT criteria. The registry description says adverse events are considered in the DLT assessment unless an event is clearly unrelated to trial treatment.

DLT assessment is used in early-phase research to characterize tolerability within a specified dose and observation window. It can help describe whether events meeting the study's criteria occurred after treatment exposure, but it is not a complete summary of all possible harms or longer-term safety. The meaning of the count depends on the protocol's operational rules, including how events are graded, attributed, and handled when assessment is incomplete. The registry record gives the CTCAE version and the general attribution rule, but does not report the full DLT decision procedure.

The primary endpoint is a count, not a direct estimate of comparative benefit. The registry describes the primary outcome population as Phase I participants enrolled to the 40mg and 60 mg cohorts. It does not report a numerical DLT result for this endpoint in the record. The serious adverse-event counts listed elsewhere are safety information, but should not be treated as the DLT result: the registry describes DLT assessment as a specific endpoint with its own criteria and timeframe.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The posted results are primarily descriptive. Understanding them requires distinguishing the Phase I DLT endpoint from the separate serious and other adverse-event summaries, and distinguishing a count within a cohort from a comparative treatment effect. The registry lists a sequential Phase I and Phase II study, no masking, and allocation as NA. It does not report a randomized allocation process or a control treatment.

What does a participant count convey?

A count records how many participants meet an event definition in a specified group. For the primary endpoint, the event is a DLT according to the study assessment and CTCAE, Version 4.03, over up to 21 days. For serious adverse events, the posted summaries identify participants at risk and participants affected. These definitions are not interchangeable: a serious adverse event is not automatically a DLT, and the registry does not state that every serious event met the DLT criteria.

Counts are easiest to interpret alongside the size of the relevant group and the outcome definition. A count alone does not describe how events were distributed over time, whether participants experienced more than one event, or whether an event was judged related to treatment. The group summaries report affected participants, not a detailed event history. The record gives group-specific counts, but not enough information to reconstruct every event or determine how repeated events were handled.

Why does the design matter?

Randomization can make treatment groups more comparable, on average, by assigning participants without selecting treatment according to their characteristics. This study is not recorded as randomized and has no like-for-like control comparison. Its sequential dose and phase groups therefore cannot be treated as exchangeable arms. Differences in their counts could reflect group composition, the timing and purpose of enrollment, follow-up, or other factors, as well as dose. The registry does not report an adjusted analysis that could address these issues.

Masking is recorded as none. In general, masking can reduce the influence of treatment knowledge on some measurements and decisions, although it does not remove all sources of bias. The registry record does not describe additional safeguards for outcome assessment. That limitation is especially relevant to assessments that involve judgment, but the specific measurement procedures are not reported in the registry record.

What can confidence intervals and tests add?

In general, a confidence interval expresses the statistical uncertainty around an estimate under a specified analysis. A narrow interval indicates greater precision than a wide interval, while an interval that spans a broad range can leave substantial uncertainty about the underlying value. A hypothesis test and its p-value address a different question: how compatible the observed data are with a specified null model and its assumptions. Neither a confidence interval nor a p-value establishes causation, and neither replaces attention to design, outcome definition, and missing information.

The registry record lists no analyses and does not report a confidence interval or p-value for the primary endpoint or posted safety counts. Consequently, the reported counts should be described as observed descriptive information, not as statistically tested differences. The absence of an analysis record does not show that analyses were never performed; it means such analyses are not reported in the registry record.

How should the planning illustrations be read?

In general, sample-size calculations are planning tools: they describe the number of participants or events a future design might require under stated assumptions. They are not observed power, and they do not establish that the completed study had sufficient precision. The registry's stated illustration assumptions use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is expressed as a fraction of the observed effect. For time-to-event outcomes, the illustrations use a total number of events; for other outcomes, they use participants per arm.

Those assumptions describe a planning illustration, not the design of this sequential study. In particular, equal allocation between two arms is not a description of the posted dose cohorts, and an effect based on observed data is not independent confirmation of that effect. The record does not report the trial's own sample-size justification. No additional sample-size values or planning results should be inferred from the assumptions alone.

Over-dispersion is a general concern when count variability exceeds what a simple model assumes. Ignoring it can make uncertainty look smaller than it is. The posted primary endpoint is a participant count with an event definition, not a detailed event-rate model, and the registry record does not report a fitted model or a dispersion assessment. More broadly, statistical methods cannot recover missing outcome data, create a control comparison, or clarify event attribution when those details are absent.

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 I: Cabozantinib and Pembrolizumab (40mg)", "Phase I: Cabozantinib and Pembrolizumab (60mg)")
n      <- c(6, 2)
events <- c(1, 0)
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 I: Cabozantinib and Pembrolizumab (40mg)  6  1  16.7  0.4  64.1  3.0  56.4  3.7  57.9
ARM  Phase I: Cabozantinib and Pembrolizumab (60mg)  2  0  0.0  0.0  84.2  0.0  65.8  0.8  70.8

11. Multiplicity, Missing Data and Analysis Populations

The record lists five outcome measures and names Disease Control Rate, Progression-Free Survival, and Overall Survival at 2 Years among the secondary outcomes. It does not report results for those efficacy outcomes, nor does it report a statistical analysis plan specifying which outcomes were confirmatory, how they were ordered, or how multiple testing was handled. When several outcomes or comparisons are tested, the chance of at least one apparently unusual result can increase. General methods for addressing this include prespecifying a hierarchy or adjusting inference, but no such approach is reported in the registry record.

The registry record lists no analyses. It therefore does not provide the analysis populations, missing-data methods, censoring rules for time-to-event outcomes, or a description of any sensitivity analyses. These details matter because incomplete follow-up or unavailable measurements can affect how outcome estimates are interpreted. For survival outcomes, censoring and follow-up definitions are especially important; the registry record does not report them.

The Phase I primary outcome specifies participants enrolled to the 40mg and 60 mg cohorts and an assessment period of up to 21 days, but the record does not report how missing or unevaluable assessments were treated. It also does not state whether the reported enrollment represents the population included in each endpoint analysis. No intention-to-treat, per-protocol, or safety population definition is reported in the registry record.

These omissions limit the ability to assess selective reporting, multiplicity, and potential bias from missing observations. They do not establish that the study lacked a prespecified plan or that data were missing; they mean those details are not reported in the registry record. Any interpretation should remain tied to the specific posted counts and outcomes rather than assuming unreported analysis choices.

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: vomiting (50.0%), myocarditis (5.0%), abdominal pain (5.0%), dysphagia (5.0%), small intestinal perforation (5.0%).

The serious adverse-event summaries are presented by study phase and cabozantinib dose group. The Phase I 40mg group lists 0 of 6 participants affected, the Phase I 60mg group lists 1 of 2, and the Phase II 40mg group lists 7 of 20. These figures are counts of participants recorded as affected within each group, not a randomized comparison. The group sizes and study phase differ, so the counts should not be used to rank doses or phases as if the groups were directly comparable.

The listed top serious-event terms are Vomiting at 50.0%, Myocarditis at 5.0%, Abdominal pain at 5.0%, Dysphagia at 5.0%, Small intestinal perforation at 5.0%, and Death NOS at 5.0%. The registry information does not identify the denominator or explain whether percentages refer to participants, events, or a particular group. It also does not report whether these terms were judged related to treatment. They should be read as listed terms and percentages, not as causal findings.

Recorded deaths are 2 for Phase I at 40mg, 1 for Phase I at 60mg, and 10 for Phase II at 40mg. Timing, cause, and relationship to treatment are not reported in the registry record. The separate DLT endpoint has its own definition and timeframe; serious adverse events and deaths should not be substituted for its result.

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

The registry record establishes that a sequential Phase I and Phase II study evaluated cabozantinib with pembrolizumab in advanced metastatic melanoma, with safety and preliminary efficacy as its stated aims. It reports adverse-event counts for the listed dose and phase groups and names a Phase I DLT endpoint. These details describe what the study record contains; they are not, by themselves, evidence that the combination caused or prevented an outcome.

The posted counts can indicate that serious adverse events and other adverse events were recorded among participants in the described groups. They cannot establish comparative safety because there is no randomized control group and no like-for-like comparison. Nor can the serious-event terms establish treatment causality: attribution is not reported for those terms. The recorded deaths likewise do not identify cause or relationship to treatment.

The primary endpoint is the number of participants with DLT under the study's criteria during up to 21 days, but its numerical result is not reported in the registry record. Results for the listed secondary outcomes are also not reported. The record therefore does not provide an efficacy estimate, a measure of its precision, or a statistical test supporting a treatment comparison. The registry's stated preliminary-efficacy aim should not be mistaken for a posted efficacy finding.

The absence of an analysis entry should be interpreted narrowly. It indicates that statistical analyses are not listed in the registry record, not that no analyses were conducted elsewhere. Similarly, the terminated status and completion date are the registry entries, but the reason for termination and its effect on outcome collection are not reported.

Overall, the available information supports a limited descriptive reading of the study design and posted safety data. It does not support conclusions about benefit, comparative risk, treatment attribution, or the precision of the primary endpoint. Further interpretation would require reported outcome results and analysis details, including definitions, follow-up, missing-data handling, and the basis of the serious-event percentages.

14. Why This Trial Matters Statistically

This study illustrates why early-phase safety outcomes and later efficacy questions should be kept separate. The Phase I primary endpoint is a DLT count defined over up to 21 days, while the listed secondary outcomes include disease control and survival. A safety count from an early-phase cohort does not answer a survival question, and a list of intended outcomes is not a report of their results.

It also highlights the limits of reading dose cohorts as if they were treatment arms in a randomized trial. The registry records a sequential model and allocation as NA, with no like-for-like control comparison. The phase and dose groups have different purposes and group sizes. Their counts can be described, but cross-group contrasts are vulnerable to differences other than the dose itself. A statistical calculation cannot eliminate this design limitation.

The record makes clear that outcome labels and definitions matter. Serious adverse events, other adverse events, deaths, and DLTs are not interchangeable categories. DLT assessment uses a stated CTCAE version and observation window, while the serious-event listing provides group counts and event terms. Without a reported link between these measures, they should remain separate in interpretation.

Finally, the record underscores the distinction between an analysis and a planning illustration. Sample-size illustrations based on assumptions are not observed power and do not substitute for estimates with uncertainty. Because the registry lists no analyses and no primary endpoint result, the data available in the record support descriptive reporting but not a precision-based or causal conclusion. These are important distinctions when communicating early-phase evidence without overstating what the record establishes.

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 record identifies the study as terminated but does not report the reason for termination or whether it affected recruitment, follow-up, or endpoint completion. Although enrollment is recorded as 28, endpoint-specific analysis populations are not reported. The registry does not state whether all enrolled participants were evaluable for the primary endpoint or how incomplete assessments were handled.

The study has no randomized control group, and allocation is listed as NA. Therefore, the posted dose and phase groups cannot provide a controlled estimate of comparative treatment effects. The sequential design also means cohorts may differ in context or participant characteristics; the registry record does not report methods to make them comparable.

Outcome reporting is incomplete in the information available. The Phase I DLT endpoint is defined, but its result is not reported. The registry names secondary disease-control and survival outcomes without reporting their results. It also does not provide confidence intervals, p-values, detailed time-to-event methods, follow-up distributions, or missing-data procedures. The registry record lists no analyses, so the statistical basis for any broader inference is unavailable.

Safety interpretation has further constraints. The serious-event percentages are listed without a stated denominator in the registry information, and attribution, timing, and event-level details are not reported. Death counts are recorded, but their causes and relationships to treatment are not specified. These limitations prevent attributing the listed events to study treatment or comparing event frequencies reliably across cohorts.

The regimen description includes cabozantinib levels of 40, 20 and 60 mg per day, while the posted adverse-event groups identify Phase I at 40mg and 60mg and Phase II at 40mg. The record does not explain the status of a 20 mg cohort in the posted safety information. This should be treated as an information gap, not evidence that the dose was or was not administered.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Cabozantinib and Pembrolizumab for Advanced Metastatic Melanoma — NCT03957551.
• Registry record last updated 2026-01-29; status last verified 2025-10.
• Results first posted 2025-08-24.
• 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 record describes a sequential Phase I and Phase II study of cabozantinib plus pembrolizumab for advanced metastatic melanoma. It records an enrollment of 28, no masking, allocation as NA, and terminated status. The recorded start date is 2019-08-27, with primary completion on 2024-06-10.

The Phase I primary endpoint is the number of participants with a DLT assessed using CTCAE, Version 4.03, over up to 21 days. The study description lists cabozantinib dose levels of 40, 20 and 60 mg per day and pembrolizumab at 200 mg once every 3 weeks. Secondary outcomes include Disease Control Rate, Progression-Free Survival, and Overall Survival at 2 Years.

The record lists serious and other adverse-event counts by phase and dose group, but does not report the primary endpoint result, results for the secondary outcomes, or statistical analyses. The design has no randomized control comparison. The posted information therefore supports descriptive reporting of the study record and safety counts, not a comparative efficacy or causal conclusion.