A Study Of PF-05212384 In Combination With Other Anti-Tumor Agents and in Combination With Cisplatin in Patients With Triple Negative Breast Cancer in an Expansion Arm (TNBC): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Of PF-05212384 In Combination With Other Anti-Tumor Agents and in Combination With Cisplatin in Patients With Triple Negative Breast Cancer in an Expansion Arm (TNBC) (NCT01920061), a non-randomized phase 1 study. The primary outcome is Number of Participants With Dose Limiting Toxicities (DLTs) - Arms A, B and C.

2. Clinical Question

In practical terms, the study’s primary question was whether toxicities meeting the protocol’s DLT definition occurred during the specified early assessment period, among participants included in its per-protocol analysis set. This kind of question is central to an early-phase combination study: it characterizes tolerability at candidate dose levels and can inform later dose selection. It does not by itself establish comparative benefit, because treatment assignment was not randomized and the primary measure is toxicity rather than a clinical efficacy outcome.

The study also included a cisplatin combination expansion in TNBC, with separate arms described by metastatic treatment line. Its stated purpose was to evaluate anti-tumor activity in that population, but the registry information presented does not include an anti-tumor activity result. Nor does the primary DLT outcome answer whether the combination improves outcomes relative to another treatment. The available record therefore supports describing the study question and outcome definitions, but not drawing a conclusion about the frequency of DLTs or the efficacy of the combinations.

3. Trial Design

This was a phase 1b, open-label, multicenter study, described in the official title as a three-arm study of PF-05212384 with other anti-tumor agents. The registry categorizes the phase as phase 1. Allocation was non-randomized, masking was absent, and the model was sequential. The study status is completed, with an enrollment of 110 recorded. The registry dates are 2013-09-10 for study start and 2020-01-08 for primary completion.

The combination arms assessed PF-05212384 with docetaxel, cisplatin, or dacomitinib. The adverse-event groups identify dose levels within these combinations, including multiple PF-05212384 doses and, for dacomitinib, more than one dacomitinib dose. The cisplatin expansion had two TNBC groups, described as Arm 1: 1L metastatic and Arm 2: 2L/3L metastatic. Arm descriptions for the listed arms are blank in the registry record, so further protocol detail about their conduct is not reported there.

Sequential, non-randomized allocation means participants were not assigned by randomization to comparable study groups. Dose groups may have been evaluated in sequence, but the protocol’s exact dose-escalation rules and decision process are not reported in the registry record. The different groups also represent different drug combinations, dose levels, or treatment-line categories. As a result, an observed contrast between groups would not isolate a treatment effect in the way a randomized controlled comparison is designed to do.

The primary outcome was the number of participants with DLTs in Arms A, B, and C during up to 21 days. Its stated analysis population was per protocol, with eligibility tied to receiving study medication and remaining within the specified Cycle 1 dose range. The registry also lists safety outcomes, laboratory abnormalities, vital-sign criteria, and a pharmacokinetic measure. Although the overall study description refers to safety, pharmacokinetics, pharmacodynamics, and maximum tolerated dose assessment, detailed analysis procedures and the results for those aims are not reported in the registry record.

Open-label conduct means masking was not used. The registry record does not report blinding of participants, clinicians, or assessors, and it does not provide a randomized comparator. These features are important context for interpreting safety reporting and any outcomes that involve judgment. They do not invalidate descriptive safety counts, but they limit the basis for causal comparisons across arms.

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 110 subjets were enrolled in this study and 3 of them didn't received any study treatment.

Only the first 6 of 20 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The per protocol analysis set included all enrolled participants who received at least one dose of study medication and who did not receive less than 75% or more than 125% of the planned Cycle 1 dose of either study drug in the combination.

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: All enrolled participants who received at least one dose of study medication.

Only the first 6 of 21 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Dose Limiting Toxicities (DLTs) - Arms A, B and C, measured over Up to 21 days.

The primary endpoint is the number of participants with dose-limiting toxicities in Arms A, B, and C during up to 21 days. A DLT was defined using specified hematologic and non-hematologic adverse events attributable to the combination, along with persistent toxicity or treatment-delivery conditions. The definition includes such events as severe neutropenia or thrombocytopenia, specified pneumonitis or other toxicities, inability to deliver planned treatment because of persistent intolerable toxicity, delays to the next cycle, and persistent QTc prolongation in an asymptomatic participant after correction of reversible causes.

Counting participants who meet a prespecified DLT definition is useful in early-phase dose assessment because the outcome focuses on clinically consequential toxicities within an explicitly defined observation window. It also gives a consistent basis for describing tolerability across candidate dose groups. The stated per-protocol population excludes participants who received no study medication and those whose Cycle 1 dose of either combination drug fell outside the specified range, so the endpoint is not necessarily a count among every enrolled participant.

The registry record does not report the DLT counts, the number analyzed in each arm for this endpoint, or a dose-selection result. Consequently, the definition and intended assessment can be described, but the frequency of DLTs and the evidence supporting any maximum tolerated dose cannot be evaluated from the outcome information recorded.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a participant count: it records how many people met the study’s definition of a DLT, rather than measuring a continuous quantity. Such outcomes are often called binary at the participant level because each participant can be classified as having or not having the event during the defined period. A count can be informative for early dose evaluation, but a count alone is easiest to interpret alongside the number of participants assessed. The registry information does not report primary endpoint counts or arm-specific denominators, so neither an event frequency nor its precision can be described.

What does the DLT definition do?

A prespecified definition makes the safety outcome more reproducible by setting out which adverse events count and how treatment-delivery consequences are handled. In this study, the definition covers hematologic and non-hematologic events attributable to the combination, as well as certain persistent toxicities, dose-delivery failures, cycle delays, and QTc prolongation. It is more specific than simply counting every adverse event. The assessment period is up to 21 days, so the endpoint concerns DLTs in that window rather than all toxicity over the full treatment course.

Attribution to the combination is part of the DLT definition. In an open-label study, the attribution process can involve clinical judgment, and masking is not used to reduce the possibility that knowledge of treatment influences assessment. The registry does not report adjudication procedures or further detail about how attribution was determined. The DLT definition also does not replace broader safety reporting: an event may be important to safety even if it does not meet the protocol’s DLT criteria.

Why does the analysis population matter?

The per-protocol population for the primary endpoint includes enrolled participants who received at least one dose and did not receive less than 75% or more than 125% of the planned Cycle 1 dose of either study drug in the combination. This restriction focuses the analysis on participants who received treatment within the specified dose range. It also means the endpoint population can differ from the overall enrolled population. If people are excluded because of dose deviations or lack of treatment, the resulting count describes the defined per-protocol set, not automatically everyone enrolled.

The registry record does not report the number included in that population for each arm, the reasons for any exclusions, or how missing endpoint assessments were handled. Without those details and the actual DLT counts, it is not possible to assess how much the analysis population differs from enrollment or whether incomplete follow-up could affect the findings.

How should uncertainty be understood?

A confidence interval is a general statistical tool for expressing the precision of an estimate. For an event frequency, it communicates a range of values reasonably compatible with the observed data under the method’s assumptions. When only a small group is assessed, estimates can be imprecise: a limited number of events may correspond to a wide range of plausible underlying frequencies. This is a general consideration for early-phase dose groups, not a reported finding for this study. The registry information does not give confidence intervals for the DLT outcome.

For participant counts, a useful presentation usually includes both the number with an event and the number assessed, sometimes with an event proportion and an interval. The count and denominator answer different questions: the count gives the observed number, while the denominator establishes how many people could have had the event. No primary DLT results are reported, so those quantities cannot be evaluated or reconstructed. It would be inappropriate to calculate a result from the enrollment figure, because enrollment is not the same as the endpoint-specific analysis population.

What does the non-randomized sequence imply?

Randomization is a general design method that helps make groups comparable, on average, with respect to both measured and unmeasured participant characteristics. In this study, allocation was non-randomized and the model was sequential. The groups therefore cannot be treated as randomized controls for one another. Dose groups may also have been evaluated at different points in the study, and the registry does not report the exact sequencing rules or decision criteria.

Because the combinations and dose levels differ, cross-group comparisons mix treatment differences with potential differences in the people treated, study timing, or eligibility. Descriptive summaries remain useful for documenting observed safety in each group, but a difference in counts does not by itself show that one regimen caused more or fewer events. This caution applies even when groups appear similar in their labels or share one component of treatment.

General methods for comparing event counts can account for denominators and uncertainty, but a statistical test cannot fix the lack of random allocation or supply information that was not collected or reported. No primary-outcome analysis method, confidence interval procedure, hypothesis test, adjustment model, or missing-data method is reported in the registry record. The analysis plan should not be inferred from the endpoint name or from the study’s phase.

What can sample-size planning show?

Planning calculations can illustrate the size of a future study under stated assumptions about the effect to detect, significance level, allocation, and statistical approximation. Such calculations are not observed results and do not establish the probability that this completed study detected an effect. The study’s own sample-size justification is not reported in the registry record. Any planning illustrations associated with observed effects should be interpreted as hypothetical planning exercises, not as the trial’s design rationale or as observed power.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 69 outcome measures and 0 analyses. The outcome list includes safety events, laboratory abnormalities, vital-sign criteria, and a pharmacokinetic measure, alongside the primary DLT outcome. The information available does not report a multiplicity strategy, such as a prespecified hierarchy or adjustment for testing multiple outcomes. The protocol’s statistical analysis plan and any interim-analysis plan are not reported in the registry record.

Multiplicity matters when many outcomes or comparisons are examined: the more results considered, the greater the chance that at least one appears noteworthy by chance alone. This is a general statistical concern, not evidence that a particular result in this study is false. In the absence of reported analyses and a stated testing strategy, the registry record does not support conclusions about how results were selected or how statistical error was controlled.

Missing-data handling is also not reported in the registry record. The primary endpoint’s per-protocol definition specifies treatment exposure and Cycle 1 dose limits, but does not say how unavailable DLT assessments, early discontinuations, or other incomplete observations were handled. The record does not provide endpoint-specific analysis counts or reasons for exclusion. Enrollment therefore cannot be used as a substitute for the primary endpoint denominator.

The analysis population is specifically described for the primary DLT outcome. The record does not provide comparable population definitions for every listed secondary outcome, nor does it report whether analyses used additional populations. These gaps limit assessment of selection, completeness, and comparability across outcomes. They also mean no missingness mechanism or impact can be inferred from the registry information.

11. 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: nausea (50.0%), stomatitis (50.0%), mucosal inflammation (50.0%), device related infection (50.0%), dermatitis exfoliative generalised (50.0%).

The serious adverse-event information is presented by named treatment group, with a number at risk and a number affected for each group. These are descriptive counts. They should be interpreted within each group, not pooled into a single trial-wide rate, and should not be treated as randomized comparisons because the study was non-randomized and sequential. The registry record does not provide an exposure-adjusted analysis or a common comparator group.

The serious-event terms listed with percentages are nausea, stomatitis, mucosal inflammation, device related infection, dermatitis exfoliative generalised, and febrile neutropenia. The reported percentages are 50.0% for each of the first five terms and 40.0% for febrile neutropenia. The record information does not state the denominator associated with these term percentages, so they should be quoted as reported rather than converted into counts or compared with group-level figures.

Serious adverse events are not identical to DLTs. A DLT is determined by the specific protocol definition and assessment window; a serious adverse event is a separate safety classification. The registry record also lists deaths by treatment group, but the information presented does not establish that deaths were caused by treatment. Seriousness, causality, severity, and protocol-defined dose-limiting status are distinct concepts and should not be conflated.

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

The record establishes that this completed study evaluated PF-05212384 with docetaxel, cisplatin, or dacomitinib and included a TNBC cisplatin expansion. It identifies an early safety outcome—participants with DLTs during up to 21 days—and describes the toxicity criteria and per-protocol population. It also records serious adverse-event counts by group and percentages for selected serious-event terms. These details characterize the study’s questions and the information recorded about safety.

However, the primary DLT results are not reported in the registry record. Without those results, readers cannot determine the observed DLT frequency in any arm, whether it varied across dose groups, or whether a maximum tolerated dose was identified. The listed serious adverse-event figures do not fill that gap because they are a separate outcome and do not establish DLT status or treatment causation.

The trial does not provide a randomized treatment comparison. Its groups differ in combination, dose, or treatment-line description, so their outcomes cannot be interpreted as direct evidence that one regimen is better or safer than another. No conclusion about efficacy in TNBC can be drawn from the information given, because anti-tumor activity results are not reported in the record information.

These limits are about what can be concluded from the registry record, not proof that analyses or results do not exist elsewhere. The registry does not report a statistical analysis plan, endpoint-specific missing-data approach, or complete analysis populations for all outcomes. A reader should therefore distinguish documented design and safety definitions from unreported outcome estimates. In particular, the study’s completed status and recorded enrollment do not by themselves establish what the primary analysis found.

In short, the information supports a description of an open-label, non-randomized, early-phase combination study and its intended safety assessment. It does not support a quantitative conclusion about DLT risk, a causal comparison among regimens, or a claim of anti-tumor benefit.

13. Why This Trial Matters Statistically

This trial is statistically important as an example of how early-phase combination studies organize safety questions across multiple drugs and dose groups. The study’s primary endpoint uses a protocol-defined DLT count over a stated assessment window, while its wider outcome list includes treatment-emergent adverse events, laboratory abnormalities, vital signs, and pharmacokinetic measurement. These outcomes address different aspects of tolerability and drug exposure, so they should not be treated as interchangeable measures.

The study also illustrates why design determines the scope of interpretation. Sequential, non-randomized dose groups can provide descriptive safety information at the studied regimens, but they do not create a randomized comparison. Differences between groups may reflect drug combination, dose, treatment line, participant composition, or study timing. Statistical summaries can describe observed data, but cannot remove these sources of confounding by themselves.

Another important issue is the definition of the analysis population. The primary endpoint uses a per-protocol set that depends on receiving treatment and staying within a specified Cycle 1 dose range. That definition is relevant to the question being asked, but endpoint interpretation also depends on knowing how many people were analyzed and how excluded or incomplete observations were handled. Those details are not reported in the registry record.

Finally, a detailed outcome definition is not a substitute for outcome results. The registry gives substantial information about what counted as a DLT and records serious adverse events by group, yet it does not report the primary DLT counts or analyses. This distinction matters in evidence synthesis: design and endpoint information help explain what a trial intended to measure, while estimates and their uncertainty are needed to characterize what it observed.

14. 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 registry information has several trial-specific limitations for statistical interpretation. Although the study is marked completed and an enrollment of 110 is recorded, the primary DLT counts and endpoint-specific denominators are not reported. This prevents estimation of the primary event frequency and its precision. Enrollment cannot stand in for the per-protocol analysis population, which has its own eligibility conditions.

The study is open-label, non-randomized, and sequential. Its arms include different combinations and dose levels, while the TNBC expansion groups are distinguished by metastatic treatment line. These features make cross-arm comparisons vulnerable to confounding and prevent interpreting observed differences as randomized treatment effects. The registry does not report enough detail to assess the exact dose-escalation decisions or group comparability.

The record does not report the statistical analysis plan, multiplicity approach, interim-analysis procedures, missing-data method, or endpoint-specific analysis populations beyond the primary endpoint definition. It also does not report DLT results or anti-tumor activity results for the TNBC expansion. Serious adverse-event term percentages are listed without a denominator in the information available, limiting interpretation of those percentages.

Finally, the serious-event counts do not identify causality, and serious adverse events should not be assumed to meet DLT criteria. The registry record does not provide additional event-level detail on timing, exposure, recurrence, or overlap. These limitations restrict conclusions to the study design and reported safety descriptors; they do not establish that unreported results were not collected or analyzed.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study Of PF-05212384 In Combination With Other Anti-Tumor Agents and in Combination With Cisplatin in Patients With Triple Negative Breast Cancer in an Expansion Arm (TNBC) — NCT01920061.
• Registry record last updated 2022-09-13; status last verified 2022-08.
• Results first posted 2021-10-07.
• 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.

18. Record Summary

NCT01920061 is a completed phase 1b, open-label, multicenter study of PF-05212384 in combination with docetaxel, cisplatin, or dacomitinib. The registry identifies non-randomized allocation and a sequential model. The recorded enrollment is 110. The primary outcome is the number of participants with DLTs in Arms A, B, and C during up to 21 days, using a per-protocol analysis population defined by treatment receipt and Cycle 1 dosing.

The study description includes a cisplatin expansion in TNBC with two metastatic treatment-line groups. The record also lists safety, laboratory, vital-sign, and pharmacokinetic outcomes. Serious adverse-event counts are recorded by group, with selected event terms reported as percentages. The primary DLT results, anti-tumor activity results, and statistical analysis details are not reported in the registry record. Accordingly, the available information supports describing the design and outcome definitions, but not a quantitative conclusion about the primary endpoint or comparative treatment effects.