Neratinib and Paclitaxel With or Without Pertuzumab and Trastuzumab Before Combination Chemotherapy in Treating Patients With Metastatic or Locally Advanced Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Neratinib and Paclitaxel With or Without Pertuzumab and Trastuzumab Before Combination Chemotherapy in Treating Patients With Metastatic or Locally Advanced Breast Cancer (NCT03101748), a non-randomized phase 1, phase 2 study. The primary outcome is Dose-Limiting Toxicities (DLTs) During Dose Escalation.

2. Clinical Question

3. Trial Design

NCT03101748 is registered as a Phase 1b and Phase II study sponsored by M.D. Anderson Cancer Center. The registry lists a non-randomized allocation, no masking, and a parallel model. Enrollment is recorded as 34. The study began on 2018-02-19, its primary completion date is 2025-05-13, and its status is terminated.

The described groups cover dose escalation and later treatment cohorts. Group A is Cohort 1 Phase Ib, and its description includes neratinib, paclitaxel, pertuzumab, and trastuzumab. Group B is Cohort 1 Phase II and is described as receiving those agents followed by doxorubicin and cyclophosphamide. Group C is Cohort 2 and is described as receiving neratinib and paclitaxel, followed by surgery and further treatment. The registry record does not report the complete treatment sequence for every group.

The primary endpoint is limited to Cohort I Phase Ib and concerns dose-limiting toxicities during the first two cycles. The adverse-event summaries also include Cohort 1 Phase Ib dose levels 0 and 1, Cohort 1 Phase II, and Cohort 2. No participants were enrolled on Cohort 1 Phase Ib dose levels -1, 2, or 3, according to the primary endpoint description.

Because allocation was not randomized, differences between groups may reflect differences in the participants, treatment stage, or other factors as well as treatment. The record does not report a randomized comparison or an analysis that adjusts for such differences. It also reports no outcome analyses. Consequently, the registry describes the design and some safety counts, but does not provide an analyzed primary efficacy or toxicity result.

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: 43 patients were accrued and assessed for eligibility and 34 patients were assigned to the cohort.

Only the first 6 of 7 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: No participants enrolled on Cohort 1 Phase Ib Dose Level -1, Cohort 1 Phase Ib Dose Level 2, or Cohort 1 Phase Ib Dose Level 3

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: No participants enrolled on Cohort 1 Phase Ib Dose Level -1, Cohort 1 Phase Ib Dose Level 2, or Cohort 1 Phase Ib Dose Level 3

Only the first 6 of 8 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Dose-Limiting Toxicities (DLTs) During Dose Escalation, measured over From treatment initiation through completion of 2 cycles (approximately 6 weeks).

The primary endpoint is the count of participants with at least one dose-limiting toxicity (DLT) during dose escalation. The stated observation period runs from treatment initiation through completion of the first two cycles, approximately 6 weeks. It applies to participants receiving escalating doses of neratinib in combination therapy in Cohort I Phase Ib.

A participant-level count is useful in dose escalation because it records whether a participant experienced a toxicity meeting the study’s DLT definition during the specified observation period. Such information can contribute to decisions about dose levels in early-phase research. That general purpose does not establish which dose is appropriate in this study; the registry record does not report the DLT definition in detail, dose-level outcome counts, or a dose-selection result.

The endpoint is a count, not a toxicity rate over person-time and not a measure of tumor response. The number of participants at risk at each dose level is important for interpreting a count, because a count without its denominator does not indicate how common the event was among participants assessed. The primary endpoint description states that no participants enrolled at dose levels -1, 2, or 3. It does not give the DLT count for the enrolled dose levels. No primary endpoint analysis is reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a participant count represent?

General statistical explanation: a participant count records how many people meet an outcome definition. For this endpoint, the definition is having at least one DLT during the stated period. A count is easier to interpret when accompanied by the number of participants assessed. The denominator clarifies the scale of the count, while the observation window clarifies when events qualify. Here, the endpoint description gives the window as the first two cycles, approximately 6 weeks, but does not report the primary endpoint counts or denominators by enrolled dose level.

A count is not itself a percentage or a rate. A percentage would express the count relative to a denominator, and a rate would also require a defined time-at-risk approach. No such primary endpoint calculation is reported. The adverse-event summaries include counts at risk and counts affected, but they should be read as the registry’s distinct adverse-event summaries, not as a replacement for the DLT endpoint analysis.

Why does the study design matter?

General statistical explanation: random assignment can help balance both measured and unmeasured participant characteristics across treatment groups, on average. In a non-randomized study, treatment groups may differ for reasons other than treatment. A numerical difference between groups therefore does not by itself estimate a causal treatment effect. This study is listed as non-randomized, unmasked, and parallel. The registry explicitly cautions that its arms are not a like-for-like control comparison.

These design features affect the interpretation of any comparison, but they do not mean that descriptive results have no value. Counts can describe what was recorded within a cohort. They cannot, without an appropriate design and analysis, isolate the effect of a regimen from the other factors that distinguish cohorts. The registry record does not report an adjusted comparative analysis.

How should precision be understood?

General statistical explanation: an estimate is more informative when its uncertainty is reported. A confidence interval is one way to describe the range of values reasonably compatible with the observed data under a specified statistical model. Wider intervals generally indicate less precision. For a participant-level binary outcome such as at least one DLT, uncertainty depends in part on the number assessed and the number experiencing the event. The registry record does not report a confidence interval or other precision estimate for the primary endpoint.

The adverse-event groups have relatively small listed denominators, so their counts should be read as descriptive observations rather than precise comparative estimates. This does not justify calculating an unreported interval or treating the groups as exchangeable. No recalculated primary endpoint result is available in the registry record.

What do the planning illustrations mean?

General statistical explanation: sample-size planning calculations describe how many participants or events a future study might require under stated assumptions. They are not observed results and do not measure the power of a completed study. The planning assumptions recorded for this study specify a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They also define the target effect as a stated fraction of the observed effect. For time-to-event outcomes, the assumptions describe total events; for other outcomes, they describe participants per arm.

The registry record does not report the numerical planning counts or an observed effect estimate for these illustrations. They therefore cannot be interpreted as an enrollment target established by the study team or as evidence that the trial had a particular probability of detecting an effect. The registry record does not report the trial’s own sample-size justification.

What analysis information is missing?

The record reports no outcome analyses. It does not report the statistical model for the primary endpoint, any adjustment variables, the handling of missing endpoint observations, or an analysis population. Those omissions limit assessment of how the endpoint was intended to be summarized or how incomplete follow-up would affect the result. General statistical methods cannot fill in those study-specific details.

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("Cohort 1 Phase Ib Dose Level 0", "Cohort 1 Phase Ib Dose Level 1")
n      <- c(2, 2)
events <- c(0, 2)
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  Cohort 1 Phase Ib Dose Level 0  2  0  0.0  0.0  84.2  0.0  65.8  0.8  70.8
ARM  Cohort 1 Phase Ib Dose Level 1  2  2  100.0  15.8  100.0  34.2  100.0  29.2  99.2

11. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures and identifies Event-Free Survival (EFS) Rate at 2 Years as a secondary outcome. It does not report outcome analyses. The statistical plan for handling multiple outcomes or repeated statistical tests is not reported in the registry record. Without that information, it is not possible to determine whether any reported inferential results would have been adjusted for multiplicity, or how the primary and secondary outcomes were prioritized analytically.

The record does not report missing-data counts, reasons for missing observations, or methods for handling missing endpoint information. It also does not describe an analysis population for the primary endpoint or secondary outcome. These details matter because exclusions and incomplete follow-up can change which participants contribute to an estimate and how representative that estimate is of those enrolled.

The primary endpoint description specifies an observation window through completion of two cycles, approximately 6 weeks, but does not state how participants who stopped treatment early or lacked complete observation were handled in analysis. No method for censored or incomplete EFS observations is reported. The registry record also does not report interim analyses, stopping rules, or a statistical analysis plan. These items should be treated as unreported rather than inferred from the Phase 1b/Phase II design or the study’s terminated status.

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: fever (50.0%), febrile neutropenia (50.0%), urinary tract infection (50.0%), nausea (50.0%), diarrhea (50.0%).

The adverse-event summaries list, for each group, the number at risk and the number affected by serious events and by other events, along with deaths. Cohort 1 Phase Ib Dose Level 0 lists 2 at risk and 1 with serious adverse events; Cohort 1 Phase Ib Dose Level 1 also lists 2 at risk and 1 with serious adverse events. Cohort 1 Phase II lists 14 at risk and 6 with serious adverse events. Cohort 2 lists 16 at risk and 3 with serious adverse events.

The corresponding death counts are 1 for Dose Level 0, 0 for Dose Level 1, 2 for Cohort 1 Phase II, and 5 for Cohort 2. The listed other-event counts are 2, 2, 14, and 16 for those groups, respectively. These are group-level descriptive counts and should not be read as comparisons of treatment safety, especially because allocation was non-randomized and the groups differ in cohort and treatment context.

The serious-event terms listed most often include Fever, Febrile neutropenia, Urinary tract Infection, Nausea, Diarrhea, and Vomiting, each shown as 50.0%. The registry record does not clarify the denominator underlying those term percentages in the information. The adverse-event summaries do not replace the primary DLT endpoint, and the registry record does not report a statistical comparison of these safety figures.

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

The record describes a clinical study and reports some descriptive safety counts, but it does not report analyses of the primary endpoint or the listed secondary outcome. The primary question concerns DLTs during the first two treatment cycles in Cohort I Phase Ib. Because the DLT outcome counts and their denominators are not reported, the primary result cannot be quantified from the registry record. No estimate of the primary endpoint’s precision is reported either.

The serious adverse-event counts and death counts describe what is recorded for the named groups. They do not establish that one regimen caused more or fewer events than another. The study was non-randomized, and the registry cautions that the arms are not a like-for-like control comparison. The counts also do not establish comparative effectiveness or show whether a treatment changed EFS.

The registry status is terminated, but status alone does not explain why the study ended or what conclusions the investigators drew. The reason for termination is not reported in the registry record. Nor does the listed overall enrollment substitute for the number assessed for a particular endpoint. The endpoint-specific analysis population and missing-data handling are not reported.

In general, trial results are most interpretable when the outcome definition, denominator, analysis method, and uncertainty are all available. For this record, the primary endpoint definition and time window are given, but the primary result and analysis details are not. The evidence presented in the registry therefore supports a description of the study design and certain adverse-event counts, not a firm conclusion about efficacy, comparative safety, or the dose-escalation outcome.

14. Why This Trial Matters Statistically

This study is statistically notable because it combines a dose-escalation question with later treatment cohorts, while using a non-randomized parallel design. Those features make it important to distinguish a descriptive cohort summary from a comparative treatment estimate. The groups have different labels and treatment sequences, and the registry does not identify them as randomized controls.

The primary endpoint is a participant-level safety outcome within a defined early treatment period. In early-phase research, such outcomes can inform evaluation of treatment tolerability and dose escalation. To interpret them, readers need the number of participants assessed at each relevant dose level, the number meeting the endpoint, and a clear account of the endpoint definition. The registry gives the observation period and notes dose levels without enrollment, but does not report the primary outcome counts.

The record also illustrates why registry status and enrollment figures cannot stand in for analysis. Enrollment is recorded as 34, and the study is marked terminated, yet the registry reports no outcome analyses. Without the endpoint results and uncertainty estimates, readers cannot assess the precision of the primary safety outcome or draw efficacy conclusions from the listed secondary outcome.

Statistically, the most useful reading is therefore bounded: the record documents a study structure, treatment groups, a defined DLT question, and descriptive adverse-event information. It also makes clear that non-randomized group comparisons are not like-for-like controls. That distinction helps prevent descriptive counts from being treated as proof of a causal difference.

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.

A key limitation is that the primary endpoint is named and defined, but its analyzed results are not reported in the registry record. No DLT counts or endpoint-specific denominators are given for the enrolled dose levels. The record notes that no participants enrolled on Cohort 1 Phase Ib dose levels -1, 2, or 3, but does not give the endpoint outcome for the other dose levels.

The study is non-randomized, and its groups are not a like-for-like control comparison. Any apparent differences in adverse-event counts across groups may reflect cohort composition, treatment sequence, or other factors, and cannot be attributed to treatment on the basis of these summaries. No adjusted comparison is reported.

Further study-specific details are not reported in the registry record: the full DLT definition, analysis populations, missing-data approach, statistical analysis plan, interim analyses, and the reason for termination. The registry lists EFS Rate at 2 Years as a secondary outcome but does not report an analysis result for it. It also does not provide confidence intervals or other precision measures for the reported safety counts. The treatment descriptions do not give every detail of the treatment sequence. These limitations restrict interpretation to the design information and descriptive counts that are recorded.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Neratinib and Paclitaxel With or Without Pertuzumab and Trastuzumab Before Combination Chemotherapy in Treating Patients With Metastatic or Locally Advanced Breast Cancer — NCT03101748.
• Registry record last updated 2026-02-12; status last verified 2026-02.
• Results first posted 2026-02-12.
• 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

NCT03101748 is a non-randomized, unmasked Phase 1b/Phase II study sponsored by M.D. Anderson Cancer Center. It concerns neratinib-based treatment, including combinations with paclitaxel and, in some groups, pertuzumab and trastuzumab, followed in some cohorts by doxorubicin and cyclophosphamide. The record lists enrollment of 34 and status as terminated.

The primary endpoint is the count of participants with at least one DLT during the first two cycles, approximately 6 weeks, in Cohort I Phase Ib. No primary endpoint analysis is reported. The record includes group-level adverse-event counts and lists EFS Rate at 2 Years as a secondary outcome, but reports no outcome analyses. Because the study is non-randomized, the groups do not form a like-for-like control comparison.