Tucatinib, Palbociclib and Letrozole in Metastatic Hormone Receptor Positive and HER2-positive Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Tucatinib, Palbociclib and Letrozole in Metastatic Hormone Receptor Positive and HER2-positive Breast Cancer (NCT03054363), a randomized phase 1, phase 2 study. The primary outcome is Phase Ib Primary Outcome: Proportion of Patients Who Experienced DLTs Attributable to Palbociclib, Tucatinib, or Both.

2. Clinical Question

3. Trial Design

This is a multicenter, open-label study conducted across phase Ib and phase II. The registry record describes a phase Ib run-in followed by a phase II rollover, with the combination intended for people with metastatic hormone receptor-positive and HER2-positive breast cancer. The study is marked completed. The listed enrollment is 42; the safety summaries separately identify 20 phase Ib participants and 22 phase II participants.

The intervention combines tucatinib, palbociclib and letrozole. The phase Ib regimen lists tucatinib 300 mg twice daily, palbociclib 125 mg daily on a 21-days-on, 7-days-off schedule, and letrozole 2.5 mg daily on a 28-day cycle. The phase II regimen lists tucatinib 300 mg twice daily, palbociclib 75 mg daily on the same on-off schedule, and letrozole 2.5 mg daily. Thus, the phase descriptions reflect different palbociclib doses, not a comparison of the combination against a separate control therapy.

There is an important design-label tension in the record: allocation is listed as randomized, while the model is single group and masking is listed as none. No randomized comparator arm is described. The posted comparison instead contrasts starting-dose phase Ib patients with dose-reduced phase II patients. Because those groups belong to different study phases and have different listed palbociclib doses, the contrast should not be read as a randomized estimate of treatment effect. The registry record does not report further details explaining the allocation label or the relationship between the phase groups.

The study dates are listed as starting on 2017-11-27 and reaching primary completion on 2023-01-17. The registry gives no further account of changes to the study over time. Its stated primary endpoint is a safety outcome in phase Ib; phase II secondary outcomes include clinical benefit rate, overall response rate and median duration of response, as well as palbociclib exposure measures.

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

Analysis population for the primary outcome, as defined in the registry: Among the 20 phase Ib patients, 1 (5%) had a dose reduction of tucatinib, and 9 (45%) had a dose reduction of palbociclib for DLTs. Two patients (10%) discontinued palbociclib because of DLTs (neutropenia) and continued tucatinib and letrozole. One patient (5%) discontinued letrozole for toxicity, and continued tucatinib and palbociclib. The pre-defined safety boundaries were not crossed.

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 Ib Primary Outcome: Proportion of Patients Who Experienced DLTs Attributable to Palbociclib, Tucatinib, or Both, measured over 15 months (From date of first consent until final safety analysis of Phase Ib).

The primary endpoint is the proportion of phase Ib participants who experienced dose-limiting toxicities potentially attributable to palbociclib, tucatinib, or both. A dose-limiting toxicity, or DLT, is a prespecified safety event used in early-phase studies to assess whether a regimen can be given at the planned dose. The registry definition includes any grade at least 3 nonhematologic adverse event, grade 3 neutropenia with fever, grade 3 thrombocytopenia with bleeding, or hematologic toxicity of grade at least 4, using NCI CTCAE version 4.03.

For this study, the endpoint was intended to inform safety and tolerability of the three-drug combination and to be interpreted against predefined safety thresholds. The registry states that doses of tucatinib, palbociclib, or both would be decreased if toxicity crossed those boundaries, while letrozole dose remained constant. This makes the endpoint relevant to dose selection and safety monitoring during the phase Ib run-in.

The posted endpoint percentages require caution. The starting-dose phase Ib arm is listed as 100.0%, and the overall cohort is also listed as 100.0%; the phase II dose-reduced arm is listed as 65.0%. Yet the comparison calculation uses 20 of 20 for phase Ib and 8 of 13 for phase II. The record does not explain why the phase II percentage differs from the proportion represented by those counts. Nor does it clarify how the arm-level 100.0% values relate to the narrative about individual dose reductions and discontinuations. Therefore, the outcome's definition is clear, but the precise meaning of every posted arm-level value is not.

7. Results Posted in the Registry

• Dose-reduced Patients for Phase II (Tucatinib 300mg, Palbociclib 75mg, Letrozole 2.5mg): 61.5% (8/13) versus 100.0% (20/20) on control; risk difference -38.5 percentage points (95% CI -64.9 to -12.0), risk ratio 0.62 (0.40–0.95); unadjusted two-proportion z-test p = 0.003.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a proportion measure? The primary endpoint is expressed as a proportion of participants with an event. In general, a proportion divides the number experiencing the specified outcome by the number assessed for that outcome. Here, the intended event is a DLT under the registry's clinical definition. It is important to keep this endpoint separate from dose reductions, treatment discontinuations, and serious adverse events: these may be related safety observations, but they are not automatically the same outcome.

What is the comparison estimating? The posted comparison is an unadjusted risk difference between the phase Ib starting-dose group and the phase II dose-reduced group. A risk difference subtracts one group's event proportion from the other's. The reported estimate is -0.385, and its confidence interval is -0.649 to -0.120. The negative direction corresponds to a lower event proportion in the phase II group under the counts used in that calculation. This comparison is based on 20 of 20 and 8 of 13, rather than directly on the posted arm-level percentages of 100.0% and 65.0%.

The comparison also reports p = 0.003. In general, a p-value measures how incompatible the observed data are with a specified null hypothesis under the analysis assumptions; it is not the probability that the null hypothesis is true. Nor does a small p-value establish that groups were comparable or that a dose change caused an outcome difference. In this case, the registry describes phase groups, not a separate randomized control arm. The p-value cannot account for unreported differences between phases or explain the mismatch in outcome reporting.

Why does study structure matter? The record labels allocation as randomized but lists a single-group model, no masking, and no distinct comparator treatment. The apparent comparison is across phase Ib and phase II, which differ in their listed palbociclib doses. A randomized contrast would generally support a different interpretation from a comparison of groups assembled in different phases. Because the record does not clarify the allocation label, the comparison should be treated as descriptive and unadjusted rather than as a causal treatment estimate.

What is a confidence interval? In general, a confidence interval describes statistical uncertainty around an estimate under the assumptions of the method. A narrower interval indicates greater precision, while a wider interval indicates less precision. It does not account automatically for selection bias, inconsistent reporting, or differences between study phases. The reported interval for the risk difference lies entirely on the negative side, but the underlying groups and endpoint reporting limit what that numerical precision can establish.

Why can small groups be difficult to interpret? With relatively small group counts, each participant's outcome can have a noticeable influence on a proportion. General statistical reasoning also cautions that uncertainty in a proportion depends on both the event count and the denominator. The phase Ib comparison count is 20 of 20, while the phase II comparison count is 8 of 13. These comparison counts do not match the phase II arm-level value exactly, so even a correctly calculated interval around the comparison cannot resolve which reported outcome representation should be interpreted as primary.

What do the sample-size illustrations mean? The planning illustrations are based on the observed effect, not on the trial's original design. Their 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 defined as a stated fraction of the observed effect. Such calculations are planning illustrations, not observed power and not proof that the original study had a particular design rationale. The registry record does not report its sample-size justification. The illustrations also do not turn the phase groups into randomized arms.

How should the phase Ib narrative be used? The narrative reports tucatinib and palbociclib dose reductions for DLTs, discontinuation of palbociclib for neutropenia, and a letrozole discontinuation for toxicity. These details can describe treatment changes, but they cannot be substituted for the primary endpoint proportion without knowing the endpoint's actual participant-level classification and the analysis denominator. The registry record does not report a participant-level dataset or a complete reconciliation of those events with the arm-level values.

Overall, the statistical output provides an unadjusted contrast with a confidence interval and p-value, while the registry's study structure and reporting leave important interpretive questions open. The estimate is a summary of the counts used in the calculation; it does not remove uncertainty about endpoint coding, phase differences, or causal interpretation.

10. Confidence Intervals and Precision

The recalculated risk difference is reported as -0.385, with a confidence interval from -0.649 to -0.120. The interval lies entirely on the negative side of zero, so the unadjusted calculation is consistent with a lower event proportion in the phase II group than in the phase Ib group, using the counts entered for that comparison. The reported p-value is 0.003. These measures describe statistical uncertainty under the calculation's assumptions; they do not address whether the groups are comparable or whether the outcome values were entered consistently.

The interval is broad enough to allow a range of potentially different magnitudes for the contrast. Its endpoints are far apart relative to the point estimate, so the estimate should not be treated as highly precise. General statistical interpretation: a confidence interval reflects sampling uncertainty under a statistical model, while real-world uncertainty can also arise from study design, outcome classification, missing observations, or data inconsistencies. Those sources are not automatically incorporated into the interval.

Interpretation is further limited by the phase II reporting difference: the arm-level percentage is 65.0%, while the comparison calculation uses 8 of 13. The registry record does not explain the difference. The phase Ib comparison uses 20 of 20, and the arm-level percentage is 100.0%. Although the interval may look numerically specific, it is attached to the comparison counts, not to a reconciliation of all posted outcome entries.

The interval also concerns groups from different study phases, with different listed palbociclib doses, rather than two randomized arms. Its precision cannot compensate for that structural limitation. It quantifies uncertainty in the unadjusted contrast represented by the calculation, not uncertainty about a causal effect of dose or the combination relative to another therapy.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists five secondary outcomes: pharmacokinetic exposure measures, clinical benefit rate, overall response rate, and median duration of response. The primary endpoint is the phase Ib proportion experiencing DLTs. The record does not report a multiplicity adjustment, a hierarchy for testing outcomes, or a prespecified approach to interpreting multiple comparisons. Without such a plan, any statistical findings across several outcomes may carry additional risk of chance-positive results. The registry record does not report whether multiplicity was addressed in a separate analysis plan.

The record reports no posted analyses in its analysis count. This does not establish that no analyses were conducted; it means the registry record does not report analyses in that listing. No analysis population definitions are reported, such as whether analyses were based on all enrolled participants, participants who received treatment, or participants with evaluable safety data. The denominator for each endpoint should therefore be interpreted from the posted outcome entry rather than assumed to follow a particular population rule.

Missing-data handling is not reported in the registry record. There is no stated method for handling participants without endpoint assessment, incomplete follow-up, or missing outcome classifications. The record also does not describe a sensitivity analysis for missing data. These omissions matter because proportions can change depending on who is included in the denominator and how unobserved outcomes are treated.

The primary outcome entries themselves raise a reporting issue: the phase II arm-level value is 65.0%, whereas the comparison uses 8 of 13. The registry record does not explain whether this reflects different endpoint definitions, analysis populations, or a data-entry issue. The reported phase Ib narrative gives dose-reduction and discontinuation details, but it does not provide a complete participant-level reconciliation with the posted percentages. As a result, multiplicity, missingness, and analysis-population questions cannot be resolved from the registry record alone.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pain (10.0%), sepsis (5.0%), abdominal distension (5.0%), abdominal pain (5.0%), ast increase (5.0%).

The serious adverse event summary separates phase Ib and phase II. In phase Ib, 9 of 20 participants are listed as affected by serious events; in phase II, 7 of 22 are listed. The record lists no deaths in either group. The separate “other” adverse-event figures list 20 of 20 in phase Ib and 22 of 22 in phase II. Serious events and other adverse events are different categories in this summary and should not be combined into one rate.

The listed leading serious-event terms include Pain at 10.0%, with Sepsis, Abdominal Distension, Abdominal Pain, AST Increase, and ALT Increase each at 5.0%. These are registry percentages for the terms, not estimates of comparative treatment risk. The record does not provide a control group for comparison, and the phase groups have different listed regimens.

Adverse-event summaries describe events recorded during the study; they do not by themselves establish that a treatment caused an event. The registry's primary endpoint uses a specific DLT definition, whereas serious adverse events are presented separately. The record does not explain how these categories overlap or how the serious-event term percentages relate to participants with DLTs. The counts should therefore be read as descriptive safety reporting, not as a causal comparison.

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

The registry record documents a phase Ib/II evaluation of a three-drug combination and identifies DLTs as the phase Ib primary safety endpoint. Its narrative records dose reductions and treatment discontinuations attributed to toxicity, while its serious adverse-event summary reports events in both phase groups and no deaths. These details describe safety observations in the registry; they do not establish a comparative benefit or harm relative to another treatment.

The posted comparison calculation shows a negative unadjusted risk difference, with a confidence interval that does not include zero and p = 0.003. It compares the phase Ib starting-dose group with the phase II dose-reduced group using 20 of 20 and 8 of 13. The phase II arm-level percentage is separately listed as 65.0%, and the registry record does not explain why it differs from the comparison count. That reporting uncertainty limits confidence in a single numerical reading of the primary outcome.

The study design also constrains interpretation. The record gives a randomized allocation label, but describes a single-group model and does not identify a distinct randomized comparator. The displayed comparison is across study phases, with different listed palbociclib doses. Consequently, the observed contrast should not be treated as proof that dose reduction caused a change in DLT risk. Differences in phase, participant selection, follow-up, or other unreported factors could contribute.

Nor does this registry entry establish whether the regimen is effective in treating cancer. Although secondary outcomes include response and clinical benefit measures, the registry record does not report their numerical results in the listed details. It also does not report a statistical analysis plan, missing-data procedures, analysis populations, or the rationale for the study's enrollment target. Those omissions prevent broader claims about efficacy, precision, or generalizability.

The appropriate reading is narrow: the record contains safety summaries and an unadjusted phase-group contrast, but important inconsistencies and design limitations remain. It supports description of what was recorded, not medical advice or a conclusion that the combination works or does not work.

14. Why This Trial Matters Statistically

This trial is statistically instructive because it shows why design labels, endpoint definitions, and numerical summaries must be read together. The record labels allocation as randomized, yet also identifies a single-group model and does not describe a separate control treatment. The comparison is between a phase Ib starting-dose group and a phase II dose-reduced group. Those details matter: a statistical test cannot make a nonrandomized phase comparison equivalent to a randomized treatment contrast.

The endpoint also illustrates the difference between an event definition and related clinical actions. DLTs are defined using adverse-event severity and specific hematologic criteria, while dose reductions and discontinuations are reported as treatment changes. These outcomes are related in the narrative but are not interchangeable. A reader needs a clear denominator and endpoint classification to interpret a proportion.

Here, the phase II arm-level value of 65.0% does not align exactly with the comparison count of 8 of 13. The registry record gives no reconciliation. This makes data consistency a central statistical issue, not a minor detail. A confidence interval and p-value can be calculated precisely from counts, but they cannot resolve ambiguity in the underlying outcome entry.

The example also reinforces that statistical significance and causal evidence are different concepts. The recalculated comparison reports p = 0.003, but interpretation still depends on study structure, group comparability, outcome coding, and missing-data handling. Those limitations remain even when a confidence interval excludes zero.

Finally, the registry lists multiple secondary outcomes but does not report a multiplicity plan or analysis populations. That gap highlights the value of prespecified methods and complete reporting for interpreting results across several outcomes. The record offers enough information to identify important questions, but not enough to answer all of them.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

Several limitations are specific to this record. First, the design information is internally difficult to reconcile: allocation is labeled randomized, but the model is single group, masking is none, and no distinct comparator treatment is described. The displayed comparison is between phase Ib and phase II groups rather than clearly randomized arms. The registry record does not explain the apparent allocation inconsistency.

Second, outcome reporting is not fully aligned. The phase II arm-level entry is 65.0%, while the comparison calculation uses 8 of 13. The record does not say whether these entries reflect different analysis populations, endpoint definitions, or a reporting error. The phase Ib narrative gives details of dose reductions and discontinuations, but does not reconcile those events with the listed 100.0% primary outcome value. Its narrative also ends before the account is complete.

Third, group membership is tied to study phase and listed palbociclib dose. The comparison therefore cannot isolate the effect of dose from differences associated with phase or participant selection. No adjustment strategy is reported. The record also does not report analysis populations, missing-data methods, a multiplicity plan, or a sample-size justification.

Finally, although secondary outcomes are named, their numerical results are not included in the listed record details. Safety events are descriptive and have no separate control group for causal comparison. These limitations mean that the posted statistics should be interpreted narrowly as registry summaries, with the outcome mismatch and phase-based comparison kept in view.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Tucatinib, Palbociclib and Letrozole in Metastatic Hormone Receptor Positive and HER2-positive Breast Cancer — NCT03054363.
• Registry record last updated 2023-11-13; status last verified 2023-11.
• Results first posted 2023-10-27.
• 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 completed, open-label phase Ib/II study of tucatinib, palbociclib and letrozole for metastatic hormone receptor-positive and HER2-positive breast cancer. Enrollment is listed as 42. The primary phase Ib outcome is the proportion of participants experiencing DLTs potentially attributable to palbociclib, tucatinib, or both. The record lists a 100.0% phase Ib arm value and a 65.0% phase II arm value, but its comparison calculation uses 20 of 20 and 8 of 13. The reason for this mismatch is not reported in the registry record.

The comparison is between phase groups with different listed palbociclib doses, and no separate control treatment is described. The record reports a recalculated risk difference of -0.385, confidence interval -0.649 to -0.120, and p = 0.003. These results are descriptive of the comparison calculation and do not establish a causal dose effect. The registry record does not report the statistical plan, analysis populations, missing-data methods, or sample-size justification.