This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.
1. Trial at a Glance
PALTAN was a phase 2, single-group, unmasked treatment study of neoadjuvant palbociclib in combination with letrozole and trastuzumab in stage II-III ER+ HER2+ breast cancer. The registry records an enrollment of 26 participants, one study arm, and a primary endpoint of pathologic complete response after completion of 4 cycles of treatment, approximately 16 weeks.
| Feature | PALTAN |
|---|---|
| Study acronym | PALTAN |
| ClinicalTrials.gov identifier | NCT02907918 |
| Phase | Phase 2 |
| Status | Terminated |
| Start date | 2017-06-30 |
| Primary completion date | 2020-08-24 |
| Lead sponsor | Washington University School of Medicine |
| Condition | Breast Cancer; Cancer of Breast; Breast Carcinoma |
| Population described in study title | Stage II-III ER+ HER2+ breast cancer |
| Allocation | NA |
| Design model | Single group |
| Masking | None |
| Primary purpose | Treatment |
| Enrollment | 26 |
| Number of arms | 1 |
2. Clinical Question
The clinical question represented by the registry is whether a neoadjuvant treatment regimen combining palbociclib, letrozole, and trastuzumab can be evaluated in patients with stage II-III ER+ HER2+ breast cancer using pathologic complete response after treatment as the primary endpoint.
Population
Patients with stage II-III ER+ HER2+ breast cancer, as specified in the study title.
Intervention
Palbociclib in combination with letrozole and trastuzumab, with additional procedures and research sampling specified in the intervention record.
Comparator
The study is registered as a single-group design with one arm; there is therefore no randomized comparator arm.
Primary question
How many participants have a pathologic complete response after completion of 4 cycles of treatment, approximately 16 weeks?
3. Trial Design
The registry describes PALTAN as a phase 2, single-group, unmasked treatment study. Allocation is recorded as NA, and the design model is SINGLE_GROUP. With one treatment arm, the principal statistical structure differs substantially from a randomized controlled trial: there is no between-arm treatment effect estimate to calculate.
Treatment and study procedures
The registry lists palbociclib, letrozole, trastuzumab, and goserelin as drug or biologic interventions. It also lists breast surgery, research tumor biopsy, optional research bone marrow, research blood sample, blood for germline DNA, and blood for detection of circulating tumor cells as study procedures or research activities.
Single treatment group
- Palbociclib
- Letrozole
- Trastuzumab
- Goserelin
- Breast surgery
- Research tumor biopsy
Biologic and translational assessments
- Research bone marrow, optional
- Research blood sample
- Blood for germline DNA
- Blood for detection of circulating tumor cells
4. Primary Endpoint
| Endpoint | Registry definition | Time frame |
|---|---|---|
| Number of Participants With Pathologic Complete Response (pCR) | A pathologic complete response (pCR) is defined as no histology evidence of invasive tumor cells in the surgical breast specimen and sentinel or axillary lymph nodes. | Completion of 4 cycles of treatment (approximately 16 weeks) |
The endpoint is a binary pathologic outcome at a defined treatment milestone. Each participant can be classified according to whether the registry's pCR definition is met after the specified treatment period and surgical assessment.
For a single-group study, the central descriptive quantity is the observed proportion of participants meeting the pCR definition. A confidence interval for that proportion can then quantify sampling uncertainty. The appropriate denominator must be defined by the prespecified analysis population and endpoint assessment rules.
5. Statistical Methodology
Single-group binary endpoint analysis
Because PALTAN is registered as a single-group study, the primary endpoint is naturally expressed as a proportion rather than a treatment-versus-control effect estimate. The observed pCR rate describes the fraction of participants satisfying the pathologic definition at the specified assessment.
For a binary endpoint with a relatively small study population, an exact binomial confidence interval is often preferable to relying on a large-sample normal approximation. With 26 enrolled participants, the distinction between an exact and approximate interval can be material, particularly when the observed number of responses is close to zero or close to the total.
Binomial framework
Here, X represents the number of participants achieving pCR, n represents the number of participants contributing to the endpoint analysis, and p represents the underlying probability of pCR under the population represented by the study.
The registry records enrollment of 26 participants, but the information available in the record does not provide a posted statistical analysis identifying the number with pCR, the analyzed denominator, a confidence interval, or a p-value. Those quantities therefore cannot be treated as reported results.
What a confidence interval would contribute
A confidence interval around a single-group response proportion would describe the statistical precision of the observed pCR estimate. This is particularly important in a small phase 2 study, because the observed proportion can change substantially when only a small number of participants contribute to the endpoint.
A confidence interval is not a prediction interval for individual patients. It describes uncertainty around the population parameter under the statistical model and sampling framework used to construct it.
No between-group hazard or risk comparison
PALTAN's single-group structure means that concepts such as a hazard ratio, randomized risk ratio, between-arm risk difference, or stratified log-rank comparison are not the primary statistical tools for this endpoint. Those methods require a comparative structure that is not present in the registered design.
6. Planned Analysis
The registry identifies pCR as the primary endpoint and defines the endpoint as the number of participants with pathologic complete response after completion of 4 cycles of treatment, approximately 16 weeks. For this type of single-group binary endpoint, the usual analysis would report the number and proportion achieving pCR, together with an appropriate confidence interval for the proportion.
Why the distinction matters
A registry can indicate that results have been posted without necessarily supplying a formal inferential analysis for every registered endpoint. For a binary single-group endpoint, the distinction is important because a raw count, an observed proportion, and a hypothesis test answer different statistical questions.
Descriptive question
What proportion of evaluated participants achieved the registered definition of pCR?
Precision question
How much statistical uncertainty surrounds that observed proportion?
Inferential question
Is the observed pCR proportion sufficiently different from a prespecified reference value?
Comparative question
Would the pCR proportion differ between treatment groups? This question is not represented by the registered single-group design.
7. Statistical Methods Explained
Why is pCR analysed as a binary endpoint?
The registry defines pCR categorically: a participant either meets the definition of no histology evidence of invasive tumor cells in the surgical breast specimen and sentinel or axillary lymph nodes, or does not. That structure makes a binomial model a natural statistical framework.
What does a single-group pCR proportion mean?
The pCR proportion is the fraction of participants in the defined analysis population who satisfy the pCR criterion. It describes the observed experience of the treatment group; it is not itself a randomized estimate of comparative treatment efficacy.
Why is the denominator important?
A proportion depends on both the numerator and denominator. Enrollment is 26, but endpoint analysis can require an explicitly defined evaluable population depending on prespecified rules for surgery, pathology assessment, withdrawal, or other endpoint-related circumstances. The registry information available here does not provide the analyzed denominator.
Why use an exact binomial confidence interval?
With a small study population, the normal approximation to a binomial distribution can perform poorly, especially for proportions near 0 or 1. An exact binomial interval avoids relying on that approximation and is commonly used for small single-arm phase 2 studies.
What would a p-value mean in a single-group study?
A p-value would require a null hypothesis specifying a reference pCR probability. It would answer whether the observed data are inconsistent with that particular null value under the chosen statistical model. It would not measure the magnitude of the pCR rate and would not establish superiority over a randomized comparator.
Why is a hazard ratio not appropriate for the primary endpoint?
The registered primary endpoint is a pathologic response assessed after a defined treatment period, not a time-to-event endpoint. A hazard ratio is designed for comparing event rates over time and therefore does not correspond naturally to the primary pCR analysis.
8. Results
The ClinicalTrials.gov record indicates that results are posted, but no formal statistical analyses are posted for the registered primary endpoint. The numerical pCR result, confidence interval, and p-value are not reported in the statistical-analysis information represented here.
Accordingly, the statistically meaningful information available from the registry is the endpoint definition and its timing rather than an estimated treatment effect. A pCR count or percentage cannot be substituted for a formal comparative effect because the study has no randomized comparator arm.
9. Safety Results
The registry reports serious adverse events by arm as 2/26 for the single treatment group identified as Palbociclib + Letrozole + Trastuzumab +/.
Serious adverse events
Affected participants / participants at risk
Single registered treatment group
| Safety measure | Single treatment group |
|---|---|
| Serious adverse events | 2/26 |
The reported 2/26 means that 2 participants were identified as affected by a serious adverse event among 26 participants at risk in the registered arm. It does not describe the number of individual serious adverse events, their severity distribution, causality, timing, or clinical outcomes.
Because there is only one registered treatment group, the figure cannot be interpreted as a randomized treatment-group difference. It is a descriptive safety count within the study population.
10. Statistical Interpretation of the Single-Group Design
The design determines what conclusions the statistical analysis can support. In a randomized trial, treatment groups can be compared because randomization creates a framework for estimating a causal contrast between assigned treatments. In PALTAN, the registry identifies one treatment group, so the principal statistical question is descriptive or single-arm inferential rather than comparative.
| Statistical quantity | Role in PALTAN |
|---|---|
| pCR count | Describes the number meeting the registered pathologic definition. |
| pCR proportion | Describes the fraction meeting the definition among the specified analysis population. |
| Confidence interval | Quantifies uncertainty around the estimated single-group pCR probability. |
| Reference-value hypothesis test | Could be used if a prespecified null pCR probability were defined. |
| Risk ratio | Not a primary quantity for a single-group design without a comparator. |
| Hazard ratio | Not appropriate for the registered binary pCR endpoint. |
| Log-rank test | Not appropriate for the registered primary endpoint. |
Why the enrollment size matters
With 26 enrolled participants, each observed participant represents a substantial portion of the study population. This makes the precision of any single-group response estimate especially important. A high or low observed proportion in a small study should therefore be distinguished from the uncertainty surrounding the corresponding underlying population probability.
This is one reason phase 2 single-arm studies often function as screening or signal-detection studies rather than definitive comparative efficacy trials. The design can characterize whether a treatment regimen produces an outcome at a potentially meaningful frequency, while a randomized study is generally required when the question is whether that outcome is better than an alternative treatment.
11. Pathologic Complete Response as an Endpoint
pCR is a pathology-based endpoint assessed after neoadjuvant treatment and surgery. The PALTAN registry definition is specific: no histology evidence of invasive tumor cells in the surgical breast specimen and sentinel or axillary lymph nodes.
Timing
The endpoint is assessed after completion of 4 cycles of treatment, approximately 16 weeks.
Breast specimen
The registry definition requires no histology evidence of invasive tumor cells in the surgical breast specimen.
Lymph nodes
The definition also incorporates the sentinel or axillary lymph nodes.
Binary classification
The endpoint creates a categorical response assessment suitable for a binomial statistical framework.
The precise endpoint definition is statistically important because small changes in the response definition can change the numerator of the pCR proportion. Preserving the registry wording therefore matters when interpreting or comparing pCR results across studies.
12. Study Timeline
Study start
The registry lists June 30, 2017 as the start date.
Single-group treatment study
The study evaluates neoadjuvant palbociclib in combination with letrozole and trastuzumab in stage II-III ER+ HER2+ breast cancer.
Primary endpoint assessment
The registered pCR endpoint is assessed after completion of 4 cycles of treatment.
Primary completion
The registry lists August 24, 2020 as the primary completion date.
Terminated
The ClinicalTrials.gov record lists the study status as terminated.
13. Research Procedures and Translational Measurements
In addition to the therapeutic interventions, the registry lists several procedures and biological samples. These include research tumor biopsy, optional research bone marrow, research blood sample, blood for germline DNA, and blood for detection of circulating tumor cells.
| Research activity | Registry-listed role |
|---|---|
| Research tumor biopsy | Procedure |
| Research bone marrow | Optional procedure |
| Research blood sample | Procedure |
| Blood for germline DNA | Genetic assessment |
| Blood for detection of circulating tumor cells | Procedure |
These measurements broaden the study beyond the primary pCR endpoint, but the registry information represented here does not provide numerical statistical results for these research measurements. They should therefore be viewed as study procedures rather than as reported efficacy estimates.
14. What the Single-Arm Analysis Does — and Does Not — Mean
A single-group pCR estimate describes the observed frequency of pathologic complete response in the treated study population. It does not, by itself, quantify how much better or worse the regimen is than another treatment.
Without randomization to a comparator, differences between the study population and another population can reflect differences in patient characteristics, disease characteristics, eligibility, treatment pathways, pathology assessment, or other factors rather than treatment effects.
A confidence interval around a single-group proportion would quantify uncertainty around the estimated underlying pCR probability. In a study with 26 enrolled participants, that uncertainty is an important part of the statistical interpretation.
If a single-arm study tests its response rate against a prespecified reference value, the resulting p-value measures the compatibility of the data with that null hypothesis. It does not measure the size of the observed pCR rate and does not establish comparative superiority without a randomized comparator.
15. Limitations
- Single-group design: the registry identifies one treatment arm, so there is no randomized concurrent comparator for estimating a treatment effect.
- Small enrollment: 26 participants means that the precision of a single-group pCR estimate is an important consideration.
- Endpoint denominator: the registry information represented here does not identify a posted statistical analysis population or a separate evaluable denominator for pCR.
- No posted formal statistical analysis: although the registry indicates that results are posted, no formal statistical analyses are included in the statistical-analysis information represented here.
- Binary endpoint: pCR captures a specific pathologic state at a defined time point and does not by itself describe all subsequent clinical outcomes.
- Safety interpretation: the reported 2/26 serious-adverse-event figure is descriptive and cannot be compared against a randomized control rate within this study.
- Generalizability: the applicability of a single-group phase 2 study depends on how closely patients outside the study resemble the enrolled population with stage II-III ER+ HER2+ breast cancer.
- Research measurements: translational procedures are listed in the registry, but corresponding statistical results are not provided in the information represented here.
16. Why This Trial Matters Statistically
PALTAN is a useful statistical teaching case because it illustrates a fundamentally different analysis problem from the large randomized trials that dominate clinical-trial reporting. The key issue is not how to estimate a hazard ratio between treatment groups, but how to interpret a small single-group binary endpoint with an explicitly defined pathologic criterion.
| Concept | How it appears in PALTAN |
|---|---|
| Single-arm design | One registered treatment group |
| Binary endpoint | pCR is defined according to a yes/no pathologic criterion |
| Fixed assessment time | After completion of 4 cycles, approximately 16 weeks |
| Binomial inference | Natural framework for a single-group response proportion |
| Confidence interval | Important for quantifying uncertainty in a small study |
| Reference-value testing | Requires a prespecified null pCR probability |
| Absence of randomized comparator | Prevents a randomized treatment-effect contrast |
| Safety description | Serious adverse events reported as 2/26 |
| Translational research | Includes tumor, blood, germline DNA, circulating tumor cell, and optional bone-marrow procedures |
17. Statistical Interpretation vs Clinical Interpretation
Statistical interpretation
The registered design supports a single-group analysis of a binary pathologic endpoint. The principal descriptive measure is the pCR proportion, with a confidence interval providing information about precision.
Clinical interpretation
The study evaluates a neoadjuvant regimen consisting of palbociclib, letrozole, and trastuzumab in stage II-III ER+ HER2+ breast cancer, with pathologic response assessed after approximately 16 weeks.
These perspectives should remain separate. Statistical estimation describes what was observed and how precisely it was estimated. Clinical interpretation considers what the endpoint represents in the context of the treatment strategy and disease setting. Neither perspective converts a single-group result into a randomized comparative treatment effect.
18. Serious Adverse Events: Statistical Perspective
The registry reports 2/26 serious adverse events for the single treatment group. The simplest interpretation is a descriptive count: two affected participants among 26 participants at risk.
This quantity describes affected participants rather than necessarily counting every adverse-event episode. The registry wording posted on ClinicalTrials.gov for the safety field is "affected / at risk."
Because no second randomized group exists, there is no internal estimate of a treatment-associated excess risk. External comparisons would require careful consideration of differences in patient populations, follow-up, event definitions, ascertainment, and treatment exposure.
19. Sources
- ClinicalTrials.gov: PALTAN, NCT02907918.
- PubMed: PubMed record associated with the PALTAN study.
Continue through the Clinical Biostats knowledge graph
Clinical trial analysis connects study design, endpoint definition, statistical estimation, uncertainty, and interpretation into a coherent statistical framework.
20. Record Summary
PALTAN is registered as a phase 2, single-group, unmasked treatment study with an enrollment of 26 participants and one treatment arm evaluating palbociclib, letrozole, and trastuzumab in stage II-III ER+ HER2+ breast cancer. Its primary endpoint is the number of participants with pathologic complete response, defined as no histology evidence of invasive tumor cells in the surgical breast specimen and sentinel or axillary lymph nodes, assessed after completion of 4 cycles of treatment, approximately 16 weeks.
The statistical structure is therefore centered on a single-group binary endpoint rather than a randomized treatment comparison. The natural framework is a binomial response analysis, with the pCR proportion providing the descriptive estimate and a confidence interval providing information about precision. The registry indicates that results are posted, but no formal statistical analyses are posted for the primary endpoint in the statistical-analysis information represented here. The safety record reports 2/26 participants affected by serious adverse events in the single treatment group.
For statistical interpretation, the most important feature of PALTAN is the distinction between observing a response rate in one treatment group and estimating a randomized treatment effect. The former is supported by the registered design; the latter requires a comparative framework that is not present in this study.