Neoadjuvant Weekly Paclitaxel and Biomarkers of Therapy Response: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Neoadjuvant Weekly Paclitaxel and Biomarkers of Therapy Response (NCT03096418), a na phase 4 study. The primary outcome is Response to Paclitaxel.

2. Clinical Question

The study hypothesis is that paclitaxel increases chromosomal instability in tumors and that this increase is lethal to tumors with pre-existing chromosomal instability. The primary question therefore links a tumor characteristic to change in tumor size during treatment. It is not a randomized test of paclitaxel against another therapy or against no treatment. Because all participants are assigned to the single listed group, any observed relationship between chromosomal instability and tumor-size change would be an association within the treated group, not a treatment-versus-control effect.

The registry describes treatment as outpatient paclitaxel before surgery. It also lists secondary assessments involving drug levels in tumor and non-tumor tissue, antimitotic effects, mitotic index, aneuploidy, chromosomal instability, and Ki67. These measures give context to the biological hypothesis, but numerical results for them are not reported in the registry record. The number of participants evaluable for the primary endpoint is reported as 12 because of insufficient specimen.

3. Trial Design

The record describes a Phase 4, single-group study with no masking. Allocation is listed as not applicable. These features mean the record does not describe randomized assignment or a blinded comparison. The intervention is weekly paclitaxel, given by standard infusion on days 1, 8, and 15 of each 21-day cycle. The arm description specifies 80 mg/m2 and continuation through cycles 2-4 before surgery. Treatment is described as outpatient.

The registry lists an enrollment of 50, but that is not the number reported as evaluable for the primary endpoint. The primary endpoint description says that only 12 participants were evaluable because of insufficient specimen. Whether the enrollment figure represents an achieved enrollment count or a target is not established by that figure alone; actual enrollment status is not reported in the registry record. The study status is recruiting. Its start is recorded as 2017-03-13 and its primary completion as 2027-06.

The study's stated purpose is to test whether tumors with high chromosomal instability respond to paclitaxel better than tumors with low chromosomal instability. Since there is no randomized comparator, the design can describe tumor response during the treatment regimen and explore relationships with tumor biomarkers, but cannot isolate paclitaxel's effect relative to an alternative course of care. The registry does not report detailed eligibility criteria, the precise biomarker threshold defining high versus low chromosomal instability, or a detailed statistical analysis plan. These details are not reported in the registry record.

The record lists 10 outcome measures and 0 analyses. The listed secondary outcomes include paclitaxel levels and tumor biomarker measures. The meaning of each measure, its analysis population, and the planned handling of multiple measures are not fully specified. The registry record also does not provide a sample-size justification for the trial itself.

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: Only 12 participants were evaluable because of insufficient specimen.

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 Response to Paclitaxel, measured over Up to 3 months.

The primary endpoint is response to paclitaxel, expressed as the percent decrease in tumor size. Tumor size is represented by the linear measurement of the greatest dimension, using RECIST-like criteria. The timeframe is up to 3 months. The endpoint description says that the purpose is to assess whether cancers with high chromosomal instability respond better than low-CIN cancers.

This endpoint turns a change in a measured tumor dimension into a continuous quantity. In general, a continuous measure can retain more information about the amount of change than a simple label such as response versus no response. The registry, however, does not report a response threshold or a numerical endpoint value. It identifies the summary parameter as a mean and its dispersion as a full range, but the actual mean and range are not reported in the registry record.

The stated measurement is a linear dimension, not a comprehensive measure of every aspect of tumor burden. The RECIST-like wording signals that the measurement follows a response-assessment approach, but the registry does not give more detail about measurement procedures, assessor blinding, or how imaging observations were selected. The endpoint is relevant to the biological question because it is intended to capture tumor-size change during preoperative paclitaxel treatment. It does not by itself establish whether chromosomal instability caused a difference in response.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary endpoint summary as a mean with a full range. A mean is a measure of the center of a set of numerical observations. The full range describes the span from the smallest to the largest observed value. In general, reporting both can indicate the typical observed response and the extent of variation, although the range can be strongly affected by extreme observations and does not describe the distribution between its endpoints. The registry record does not report the mean or the full range for this endpoint.

What does the endpoint quantify?

The endpoint is percent decrease in tumor size, based on change in the greatest linear measurement under RECIST-like criteria. Because it is expressed as a percent decrease, the endpoint is intended to describe relative change rather than only the measured dimension at a single assessment. The registry record does not specify the exact calculation convention, the baseline and follow-up measurement procedures, or the handling of measurements that increase rather than decrease. Those details are not reported in the registry record.

In general, a continuous response measure can distinguish different amounts of change that would be grouped together by a binary response category. Its interpretation depends on consistent measurement and on the definition of the observation time. The stated timeframe is up to 3 months, but the record does not state whether all measurements were made at a common scheduled time or how variation in assessment timing was handled. This limits interpretation of comparisons across participants.

How should the CIN question be understood?

The question compares response between cancers described as high and low in chromosomal instability. Such a comparison requires a rule for classifying CIN, and the registry record does not report the threshold or classification method. It also does not report the number of evaluable participants in either CIN group, their separate response summaries, or the statistical test used. Without those items, the primary comparison cannot be reconstructed or its uncertainty assessed.

In general, an observed association between a biomarker and response within a treated group is not the same as evidence that the biomarker predicts a treatment effect. A predictive claim about differential benefit generally relies on comparing treatment effects across biomarker groups. This trial is described as single-group, with no randomized comparator, so the registry record does not establish such a treatment-by-biomarker contrast. Other tumor characteristics could also be associated with both CIN and tumor-size change.

What do precision measures usually add?

In general, a confidence interval gives a range of values compatible with the data under a specified statistical model and indicates the precision of an estimate. A narrow interval implies greater precision than a wide interval, while neither alone establishes that a result is clinically important. No confidence interval for the primary response or the CIN comparison is reported in the registry record. No p-value or test statistic for the primary outcome is reported either. The available information therefore does not permit a formal judgment about statistical uncertainty.

Sample-size calculations are planning tools, not evidence about the strength of an observed result. The registry's stated planning illustration uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect defined as a fraction of the observed effect. It also distinguishes participant-per-arm calculations from total event counts for time-to-event outcomes. These are assumptions for an illustration, not a description of the trial's own design or its achieved power. The actual study is registered as single-group, and the record does not report the resulting planning values or a trial-specific sample-size rationale. The illustration cannot substitute for a reported endpoint estimate and its uncertainty.

What is not specified about the analysis?

The registry lists 10 outcome measures and 0 analyses. That does not identify which statistical model, if any, was used for each outcome. The primary endpoint is described as a mean and full range, but the record does not state whether the CIN relationship was assessed using a model, a direct group comparison, or another method. Adjustment for other prognostic variables, assumptions about the outcome distribution, and any sensitivity analyses are not reported in the registry record.

In general, model choice should reflect the outcome type, the study design, and the way observations were collected. Here, a continuous tumor-size response is being related to a biomarker classification, but group-specific data and analysis details are unavailable. It is therefore not appropriate to infer a particular regression method, a normality assumption, a covariate adjustment, or an inferential result. The endpoint and design can be described; a full statistical interpretation requires details that the registry record does not report.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 10 outcome measures and 0 analyses. The named secondary measures span drug levels, antimitotic effects, mitotic index, aneuploidy, chromosomal instability, and Ki67. Considering several outcomes or biomarker relationships can create multiple opportunities for findings to appear notable by chance. In general, multiplicity methods can control or describe that issue, but the registry record does not report a multiplicity adjustment, a hierarchy of outcomes, or which comparisons were intended to be confirmatory. It is not possible to infer from the listed outcomes that any particular correction was or was not used in an analysis.

Missingness is directly relevant to the primary endpoint: only 12 participants were evaluable because of insufficient specimen. The registry does not report how many participants lacked adequate specimens, whether missingness differed by participant or tumor characteristics, or whether another analysis population was used. It also does not describe imputation, sensitivity analyses, or other methods for handling missing endpoint data. Without this information, it is difficult to judge whether the evaluable participants represent the broader enrolled population.

The record lists an enrollment of 50, but does not establish in the information available whether this is achieved enrollment or a target. It should not be treated as the primary endpoint denominator. The number evaluable is the explicit figure for the primary outcome, and its stated limitation is specimen insufficiency. Details about exclusions, protocol deviations, and analysis-population definitions are not reported in the registry record. The absence of analysis entries also means that the registry record does not document how the endpoint data were analyzed.

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 0% reporting threshold.

The registry reports one Weekly Paclitaxel group. For that group, 24 are listed at risk for serious adverse events and 0 are listed as affected. It also lists 24 at risk for other adverse events and 0 affected, with 0 deaths. No top serious adverse event terms are listed.

These counts describe what is recorded, not a guarantee that no adverse events occurred. A zero count can reflect the recorded event category and reporting scope; it does not establish that an event is impossible or that every event was captured. The registry record does not explain the adverse-event collection period, the criteria for event severity, or the completeness of safety reporting. It also does not provide event-level descriptions or a safety comparison with another treatment group.

The at-risk figures are safety-reporting denominators and should not be substituted for the primary endpoint's evaluable population. The endpoint description separately states that 12 participants were evaluable because of insufficient specimen. Since the study has no randomized comparator, the reported safety counts cannot establish comparative safety. They should be read narrowly as the counts recorded for the listed group.

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

The record describes a study of preoperative weekly paclitaxel that asks whether tumor chromosomal instability is related to tumor-size response. It specifies the primary endpoint, its measurement basis, the treatment schedule, and a reason that limited primary endpoint evaluability. These details clarify the scientific question, but they are not a numerical answer to it. The record does not report the primary mean, full range, or a comparison of response by CIN status.

As a result, the record does not establish whether high-CIN cancers had a larger or smaller size decrease than low-CIN cancers, nor how precise such a difference would be. It also does not report an inferential test, confidence interval, or p-value for the primary question. No statement about statistical significance or magnitude of response can be supported from the endpoint information described in the registry.

Even if an association between CIN and response were available, this single-group study would not provide a randomized comparison of paclitaxel against another treatment or no treatment. An association among paclitaxel-treated participants could reflect the biomarker, other tumor characteristics, or factors not described in the registry. It would not by itself show that CIN identifies who benefits more from paclitaxel than from another option.

The 12 evaluable participants and the stated specimen insufficiency matter when considering the scope of any endpoint result. The record does not explain how missing specimens relate to patient or tumor characteristics, so the extent of possible selection bias cannot be assessed. The enrollment figure of 50 is not interchangeable with the stated evaluable count, and the record does not clarify whether enrollment is an achieved count or a target.

The appropriate reading is therefore descriptive and cautious: the registry records a biological hypothesis and a measurement strategy, but numerical primary outcome findings and analysis details are not reported in the registry record. The safety counts likewise describe reported events in one group and do not provide a comparative risk estimate.

13. Why This Trial Matters Statistically

This study is statistically interesting because it connects a tumor biomarker hypothesis with a continuous measure of tumor-size change. A biological marker may be associated with response even when a study is not designed to compare treatments. Separating those questions is important: the within-group association and the causal effect of treatment are different targets of inference.

The single-group design makes the distinction especially clear. Participants receive the listed paclitaxel regimen, and the stated primary question concerns differences between high- and low-CIN cancers. Without a randomized comparator, the design cannot determine whether any response pattern is specific to paclitaxel rather than reflecting general tumor characteristics or another influence. A biomarker-response relationship may motivate further investigation, but it is not itself a comparative treatment effect.

Specimen availability is also consequential. The record says only 12 participants were evaluable for the primary outcome because of insufficient specimen. Biomarker studies depend on adequate tissue and measurable characteristics, so the analyzed set can differ from the participants originally considered for the study. The registry does not report enough information to determine whether that limitation could change the observed relationship.

Finally, the record identifies multiple outcome measures, while listing 0 analyses. This highlights why a complete account of analysis choices matters when studies assess several biological signals. Outcome definitions, analysis populations, missing-data handling, and multiplicity plans help readers distinguish planned questions from exploratory patterns. Those details are not reported in the registry record, leaving the statistical interpretation narrower than the scientific hypothesis.

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 principal limitation for interpreting the primary question is the absence of numerical outcome results. The record defines the endpoint as mean percent decrease in tumor size with a full range, but reports neither summary. It also does not report response summaries by CIN group, the CIN classification rule, or an inferential comparison. The size and precision of any association therefore cannot be judged from the registry information.

Only 12 participants are reported as evaluable for the primary endpoint because of insufficient specimen. The registry does not state how many specimens were insufficient or whether the evaluable participants differed from those without usable specimens. Missing-data methods and sensitivity analyses are not reported. These gaps limit assessment of selection effects and generalizability of the endpoint results.

The study is single-group, with no randomization, masking, or separate control arm described. Any biomarker-response relationship would be observational within the treated group and could be affected by other tumor characteristics. The registry also does not provide detailed eligibility criteria, assessment procedures, or a complete statistical analysis plan. It lists 10 outcome measures and 0 analyses, while multiplicity procedures and outcome hierarchy are not reported.

The enrollment figure of 50 should not be mistaken for the number evaluable for the primary outcome, and the record does not establish whether it is achieved enrollment or a target. Safety counts are also limited to the listed group and lack details on reporting windows and ascertainment. These constraints mean the record is most informative about the trial's question and design, not its outcome magnitude or comparative clinical implications.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Neoadjuvant Weekly Paclitaxel and Biomarkers of Therapy Response — NCT03096418.
• Registry record last updated 2026-06-16; status last verified 2026-06.
• Results first posted 2023-10-13.
• 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

The registry record describes a recruiting Phase 4, single-group study of weekly paclitaxel for female breast neoplasm. Paclitaxel is listed at 80 mg/m2 on days 1, 8, and 15 of a 21-day cycle, continuing through cycles 2-4 before surgery. The primary question is whether high-CIN cancers respond better than low-CIN cancers, with response measured as percent decrease in the greatest linear tumor dimension over up to 3 months.

The record lists enrollment of 50 and reports 12 participants evaluable for the primary endpoint because of insufficient specimen. It does not report the primary mean, full range, or comparison by CIN status. The listed safety counts are 0 serious adverse events, 0 other adverse events, and 0 deaths in the reported group. The study has no randomized comparator, and analysis and missing-data details are not reported in the registry record.